Xem Trang 35 Của Cái Này Có Bảng Các Định Nghĩa Nhé

Accelerat ing t he world's research.
Principle of the Cognitive Grinding

of Reuse Materials
Adam Mroziński
Principle of the Cognitive Grinding of Reuse Materials. In: Golinska-Dawson P. (eds) Logistics
Operations and Management for Recycling and Reuse. EcoProduction (Environmental Issues in
Logistics and Manufacturing). Springer, Berlin, Heidelberg.
Cite this paper Downloaded from Academia.edu 
Get the citation in MLA, APA, or Chicago styles
Related papers Download a PDF Pack of t he best relat ed papers 
CIRCULAR ECONOMY APPLICAT ION OF REVERSE LOGIST ICS IN T HE RECYCLING OF URBAN T R…

Andrey Sart ori
Valuing Urban Wast e: T he need for comprehensive mat erial recovery and recycling policy.
Pinky Hasirudala, Pinky Chandran
Wast e generat ion, Wealt h and GHG emissions from t he wast e sect or: Is Denmark on t he pat h t oward…
Cosimo Magazzino, Nicolas Schneider
EcoProduction.
Environmental Issues in Logistics and Manufacturing
Paulina Golinska-Dawson Editor
Logistics
Operations and
Management
for Recycling and
Reuse
EcoProduction
Environmental Issues in Logistics and Manufacturing
Series Editor
Paulina Golinska-Dawson, Poznań, Poland
The EcoProduction Series is a forum for presenting emerging environmental issues
in Logistics and Manufacturing. Its main objective is a multidisciplinary approach
to link the scientific activities in various manufacturing and logistics fields with the
sustainability research. It encompasses topical monographs and selected conference
proceedings, authored or edited by leading experts as well as by promising young
scientists. The Series aims to provide the impulse for new ideas by reporting on the
state-of-the-art and motivating for the future development of sustainable manufac-
turing systems, environmentally conscious operations management and reverse or
closed loop logistics.
It aims to bring together academic, industry and government personnel from
various countries to present and discuss the challenges for implementation of
sustainable policy in the field of production and logistics.
More information about this series at http://www.springer.com/series/10152

Paulina Golinska-Dawson
Editor
Logistics Operations
and Management
for Recycling and Reuse
Editor
Faculty of Engineering Management
Poznań University of Technology
Poznań, Poland
ISSN 2193-4614 ISSN 2193-4622 (electronic)

EcoProduction
ISBN 978-3-642-33856-4 ISBN 978-3-642-33857-1 (eBook)
https://doi.org/10.1007/978-3-642-33857-1
© Springer-Verlag GmbH Germany, part of Springer Nature 2020

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer-Verlag GmbH,

DE part of Springer Nature.
The registered company address is: Heidelberger Platz 3, 14197 Berlin, Germany
Preface
The increasing awareness of the circular economy concept and the sustainability
challenges are reshaping the logistics operations and the supply chain management
principles. The emerging new resource-efficient business models allow reducing,
reusing and recycling and therefore they support transition towards a “zero waste”
economy. The focus on the resource efficiency and waste reduction is translated into
searching new business opportunities through closing and narrowing the materials
loops in the economy. The aim of this book is to present quantitative and qualitative
aspects of logistics operations supporting 3R (reduce, recycle and reuse) policy.
In individual chapters, the authors address various challenges related to reverse
network configuration and the organization of collection, transportation and recovery
activities.
The authors in the individual chapters discuss the original methods and tools, as
well as practical case studies on topics, as follows:
• Circular Economy,
• Reverse Logistics Flows and Network’s Configuration Problems,
• Practical Aspects of Application the 3R (Reduce, Recycle, Reuse).
In the first part the challenges of the circular economy with regard to the logistics
sector and supply chain management have been discussed. The environmental impact
of the logistics operations on the global scale (especially transportation) has been
widely discussed in the scientific literature. However, the logistics providers link
different companies in the supply chain therefore they have transformational potential
towards circular economy at meso-scale. Most of the studies on circular business
models focus on manufacturing and product design. In the chapter “Towards Circular
Economy Transition—Developing the Innovative Sustainable Practices in Logistics
Industry” the sustainable practices in the logistics sector are discussed and classified.
The bottom-up approach is taken, in order to analyse how sustainable practices at
micro-level have transitional potential towards circular economy.
In chapter “A Literature Analysis of Definitions for a Circular Economy” are
reviewed 28 different definition of the CE. The authors have highlighted the impor-
tance of stakeholder perspective in scientific discourse on the transition towards the
circular economy.
v
vi Preface
In the second part of this book the focus is placed on the reverse logistics flows and
network’s configuration problems. In the chapter “Robust Reverse Logistics Network
Design” is presented a practically applicable optimization method for waste wood
reverse logistics. The authors extend the local search heuristic for the facility location
problem with nonlinear objective function that allows minimizing the costs. They
discuss the results of numerical experiments with regard to the impact of the waste
wood recycling on the CO2 emission.
The authors of the subsequent chapter discuss “Drivers and Barriers for Cooper-
ation Between Municipalities in Area of Municipal Solid Waste Management”. The
empirical data from over 230 Polish municipalities is analyzed in order to conclude
how the decisions to engage in such cooperation are facilitated.
The authors of chapter “A Novel Formulation for the Sustainable Periodic Waste
Collection Arc-Routing Problem: A Hybrid Multi-objective Optimization Algo-
rithm” develop an effective methodology to determine the optimal plans for waste-
collection routes and the required number of vehicles They propose mixed-integer
linear programming model in order to obtain a multi-objective multi-trip sustainable
plan for waste collection.
In the subsequent chapter is presented “A Perishable Product Sustainable Supply
Chain Network Design Problem with Lead Time and Customer Satisfaction using
a Hybrid Whale-Genetic Algorithm”. The authors propose an integrated mathe-
matical model that allows minimizing the production, distribution, and customer
satisfaction related costs, minimizing total CO2 emissions, and maximizing social
responsibility.
The third part of the book includes selected practical aspects of application the
3R (reduce, recycle, reuse).
The authors of chapter “How to Assess Internal Transport in Terms of Sustain-
ability in the Recycling Industry?—Case Study” analyse the internal transport
process in a company which specializes in End-of-Life Vehicles disassembling. They
develop a heuristic to assess and improve the intralogistics operations.
The chapter “Principle of the Cognitive Grinding of Reuse Materials” presents a
multi-disc grinder of environmentally enhanced design. The authors propose for eval-
uation of structural system operational parameters, the criteria: energy consumption,
material consumption, operational efficiency and effectiveness, the overall waste
balance (waste balance, waste produced).
In the final chapter “Smart-Tracking Systems Development with QR-Code
and 4D-BIM for Progress Monitoring of a Steel-Plant Blast-Furnace Revamping
Project in Korea”, the authors develop a smart tracking system with wireless tagging
technology. They discuss the outcome of the application of the proposed solutions
with regard to the reduction of costs and work load of workers.
This book might be a valuable resource for both academics and practitioners who
want to deepen their knowledge of logistics operations and management for recycling
and reuse.
This monograph is edited in association with the 15th International Congress on
Logistics and SCM Systems (ICLS 2020). The ICLS 2020 is organized by the Faculty
of Engineering Management, Poznań University of Technology and the International
Preface vii
Federation of Logistics & SCM Systems (IFLS). I would like to express my gratitude
to the Board of the International Federation of Logistics and SCM Systems (IFLS)
for the invaluable inspiration and motivation to prepare this volume:
• Honorary Chairman—Prof. Karasawa, Yutaka, Kanagawa University, Japan.
• Advisors—Prof. Kachitvichyanukul, Voratas, Asian Institute of Technology,
Thailand; Prof. Katayama, Hiroshi, Waseda University, Japan.
• Chairman—Prof. Tsai, Kune-Muh, National Kaohsiung University of Science
and Technology, Taiwan.
• Vice Chairmen: Prof. Lai, Kin Keung, City University of Hong Kong; Prof. Liu,
Xiaohong, Central University of Finance and Economics, China; Prof. Rim, Suk-
Chul, Ajou University, Korea; Sethanan, Kanchana, Khon Kaen Univesity, Thai-
land; Prof. Wakabayashi, Keizo, Nihon University, Japan and Wu, Yenchun Jim,
National Taiwan Normal University, Taiwan.
• The Board Members.
This scientific monograph has been blind reviewed. I would like to thank all
reviewers whose names are not listed in the volume due to the confidentiality of the
process. Their voluntary service and comments helped the authors to improve the
quality of the manuscripts.
Although not all of the received chapters appear in this book, the efforts spent and
the work done for this book are very much appreciated.
Poznań, Poland Paulina Golinska-Dawson

Contents
Circular Economy
Towards Circular Economy Transition—Developing the Innovative
Sustainable Practices in Logistics Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
A Literature Analysis of Definitions for a Circular Economy . . . . . . . . . . . 19
Usama Awan, Narmeen Kanwal, and Mohammed Khurrum S. Bhutta
Reverse Logistics Flows and Network’s Configuration Problems

Robust Reverse Logistics Network Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Péter Egri, Balázs Dávid, Tamás Kis, and Miklós Krész
Drivers and Barriers for Cooperation Between Municipalities
in Area of Municipal Solid Waste Management . . . . . . . . . . . . . . . . . . . . . . . 55
Paulina Golinska-Dawson, Arkadiusz Kawa, and Piotr Januszewski
A Novel Formulation for the Sustainable Periodic Waste Collection
Arc-Routing Problem: A Hybrid Multi-objective Optimization
Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Erfan Babaee Tirkolaee, Alireza Goli, Gerhard-Wilhelm Weber,
and Katarzyna Szwedzka
A Perishable Product Sustainable Supply Chain Network Design
Problem with Lead Time and Customer Satisfaction using
a Hybrid Whale-Genetic Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Alireza Goli, Erfan Babaee Tirkolaee, and Gerhard-Wilhelm Weber
Practical Aspects of Application the 3R (Reduce, Recycle, Reuse)

How to Assess Internal Transport in Terms of Sustainability
in the Recycling Industry?—Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Izabela Kudelska and Monika Kosacka-Olejnik
ix
x Contents
Principle of the Cognitive Grinding of Reuse Materials . . . . . . . . . . . . . . . 145

Adam Mroziński, Józef Flizikowski, Kazimierz Bieliński, and Marek Macko
Smart-Tracking Systems Development with QR-Code and 4D-BIM
for Progress Monitoring of a Steel-Plant Blast-Furnace Revamping
Project in Korea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
H. Y. Roh, E. B. Lee, I. H. Jung, and C. Y. Kim
Circular Economy
Towards Circular Economy
Transition—Developing the Innovative
Sustainable Practices in Logistics
Industry
Abstract The circular economy concept encourages the redesign of industrial

activities and social practices in order to decouple the economic growth from negative
environmental impact. The focus is placed on establishing more resource-efficient
business models that allows reducing, reusing and recycling and therefore they
support transition towards a “zero waste” economy. The environmental impact of
logistics operations is formidable due to the GHG emissions, noise, and conges-
tion and material waste. The logistics providers act, as the link between different
companies in the supply chain therefore they have transformational potential towards
circular economy at meso-scale. Most of the studies on circular business models
focus on manufacturing and product design. The studies on logistics sector are very
limited and this chapter addresses that research gap. The aim of this paper is to link
the described in the literature circular economy business models with innovative
sustainable practices in the logistics industry. The main contribution is the classifica-
tion of the innovative technological, organizational and social sustainable practices
in logistics sector. The bottom-up approach is taken, as it describes how sustainable
practices at micro-level have transitional potential towards circular economy.
Keywords Logistics service providers · Logistics infrastructure · Circular

economy · Sustainability · Supply chain · Logistics
1 Introduction
The Circular Economy (CE) objective is to highlight the interplay between the envi-
ronment and the economic system (Ghisellini et al. 2016). It is rooted in a diverse
theoretical background, such as: industrial ecology (Ayers et al. 1989), environmental
economics (Pearce and Turner 1990) and industrial economics (Stahel 1982). It is
influenced by different environmentally conscious theoretical concepts, such as:
P. Golinska-Dawson (B)
Faculty of Engineering Management, Poznań University of Technology,
Jacka Rychlewskiego 2 str., 60965 Poznań, Poland
e-mail: paulina.golinska@put.poznan.pl
© Springer-Verlag GmbH Germany, part of Springer Nature 2020 3

P. Golinska-Dawson (ed.), Logistics Operations and Management for Recycling
and Reuse, EcoProduction, https://doi.org/10.1007/978-3-642-33857-1_1
4 P. Golinska-Dawson
• cradle-to-cradle (McDonough and Braungart 2002),

• performance economy (Stahel and Clift 2016),
• eco-efficiency (Huppes and Ishikawa 2005),
• life cycle management (Kirkke et al. 2004).
The CE approach contrasts with the linear economy model (take-make-use-
dispose) and provides new approaches to achieve benefits, as it buffers against the
scarcity of fossil fuels and raw materials by recapturing and recovering resources
and energy which are already embodied in the existing techno-social-economical
systems. Circular Economy is “restorative and regenerative by design and aims to
keep products, components, and materials at their highest utility and value at all
times” (Ellen McArthur 2015, p. 2).
The concept of Circular Economy (CE) in the last few years has been promoted
by the European Commission (COM/2015/0614), as a new pathway for a more
resource efficient and competitive economy. The circular economy concept encour-
ages the redesign of industrial activities and social practices in order to decouple
the economic growth from environmental burden and trigger a transition towards a
“zero waste” economy. The European Commission has stated that “the EU has no
choice but to go for the transition to a resource-efficient and ultimately regenerative
circular economy’ (EC 2011 p. 1). The implementation of the CE at the local (called:
micro-level, for example at a company) and regional level (called: meso-level, for
example a supply chain) is challenging. At the same time the companies, (as the
singular actor) own most resources and capabilities, therefore can stimulate CE tran-
sition by creating added value through an extended and more proactively managed
stakeholders’ network (Geissdoerfer et al. 2016). The logistics providers act, as the
link between different companies in the supply chain therefore they have transfor-
mational potential towards circular economy at meso-scale. However studies on the
logistics sector with regard to the logistics sector are very limited.
The circular economy can be analyzed, as a part of ecological modernization
theory (EMT). The EMT reconciles the conflict between industrial development and
environmental protection (Murphy and Gouldson 2000) by “ecologizing economy”
and “economizing ecology”. The EMT stipulates that sustainability transition can be
achieved by increasing resource productivity, improving environmental sustainability
while retaining production and consumption. The EMT “can be used to help corporate
managers understand and guide ecologically oriented management innovation and
change, at both the firm and supply chain level of analysis” (Park et al. 2010. p. 1495).
The technological and organizational innovations can result in the redesign of
products and services for reuse and easier value recovery in multiple life-cycles,
resulting in new relations between stakeholders in the supply chain. The transition
towards more sustainable practices requires to identify and address opportunities and
concerns, and to encourage “leadership thinking and best practice and to provide a
forum for policy innovation” (Hobson 2016, p. 94).
The innovative practices can allow for the reduction in resource consumption
and lead to resource-conservative (Rashid et al. 2013), or resource-efficient business
models. The critics of that approach postulate that it is necessary to move ‘beyond
Towards Circular Economy Transition—Developing … 5
incremental efficiency gains to deliver transformative change’ (Preston 2012, p. 1).

They suggest sufficiency or de-growth, as main drivers for change.
In the last five years, volume of CE related literature has been rapidly growing but
the studies on the logistics and supply chain management are still underrepresented
(Govindan and Hasanagic 2018). The supply chain perspective with regard to the
circular economy is addressed by relatively few authors (e.g. Geissdoerfer et al.
2018; Govindan and Hasanagic 2018; Ghisellini et al. 2016; Genovese et al. 2017;
De Angelis et al. 2018; Hussain and Malik 2020). De Angelis et al. (2018) have
stated that, there are very limited studies which investigate the relevance of supply
chain innovations and their transitional potential towards a more resource efficient
and circular economy.
The role of the logistics sector in transition towards CE is even less addressed in
the literature. Van Buren et al. (2016) have presented a pioneering work in that field
and have discussed the role of the Dutch logistics sector in transition towards CE.
The aim of this paper is to link the described in the literature circular economy
business models with innovative sustainable practices in the logistics industry. Its
main contribution is the classification of the innovative technological, organizational
and social sustainable practices in logistics sector. The bottom-up approach is taken,
as it describes how sustainable practices at micro-level have transitional potential
towards circular economy.
The research questions are as follows:
RQ 1. What sustainable practices are implemented by the logistics providers (from
1PL to 5PL)?
RQ 2. How the sustainable logistics practices can be linked with regard to the
circular economy business models?
The remainder of the paper is organized, as follows, first are discussed the related
studies on the circular economy, innovative practices and circular business models.
Then the role of the logistics services providers is discussed in transition towards
CE. Finally the innovative sustainable practices in the logistics industry are classified
and linked with the described in the literature circular economy business models.
2 Related Studies
The differences and similarities between sustainable development and the circular
economy approach are still not clearly stated in the body of literature. Both concepts
focus on the importance of integration of environmental and social aspects with
economic progress. They emphasize the transition at the system-level and call for
innovations. The CE can be treated, as a precondition for sustainable development
(Rashid et al. 2013), or the main solution for transformation (Jawahir and Bradley
2016). Geissdoerfer et al. (2017) have conducted an extensive literature review to
investigate the relationships between sustainability and the CE. They classify those
relations in three categories, as follows:
• conditional,
• beneficial or
• trade-off.
Beneficial relationship means that the CE is an option to foster sustainable devel-
opment and establish more sustainable business models by increasing resource effi-
ciency and dematerialization (Bocken et al. 2014; Weissbrod and Bocken 2017).
Murray et al. (2017) provide some critical assessments of the relation between the
CE and sustainability and identified the trade-offs. They have stated that the CE has
a positive impact on some aspects of sustainability but at the same time it underes-
timates the importance of the social dimension of sustainability. Some authors also
highlight that focus on the 3R or 6R (reduce, reuse, recycle, remanufacture, repair,
refurbish) can lead to additional costs and environmental burden by low technical
feasibility of closing the materials loop, increased transport operations in reverse
logistics, or energy inefficiencies in the recycling process (Allwood 2014).
The most common definition of sustainable development (from Brundtland 1987)
provides an interpretative flexibility of its paradigms. The critics often state that the
sustainable development concept is not precisely defined and it lacks the implemen-
tation framework, which influences its operationalization, especially at the company
level (Geissdoerfer et al. 2017; Golinska and Kuebler 2014). The CE concept refers
mostly to the individual economic benefits “through input reduction, efficiency gains,
and waste avoidance with relatively immediate results compared to sustainability”
(Geissdoerfer et al. 2017, p. 766). The economic benefits (economizing of ecology)
attract the participation of the stakeholders in the CE transition.
The CE derives from the industrial ecology (IE) paradigm that emphasizes the
benefits of recycling of waste materials and by-products in order to minimize resource
utilization (Andersen 2007). The IE’s “industrial metabolism” promotes the close
cycle of materials and energy flows (Ayres et al. 1989). The studies on the application
of the CE concept in practice have highlighted the material flow perspective, as:
• implementation of closed loop materials flow in the whole economic system (Geng
and Doberstein 2008);
• implementation of the circular (closed) flow of materials and the use of raw
materials and energy through multiple phases (Yuan et al. 2006);
• application of the 3R principles: reduce, reuse, and recycle (Yuan et al. 2006;
Sakai et al. 2011; Preston 2012) in production, consumption and services.
Ghisellini et al. (2016, p. 11) have state that the adoption of the CE concept at a
company level effects increased stakeholders (producers and consumers) responsi-
bility and awareness, with regard to “the use of renewable technologies and materials
(wherever possible), as well as the adoption of suitable, clear and stable policies
and tools”. Moreover stakeholders’ involvement allows for creating additional value
through collaboration and contributes new exchange patterns in the supple chain
(Ghisellini et al. 2016).
Fig. 1 The elements of the

CE transition (micro and
meso level)
Sustainable
innovative
practices
Business
models for
closing
materials
loops &
resource
efficency
In Fig. 1 is presented the summary of the elements of the bottom-up CE transition

from a company perspective. The aspects related to the CE governmental policy
(top-down transition) are not covered in the scope of this paper.
2.1 Innovative Practices
The circular economy pushes the frontiers of environmental sustainability at a

company and in a supply chain. This is achieved by a redesign in materials flows by
innovations. The transition beyond the delayed cradle-to-grave material flow can be
established by transforming business models into innovative self-sustaining supply
chains (Genovese et al. 2017). The challenges of the transition towards CE are
heterogeneous and in order to meet them a multi-perspective approach is required.
The multi-level model of innovation (Rip and Kemp 1998) covers a macro level
of the sociotechnical landscape, the meso level regime and the micro level niche. It
advocates that transition processes of co-evolution happen within and between these
layers. The transition might happen by different agents who are “shaping or making
niches and paths” (Berkhout et al. 2004, p. 50).
According to Kemp et al. (2007, p. 1) transition management is “a multilevel
model of governance which shapes processes of co-evolution using visions, transi-
tion experiments and cycles of learning and adaptation”. The existing policy frame-
works are often fragmented and for that reason are suitable to deal with complex
sustainable development problems. Transition management creates space for inno-
vative governance at local, regional and national level. It combines top-down and
bottom-up approaches to fostering the transition.
The transition is often triggered by forerunners, who are acting autonomously from
the current dominant regime (meso-level). The empowerment of the first movers is
crucial and can be achieved by a joint policy agenda; financial incentives, innovations,
and small-scale experiments (stakeholders’ practices). Transition trajectories can
also be triggers by governance practices at the local, regional and national level. The
transition towards the CE is “predominantly issues of innovation, technical systems,

and fiscal and reformulated business models” (Hobson 2016, p. 89).
The transitions towards Circular Economy in logistics industry can be perceived
as an outcome of “multi-dimensional interactions between radical niche-innovations,
an incumbent regime, and an external landscape” (Verbong and Geels 2010, p. 1215).
The multi-level perspective on social-technological transition assumes that innova-
tion is a pre-condition of systemic change. The innovation starts at isolated innovative
initiatives at local level (so called niches), spreading to society as a whole and, finally,
acquiring transformative scale (so called regime level). The innovative sustainable
practices are created by the pioneers in the niches (micro-level). The innovation
results from market niches or technological niches. The innovative practices usually
are forested by the “protection from small networks of actors who are willing to
invest in the development of new technologies” (Verbong and Geels 2010, p. 1215).
The scaling up of the local sustainable practices results in transition at meso-level
and finally macro-level (the socio-technical landscape).Those levels with regard to
the logistics industry can be defined:
1. Micro-level: single logistics services providers (niches)
2. Meso-level: spreading of initiatives resulting in wider changes in the sector
(patchwork of regimes)
3. Macro-level: paradigm-shift penetrating the entire “landscape” of society
contributing towards circular economy (Fig. 2).
There are different pathways for transition based on the reinforcing or disruptive
relations between “niche innovations” and the existing “socio-technological regime”,
they can be classified as follows (Geels and Schot 2007):
• transformation, when the existing regimes are subject to external pressure (by
different stakeholder from the external landscape), therefore there is need for the
adjustment and reorientation of existing regimes. It is caused mainly by pressure
on the new regulations and new systematic solutions. The innovations are limited
to the niches;
Fig. 2 The multilevel—perspective, developed based on Verbong and Geels (2010, p. 1215)
• reconfiguration, when selected niche-innovations are developed (as regimes face

problems and external pressures) and incorporated, as add-ons or component’s
substitutions, in order to allow the gradual reconfiguration towards “new regime”;
• technological substitution, when new technologies gain more momentum (so
called “niche-accumulation’), and penetrate bigger markets, eventually replacing
the existing “regime”.
• de-alignment and re-alignment, when the uncertainty rises due to the destabiliza-
tion of the existing regimes. The multiple niche-innovations are observed and
after some period of time one option becomes dominant.
The circular economy transitions requires simultaneous shift in the consumers’
behavior, governmental policies and business practices (Van Buren et al. 2016). In
this paper the focus is places on the business model and innovation perspective.
Bidmon and Knab (2018) have stated that the business model can contribute in
transitions, as:
• business models facilitate the process of technological innovation and allow to
expend them from niche to “regime level”,
• novel business models drive transitions without relying on technological innova-
tion.
2.2 Circular Business Models
Studies on circular business model gain a lot of popularity among academia in the last
decade. A business model describes how a company creates, delivers, and captures
value (Osterwalder 2004). The business model serves the company to commercialize
the product and technology innovations (Chesbrough 2010), as they facilitate better
planning, and communicating in face of the increasing complexity of organizational
configurations and activities (Osterwalder and Pigneur 2010). It is a “like a blueprint
for a strategy to be implemented through organizational structures, processes, and
systems” (Osterwalder and Pigneur 2010, p. 10).
Boons and Lüdeke-Freund (2013) have stated that business models allow bringing
the sustainable innovation to the market by combining a value proposition, with the
upstream and downstream value chain and a financial model. They have distinguished
the elements of the business model, as follows (Boons and Lüdeke-Freund 2013):
• value proposition, that defines value, which is embedded in the offered
product/service;
• supply chain, that describes how are structured and managed the upstream
relationships with suppliers;
• customer interface, that describes how are structured and managed the relation-
ships with customers;
• financial model: that covers costs and benefits and their distribution between
stakeholders.
Summarizing, the business model is implemented:

• to provide the value for stakeholders by delivering products and or services;
• to create cooperation downstream and upstream in the supply chain;
• to capture value and sustain the financial stability.
Circular business model (CBM) describes business model that supports transition
towards the Circular Economy at a company level taking into consideration the
resources and capacities. Bocken et al. (2016) have stated that such a business model
allows to: slow, narrow, and close resource loops, therefore it is possible to reduce
resource input into the company and to minimize waste and emissions leakage.
Geissdoerfer et al. (2018) have extended the above mentioned classification
by adding two more circular business models. They have highlighted to impor-
tance of intensifying the usage phase of the existing resources, and the substitution
(where possible) of product utilization by service and software solutions (so called
dematerializing).
The CBM aims to create the monetary and non-monetary value by innovations
(technological, organizational and social) and pro-active management of a multiple
stakeholders.
The next section of the paper present the relation between the circular business
model and transition research with regard to the logistics industry.
3 Transition Framework
The sustainable practices are described usually at a single company or in the supply
chain context. The sustainable practices in the supply chain can be divided into:
• sustainable procurement (sustainable purchasing);
• sustainable warehousing;
• sustainable transport;
• sustainable packing;
• reverse logistics.
The implementation of the circular economy principles in the supply chain can
be translated into different practices, for example such as:
• reverse logistics,
• creating suppliers’ communities (industrial symbiosis),
• local sourcing or dual sourcing with secondary materials,
• product service systems (PSS),
• sharing of infrastructure.
The existing body of literature provides very limited discussion on the perspective
of the logistics service providers (LSPs). The logistics service providers’ (LSPs) link
different stakeholders (manufacturers, retailers, consumers etc.) in the supply chain,
therefore they have potential to facilitate the diffusion of the sustainable practices and
transition towards the CE. In the logics industry the logistics services providers (3
PLs -5 PLs) coordinate and orchestrate the downstream and upstream relationships
in the supply chain, therefore enabling the diffusion of the innovative practices at the
intra-organizational level.
PLs face a lot of challenges related to: globalization, pressure on the increased
business-to-business relations in production–consumption networks, development
of new disruptive technologies that may reconfigure the existing business networks
(e.g., 3D printing, IoT), increasing importance of after sales services and reverse
logistics. Moreover the expanding e-commerce has resulted in a complex last mile
distribution and related reverse flow of commercial products, so called first mile
distribution (Van Buren et al. 2016).
The first research question in this paper is stated, as:
RQ1: What sustainable practices are implemented by the logistics providers (from
1PL to 5PL)?
The LPSs can be characterized with regard to sustainability, as follows (based on
Gruchmann 2019):
• Single Service Provider (1PL)—a company offers a single logistics service (trans-
port, warehousing etc.) usually as subcontractor. It implements the sustainable
practices in order to meet the CO2 emissions target by reduction of its environ-
mental impact of logistical assets (e.g. using HGVs with cleaner drive, increased
energy efficiency, etc.).
• 2nd Party Logistics Provider (2PL)—a company offers all classical logistics func-
tions (transportation, handling and warehousing) and usually it uses more than
single transport mode. It implements the sustainable practices by using multi-
modal transport, and the selection of the best modal split allows reducing the
environmental impact of its logistical activities.
• 3rd Party Logistics Provider (3PL)—a company offers extended services in
comparison to the 2PL by creating the added value to customers by cross docking,
inventory management, co-packaging etc. Usually it is a globally acting company.
It has got sufficient financial resources to invest in the advanced decision support
systems dedicated to lower environmental impact of its operations (GDSS) by
optimizing transport mode, route and capacity usage. Usually 3PL also imple-
ments and communicates to the customers the social corporate responsibility
standards (CSR).
• 4th Party Logistics Provider (4PL)—a company acts, as an integrator in the supply
chain and manages the network of 3PLs. It acts, as one-stop-shop and often it oper-
ates without owning logics infrastructure (so called non-asset-owning service
providers). The sustainable practices focuses on the achieving greener supply
chain configurations, and ensuring meeting the CSR standards among the partici-
pants. It also can integrate all reverse logistics flow in the supply chain (first mile
logistics).
• 5th Party Logistics Provider (5PL)/Lead Logistics Provider—a company offers
services that focus on the design and management of the supply networks. It
focuses on the e-business support. It implements the sustainable practices by

integration of the commodity flows and reverse flows by implementing the sharing
and circular economy principles.
Gruchmann et al. 2018 has also proposed to add to the above classification the Lead
Sustainability Service Provider (6PL). That terminology is not yet well established
in the literature or practice and for that reason it is not included in this paper in the
further considerations.
The previous studies (e.g. Gruchmann 2019; Gruchmann et al. 2018; Govindan
and Hasanagic 2018; Genovese, et al. 2017; Van Buren et al. 2016) have identified
the number of the practices, drivers and barriers for the development of the sustain-
able practices in the supply chain. Table 1 presents the summary of the sustainable
practices from the literature review. The practices are clustered into technological,
organizational and social category. As, the PLs link other in the supply chain it is
assumed that the scaling up of sustainable practices in logistics sector is related to
the enablers that are classified four dimensionally, as:
• HI—horizontal integration (with other companies in logistics sector);
• VI—vertical integration (upstream the supply chain);
• CI—customers integration (downstream the supply chain);
• RI—reverse flow integration (RL).
The above mentioned practices can be summarized, as:
• avoidance of the unnecessary transport, increasing utilization of existing resources
and usage of cleaner (low emission) technologies (practices: T.1–T.6, T.10–T.12,
O.1–O.6, O.8, S.4);
• city logics (practices: O.7; T.9, S.1–S.3); after sales services and reverse logistics
(practices: T.7, T.8);
• after sales services and reverse logistics (practices: T.7, T.8);
• recovering and re-using resources (practices: O.9, O.10).
The avoidance of unnecessary transport can be achieved by optimization of
product flows towards the consumer/end user, using advanced IT systems in order to
optimize routes and drop offs, as well as environmental friendly and energy efficient
transport modes (and combination of them). The infrastructure sharing and inno-
vative ownership models (e.g. pay for use, PPS etc.) allow for increasing the load
factor of fleet and higher utilization rate of the other infrastructure (e.g. distribution
centers). The LPs can act also as facilitator towards industrial symbiosis (relocation
of the companies into industrial/eco parks) and local sourcing. The material flow
perspective (through the logistics hubs) allows relating the local infrastructure with
the meso-level perspective (the supply chain). The innovation in the material flows
can lead to the spatial reconfiguration of the stakeholders in supply chain (e.g. moving
from current location to the industrial park). The industrial park can contribute to
the sustainable development in the future as “as companies can co-locate to facilitate
the material interchange and ‘energy cascading’ necessary for closed production-
consumption loops (Gibbs et al. 2005, p. 174). According to the critics of the CE
Table 1 The sustainable practices in logistics industry, developed based on Gruchmann (2019),
Gruchmann et al. (2018), Genovese et al. (2017), Van Buren et al. (2016)
Sustainable practice Enabler 1PL 2PL 3PL 4PL 5PL
Technological
T.1 Multimodal transportation HI – + + + +
T.2 E-mobility HI – – – – +
T.3 Implementation of cleaner technologies (e.g. hybrid, VI + + + – –
bio fuels, hydrogen, LNG, Power-to gas etc.)
T.4 Advanced digital solutions (e.g. Green Decision VI/HI – – + + +
Support Systems GDSS)
T.5 Last-mile bundling of cross company commodity VI – – – + +
flow
T.6 Last-mile DSS (for decreased travel distance and VI – + + + +
increased the drop-off rate)
T.7 First-mile integration of reverse logistics RI – – – + +
T.8 Sustainable packing VI/RI – – + + +
T.9 Alternative transport modes in urban logistics (e.g. VI + + + + +
cargo bikes)
T.10 Join development of new technologies (e.g. VI/HI – – – + +
co-financing)
T.11 Subcontracting platforms HI + + + + +
T.12 Increased energy efficiency VI/HI + + + + +
Organizational
O.1 Shared used infrastructure (e.g. warehouses) HI + + + + +
O.2 Reconfiguration of the stakeholders (e.g. relocation VI – – – + +
to industrial/eco-parks)
O.3 Product Service Systems PSS (leasing, renting, pay HI + + + – –
per unit, etc.)
O.4 Cross-company use of vehicles HI + + + – –
O.5 Eco-driving training for drivers HI + + + – –
O.6 Consolidation commodity flow HI – – – + +
O.7 Micro hubs and depots HI/VI – – + + +
O.8 Implementation of environmental standards HI/VI – + + + +
O.9 Dual sourcing (e.g. purchase of remanufactured, VI + + + – –
recycled products for servicing purpose)
O.10 Support for local sourcing
Social
S.1 Crowd logistics solutions CI – + + + +
S.2 Consumers empowerment CI – – + + +
S.3 Building customers’ networks/communities CI – – + + +
S.4. Implementation of CSR HI/VI – + + + +
approach e.g. (Allwood 2014), the CE practices in material flow management in a

company or group of companies (a supply chain) might lead to an increased nega-
tive environmental impact. Therefore the impact of the local and regional transport
on the environment can be increased, so use of eco-friendly and energy efficient
technologies should be ensured.
The sustainable practices in city logics are focused mainly on the last mile alter-
native transport modes (e.g. cargo bikes), innovative (however still unregulated and
therefore controversial) crowd logistics solutions. The underdeveloped potential is
laid in the emerging integration of the first and last mile logistics with regard to the
e-commerce in urban/metropolitan areas.
The new sustainable practices are emerging that allow for recovery and re-using
the resources (e.g. usage of the remanufacturing components in the fleet servicing).
Moreover LSPs can play a vital role in providing services in order to become the a
“focal point” for return flows, facilitating economy of scale in recycling, re-use, the
supply of new biodegradable raw materials or raw materials for 3D printing (Van
Buren 2016).
The shift towards circular business models and related innovations (technological,
social, organizational) may allow to change the existing practices in the utilization
of logistics infrastructure (like for example: networking, infrastructure sharing, shift
from ownership to availability models, dual sourcing with primary and recycle mate-
rials) and management of materials flow (closing, narrowing or slowing the materials
loop).
Van Buren et al. 2016 has stated that logistics sector should change the focus
from the services dedicated to moving goods (from A to B) to innovative services
that becoming an indispensable enabler of circular economy through adding value
service to production–distribution–consumption chains of other industries.
The current body of literature doesn’t sufficiently link the sustainable practices and
circular business models with regard to the logistics industry. Most of the CBM which
are described in the literature focus on the products’ perspective not services’ perspec-
tive. Therefore the second research question aims to fill that gap. It is formulated
as:
RQ 2. How the sustainable logistics practices can be linked with regard to the
circular economy business models?
The CBM models are considered, that are described in the previous section. In
Table 2 are linked the sustainable practices in logistics industry with the CBMs. The
codding of the sustainable practices is the same, as in Table 1.
The majority of the proposed practices support CBMs for the resource efficiency
and intensifying usage of the existing resources. Currently the logistics infrastructure
is used at relatively low rate. The material flow perspective (flow of products through
the hubs) allows relating the local infrastructure with the meso-level perspective
(the supply chain). The technological, organizational and social innovations can
result in the redesign of services for reuse and easier value recovery in multiple
life-cycles, resulting in new relations between stakeholders in the supply chain. The
Table 2 Sustainable logistics practices with regard to the CBM, developed based on Bocken et al.
(2016), Geissdoerfer et al. (2018)
Circular business model The goal for logistics industry Sustainable practice
Slowing the resource loop The focus is placed on the O.3; O.9
business practices that allows for
extended useful life period (e.g.
through repair, remanufacturing)
of logistics infrastructure and
fleet
Closing resource loops The focus is placed on the T.7; T.8
business practices that allow for
recycling and reuse of the
recycled materials in next life
cycles
Resource efficiency or The focus is placed on the T.2; T.3; T.6; T.10; T.12; O.2;
narrowing resource flows business practices that allow to O.3; O.5; O.8 O.10; S.4
reduce the utilization of resource
needed with regard to the
providing services for the
customers
Intensifying the usage The focus is placed on the T.1; T.5; T.9; T.11; O.1; O.4;
phase of the resources business practices that allow to O.6; O.7
eliminate the inefficiencies in
utilization the use of resource
(e.g. by sharing, subcontracting)
needed with regard to the
providing services for the
customers
Dematerializing The focus is placed on the T.4; T.6; S.1; S.2; S.3
business practices that allow for
decoupling the resources from
service provided for the
customers
transition towards CE (at meso-level) might leads to increased role of the local
sourcing therefore reducing long-haul transportation.
Taking into consideration the “multi-perspective transition theory (as described
in the Sect. 2) two pathways towards CE are supported by the existing sustainable
practices in the logistics sector, namely:
• reconfiguration, when selected innovative sustainable practices (niche-
innovations) will be incorporated, as add-ons or component’s substitutions, in
order to allow the gradual reconfiguration towards circular business models in
the industry, especially with regard to sharing the logistics infrastructure, moving
from infrastructure ownership towards pay-for-use (PSS), therefore leading to
resource efficiency and eco-friendly intensification in usage of the existing vehi-
cles fleet (increased load factor) by implementation of the advanced GDSS and
bundling cross company product flow and/or bundling the reverse logics flow
from consumers.
• technological substitution, when new technologies (e.g. e-mobility, 3-D printing)
gain more momentum (so called “niche-accumulation’), and allow penetrate
bigger markets, eventually replacing the existing “regime” in logistics industry.
4 Conclusions and Further Research
In this paper the focus is placed on the discussion on the role of the logistics service
providers with regard to the CE. This topic is underrepresented in the literature. The
logistics services providers play dual role with the regard to the CE. First they may
implement the CBM on the organizational level (micro-level), secondly they may
facilitate (act as focal point, or coordinator) for the supply chain transition towards
CE (meso-level).
The main contribution of this paper the classification of the innovative techno-
logical, organizational and social sustainable practices in logistics sector and then
linking them to the circular business models. The multi-perspective transition theory
has been used in order to allow for discussion on the potential transition of the logics
sector towards CE. The bottom-up approach has been used in order to highlight how
local innovations in the niches (at micro-level) can contribute to scaling up transition
in the sector.
Limitation of this research is the focus on the bottom-up approach. The impact
of the governmental policy and tools on the development of the sustainable logistic
practices is not investigated in this paper. The considered enablers are limited to the
integration with other stakeholders (other companies in logistics sector, customers,
suppliers and manufacturers).
The further research will focus on the empirical studies among the logistics service
providers in order to verify to importance of link between the sustainable business
practices and CBMs. Moreover the enablers for the logistics sector towards circular
economy will be further investigated.
References
Allwood JM (2014) Squaring the circular economy: the role of recycling within a hierarchy of
material management strategies. In: Handbook of recycling, pp 445–477
Andersen MS (2007) An introductory note on the environmental economics of the circular economy.
Sustain Sci 2(1):133–140
Ayres RU, Norberg-Bohm V, Prince J, Stigliani WM, Yanowitz J (1989) Industrial metabolism, the
environment, and application of materials-balance. In: Principles for selected chemicals
Berkhout F, Smith A, Stirling A (2004) Socio-technical regimes and transition contexts. In: Elzen
B, Geels FW, Green K (eds) System innovation and the transition to sustainability. Edward Elgar,
Cheltenham
Bidmon CM, Knab SF (2018) The three roles of business models in societal transitions: new linkages
between business model and transition research. J Clean Prod 178:903–916
Bocken NM, de Pauw I, Bakker C, van der Grinten B (2016) Product design and business model
strategies for a circular economy. J Ind Prod Eng 33(5):308–320
Bocken NM, Short SW, Rana P, Evans S (2014) A literature and practice review to develop
sustainable business model archetypes. J Clean Prod 65:42–56
Boons F, Lüdeke-Freund F (2013) Business models for sustainable innovation: state-of-the-art and
steps towards a research agenda. J Clean Prod 45:9–19
Brundtland (1987) World commission on environment and development. Our common future.
OxfordUniversityPress, Oxford
COM/2015/0614: Communication from the Commission to the Europ. Parl., the Council, the Euro-
pean Economic and Social Com. and the Com. of the Regions, Closing the loop—An EU action
plan for the Circular Economy, COM/2015/0614 final
Chesbrough H (2010) Business model innovation: opportunities and barriers. Long Range Plan
43:354–363. https://doi.org/10.1016/j.lrp.2009.07.010
De Angelis R, Howard MB, Miemczyk J (2018) Supply chain management and the circular
economy: towards the circular supply chain, production planning and control (n print) available
from http://ore.exeter.ac.uk
EC (2011) Roadmap to a resource efficient Europe. Available at: http://www.europarl.europa.
eu/meetdocs/2009_2014/documents/com/com_com(2011)0571_/com_com(2011)0571_en.pdf.
Accessed 3 Apr 2018
Ellen MacArthur Foundation (2015) Towards the circular economy: economic and busi-
ness rationale for an accelerated transition. Isle of Wight: Ellen MacArthur Founda-
tion. Available http://www.ellenmacarthurfoundation.org/assets/downloads/TCE_Ellen-MacArt
hur-Foundation_9-Dec-2015.pdf. Accessed 4 Apr 2018
Geels FW, Schot J (2007) Typology of sociotechnical transition pathways. Res Policy 36(3):399–417
Geissdoerfer M, Bocken NM, Hultink EJ (2016) Design thinking to enhance the sustainable business
modelling process–a workshop based on a value mapping process. J Clean Prod 135:1218–1232
Geissdoerfer M, Morioka SN, de Carvalho MM, Evans S (2018) Business models and supply chains
for the circular economy. J Clean Prod 190:712–721.T
Geissdoerfer M, Savaget P, Bocken NM, Hultink EJ (2017) The circular economy–a new
sustainability paradigm? J Clean Prod 143:757–768
Geng Y, Doberstein B (2008) Developing the circular economy in China: challenges and
opportunities for achieving ‘leapfrog development’. Int J Sustain Dev World Ecol 15(3):231–239
Genovese A, Acquaye AA, Figueroa A, Koh SL (2017) Sustainable supply chain management and
the transition towards a circular economy: evidence and some applications. Omega 66:344–357
Ghisellini P, Cialani C, Ulgiati S (2016) A review on circular economy: the expected transition to
a balanced interplay of environmental and economic systems. J Clean Prod 114:11–32
Gibbs D, Deutz P, Proctor A (2005) Industrial ecology and eco-industrial development: a potential
paradigm for local and regional development? Reg Stud 39(2):171–183
Golinska P, Kuebler F (2014) The method for assessment of the sustainability maturity in
remanufacturing companies. Procedia Cirp 15:201–206
Govindan K, Hasanagic M (2018) A systematic review on drivers, barriers, and practices towards
circular economy: a supply chain perspective. Int J Prod Res 1–34
Gruchmann T (2019) Advanced green logistics strategies and technologies. In: Operations, logistics
and supply chain management. Springer, Cham, pp 663–686
Gruchmann T, Melkonyan A, Krumme K (2018) Logistics business transformation for sustain-
ability: assessing the role of the lead sustainability service provider (6PL). Logistics 2(4):25
Hobson K (2016) Closing the loop or squaring the circle? Locating generative spaces for the circular
economy. Prog Hum Geogr 40(1):88–104
Huppes G, Ishikawa M (2005) A framework for quantified eco-efficiency analysis. J Ind Ecol
9(4):25–41
Hussain M, Malik M (2020) Organizational enablers for circular economy in the context of
sustainable supply chain management. J Clean Prod 256:120375
Jawahir IS, Bradley R (2016) Technological elements of circular economy and the principles of
6R-based closed-loop material flow in sustainable manufacturing. Procedia Cirp 40:103–108
Kemp R, Loorbach D, Rotmans J (2007) Transition management as a model for managing processes
of co-evolution towards sustainable development. Int J Sustain Dev World Ecol 14(1):78–91
Krikke H, Blanc IL, van de Velde S (2004) Product modularity and the design of closed-loop supply
chains. Calif Manag Rev 46(2):23–39
McDonough W, Braungart M (2002) Cradle to Cradle: remaking the Way We Make Things, 1st
edn. North Point Press, New York
Murphy J, Gouldson A (2000) Environmental policy and industrial innovation: integrating
environment and economy through ecological modernisation. Geoforum 31(1):33–44
Murray A, Skene K, Haynes K (2017) The circular economy: an interdisciplinary exploration of
the concept and application in a global context. J Bus Ethics 140(3):369–380
Osterwalder A (2004) The business model ontology: aproposition in a design science approach.
PhD thesis. Universite de Lausanne – Ecole des Hautes Etudes Commerciales
Osterwalder A, Pigneur Y (2010) Business model generation: a handbook for visionaries, game
changers, and challengers. Wiley
Park J, Sarkis J, Wu Z (2010) Creating integrated business and environmental value within the
context of China’s circular economy and ecological modernization. J Clean Prod 18(15):1494–
1501
Pearce DW, Turner RK (1990) Economics of natural resources and the environment. JHU Press
Preston F (2012) A global redesign? Shaping the circular economy. Energy, Environ Res Gov 2:1–20
Rashid A, Asif FM, Krajnik P, Nicolescu CM (2013) Resource conservative manufacturing: an
essential change in business and technology paradigm for sustainable manufacturing. J Clean
Prod 57:166–177
Rip A, Kemp R (1998) Technological change. Hum Choice Clim Change 2:327–399
Sakai SI, Yoshida H, Hirai Y, Asari M, Takigami H, Takahashi S, Douvan AR (2011) International
comparative study of 3R and waste management policy developments. J Mater Cycles Waste
Manage 13(2):86–102
Stahel W (1982) The product life factor. In: Orr GS (ed) An inquiry into the nature of sustainable
societies. The role of the private sector. Houston Area Research Centre, Houston, pp 72–105
Stahel WR, Clift R (2016) Stocks and flows in the performance economy. In: Taking stock of
industrial ecology, pp 137–158. Springer International Publishing
Van Buren N, Demmers M, Van der Heijden R, Witlox F (2016) Towards a circular economy: the
role of Dutch logistics industries and governments. Sustainability 8(7):647
Verbong GP, Geels FW (2010) Exploring sustainability transitions in the electricity sector with
socio-technical pathways. Technol Forecast Soc Chang 77(8):1214–1221
Weissbrod I, Bocken NM (2017) Developing sustainable business experimentation capability–a
case study. J Clean Prod 142:2663–2676
Yuan Z, Bi J, Moriguichi Y (2006) The circular economy: a new development strategy in China. J
Ind Ecol 10(1–2):4–8
A Literature Analysis of Definitions
for a Circular Economy
Usama Awan, Narmeen Kanwal, and Mohammed Khurrum S. Bhutta
Abstract This chapter aims to identify and analyze the published definitions of
circular economy (CE). Twenty-eight definitions were gathered through intensive
critical literature review, using both Scopus and Web of Science. The definitions
developed from peer-reviewed literature analysis covered a period from 1999 through
2019, although most definitions were published from 2011 onwards. CE received
significant attention in the early 90s and now is considered a mainstream strategy
for product design and social, economic and environmental sustainability. Given
that research is still relatively new in the sustainable circular economy. CE offers
a reverse resource regenerative idea to eliminate the linearity of production and
consumption system to support sustainability objectives. The CE definition analysis
revealed that so far, resilience and stakeholder perspective is not explicitly included in
the definition, although stakeholder is considered to be part of a natural and ecology
system. CE has an impact on different aspects of the business throughout the entire
supply chain. The concept of a CE is a value-orientated resource transformational
process. CE considers both upstream and downstream production and consumption
patterns to promotes the resilience orientation of resources. Currently, CE practices
are carried out—meso, micro and macro. This chapter highlights that suggest that
it is imperative to consider exosystem and chronosystem to better move away from
linear to circular economy.
Keywords Circular economy · Resource regenerative · Product design ·

Stakeholder’s perspective · Supply chain
U. Awan (B)
Industrial Engineering and Management, The Lappeenranta-Lahti University of Technology LUT,
Lappeenranta 5385, Finland
e-mail: Usama.awan@Lut.fi
N. Kanwal
University of Trier, Weidengraben 90, A18, 54296 Trier, Germany
e-mail: narmeenkanwal786@yahoo.com
M. K. S. Bhutta
College of Business, Ohio University, 522 Copeland Hall, Athens, OH 45701, USA
e-mail: bhutta@ohio.edu

20 U. Awan et al.
1 Introduction
Global warming and climate change issues are gaining popularity, thus the managers
of supply chains are faced with the challenge of how to integrate the circular economy
(CE) practices into the production and management of their organisations. The CE
practices are a way forward to lowering global warming, greenhouse gases, and
minimizing air pollution by integrating sustainability objectives into the design of
operations and products. There is a growing realization that companies must direct
the issue of resource scarcity and dematerialization in their product physical life
cycle. Since the inception of the sustainability concept in the 1980s, debates on
the application circular economy has grown significantly (Awan et al. 2020). CE
is commonly defined as resource reduction to meet the sustainability agenda of
the existing and future generation. Recently, CE has been the focus of attention
from practitioners and academicians. There is increasing recognition of CE on the
political agenda of nations and particularly in Europe (European Commission 2015).
CE is expected to play a vital role in the move towards economic development by
generating new business opportunities, saving material cost, improving the security of
future supply, while at the same time move towards a sustainable future (Kalmykova
et al. 2017). CE is commonly considered as a solution for environmental protection
and economic growth by utilizing resources in circularity. CE is understood as the
“realization of closed-loop material flow in the whole economic system” (Geng and
Doberstein 2008).
In CE definitions confusion arises frequently surrounding the principles of 3R
(reuse, reduction and repair). The term CE has been interpreted in a variety of
different ways, ranging from closed flow of material to a “spiral-loop system” (Stahel
and Reday-Mulvey 1981; Yuan et al. 2006). Previous CE definition tends to focus
on transformation function of resources in the value chain (EllenMacArthur Foun-
dation 2013), but as research expands, they are adopting CE concepts to sustainable
development (SD) (Kirchherr et al. 2017b). There has been an attempt to establish a
preferable grasp understanding of the CE and the application of SD in an industrial
ecology context. CE has been defined in various ways, with one possibility “opti-
mising consumption of resources and patterns, and redesigns the industrial system at
the system level” (European Commission 2014). In a broader perspective, CE signi-
fies the resilience, resource efficiency, and a restorative economic system. Perhaps the
best known was by Stahel and Reday-Mulvey (1981), who defined CE in academic
literature, as a “closed-loop economy”. The management of the CE is receiving
increasing recognition. A number of recent literature reviews on the CE definition
have been published including Ghisellini et al. 2016; Lewandowski 2016; Lieder
and Rashid 2016; Sauvé et al. 2016; Blomsma and Brennan 2017; Geissdoerfer et al.
2017; Kirchherr et al. 2017b; and Murray et al. 2017. These literature reviews are
indeed necessary in order to further develop the concept and transparency regarding
the existing understanding of CE (Blomsma and Brennan 2017; Kirchherr et al.
2017a). However, they merely present an account of available definitions, basis to
understand how CE is currently constructed at the macro, meso and micro level.
A Literature Analysis of Definitions for a Circular Economy 21
There is an ongoing discussion about the significance and application of CE in a

business context. However, the CE definition has not been explored in-depth, there
are numerous and varied definitions of CE (Lieder and Rashid 2016). There is no
broadly agreed definition of CE in the literature (Yuan et al. 2006). While Kirchherr
et al. (2017b) write that no one single definition of CE has gained international
acceptance. Merli et al. (2018) have pointed out the need for developing broader and
more dynamic definitions and associated practices, in which they viewed CE as a
system in which the value of the materials, products and resources is maintained in
the entire product lifecycle.
A specific conceptual and theoretical development might be required for a better
understanding of the concept in discourse. As far as, there is no comprehensive
definition of the CE, however, CE is often studied with an explicit definition of the
concept (Blomsma and Brennan 2017). A recent study on the conceptualization of
CE through analyzing definitions by Kirchherr et al. (2017b) focusing on broadening
the concept of the CE among scholars and practitioners. CE is defined in so many
different perspectives that it is difficult to come up with a consensus single definition.
One possible explanation as to why there is no consensus regarding the definition
of the CE could be due to the differences of conceptualization of various members
about the implementation of CE in the context of reducing, reuse and recycling.
These challenges potentially identify a need for a new definition of CE, as reported
by the previous discussion in the literature. Firms employing CE approaches would
also want to seize the potential benefits and opportunities offered by making use of
waste prevention strategies before the design of the product. It is therefore important
to discover what has not been addressed in the previous CE definitions and identify
the knowledge gaps to suggest a new definition in a system perspective. This gap
in the literature provided the impetus for the analysis of CE definitions by specific
characteristics and measures, which is imperative for achieving sustainable growth.
This study is distinct from previous research studies in two aspects. First, the study
used keywords search from Scopus and Web of Science from the period of 1999–
2019. Second, the study addressed key CE characteristics assessed by the definitions
in the literature. This study provides an overview of CE approaches development and
analyzes the previous literature on circularity in firms to provide a shared definition
and provides a series of future research avenues. Thus, the purpose of this paper is to
identify and analyze how CE is defined in the existing literature and suggest future
research direction. We have chosen the systematic literature review approach and
summarize the salient findings of previous studies. The contribution of this study is
two-fold: first to provide an update to literature review on CE, as well as to document
the strengths and weakness of definitions in said literature. Second, to structure
the literature in a way to provide a basis for the development of a comprehensive
definition for a CE and point for future research in these areas.
22 U. Awan et al.
1.1 Systematic Literature Review Methodology
The systematic literature review was limited to articles that were published in English
peer-reviewed journals during the last 20 years (1999–2019). We consider this
timespan to the possible inclusion of the articles published during this time. We
include Scopus and Web of science our main identification source of articles. We
selected both databases due to its broad coverage in engineering and technology
journals. Since Scopus does not cover every peer-reviewed publication, it is possible
to relevant papers published in the domain of circular economy might not miss. So, to
avoid this, we also consider the web of science to search the related published work.
To identify the relevant articles, we followed a structured pre-defined keywords
search. Accordingly, the terms, “circular economy” and “Cradle to cradle” and
“Closed-loop” were separately searched. The search word conducted along with
the key terms “definition”, “defined”, define” and “refers” by utilizing “all field cate-
gory” search in title, abstract and keywords. The data range to search the key terms
were set for the papers published from 1999 to 2019 by utilizing the subject areas
category. As of October 7, 2019.
The systematic literature review approach using keywords has been used previ-
ously in many studies. In this present study, to identify the most relevant articles, we
decided to reduce the number of articles for the review following three additional
criteria for our literature review on circular economy and cradle to cradle approach.
First, to be included in our review, the title of the article must deal with the circular
economy concept. Second, an abstract of the article must reflect the circular economy
as the main concept related to the reuse, repair and recycling. Finally, the journal in
which article appeared must have been ranked in the source database. After applying
such criteria and removing duplicates articles that were appeared in a web of science
and Scopus data sources. A total of 1134 articles identified in the engineering, envi-
ronmental and management science. As of 347 duplicates articles were excluded
from the review at the initial stage. we then eliminated 411 articles that did not fit
these criteria, it led to an initial number of 376 articles. As a result of this approach,
our overall sample included 177 articles.
After reading this article in-depth, we became familiar with this concept. We
also noted that circular economy approaches have also appeared in published reports
which have not been appeared in our keyword research. We carefully read these
published working papers and decided to include in the references section. Given
the scope and space consideration for this published working papers, however, we
present a brief view of definitions that have been appeared in this literature.
1.2 The Emergence of the Circular Economy Concept

and Definitions
The idea of the circular economy (CE) was first introduced by British environmental
economists. They described the environment as a waste reservoir without a built-in
propensity to recycle (Pearce and Turner 1990). In 1976, Stahel and Ready-Mulvey
encouraged scholars to contribute to the natural resources and referred to CE as a
“closed-loop economy”. Stahel and Reday-Mulvey (1981) In a regenerative ecology
system, the origin of the word CE was not known. The term CE comes from western
literature in the 1980s, to describe a closed-loop system concept to the recognition
of the importance of quality of natural resources (Pearce and Turner 1990). In litera-
ture, the term gained prominence in the late ‘90s after prominence of the German and
Swedish industrial ecological paradigm. CE was viewed as activity resource produc-
tivity and efficiency in the perspective of industrial ecology. Some claims that it
first appears in China (Yuan et al. 2006). A circular economy is a multi-dimensional
concept root in a product designed in such a way that can fully be recycled (Yap
2005). Several previous studies investigated CE definition in the engineering disci-
pline based on the categorization of concepts in which CE appeared. CE received
significant attention in the early 90s and now is considered a mainstream strategy for
product design and social, economic and environmental sustainability.
The concept of CE was narrow and the focus was on a healthy economy and envi-
ronment co-existing as a way to integrate environmental management throughout the
twentieth century (Chertow 2000). The German and Swedish environmental policy
on a closed-loop originated from the industrial paradigm, and it has been followed
by China’s policymakers as a prospective strategy to furnish a solution to existing
environmental problems. The concept of CE comes out from the industrial ecology
and has a closed-loop of product flow. Industrial ecology (IE) is referred Awan (2020)
to as “a subdiscipline of the natural ecosystem which aims to restructure the indus-
trial ecosystem in ways of managing and designing linear to closed-loop industrial
production and consumption system. Industrial ecology seeks to form harmonized
relations between ecological and human systems to provide sustainable benefits of
all aspects of sustainability including social, environmental, and economic” (p. 370).
The focus of CE is on eco-efficiency and resource productivity to achieving improve-
ments in resources and achieve. The government of China that enforced the policies
related to “circular economy”, these were referred to as “Circular Economy Promo-
tion Law of the People’s Republic of China”. The concept of CE first to come into
force in January 2009. CE was initially launched at three-levels in China, individual
business level, eco-industrial parks, and at the eco-cities municipalities. Under these
principles, CE aims to promote economic growth and at the same time bring about
an economic development strategy (minimise material consumption and energy use)
(Yuan et al. 2006). Industrial networks of China advance the initiative of CE and its
practices at a country level.
Another phenomenon related to macro-level or national level has been described
as the promotion of consumption and production of efficient resources, promote
24 U. Awan et al.
eco-cities with the aim of developing a recycling-oriented society (Geng et al. 2012).
The term micro level means implementing cleaner production initiatives at the firm
level. Their term Meso level means the study of inter-firm at the supply chain level
represented firms in cluster use waste of another firm (Yuan and Shi 2009). The
word circular related to the concept of cycle and economy in terms of production
and consumption of two cycles, (1) bio-geochemical cycles and (2) idea of recy-
cling of products (Murray et al. 2017). The idea of the circular economy has been
expressed by (Kirchherr et al. 2017b) in 3R (i.e., reduce, reuse and recover) and
4R (European Commission 2008), reduce, reuse, recycle and recover. CE is defined
as “an economic model wherein planning, resourcing, procurement, production and
reprocessing are designed and managed, as both process and output, to maximize
ecosystem functioning and human well-being” (Murray et al. 2017).
Yuan et al. (2006) concept of CE was the foundational step in developing CE as a
path to achieving eco-efficiency and resource productivity. While existing research
has in a common theme “cyclical closed-loop system” (Murray et al. 2017). The
basic premises of the CE appear to be closing and slowing loops. Closing loops
refers to (post-consumer waste recycling, slowing is about retention of the product
value through 3R principals (Murray et al. 2017). Recently, Blomsma and Brennan
(2017) term CE as prolonging resource productivity. However, Slowing resource
loops referred as “Through the design of long-life goods and product-life extension
(i.e. service loops to extend a product’s life, for instance through repair, remanufac-
turing), the utilization period of products is extended and/or intensified, resulting in
a slowdown of the flow of resources”. Closing resource loops referred to as “through
recycling, the loop between post-use and production is closed, resulting in a circular
flow of resources” (Bocken et al. 2016, p. 309). Stahel (2016) referred good use of
the utilization of end of life products as Loop (or circular) Economy.
This unavailability of definitional comprehensibility brings forth the potential
approach of misunderstanding, supporting dispersion than the convergence of views
and additionally predicting cumulative examination of development on circular
economy approaches (Table 1).
It is interesting to note that no definitions define the stakeholder aspects except
(Murray et al. 2017), but rather described what CE constitutes for human well-
being. There are a variety of available definitions of CE and they are frequently
referring to few dimensions, although they relate to different phrases, they fail to
represent any guidance on how to manage stakeholders perspectives within these
CE definitions. Circular economy order origin as a normative closed-loop strategy
discipline becoming popular in 2003. Some scholars presented an inclusive treatment
of CE definition which positions CE as a hybrid field intertwined from industrial
ecology, natural resources and regenerative ecology system. Eco-effectiveness has
been couched by CE to a great extent in which focus is to material extraction to
continue use at the end of life and recycle.
Literature analysis reveals that there is still a need to realise benefits from industry
4.0 including co-creation and expansion of the circular economy. Additionally, there
are no research studies which investigated what artificial intelligence management
capabilities needed for successful industry 4.0 transformations. Internet of things
A Literature Analysis of Definitions for a Circular Economy
Table 1 Representative definitions of circular economy
S. No. Authors Key definition Key terms
1 Cooper “A circular economy is proposed, in which the throughput of energy and raw materials is reduced” (p. 10) Reduction of use of
(1999) energy and raw material
consumption
2 Yap (2005) “Circular economy is described as a scientific development model where resources become products, and the Emphasize on recycling
products are designed in such a way that they can be fully recycled” (p. 13)
3 Yuan et al. “CE is the circular (closed) flow of materials and the use of raw materials and energy through multiple phases” Focus on the closed flow
(2006) (p. 5) of material
4 Geng and “A circular economy approach encourages the organisation of economic activities with feedback processes Reuse of the materials
Doberstein which mimic natural ecosystems through a process of ‘natural resources → transformation into manufactured
(2008) products → byproducts of manufacturing used as resources for other industries” (p. 232)
5 Liu et al. “Circular economy defines its mission as solving the problems from the perspective of reducing the material flux Reduction of material use
(2009) and making the material flow balanced between the ecosystem and the socioeconomic system” (p. 265)
6 Hu et al. “Circular economy (CE) focuses on resource-productivity and eco-efficiency improvement in a comprehensive Eco-efficiency and
(2011) way, especially on the industrial structure optimization of new technology development and application, resource productivity
equipment renewal and management renovation” (p. 221)
7 Bilitewski “A circular system when the connection between resource use and waste residuals” (p. 1) Use of waste
(2012)
8 MacArthur “A circular economy is an industrial system that is restorative or regenerative by intention and design. It replaces Regenerative and
(2013) the ‘end-of-life’ concept with restoration, shifts towards the use of renewable energy, eliminates the use of toxic restorative of resources
chemicals, which impair reuse, and aims for the elimination of waste through the superior design of materials,
products, systems, and, within this, business models” (p. 07)
9 Thomas and “3R’s principles (Reduce, Reuse, Recycle) or Sustainable Design are concepts that should lead to this idea of a Focus on closed-loop
Birat (2013) circular or closed-loop economy” (p. 5)
(continued)
25
26
Table 1 (continued)
10 Stahel (2013) “Reducing the economic importance of resource extraction and waste management, and also reducing the Reducing environmental
environmental impairment caused by these industrial sectors” (p. 4) damage
11 Geng et al. “A circular economy is an industrial system focused on closing the loop for material and energy flows and Closing the loop for
(2013) contributing to long-term sustainability” (p. 1256) material
12 Webster “A circular economy is one that is restorative by design, and which aims to keep products, components and Restorative by design
(2015) materials at their highest utility and value, at all times” (p. 16)
13 Haas et al. “The circular economy (CE) is a simple, but convincing, strategy, which aims at reducing both inputs of virgin Closing economic and
(2015) materials and output of wastes by closing economic and ecological loops of resource flows” (p. 765) ecological loops
14 Sauvé et al. “Production and consumption of goods through closed-loop material flows that internalize environmental Closed-loop of material
(2016) externalities linked to virgin resource extraction and the generation of waste (including pollution)” (p. 49) flow
15 Stahel (2016) “Loop (or circular) economy is to bring goods and molecules back into new use in a grave-to-cradle approach” Resource effectiveness
(p. 6)
16 Murray et al. “The circular economy is an economic model wherein planning, resourcing, procurement, production and Maximize ecosystem
(2017) reprocessing are designed and managed, as both process and output, to maximize ecosystem functioning and functioning
human well-being” (p. 377)
17 Geissdoerfer “Circular economy as a regenerative system in which resource input and waste, emission, and energy leakage are Regenerative system
et al. (2017) minimised by slowing, closing, and narrowing material and energy loops. This can be achieved through
long-lasting design, maintenance, repair, reuse, remanufacturing, refurbishing, and recycling” (p. 759)
18 Blomsma and “An emergent framing around waste and resource management that aims to offer an alternative to prevalent linear Take-make-dispose
Brennan take-make-dispose practices by promoting the notion of waste and resource cycling” (p. 603) practices
(2017)
19 Cullen (2017) “A circular economy is one that is restorative and regenerative by design and aims to keep products, components, Restorative and
and materials at their highest utility and value at all times” (p. 483) regenerative by design
U. Awan et al.
(continued)
A Literature Analysis of Definitions for a Circular Economy
Table 1 (continued)
20 Den “The economic and environmental value of materials is preserved for as long as possible by keeping them in the Persevered material for
Hollander economic system, either by lengthening the life of the products formed from them or by looping them back in the long-life
et al. (2017) system to be reused” (p. 517)
21 Murray et al. “An economy is envisaged as having no net effect on the environment; rather it restores any damage done in Restoration by design
(2017) resource acquisition, while ensuring little waste is generated throughout the production process and in the life
history of the product” (p. 371)
22 Zink and “Circular economy-the concept of closing material loops to preserve products, parts, and materials in the Closing material loops
Geyer (2017) industrial system and extract their maximum utility” (p. 1)
23 Genovese “Circular economy pushes the frontiers of environmental sustainability by emphasising the idea of transforming Restoration and value
et al. (2017) products in such a way that there are workable relationships between ecological systems and economic growth” addition in resources
(p. 354)
24 Kirchherr “A circular economy describes an economic system that is based on business models which replace the Creating environmental
et al. (2017b) ‘end-of-life’ concept with reducing, alternatively reusing, recycling and recovering materials in quality
production/distribution and consumption processes, thus operating at the micro-level (products, companies,
consumers), meso level (eco-industrial parks) and macro-level (city, region, nation and beyond), with the aim to
accomplish sustainable development, which implies creating environmental quality, economic prosperity and
social equity, to the benefit of current and future generations” (pp. 224–225)
25 Moreau et al. “A concept and practice, promoting closed material cycles by focusing on multiple strategies from material Increased resource
(2017) recycling to product reuse, as well as rethinking production and consumption chains toward increased resource efficiency
efficiency” (p. 497)
26 Homrich et al. “CE is a strategy that emerges to oppose the traditional open-ended system, aiming to face the challenge of Promoting resource
(2018) resource scarcity and waste disposal in a win-win approach with economic and value perspective” (p. 534). resilience
(continued)
27
28
Table 1 (continued)
27 Korhonen “Circular economy is an economy constructed from societal production-consumption systems that maximize the Maximizes the service
et al. (2018) service produced from the linear nature-society-nature material and energy throughput flow. Circular economy produced
limits the throughput flow to a level that nature tolerates and utilises ecosystem cycles in economic cycles by
respecting their natural reproduction rates” (p. 39)
28 Suárez-Eiroa “Circular economy is a regenerative production-consumption system that aims to maintain extraction rates of Regenerative
et al. (2019) resources and generation rates of wastes and emissions under suitable values for planetary boundaries, through production-consumption
closing the system, reducing its size and maintaining the resource’s value as long as possible within the system, system
mainly leaning on design and education, and with a capacity to be implemented at any scale” (p. 14)
U. Awan et al.
applications in the CE are underutilised, and until now, there is a lack of research
studies has been carried out regarding the implementation of Industry 4.0 in the
resource conservation and closed-loop supply chain. CE as a regenerative concept
inherently increases resource efficiency, effectiveness and encourage the continued
use of the material as new strategic resource management. Literature shows that
natural environmental system is changing as a response to activities in the last few
decades. The purpose of CE is to develop a resource management strategy to achieve
resource utilization at an optimal level and add value in the current consumption
patterns. The circular economy is about continuous using products even after the end
of life for downcycling or upcycling.
2 The Definition of Circular Economy Debate
The exponential growth of the CE concept is evident in recent literature, however,

uptake and use of CE as strategic management literature are still at infancy. According
to Geng and Doberstein (2008) ecological modernization is a central concept in the
circular economy. Kainuma and Tawara (2006) posited that environmental manage-
ment is a sub-discipline of circular economy and include recycling, reuse concepts
throughout the life cycle of services and products. Therefore, a key concern is the
CE literature is what to consider in a traditional sense in an analysis of the definition.
Rather, McArthur has extended this concept and highlight the need to keep materials
in continuous use, rather eco-efficiency which focuses on dump the material at the
end of life or recycled. CE views the firm as a resource effectiveness entity, accom-
plish straight objective to support the natural environment and meet the interest and
expectations of the various players and full fill demands. Towards this end, CE is
intended to address the key questions, which material is to use continuous and which
material is to be used. As such, management attention to resources planning and
management is at the heart of CE strategy (Bocken et al. 2017).
Yuan and Shi (2009) and Zink and Geyer (2017) address CE as continuous use
of the material and closing the loop. On the other hand, many authors explicitly
address the multiple dimensions of the CE. In these papers (Cooper 1999; Liu
et al. 2009; Stahel 2013) address the minimizing the environment impairment by
reducing the material used. The remaining definitions, (MacArthur 2013; Webster
2015; Cullen 2017; Geissdoerfer et al. 2017; Murray et al. 2017; Suárez-Eiroa et al.
2019) focused exclusively on regenerative and restorative by design. Environmental
and social issues were not solely characterized in any previous literature excepts
(Geng et al. 2013; Kirchherr et al. 2017b; Murray et al. 2017). Blomsma and Brennan
(2017) focused on the combination of taking make and dispose of and resource recy-
cling. Den Hollander et al. (2017) highlighted the combination of environmental and
economic ethics, and suggested that preserve material for the long-life. Genovese
et al. (2017) focused on environmental sustainability by highlighting the impor-
tance of transforming a product that supports sustainable development considera-
tion. Moreau et al. (2017) explicitly mentioned the importance of promoting closed
30 U. Awan et al.
material cycles and increase resource efficiency. Among the other CE characteristics
examined, only the eco-efficiency and resource productivity were addressed by some
of the definitions. None of the definitions was mentioned CE resilience as an adaptive
ability to minimize environmental impairment and transforming products in such a
notion of reusing and recovering material. The CE definition analysis revealed that
so far, stakeholder perspective is not explicitly included in the definition, although
stakeholder is considered to be part of a natural and ecology system. Stakeholder
and resilience perspective is rarely discussed in the literature. This may help explain
why these dimensions were not mentioned and incorporated in the CE published
definitions. Based on the CE definition analysis, reduction of material use, eco-
efficiency and resource productivity, regenerative and restorative of resources, focus
on closed-loop, maximize ecosystem functioning, persevered material for long-life
and promoting resource resilience are used randomly in the definitions.
In essence, the circular economy is a set of practices aimed to keep products in its
process after the end of life products. There are two main approaches, openly closed-
loop supply chain and closed-loop supply chain. A traditional re-manufacturing is
concerned with economic value. Unlike the traditional recycling process, the recy-
cling process in the circular perspective integrates the economic and environmental
concerns. The literature on CE is growing as both practitioners and researchers begin
to realise that the management of raw materials and end of life product is critically
important to meet the needs of future generations.
The characteristics of the circular economy(CE) identified in this paper provide
the ground for proposing a new definition for CE. Building on the previous discussion
in this paper, circular economy may be defined as:
Circular Economy (CE) is an activity, set of process for reducing the material used in
production and consumption, promoting material resilience, closing loops and exchange
sustainability offering in such a way that maximize the ecological system.
Altogether, this shows that the purpose of the CE is increased material utiliza-
tion and promotion of material resilience. Previous literature on CE has examined
on meso, micro and macro level (Kirchherr et al. 2017b). Ecological system theory
(Bronfenbrenner 1979) can be seen as an important theoretical lens to understand
the CE. Currently, CE indicators are measured at three-level, micro, meso and macro
level. Following (Bronfenbrenner 1979) ecological system theory dimensions, we
extend it towards a better understanding of circular economy practices. A future
research study with ecological system theory perspective may give further insights
into our understanding of exo-system (firm does not experience and affect directly
but indirectly by the external environmental forces, such as customers, competi-
tors, buyers and social forces) and Chronosystem (is concerned with the environ-
mental patterns influences on the firm circular economy practices over time). In exo-
system, A firm does not experience and affect directly but indirectly by the external
environmental forces, such as customers, competitors, buyers and social forces. In
Chrono-system, a firm experienced and concerned with the continuous changing
environmental patterns which influence on the firm circular economy practices over
time.
However, how CE is to be understood by the business and consumer marketing

still poses a challenge. CE has started to make an impact on a different aspect of
the business value chains and extends through the entire supply chain. CE has been
interpreted in a variety of ways, ranging from resource reduction to resource trans-
formation. In this situation, more investigation is needed to carefully evaluate what
specific type of digitization tools could have a significant impact on the management
decision in restorative and regenerative use of material. Early CE initiatives tend to
focus on resource reduction, but increasingly they are focusing on the adoption of
sustainable development objectives.
3 Conclusion
In this chapter, the circular economy(CE) definition analysis is viewed from the
manufacturing companies’ perspective utilising a critical literature review. It may
not possible to develop a standard definition; however, it is possible to develop
insights to expand the current literature on how CE is defined. The literature analysis
showed that there were varied CE characteristics i.e., Reduction of virgin material,
resource-productivity and eco-efficiency improvement, restorative or regenerative
system, closed-loop, the value of materials is preserved and take-make-dispose prac-
tices. Circular Economy (CE) is an activity, set of process for reducing the material
used in production and consumption, promoting material resilience, closing loops
and exchange sustainability offering in such a way that maximize the ecological
system. CE is diverging the application and use of natural resources rather than
converging it. The concept of a CE is a value-orientated resource transformational
process. CE considers both upstream and downstream production and consumption
patterns to promotes the resilience orientation of resources.
Given that research is still relatively new in the sustainable circular economy. CE
offers a reverse resource regenerative idea to eliminate the linearity of production and
consumption system to support sustainability objectives. CE is a tool of promotion
of resilient sustainability objectives. CE is a new archetype for companies to achieve
resource effectiveness and efficiency objectives by lowering their sustainability risks
and impacts while raising material resilience. This chapter highlights the need that
companies require new ways to define business models by incorporating eco-philia
thinking. Circular economy business models set out pathways to provide opportuni-
ties for the dematerialization of resources throughout the product physical life cycle.
CE is a practice of maximum utilization of material use across the whole life-cycle as
well as deliver value addition in production and consumption. Literature addresses
sustainability only in a few definitions, while resilience and stakeholder focus were
not captured clearly by any of the published definitions. CE aims to integrate a system
thinking approach as a way to achieve ecological efficiency and minimizing the envi-
ronmental impairment into the material to support restorative and regenerative system
and meets stakeholders’ requirements and improve organizational triple-bottom-line
performance. A future research study with ecological system theory perspective may
32 U. Awan et al.
give further insights into our understanding of exo-system, Chronosystem in addi-

tion to meso, macro and micro level system. CE embedded in the internet of things
(IoT) is not yet a discipline that has attracted attention in many emerging countries.
CE practices in this context are particularly difficult to achieve in the manufacturing
industry because the product shipped to various customers and difficult to keep track
of the product.
References
Awan U (2020) Industrial ecology in support of sustainable development goals. In: Leal Filho W,
Azul AM, Brandli L et al (eds) Responsible consumption and production. Springer International
Publishing, Cham, pp 370–380
Awan U, Kraslawski A, Huiskonen J (2020) Progress from blue to the green world : multilevel
governance for pollution prevention planning and sustainability. In: Hussain CM (ed) Handbook
of environmental materials management. Springer Nature Switzerland AG, pp 1–22
Bilitewski B (2012) The circular economy and its risks. Waste Manag 1:1–2
Blomsma F, Brennan G (2017). The emergence of circular economy: a new framing around
prolonging resource productivity. J Ind Ecol 21:603–614. https://doi.org/10.1111/jiec.12603
Bocken NMP, Ritala P, Huotari P (2017) The circular economy: exploring the introduction of the
concept among S&P 500 firms. J Ind Ecol 21:487–490. https://doi.org/10.1111/jiec.12605
Bocken NMP, de Pauw I, Bakker C, van der Grinten B (2016) Product design and business model
strategies for a circular economy. J Ind Ecol 33:308–320
Bronfenbrenner U (1979) The ecology of human development. Harvard University Press Cambridge,
Massachusetts, United States
Chertow MR (2000) INDUSTRIAL SYMBIOSIS : literature. Ann Rev Environ Res 25:313–337
Cooper T (1999) Creating an economic infrastructure for sustainable product design. J Sustain Prod
Des 8:7–17
Cullen JM (2017) Circular economy: theoretical benchmark or perpetual motion machine? J Ind
Ecol 21:483–486
Den Hollander MC, Bakker CA, Hultink EJ (2017) Product design in a circular economy:
development of a typology of key concepts and terms. J Ind Ecol 21:517–525
EllenMacArthur Foundation (2013)TOWARDS THE CIRCULAR ECONOMY: economic and
business rationale for an accelerated transition
European Commission (2008) Directive 2008/98/EC of the European Parliament and of the Council
of 19 November 2008 on Waste and Repealing Certain Directives
European Commission (2014) Towards a circular economy: a zero waste programme for Europe.
European Commission, Brussels
European Commission (2015) Facilitating better information on the environmental performance of
products and organisations. European Commission, Brussels
Geissdoerfer M, Savaget P, Bocken NMP, Hultink EJ (2017) The circular economy—a new
sustainability paradigm? J Clean Prod 143:757–768. https://doi.org/10.1016/j.jclepro.2016.
12.048
Geng Y, Doberstein B (2008). Developing the circular economy in China: challenges and opportu-
nities for achieving ‘leapfrog development’. Int J Sustain Dev World Ecol 15:231–239. https://
doi.org/10.3843/susdev.15.3
Geng Y, Fu J, Sarkis J, Xue B (2012) Towards a national circular economy indicator system in
China: an evaluation and critical analysis. J Clean Prod 23:216–224. https://doi.org/10.1016/j.jcl
epro.2011.07.005
Geng Y, Sarkis J, Ulgiati S, Zhang P (2013) Measuring China’s circular economy. Science 340:1526–
1527. https://doi.org/10.1126/science.1227059
Genovese A, Acquaye AA, Figueroa A, Koh SCL (2017) Sustainable supply chain management and
the transition towards a circular economy : evidence and some applications. Omega 66:344–357.
https://doi.org/10.1016/j.omega.2015.05.015
Ghisellini P, Cialani C, Ulgiati S (2016) A review on circular economy: the expected transition to
a balanced interplay of environmental and economic systems. J Clean Prod 114:11–32. https://
doi.org/10.1016/j.jclepro.2015.09.007
Groenewald J, Grandjean T, Marco J (2017) Accelerated energy capacity measurement of lithium-
ion cells to support future circular economy strategies for electric vehicles. Renew Sustain Energy
Rev 69:98–111. https://doi.org/10.1016/j.rser.2016.11.017
Haas W, Krausmann F, Wiedenhofer D, Heinz M (2015) How circular is the global economy? An
assessment of material flows, waste production, and recycling in the European union and the
world in 2005. J Ind Ecol 19:765–777. https://doi.org/10.1111/jiec.12244
Homrich AS, Galvão G, Abadia LG, Carvalho MM (2018) The circular economy umbrella: trends
and gaps on integrating pathways. J Clean Prod 175:525–543. https://doi.org/10.1016/j.jclepro.
2017.11.064
Hu J, Xiao Z, Zhou R et al (2011) Ecological utilization of leather tannery waste with circular
economy model. J Clean Prod 19:221–228. https://doi.org/10.1016/j.jclepro.2010.09.018
Kainuma Y, Tawara N (2006) A multiple attribute utility theory approach to lean and green supply
chain management. Int J Prod Econ 101:99–108
Kalmykova Y, Sadagopan M, Rosado L (2017) Circular economy—from review of theories and
practices to development of implementation tools. Res Conserv Recycling 135:1–13. https://doi.
org/10.1016/j.resconrec.2017.10.034
Kirchherr J, Bour R, Kostense-Smit E et al (2017a) Barriers to the circular economy: evidence
from the European Union (EU). Ecol Econ 150:264–272. https://doi.org/10.1016/j.ecolecon.
2018.04.028
Kirchherr J, Reike D, Hekkert M (2017b) Conceptualizing the circular economy: an analysis of
114 definitions. Res Conserv Recycling 127:221–232. https://doi.org/10.1016/j.resconrec.2017.
09.005
Korhonen J, Honkasalo A, Seppälä J (2018) Circular economy: the concept and its limitations. Ecol
Econ 143:37–46. https://doi.org/10.1016/j.ecolecon.2017.06.041
Lewandowski M (2016) Designing the business models for circular economy-towards the conceptual
framework. Sustainability 8:1–28. https://doi.org/10.3390/su8010043
Lieder M, Rashid A (2016) Towards circular economy implementation: a comprehensive review
in context of manufacturing industry. J Clean Prod 115:36–51. https://doi.org/10.1016/j.jclepro.
2015.12.042
Liu Q, Li H ming, Zuo X et al (2009) A survey and analysis on public awareness and performance
for promoting circular economy in China: a case study from Tianjin. J Clean Prod 17:265–270.
https://doi.org/10.1016/j.jclepro.2008.06.003
MacArthur E (2013) Towards the circular economy: opportunities for the consumer goods sector.
UK, Ellen MacArthur Found
Merli R, Preziosi M, Acampora A (2018) How do scholars approach the circular economy? A
systematic literature review. J Clean Prod 178:703–722. https://doi.org/10.1016/j.jclepro.2017.
12.112
Moreau V, Sahakian M, van Griethuysen P, Vuille F (2017) Coming full circle: why social and
institutional dimensions matter for the circular economy. J Ind Ecol 21:497–506. https://doi.org/
10.1111/jiec.12598
Murray A, Skene K, Haynes K (2017) The circular economy: an interdisciplinary exploration of the
concept and application in a global context. J Bus Ethics 140:369–380. https://doi.org/10.1007/
s10551-015-2693-2
Pearce DW, Turner RK (1990) Economics of natural resources and the environment. Johns Hopkins
University Press, Baltimore MD
34 U. Awan et al.
Sauvé S, Bernard S, Sloan P (2016) Environmental sciences, sustainable development and circular
economy: alternative concepts for trans-disciplinary research. Environ Dev 17:48–56. https://doi.
org/10.1016/j.envdev.2015.09.002
Stahel WR (2013) Policy for material efficiency—sustainable taxation as a departure from the
throwaway society. Philos Trans Roy Soc A: Mathe Phys Eng Sci 371:1–19
Stahel WR (2016) The circular economy. Nature 531:435–438
Stahel WR, Reday-Mulvey G (1981) Jobs for tomorrow: the potential for substituting manpower
for energy. Vantage Press, The University of California, USA
Suárez-Eiroa B, Fernández E, Méndez-Martinez G, Soto-Oñate D (2019) Operational principles
of circular economy for sustainable development: linking theory and practice. J Clean Prod
214:952–960
Thomas J-S, Birat J-P (2013) Methodologies to measure the sustainability of materials–focus on
recycling aspects. Revue de Métallurgie–Int J Metal 110:3–16
Webster K (2015) Circular economy. A Wealth of Flows Ellen MacArthur Foundation Publishing,
UK. ISBN: 9780992778422
Yap NT (2005) Towards a circular economy: progress and challenges. Greener Manag Int 50:11–24
Yuan Z, Bi J, Moriguichi Y (2006) The circular economy: a new development strategy in China. J
Ind Ecol 10:4–8
Yuan Z, Shi L (2009) Improving enterprise competitive advantage with industrial symbiosis: case
study of a smeltery in China. J Clean Prod 17:1295–1302
Zink T, Geyer R (2017) Circular economy rebound. J Ind Ecol 21:593–602
Reverse Logistics Flows and Network’s
Configuration Problems
Robust Reverse Logistics Network Design
Péter Egri, Balázs Dávid, Tamás Kis, and Miklós Krész
Abstract Recycling waste materials has become increasingly important recently

both for economic and environmental reasons. In order to efficiently operate the
backward flow of the materials, a basic challenge is to design the proper reverse
logistics network containing the collection points, test centers and manufacturing
plants. This paper studies the supply network of waste wood, which has to be collected
in dedicated accumulation centers, and transported to processing facilities. We focus
on the facility location of processing centers and propose mathematical models that
take economies of scale and robustness into account, including a novel approach
based on bilevel optimization. We also give a local and tabu search method for the
solution of the problem. Test results are presented for both the robust and non-robust
case using instances based on a real-life dataset.
Keywords Facility location · Robust optimization · Economies of scale · Reverse

logistics for wood recycling
P. Egri (B)
Centre of Excellence in Production Informatics and Control,
Institute for Computer Science and Control, Budapest, Hungary
e-mail: egri@sztaki.hu
T. Kis
Engineering and Management Intelligence Laboratory,
Institute for Computer Science and Control, Budapest, Hungary
Department of Operations Research, Eötvös Loránd University, Budapest, Hungary
e-mail: kis.tamas@sztaki.hu
B. Dávid · M. Krész
InnoRenew CoE and University of Primorska, Izola, Slovenia
e-mail: balazs.david@innorenew.eu
e-mail: miklos.kresz@innorenew.eu

38 P. Egri et al.
1 Introduction and Motivation
With the recent increase in the importance of environmental awareness, more stress
is being put to on the end-of-life recovery and reuse of resources in supply chains.
Activities that aim to recover resources from their final destination are integrated by
the field of reverse logistics (Dekker et al. 2004). The goal of the reverse logistics
is to use these end-of-life resources either to produce further value or to dispose of
them properly, usually through a complex recovery process consisting of the stages
of repair, reuse, refurbish, remanufacture, retrieve, recycle, incinerate and landfill.
Reverse logistics methods can also be integrated into the conventional process of
supply chains, forming so-called closed-loop supply chains that account for both
forward and reverse flows of resources (Kazemi et al. 2019).
Wood is an extremely versatile raw material with application fields ranging from
paper production and packaging to the building industry. Moreover, wooden products
can be reused and recycled after their original function becomes obsolete. According
to data collected by the Horizon 2020 BioReg Project (Cocchi et al. 2019), the EU
countries collectively produced between 40 and 60 million tonnes of yearly wood
waste in the past ten years. Recovery rates of this depend on both the country and the
type of wood waste, but it can be seen that there is room for improving the current
amounts (Garcia and Hora 2017).
The amount of research dealing with the management of waste wood has increased
over the past years. As an example, the interest can be seen in the furnishing sector,
where several studies have been conducted. The paper by (de Carvalho Araújo 2019)
assesses the literature of circular economy in wood panel production. They conclude
that while circular economy as a concept is being investigated with regard to waste
production in this sector, mainly LCA (life cycle assessment) studies were carried out
(Hossain and Poon 2018; Kim and Song 2014). Daian and Ozarska (2009) studied
a sample group of six SMEs in the wood furniture sector of Australia and collected
data about the current state of their wood waste and its reuse, recycle and disposal.
Based on this, they formulated suggestions on wood waste management. Evaluating
the availability of wood waste (and wood biomass in general) is also becoming more
and more important, which can be seen from the multiple recent studies that have
dealt with this question. Research by Verkerk et al. (2019) and Borzecki et al. (2018)
assessed the potential availability of forest biomass from European forests and its
spatial distribution, focusing on the hotspots of biomass. Studies comparing waste
wood management in selected European countries were also conducted by Garcia
and Hora (2017) and the BioReg Project (Cocchi et al. 2019).
Although similar studies have become more widespread over the past years, the
number of papers dealing with the mathematical modelling and optimization of
processes in the waste wood supply chain is still scarce. Network design and planning
is one of the most studied problem classes in logistics (Govindan et al. 2015). While
there have been recent studies into the combined design of the network nodes and
their possible links (Rahmaniani and Ghaderi 2013), it is usually safe to assume for
transport problems that the underlying road network already exists. In this case, the
Robust Reverse Logistics Network Design 39
most important problem to solve is facility location. The goal of this problem is to
find an optimal placement of facilities on a network in order to minimize arising
costs, which usually include transportation and opening facilities.
Mathematical models of facility location are extensively studied, see e.g., Chap. 4
in Dekker et al. (2004). Further variations of the facility location problem (not specific
to reverse logistics networks) can be found in Simchi-Levi et al. (2014). Stochastic
variations of the problem can be found in Verter and Dincer (1992), which also
considers capacity planning as the Capacity Expansion Problem once the facility
locations are established. Dasci and Laporte study facility location and capacity
acquisition by segmenting a market on the infinite continuous plain with uncertain
demand (Dasci and Laporte 2005). In a recent manuscript, Ahmadi-Javid et al. study
a combined facility location and capacity planning problem, where the facilities
should serve customers with demand modeled as Poisson processes, which results in
a nonlinear model (Ahmadi-Javid et al. 2018). Solution methods for facility location
with economies of scales are studied in Bucci (2009), Lu (2010).
Facility location problems usually consider two types of uncertainties; namely,
stochastic parameters and disruptions (Chun et al. 2017). An example for the former
one is the stochastic demand or cost parameters, see e.g., (Carrizosa and Nickel
2003). Robust models, on the other hand considers possible changes in the network
structure, e.g., expected consequences of random disruptions or targeted attacks by
malevolent attackers (Daskin 2013). Robust facility location is studied in Cheng et al.
(2018).
While general solutions designed for backward biomass streams have been studied
in the past (e.g. Nunes et al. 2020; Sharma et al. 2013), we only found a handful
of papers that focus entirely on waste wood. The reverse logistics network redesign
problem for waste wood from the construction industry is investigated in Trochu et al.
(2018), and a MILP (mixed integer linear programming model) was proposed for its
solution. A use-case on a scenario from Quebec, Canada, was also presented. Devjak
et al. (1994) formulated a mathematical model for optimizing the transportation of
wood waste produced in sawmills, but did not present any computational experiments
to back up its efficiency. Burnard et al. (2015) gave a reverse logistics model for
facility location and transportation for waste wood, and presented computational
results for a use-case in Slovenia.
In this paper, we consider the facility location problem for transporting waste
wood from accumulation centers to processing facilities. Besides transportation, we
also study economies of scale as well as the robustness of the network in case of the
breakdown of facilities. First, we formulate mathematical models for the problems,
and propose both a local and tabu search heuristic method for their solution. The
efficiency of these methods is shown on test instances generated using a real-life
dataset.
40 P. Egri et al.
2 Problem Definition
In the following subsections we formulate the uncapacitated facility location problem

and its extensions.
2.1 Uncapacitated Facility Location Problem
Let I denote the set of fixed accumulation point locations and J the set of potential
facility locations. Let f j denote the cost of opening facility j and ci j denote the
transportation cost from point i to facility j per m3 . Let u i denote the annual yield
of waste wood from accumulation point i ∈ I (in m3 ).
The formulation uses two types of binary variables: Y j is the indicator of opening
facility j ∈ J , while X i j indicates product flow from accumulation point i to facility
j. Note that due to uncapacitated facilities, an optimal solution always transports the
whole amount of wood from each accumulation point to the closest open facility.
The optimization problem is then the following binary integer problem:

min f jYj + u i ci j X i j (1)
j∈J i∈I j∈J
subject to

Xi j = 1 ∀i ∈ I (2)
j∈J
Xi j ≤ Y j ∀i ∈ I, j ∈ J (3)
Y j ∈ { 0, 1 } ∀j ∈ J (4)
X i j ∈ { 0, 1 } ∀i ∈ I, j ∈ J (5)
The objective function (1) minimizes the total opening and transportation cost, (2)
ensures that the wood is transported from each accumulation point, while (3) states
that all wood is transported to an open facility. Constraints (4) and (5) state that the
variables are binary.
2.2 Economies of Scale Problem
It is often realistic to assume that the higher the capacity of a facility, the lower its
production cost due to the economies of scale (Garcia and Hora 2017). We consider
the following production cost at facility j (based on Bucci 2009): S bj p j , where S j > 0
is the total quantity processed at facility j, p j is the unit production cost at facility
j and b is a scale factor, typically −0.35 for manufacturing facilities and between
−0.56 and −0.47 in the paper industry. With this modification the objective function
of the program becomes non-linear as follows:

min f jYj + u i X i j ci j + S bj p j S j (6)
j∈J i∈I j∈J j∈J
The constraints are the same as (2)–(5) with the following additional constraint
defining the variable S j :

X i j ui = S j ∀j ∈ J (7)
i∈I
We still consider solutions where wood from each accumulation point is trans-
ported to only one facility, since there exist an optimal solution with this property,
see Dupont (2008). However, it is no longer true that all wood should necessarily be
transported to the closest open facility, for each set of open facilities an assignment
problem should be solved to determine the optimal transportation.
2.3 Robust Optimization Problem
Robust optimization can be modeled as a multi-objective optimization problem,

where one objective is minimizing the cost in case of no disruptions, the other is
minimizing the cost in case of a disruption. However, we consider only minimizing
the cost in case of a disruption instead. More specifically, we consider a solution
optimal, if any facility breaks down—i.e., all accumulation points connected with
this facility must transport to another facilities—then the resulting cost in the worst
case is minimal.
We model this problem as a bilevel optimization: the leader determines which
facilities to open, while the follower determines which accumulation point is con-
nected to which facility. The follower’s problem assumes a given set of open and
undisrupted facilities ({ j | Y j′ = 1 }) and assign the accumulation points to these
facilities minimizing the transportation costs:

min u i ci j X i j (8)
i∈I j∈J
subject to
42 P. Egri et al.

Xi j = 1 ∀i ∈ I (9)
j∈J

X i j ≤ Y j′ ∀j ∈ J (10)
i∈I
X i j ∈ { 0, 1 } ∀i ∈ I, j ∈ J (11)
Note that the follower’s problem can be easily solved by transporting all the wood
to the closest open facility. Let G(Y ′ ) denote the optimal objective value for the
follower’s assignment problem on the input vector Y ′ .
Then the leader’s problem is to determine the set of facilities to open with the
minimal opening cost together with the transportation cost in case of the disruption
of exactly one facility:
⎧ ⎧ ⎫⎫
⎨ ⎨ ⎬⎬
min f j Y j + max G(Y ′ ) : Y j′ + 1 = Y j ∧ Y j′ ≤ Y j (∀ j ∈ J )
Y ⎩ Y′ ⎩ ⎭⎭
j∈J j∈J j∈J
(12)
This expresses that facilities { j | Y j = 1 } are opened, but then one of them cannot
be used because of a disruption, therefore the transportation has to be determined
not using the disrupted facility. The worst case is considered, i.e., when the disrupted
facility causes the highest transportation costs. This corresponds to a pessimistic
bilevel program.
3 Solution Approaches
Solving facility location problems in realistic sizes (i.e., several thousands of accu-
mulation points and possible facility locations) is computationally intractable even
without considering economies of scale or robustness. Therefore, similarly to other
works in this field, we used metaheuristic algorithms to find quasi-optimal solutions.
3.1 Determining the Worst Case Cost Effectively
If economies of scale are disregarded, the optimal solution always transports the
wood to the closest open facility. We use this observation to efficiently compute the
cost in case of disturbances. Let πi denote a permutation of the facilities for each i
such that ciπi1 < · · · < ciπin , where n = |J | is the number of facilities. If Y denote the
status of the facilities with at least two open facilities, then let Fi (Y ) = min{Yπik =
k
1} denote the closest open facility to point i, and let Bi (Y ) = min{Yπik = 1 ∧ k =
k
Fi (Y )} denote the second closest one. If there is a disruption at facility Fi (Y ), then
the wood from point i should be transported to facility Bi (Y ) instead, which means
(ci Bi (Y ) − ci Fi (Y ) )u i additional transportation cost. Therefore in case of a disruption
at an open facility j, the additional cost is CoD j (Y ) = (ci Bi (Y ) − ci Fi (Y ) )u i .
i:Fi (Y )= j
Then the cost increase of disruption in the worst case is simply max CoD j .
j:Y j =1
Therefore by maintaining the F, B and CoD vectors when the search heuris-
tics open or close a facility, the value of the objective function can be determined
efficiently.
3.2 Local Search Heuristic
We use the neighborhood defined by Korupolu et al. (2000), which represents the
solution only with the set of open facilities. Let S = { j | Y j = 1 } denote the set
of open facilities, then the neighborhood of S is { T : |S \ T | ≤ 1 ∧ |S \ T | ≤ 1 }.
From a solution S one can apply three operations to reach a neighbor: (i) open a new
facility, (ii) close a facility (in case |S > 1|), and (iii) change the status of an open
and a closed facility. This neighborhood contains O(|J |2 ) solutions, where J is the
set of potential facilities.
If one intends to solve the robust facility location problem, then instead of the
cost defined by (1), the worst case cost should be considered.
3.3 Tabu Search Heuristic
We have implemented the tabu search based on the approach described in Sun (2006)
with some modifications. In addition to seeking the minimal cost in case of a dis-
ruption, we applied a different medium term memory process as well as different
approach for updating the lengths of the tabu lists.
The short term memory process is the following. Let k denote the number of moves
since the start of the search and ∆z kj the cost change by changing facility j’s status,
i.e., closing if it is open and open if it is closed. The integer vector t is used to store the
last time when the status of the facilities changed, i.e., t j is the value of k when facility
j changed its status last. Let z 0 denote the best objective value in the current search
cycle and k0 denote the time when z 0 was last updated. Let l0 (lc ) denote the tabu sizes
for the open (closed) facilities, i.e., they cannot change status twice during this time
interval unless the aspiration criterion is satisfied. These lengths are bound by lower
limit lo1 (lc1 ) and upper limit lo2 (lc2 ). Each move is changing the status of a facility. We
44 P. Egri et al.
choose facility j̄, where ∆z kj̄ = min{ ∆z kj | facility j is not flagged }. A facility j̄ is
flagged, if the following tabu condition holds: k − t j̄ ≤ lc if Y j̄ = 0 or k − t j̄ ≤ l0
if Y j̄ = 1, but does not hold the following aspiration criteria: z + ∆z kj < z 0 , where
z represents the cost of the current solution. The short term process ends when the
solution could not be improved for a specified time, i.e., when k − k0 > α1 n, where
α1 is a parameter of the search.
After each step the lengths of the tabu lists are updated: if the current solution
improved the objective value, then l0 (lc ) is increased by one, otherwise it is decreased
by one to extend the search space.
In the medium term, we changed the frequency based memory process described
by Sun (2006) and use a wider neighborhood instead. We seek for an open and a
closed facility such that if we change their statuses, the total cost decreases the most.
Sun states that considering this operation is costly, but our algorithm only use it
when the short term process fails to improve the solution, thus providing a trade-off
between computation time and solution quality. We have found that this approach
performs better on the tested instances.
If the solution can be improved, the search continues with the short term process.
The medium term process ends when no improvement can be found.
Finally, the long term process is invoked C times and when invoking the cth
time, c moves are made changing the status of facility j̄ according to the following
criterion: t j̄ = min{ t j | j = 1, . . . , n }.
3.4 Exact Solution of the Robust Problem
The exact solution can be computed by completely enumerating all possible subsets
of open facilities, and for each combination of open facilities, a simple assignment
problem must be solved. However, we can apply the following simple bounding
procedure to reduce the search space. When the algorithm already has a solution with
objective value z, then for any subset of open facilities S where the sum of opening
costs—disregarding the transportation costs—exceeds z, the search can ignore all
supersets of S, since they cannot result in a less expensive solution.
This exact method performs well if the opening costs are high compared to the
transportation costs, because in those cases finding a good solution can restrict the
search to a small number of open facilities. Nevertheless, this exact method works
on small problem instances only.
3.5 Bilevel Integer Program Formulation
Considering the formulation of Sect. 2.3, it can be observed that once the Y ′ variables
are fixed, the X variables are easy to determine to minimize the transportation costs
by assigning each accumulation point to the closest open facility. This suggests that
a solution of the following constraints determines an optimal assignment.
X iπi1 ≥ Yπ′ i1 ∀i ∈ I (13)

k−1

X iπik ≥ Yπ′ ik − Yπ′ it ∀i ∈ I, k = 2, . . . , m (14)
t=1

Xi j = 1 ∀i ∈ I (15)
j∈J
X i j ∈ { 0, 1 } ∀i ∈ I, j ∈ J (16)
Then, the inner maximization problem of (12) takes the form

max u i ci j X i j (17)
i∈I j∈J
subject to (13)–(16) and the constraints

Y j′ + 1 = Yj (18)
j∈J j∈J
Y j′ ≤ Y j ∀j ∈ J (19)
Note that this formulation does not include non-linearity in contrast to the usual
duality-based formulation (see e.g., Cheng et al. 2018).
Using this observation, we search for the optimal solution where exactly k facilities
(ρ1 < · · · < ρk ) are open: j∈J Y j = k and Yρl = 1 (l ∈ { 1, . . . , k }). Let Y l denote
the vector that differs from Y only in its ρl th element and { X il j : i ∈ I, j ∈ J } the
optimal transportation from accumulation point i to facility j using open factories
determined by Y l . Then the optimization problem (12) becomes:

min f jYj + z (20)
j∈J
subject to
46 P. Egri et al.

z≥ u i ci j X il j ∀l ∈ { 1, . . . , k } (21)
i∈I, j∈J
l
X iπi1
≥ Yπl i1 ∀i ∈ I, l ∈ { 1, . . . , k } (22)
s−1

l
X iπis
≥ Yπl is − Yπl it ∀i ∈ I, s = 2, . . . , m, l ∈ { 1, . . . , k } (23)
t=1

X il j = 1 ∀i ∈ I, l ∈ { 1, . . . , k } (24)
j∈J
Y jl ≤ Y j ∀ j ∈ J , l ∈ { 1, . . . , k } (25)
k

Y jl = (k − 1)Y j , ∀j ∈ J (26)
l=1

Yj = k (27)
j∈J
Y j ∈ { 0, 1 } ∀j ∈ J (28)
Y jl ∈ { 0, 1 } ∀ j ∈ J , l ∈ { 1, . . . , k } (29)
X il j ∈ { 0, 1 } ∀i ∈ I, j ∈ J , l ∈ { 1, . . . , k } (30)
Constraints (22)–(24) are the constraints of the inner optimization problem.

Inequality (26) says that if Y j = 0, then all Y jl = 0, whereas if Y j = 1, then exactly
k − 1 of the Y l has a 1 in position j. This, along with (25) implies that vectors Y l
are all different, they are not bigger than Y (coordinate wise), and they have k − 1
coordinates of value 1, all other coordinates being 0.
This formulation considers a fixed number of open facilities, therefore it should
be solved for all possible (or realistic) k values.
4 Numerical Study
Based on the industrial dataset of 1839 accumulation points and possible facility
locations, we generated test sets containing 50, 100 and 500 locations, five different
test cases for each set. Then we computed the solutions assuming different facility
opening costs from the realistic 5 million to 1000. The solutions were computed
using the local search, the tabu search, and when possible, the exact solver. For the
tabu search we used the same parameters as (Sun 2006): lc1 = lo1 = 10, lc2 = lo2 = 20,
C = 5 and α1 = 2.5.
Table 1 contains the average results over the five test sets. The non-robust solutions
aim at minimizing the total opening and transportation cost indicated in the cost
column, while robust solutions aim at minimizing the worst case cost (WCC), i.e.,
the total opening cost and transportation costs in case of a facility disruption.
We have estimated the production cost for the facility location model with the
economies of scale, however, we have found that for realistic cases (large number of
accumulation points, large facility opening costs, few open facilities) the economies
of scale does not influence the solution. Therefore the non-linearity of the problem
was not considered in the results presented below, which resulted in more tractable
problems.
The following indicators are included in the table: the cost of disruption (CoD) is
the additional cost in case of a disturbance ((WCC-cost)/cost), the price of robust-
ness (PoR) is the difference between the robust solution and the non-robust one
((robust cost—non-robust cost)/non-robust cost) and the benefit of robustness (BoR)
is the difference in case of a disruption ((non-robust WCC—robust WCC)/non-robust
WCC). This latter indicator cannot be interpreted when the non-robust solution con-
tains only one opened facility, i.e., when in case of a disruption the whole network
fails.
The rows labelled with “OPT” denote the average costs of the optimal solutions.
For the non-robust problem, it is computed by the the FICO XPRESS Solver using
the formulation in Sect. 2.1, and for the robust problem the optimum is computed by
solving the bilvel programming formulation of Sect. 3.5.
Table 1 contains the results of the solutions considering 50 locations. The follow-
ing observations were made:
• For the opening costs between 1 million and 5 million, the exact solutions could
be computed for the non-robust, and the robust variants as well, and both the local
search and the tabu search could find the optimal solution in every case.
• Changing the opening costs in a wide range (above one million) did not change
the solutions. That means that the uncertainty of the exact opening cost does not
matter too much.
• For the 4 largest facility opening costs, the non-robust solutions contain only one
opened facility, therefore they are quite vulnerable for disruptions. Adding one
more factory to improve robustness is quite expensive, increasing the required
budget by 36–76%.
• Considering 1000 as the opening cost, the tabu search resulted in better solution
both for robust and non-robust cases in one case out of five, therefore the last two
rows are separate in order to differentiate the two approaches. The robust version
of the problem could not be solved with the exact solver.
• For an extremely low opening cost, large number of facilities are opened and even
the non-robust solution offer some robustness. However, the robustness can be
improved relatively inexpensively (for less than 0.4% of the budget) and in case
of a disruption this can result in more than 10% saving in the additional costs.
• In each cases, either the local search or the tabu search could find the optimal
solution for the uncapacitated facility location problem without robustness.
Table 2 contains the results of the solutions considering 100 locations. The fol-
lowing observations were made:
• For opening costs 5 million and 2.5 million local search and tabu search resulted in
the same results as the exact solver. The non-robust solutions in these cases always
48 P. Egri et al.
contain only one open facility and adding robustness by opening more facilities
are quite costly.
• For opening costs 1.6 and 1 million, the non-robust solutions contain one or two
open facilities. The WCC and BoR values are the averages over the valid values.
For these problems the solution of the local search and the tabu search often differ
and it varies which performs better.
• For opening cost 1000, the tabu search performed better in one case. It can be
observed that increasing robustness in this case is quite inexpensive, but the
achieved benefit is also lower that in the 50 facility case.
• For only one problem instance neither the local search nor the tabu search heuristics
could find the optimal solution for the uncapacitated facility location problem
without robustness.
Table 3 contains the results of the solutions considering 500 locations. The following
observations were made:
• With this size of solution space the result of the local search and the tabu search
often differ and on average the tabu search performs slightly better.
• Most of the non-robust solutions contain two or more open facilities. Optimizing
for robustness increases the cost usually under 20%, therefore as the problem size
growths, it becomes relatively less expensive to provide robustness. However, in
case of the disruption robustness can save at least 10% of the additional cost, when
the opening cost is above one million.
• For five problem instances neither the local search nor the tabu search heuristics
could find the optimal solution for the uncapacitated facility location problem
without robustness. Four of these five cases have 1000 as the facility opening cost.
As a conclusion it can be observed that including robustness is the most important

when the number of opened facilities are low.
5 Conclusions
In this paper, we studied the facility location problem in the reverse logistics network
of waste wood. This network considered the accumulation centre for waste wood as
well as the processing facilities where they have to be transported. The traditional
facility location was extended with the consideration of economies of scale and
robustness against the breakdown of facilities. We formulated mathematical models
for these problems including a novel approach based on bilevel optimization, and
also presented a local and tabu search heuristic method for their solution.
We tested the efficiency of the proposed methods on instances generated using a
real-life dataset. Different facility opening costs were considered, and robust and non-
robust solutions were examined in every case. While economies of scale seemed to
have no influence on the solutions in the case of realistic cost parameters, robustness
turned out to be significant when the number of opened facilities was low. In the case
Table 1 Average performance using 50 locations

Opening Non-robust Robust
Open Cost WCC Open Cost WCC CoD PoR BoR
(%) (%) (%)
5000000 1 6,452,010 – 2 11,349,166 11,494,142 1.28 75.98 –
(OPT)
5000000 1 6,452,010 – 2 11,349,166 11,494,142 1.28 75.98 –
(LS/TS)
2500000 1 3,952,010 – 2 6,349,166 6,494,142 2.31 60.81 –
(OPT)
2500000 1 3,952,010 - 2 6,349,166 6,494,142 2.31 60.81 –
(LS/TS)
1666666 1 3,118,677 – 2 4,682,499 4,827,476 3.14 50.34 –
(OPT)
1666666 1 3,118,677 – 2 4,682,499 4,827,476 3.14 50.34 –
(LS/TS)
1000000 1 2,452,010 – 2 3,349,166 3,494,142 4.42 36.81 –
(OPT)
1000000 1 2,452,010 – 2 3,349,166 3,494,142 4.42 36.81 –
(LS/TS)
1000 40.8 46,360 102,481
(OPT)
1000 (LS) 40.6 46,377 102,498 41.4 46,534 93,605 102.20 0.34 10.57
1000 (TS) 40.8 46,360 102,481 41.6 46,518 93,588 102.21 0.34 10.58
of a larger number of opened facilities (which usually happened with unrealistically

low facility costs) even the non-robust solutions contained some inherent robustness.
While the heuristic method gave the same solutions for instances with a smaller
number of locations (where they mostly found the optimal solution), the tabu search
had a slight edge over the local search for larger instances. However, we were not able
to obtain exact solutions for these instances with a large number of locations, and
working on mathematical programming methods to help the solution of the model
will be part of our future work.
An important limitation of our model compared to other robust facility location
models, is that we limit the possible number of disruptions to one. The reason behind
this is that we are concerned with infrequent and random failures, instead of targeted
attacks—which are not typical in the waste wood logistics. This limitation facilitates
the novel bilevel integer program formulation, which can be solved more efficiently
than general methods even for much larger problem instances, see e.g., (Cheng et al.
2018). However, it still cannot handle the huge networks typical in case of wood
recycling. For these practical problems, we applied heuristics, and found that the tabu
search using wider neighborhood in the medium term process yields significantly
better results in reasonable time than frequency-based processes, e.g., (Sun 2006).
50 P. Egri et al.

(%) (%) (%)
5000000 1 7,579,528 - 2 12,383,253 12,594,382 1.72 63.58 –
(OPT)
5000000 1 7,579,528 - 2 12,383,253 12,594,382 1.72 63.58 –
(LS)
5000000 1 7,579,528 - 2 12,383,253 12,594,382 1.72 63.58 –
(TS)
2500000 1 5,079,528 - 2 7,383,253 7,594,382 2.90 45.67 –
(OPT)
2500000 1 5,079,528 - 2 7,383,253 7,594,382 2.90 45.67 –
(LS)
2500000 1 5,079,528 - 2 7,383,253 7,594,382 2.90 45.67 –
(TS)
1666666 1 4,246,195 – 2 5,927,715
(OPT)
1666666 1.2 4,327,338 8,102,164 2.2 5,703,829 6,083,148 6.33 32.85 5.87
(LS)
1666666 1 4,246,195 - 2 5,612,645 5,998,162 6.67 32.95 –
(TS)
1000000 1.4 3,510,617 2 4,594,382
(OPT)
1000000 1.6 3,554,200 5,663,822 2.2 4,237,162 4,616,481 8.76 19.82 15.35
(LS)
1000000 1.4 3,533,251 6,062,458 2.2 4,242,211 4,628,458 8.85 20.56 17.01
(TS)
1000 77.6 90,535 118,734
(OPT)
1000 (LS) 77.6 90,535 118,734 78.2 90,703 113,882 25.67 0.18 4.05
1000 (TS) 77.6 90,535 118,734 78.6 90,829 113,859 25.49 0.32 4.07
As a future work, we intend to further study the integer program formulation of

the bilevel robust facility location model and use it for computing lower bound on
the cost. In addition, we are going to examine the delivery planning problem in the
network designed by the facility location optimization.
Acknowledgements The research of Péter Egri and Tamás Kis has been supported by the National
Research, Development and Innovation Office—NKFIH, grant no. SNN 129178, and ED_18-2-
2018-0006. Tamás Kis was supported by Project ED-18-1-2019-030 (Application-specific highly
reliable IT solutions), which has been implemented with the support provided from the National
Research, Development and Innovation Fund of Hungary, financed under the Thematic Excellence
Programme funding scheme. Balázs Dávid and Miklós Krész gratefully acknowledge the Euro-
pean Commission for funding the InnoRenew CoE project (Grant Agreement #739574) under the

(%) (%) (%)
5000000 1.6 19,242,002
(OPT)
5000000 2 19,497,959 31,080,579 2.4 23,068,454 25,278,632 9.79 18.57 17.72
(LS)
5000000 1.6 19,307,357 33,319,075 2.4 23,062,160 25,421,665 10.31 19.78 20.13
(TS)
2500000 2.4 14,263,693
(OPT)
2500000 2.2 14,372,353 25,267,673 3.8 16,318,181 19,023,002 17.26 13.69 22.33
(LS)
2500000 2.6 14,304,504 21,116,645 3.8 16,336,745 18,676,174 14.55 14.38 11.35
(TS)
1666666 3.2 11,900,321
(OPT)
1666666 3 11,952,078 18,065,126 4 13,464,150 15,561,839 15.60 12.69 13.48
(LS)
1666666 3.4 11,905,914 17,393,652 4.8 13,380,271 15,323,262 14.60 12.36 11.43
(TS)
1000000 4.4 9,496,951
(OPT)
1000000 4.4 9,496,951 13,729,015 6 10,402,176 11,881,160 14.26 9.60 12.61
(LS)
1000000 4.4 9,522,457 13,678,493 5.8 10,429,309 11,946,057 14.65 9.48 11.80
(TS)
1000 282 396,016
(OPT)
1000 (LS) 281.4 396,159 454,042 281.60396,163 453,068 14.38 0.00 0.24
1000 (TS) 282 396,054 453,937 283 396,198 452,907 14.33 0.04 0.25
Horizon2020 Widespread-Teaming program, and the Republic of Slovenia (Investment funding of

the Republic of Slovenia and the European Union of the European Regional Development Fund).
Miklós Krész is also grateful for the support of the Slovenian ARRS grant N1-0093. The authors
would like to thank Aleksandar Tosic for his useful insights regarding the problem and for providing
the real-world input dataset.
References
Ahmadi-Javid A, Berman O, Hoseinpour P (2018) Location and capacity planning of facilities with
general service-time distributions using conic optimization. (arXiv:1809.00080)
52 P. Egri et al.
Borzecki K, Rafal P, Kozak M, Borzecka M, Faber A (2018) 01). Spatial distribution of wood waste
in Europe. Sylwan 162:563–571
Bucci MJ (2009) Solution procedures for logistic network design models with economies of scale.
Unpublished doctoral dissertation. North Carolina State University
Burnard M, Tavzes Č, Tošić A, Brodnik A, Kutnar A (2015) The role of reverse logistics in recycling
of wood products. In Environmen- tal implications of recycling and recycled products. Springer
Singapore, pp 1–30. https://doi.org/10.1007/978-981-287-643-0-1
Carrizosa E, Nickel S (2003) Robust facility location. Math Methods Oper Res 58 (2):331–349.
Retrieved from https://doi.org/10.1007/s001860300294
Cheng C, Adulyasak Y, Rousseau L-M (2018) Robust facility location under disruptions (Tech.
Rep.). https://www.gerad.ca/fr/papers/G-2018-91: GERAD—Groupe d’Etudes et de Recherche
en Analyse des Décisions
Cocchi M, Vargas M, Tokacova K (2019) State of the art technical report (Tech. Rep.). Absorbing
the potential of wood waste in EU regions and industrial bo-based ecosystems—BioReg
Daian G, Ozarska B (2009) Wood waste management practices and strategies to increase sus-
tainability standards in the australian wooden furniture manufacturing sector. J Cleaner Prod
17(17):1594–1602. https://doi.org/10.1016/j.jclepro.2009.07.008
Dasci A, Laporte G (2005) An analytical approach to the facility location and capacity acquisi-
tion problem under demand uncertainty. J Oper Res Soc 56:397–405. https://doi.org/10.1057/
palgrave.jors.2601826
Daskin MS (2013) Network and discrete location: models, algorithms and applications, 2nd edn.
Wiley
de Carvalho Araújo CK, Salvador R, Moro Piekarski C, Sokulski CC, de Francisco AC, de Carvalho
Araújo Camargo SK (2019) Circular economy practices on wood panels: a bibliographic analysis.
Sustainability 11(4). https://doi.org/10.3390/su11041057
Dekker R, Fleischmann M, Inderfurth K, van Wassenhove LN (eds) (2004) Reverse logistics:
quantitative models for closed-loop supply chains. Springer, Berlin
Devjak S, Tratnik M, Merzelj F (1994): Model of optimization of wood waste processing in slovenia.
In: Bachem A, Derigs U, Jünger M, Schrader R (eds) Operations research ’93. Physica- Verlag,
pp 103–107
Dupont L (2008) Branch and bound algorithm for a facility location problem with concave site
dependent costs. Int J Prod Econ 112(1):245–254. Retrieved from http://www.sciencedirect.com/
science/article/pii/S0925527307001430 (Special Section on Recent Developments in the Design,
Control, Planning and Scheduling of Productive Systems) https://doi.org/10.1016/j.ijpe.2007.04.
001
Garcia CA, Hora G (2017) State-of-the-art of waste wood supply chain in Germany and selected
European countries. Waste Manag 70:189–197. Retrieved from http://www.sciencedirect.com/
science/article/pii/S0956053X17306931https://doi.org/10.1016/j.wasman.2017.09.025
Govindan K, Soleimani H, Kannan D (2015) Reverse logistics and closed-loop supply chain: a
comprehensive review to explore the future. Eur J Oper Res 240(3):603–626. https://doi.org/10.
1016/j.ejor.2014.07.012
Hossain MU, Poon CS (2018) Comparative lca of wood waste management strategies generated
from building construction activities. J Cleaner Prod 177:387–397. https://doi.org/10.1016/j.
jclepro.2017.12.233
Kazemi N, Modak NM, Govindan K (2019) A review of reverse logistics and closed loop supply
chain management studies published in ijpr: a bibliometric and content analysis. Int J Prod Res
57(15–16):4937–4960. https://doi.org/10.1080/00207543.2018.1471244
Kim MH, Song HB (2014) Analysis of the global warming potential for wood waste recycling
systems. J Cleaner Prod 69:199–207. https://doi.org/10.1016/j.jclepro.2014.01.039
Korupolu MR, Plaxton C, Rajaraman R (2000) Analysis of a local search heuristic for facility
location problems. J Alg 37(1):146–88. Retrieved from http://www.sciencedirect.com/science/
article/pii/S0196677400911003https://doi.org/10.1006/jagm.2000.1100
Lu D (2010) Facility location with economies of scale and congestion. Unpublished master’s thesis.
University of Waterloo
Nunes L, Causer T, Ciolkosz D (2020) Biomass for energy: a review on supply chain management
models. Renew Sustain Energy Rev 120:109658. https://doi.org/10.1016/j.rser.2019.109658
Peng C, Li J, Wang S (2017) Two-stage robust facility location problem with multiplicative uncer-
tainties and disruptions. In 2017 international conference on service systems and service man-
agement, pp 1–6. https://doi.org/10.1109/ICSSSM.2017.7996131
Rahmaniani R, Ghaderi A (2013) A combined facility location and network design problem with
multi-type of capacitated links. Appl Math Modell 37(9):6400–414. https://doi.org/10.1016/j.
apm.2013.01.001
Sharma B, Ingalls R, Jones C, Khanchi A (2013) Biomass supply chain design and analysis: basis,
overview, modeling, challenges, and future. Renew Sustain Energy Rev 24:608–627. https://doi.
org/10.1016/j.rser.2013.03.049
Simchi-Levi D, Chen X, Bramel J (2014) The logic of logistics. theory, algorithms, and applications
for logistics and supply chain management, 3rd edn. Springer
Sun M (2006) Solving the uncapacitated facility location problem using tabu search. Comput
Oper Res 33(9):2563–2589. Retrieved from http://www.sciencedirect.com/science/article/pii/
S0305054805002406 (Part Special Issue: Anniversary Focused Issue of Computers & Opera-
tions Research on Tabu Search) https://doi.org/10.1016/j.cor.2005.07.014
Trochu J, Chaabane A, Ouhimmou M (2018) Reverse logistics network redesign under uncertainty
for wood waste in the crd industry. Resourc Conserv Recycl 128:32–47. https://doi.org/10.1016/
j.resconrec.2017.09.011
Verkerk PJ, Fitzgerald JB, Datta P, Dees M, Hengeveld GM, Lindner M, Zudin S (2019) Spatial
distribution of the potential forest biomass availability in europe. Forest Ecosyst 6(1):5. https://
doi.org/10.1186/s40663-019-0163-5
Verter V, Dincer MC (1992) An integrated evaluation of facility location, capacity acquisition, and
technology selection for designing global manufacturing strategies. Eur J Oper Res 60(1), 1–18.
Retrieved from http://www.sciencedirect.com/science/article/pii/0377221792903287https://doi.
org/10.1016/0377-2217(92)90328-7
Drivers and Barriers for Cooperation
Between Municipalities in Area
of Municipal Solid Waste Management
Paulina Golinska-Dawson, Arkadiusz Kawa, and Piotr Januszewski
Abstract The growing amount of municipal solid wastes (MSW) per inhabitant is
increasing in most of the European countries. The organization of efficient logistics
for MSWs’ collection and recovery is crucial in order to improve the quality of
environment and the human wellbeing. In European Union the municipalities are
responsible to adopt measures to recycling and recovering of waste streams. The aim
of this chapter is to identify the drivers and barriers inter-municipal solid waste. The
data from over 230 Polish municipalities is analyzed. Our research contributes to the
sustainability literature by providing the empirical evidence in this area.
Keywords Inter-municipal cooperation · Solid waste management · Barriers and

drivers
1 Introduction
The European Union directive obliges member states to reduce their waste production
and to adopt measures in order to increase the level of recycling of waste stream (EC
2008). The municipal solid waste (MSW) management must be organized in a way
which is both environmentally and economically efficient. Previous research (e.g.
Bel and Warner 2008; Bel and Mur 2009; Bel et al. 2013; Blaeschke 2014; Dijkgraaf
and Gradus 2007; Warner and Hefetz 2003) have proved that collaboration between
municipalities in the area of the MSW management can lead to more cost-efficient
and environmental friendly solutions. Thus, such collaboration contributes to the
fulfilment of the goals of sustainable policy by the reduction of waste being landfilled.
P. Golinska-Dawson (B)
Faculty of Engineering Management, Poznań University of Technology,
Jacka Rychlewskiego 2 str., 60965 Poznań, Poland
e-mail: paulina.golinska@put.poznan.pl
A. Kawa
Łukasiewicz Research Network Institute of Logistics and Warehousing, Poznań, Poland
P. Januszewski
AtomScript, Poznań, Poland

56 P. Golinska-Dawson et al.
Bel et al. (2013) have stated that there is a shortage of empirical papers analyzing
the decision to engage in municipal cooperation.
Our paper contributes to this research gap by providing empirical evidence from
a big group of Polish municipalities. The aim of this chapter is first to examine the
drivers and barriers of the collaboration among municipalities and then to propose a
framework for facilitating decisions to engage in such a cooperation. In order to gain
in-depth knowledge, we have designed and conducted national survey among Polish
municipalities. This study seeks the answers for the research questions, as follows:
• RQ1: Which factors (drivers) are encouraging the municipalities to collaborate
by municipal solid waste management?
• RQ2: Which factors (barriers) are making the collaboration by MSW management
difficult?
• RQ3: How can decisions to engage in such cooperation be facilitated?
The paper is structured as follows. First a literature review is conducted in order to
identify and classify the drivers and barriers. The survey‘s methodology is presented
in Sect. 3. The results are analyzed in Sect. 4. The discussion on the empirical
studies is provided in Sect. 5 and the framework is also presented there. Finally, the
conclusions are stated in Sect. 6, including research limitations and future steps.
2 Literature Review
The literature review has been conducted in three main areas, as presented in Fig. 1.
Fig. 1 Scope of the literature review

Drivers and Barriers for Cooperation Between … 57
2.1 Legal Regulation and Its Impact on the MSW

Management
The accession of Poland to the European Union in 2004 has stimulated a series
of new environmental regulations with regard to MSW. The “Act on maintaining
cleanliness and order in communes” from 1st July 2011 (updated on 19th July 2019)
has introduced a new approach to MSW management based on the solutions used in
other EU countries. The Act puts on the municipalities the obligations to organize
a comprehensive MSW management. In accordance with that legal document the
municipality may act as individual entity in the area of MSW management or establish
a collaborative system with others. The main duties that must be fulfilled are, as
follows:
• to comprise all property owners in the commune in the MSW system,
• to supervise collection of MSW from property owners,
• to establish selective collection of MSW, including at least the following fractions:
waste: paper, metal, plastic, glass and packaging and multi-material biodegradable
municipal waste, biodegradable packaging,
• to create points of selective collection of MSW in a way, that it provides easy
access to all residents and where they can leave all waste electrical and electronic
equipment (WEEE).
Furthermore the municipalities should ensure the achievement of adequate levels
of recycling, re-use and recovery. They are responsible to constantly reduce the
weight of biodegradable MSW, which is transferred to disposal sites.
The municipality has to provide the information and educational campaigns on the
proper sorting of the MSW. The legal regulations are designed in a way which should
ensure the continuous improvement of the MSW management at the local level.
The legal regulations encourage the collaboration among the local governments in
order to build up the regional infrastructure and lower the MSW management costs.
Networks facilitate joint activities and enable the acquisition of resources, which
are unavailable to any of the individual parties. The tasks of collaborative model for
MSW management are, as follows:
• construction, maintenance and operation of common installations and equipment
for recovery or disposal of MSW,
• organizing selective collection of MSW,
• sorting and storage of MSW,
• common promoting activities.
In principle, members should jointly finance such investments. This is a favor-
able solution, especially for small and poor communities which are not able to
independently carry out the activities. The cooperation might cover different activ-
ities, from segregation at the source, through collection of waste from households,
to storage, recycling and/or disposal. Usually, such a cooperation also incudes a
common pricing and educational policy. Collaborative MSW management system

can develop common rules for waste collection and calculating related fees.
2.2 Municipal Solid Waste Management
The literature on municipal solid waste management can be divided into three
categories (Morrissey and Browne 2004):
• cost benefit analysis, which enables decision-makers to see what scenarios are
efficient in their use of resources,
• based on life cycle assessment LCA,
• based on multi-criteria decision making (MCDA), which takes into consideration
several individual and often conflicting criteria leading to more robust decision
making rather than optimizing a single dimensional objective function.
Tanskanen (2000) has stated that the first MSW management models focused on
optimization and dealt with specific aspects of that problem, for example vehicle
routing. Current theoretical models reflect a change in governmental policy where
MSW management is being transferred from a reliance on landfill, towards a
wider range of recycling and recovery options. Integrated Solid Waste Management
considers the full range of waste streams and selects the preferred recovery option
based on site specific environmental and economic factors (Morrissey and Browne
2004).
In our opinion the additional class of the literature on the municipal solid waste
management can be distinguished, namely country-specific studies. These studies
analyze different aspects of MSW at local (e.g. Bel and Mur 2009; Asase et al. 2009;
Al-Jarallah and Aleisa 2014) or national level (Visvanathan and Trankler 2003).
The literature on cooperation between municipalities by MSW management is
very limited. Bel et al. (2013) has stated that the literature presents a shortage of
empirical papers analyzing the decision taken to engage in cooperation. Warner
and Hefetz (2003), as well as Bel and Mur (2008) have suggested that cooperation
might be a good alternative to privatization, especially in smaller rural municipalities,
where the amount of potential contractors is small. Municipal cooperation by MSW
management in the Netherlands have been analyzed by Dijkgraaf and Gradus (2007).
Bel and Mur (2013) have analyzed the effects of municipal cooperation and priva-
tization on the costs of MSW management in rural areas. The results indicate that
small towns that cooperate have lower costs for their MSW service, higher collection
frequency and improved the quality of the service. Similar results has obtained by
Blaeschke (2014).
At the same time Sørensen (2007) has presented empirical results that those munic-
ipalities which cooperate have higher cost for the service than those municipalities
that don’t cooperate. Garrone et al. (2010) have showed that in Italy responsible for
managing municipal cooperation is a multi-government body, which may intensify
leadership problems. They state that inter-municipal collaboration might suffer from
higher coordination costs, which are not outweighed by the scale of achieved benefits.
In their opinion the management tensions can be a significant source of inefficiency.
Summarizing the literature review, it can be stated that the existing body of litera-
ture does not provide sufficient proof whether the cooperation by MSW management
creates more benefits or problems.
Our survey is contributing to that research gap by providing empirical evidence
about benefits and problems in the inter-municipal cooperation with regard to MSW
management.
Due to the lack of the clearly defined in the literature drivers and barriers of inter-
municipal collaboration by MSW management, the authors have applied in this work
the “supply chain partnership” theory.
In our opinion this theory is relevant, because the collaborative structure which
municipalities create to manage MSW aim to collect, transport and reprocess waste
from a big group of the citizens. The citizens create local decentralized “sourcing”
networks from which streams of waste (materials) are transported to the centralized
reprocessing facilities (for example recycling facilities, storage facilities, sorting
facilities). The processes are also typical for supply chains (especially reverse supply
chains).
2.3 Supply Chain Partnership
Cooperation between independent entities can have a formal, as well as informal

nature. Partnership in a supply chain should be understood as the formation of
economic relations between its participants on the basis of trust, shared risks and
benefits, leading to additional synergistic effects and a competitive advantage. In
principle, this is significant and long-term cooperation which is defined as strategic
partnership. It involves making joint business ventures aimed at achieving different
individual objectives subordinated to common goals of the whole supply chain
(Witkowski 2010).
The concept of supply chain management assumes that all participants obtain
benefits. Confrontation, ruthless competition and the “win-lose” market struggle
become costly and ineffective and begin to give way to the prudent cooperative
“win-win” game (Kempny 1998). At the same time, the “win-win” principle does not
necessarily mean equal distribution of benefits; however, each of the partners should
benefit from the cooperation. In case of reverse logistics processes the challenge is
to reach the “win-win” situation, where both environmental and economic goals are
met (Golinska and Kawa 2011).
Parties involved in the cooperation have their own resources, capabilities, tasks,
and objectives so there are difficulties in coordination of the constant flows of infor-
mation, materials, and funds across multiple functional areas both within and between
chain members (Golinska 2009).
Building a partnership is an expensive undertaking - good communication, excel-
lent coordination and an ability to share risk are needed. The partnership may be
justified if the results achieved within its framework are better than the effects of indi-
vidual business partners (Lambert et al. 2004). If the partnership is to be successful,
it is necessary to be able to recognize the relations of great potential and together
agree on the expectations concerning the partnership. One may wonder about the
method for selection of partners, but sometimes it is easier to reduce the number of
suppliers in order to be able to dominate these relationships. The idea is to organize
the cooperation in such a way that the parties gain as much as possible. In order to
do this, the expectations of the parties must be known, which is essential to build
strong relationships.
Leadership, goes beyond the organization, is associated with entrusting operating
units with decisions, and includes negotiations with external partners and adjustment
to constant changes (Schary and Skjøtt-Larsen 2001).
At the top of the supply chain is a leader that integrates the spatially dispersed
network entities, managing the flow of things, information and financial resources
along the network. It determines the shape of the supply chain, choosing the right
partners and mobilizing them to work, and protects it from competition from other
chains.
Govindan et al. (2013) have stated that coordination in the supply chain can
be achieved by various means of coordination mechanisms used to motivate the
members to participate in the optimization of the supply chain network. They are the
four main coordination mechanisms in the supply chain presented in the literature
(Govindan et al. 2013): contracts, information technology, information sharing, joint
decision making.
In case of inter-municipalities cooperation the focus is placed on coordination by
contracts and joint decision making.
Walker et al. (2008) have analyzed drivers and barriers to environmental supply
chain management practices in the private and public sector. Some of their findings
are also relevant to cooperation by the MSW management. They have identified the
internal and external barriers for supply chain cooperation, as follows (Walker et al.
2008):
• external: regulation, poor supplier commitment, industry specific barriers,
• internal: implementation costs, lack of intimacy.
In the same study are also identified the drivers, as follows (Walker et al. 2008):
• external: regulation, customer, competition, society, suppliers,
• internal: costs reduction pressure, organizational values.
Other studies have identified the most important drivers for building relation-
ships within business networks and cooperation, as follows (Ratajczak-Mrozek 2012;
Sarmah et al. 2006):
• new business opportunities, opportunities for growth,
• cost reduction,
• access to resources and capabilities, including knowledge,
• increasing the innovativeness of the company and its products,
• reducing operational risk,

• increasing bargaining power against other entities,
• the benefits of specialization,
• economy of scale.
In our survey, when designing research questions we have decided to not distin-
guish the barriers into external or internal categories. The barriers are defined here,
as follows:
• regulations (mainly at local government level),
• lack of suitable partners (as the equivalent to “poor supplier commitment”),
• lack of cooperative structures in the region (as the equivalent to “industry specific
barriers”),
• higher costs of cooperation,
• difficulties in a common realization of waste management tendering procedures
(as the equivalent to “lack of intimacy”).
The drivers are defined, as follows:
• possibilities for cost reduction and economy of scale resulting in lowering the
MSW fees for the inhabitants,
• more efficient promotion and PR activities,
• new possibilities for infrastructure investments by external financing (national
and international funds),
• organizational values (prestige),
• increasing the bargaining power over companies which tender for waste manage-
ment contracts.
In the next section is presented the methodology for primary data collection
in order to gain the in-depth knowledge for verifying the theoretical framework
presented in Fig. 2.
3 Methodology for Conducting the Survey
Due to the lack of sufficient secondary data to build up our model, we have decided to
conduct the descriptive survey first, to collect the relevant primary data. The descrip-
tive survey aims “understanding of the relevance of the phenomenon and describing
the incidence or distribution of the phenomenon in the population” (Karlsson 2009).
The survey aims to provide in-depth knowledge about the drivers and barriers and
barriers to establish cooperation between municipalities for management of MSW.
The following research steps have been conducted:
1. the municipalities database creation,
2. construction of the research questions and its verification by academia experts,
3. updating of the questionnaire and design of an on-line survey tool,
4. survey distribution by emails among municipalities in Poland,
New financing
possiblities
Cost reduction and Organizational values
economy of scale
More efficient Increased

promotion barganing power
and PR CMSWM
More difficult
Diffrent tendering
regulations procedure
Lack of suitable Higher costs of

partners cooperation
Lack of cooperative
structures in the region
Fig. 2 Theoretical framework—drivers and barriers to establish cooperation between municipali-

ties for management of MSW
5. data collection,
6. data analyses,
7. building the model and verification.
The next subsections present in detailed the conducted research procedures.
3.1 Research Sample
The local governance structure of Poland consists of three-levels: voivodships

(regions), counties, and municipalities (also called communes). At present, Poland is
divided into 16 voivodships, 380 counties, and 2479 municipalities (Karlsson 2009).
There are four kinds of communes:
• rural communes covering only rural areas,
• rural-urban communes which cover towns and rural areas,
• urban communes covering only towns/cities.
• cities with the district rights.
Table 1 presents the population size in each region and the amount of and munic-
ipalities per region. The regions are shown in the decreasing order regarding their
population size.
Table 1 Characteristics of regions in Poland (elaborated based on data from stat.gov.pl)

Voivodship Population [%] of Amount of [%] of total
population communes communes
1 Masovian Voivodship 5,301,760 13.8 314 12.7
(Mazowieckie)
2 Silesian Voivodship 4,615,870 12.0 167 6.7
(Śl˛askie)
3 Greater Poland Voivodship 3,462,196 9.0 226 9.1
(Wielkopolskie)
4 Lesser Poland Voivodship 3,354,077 8.7 182 7.3
(Małopolskie)
5 Lower Silesian Voivodship 2,914,362 7.6 169 6.8
(Dolnośl˛askie)
6 Lodz Voivodeship 2,524,651 6.6 177 7.1
(Łódzkie)
7 Pomeranian Voivodeship 2,290,070 5.9 123 5.0
(Pomorskie)
8 Lublin Voivodeship 2,165,651 5.6 213 8.6
(Lubelskie)
9 Subcarpathian Voivodeship 2,129,951 5.5 160 6.5
(Podkarpackie)
10 Kuyavian-Pomeranian 2,096,404 5.4 144 5.8
Voivodeship
(Kujawsko-Pomorskie)
11 West Pomeranian 1,721,405 4.5 114 4.6
Voivodeship
(Zachodniopomorskie)
12 Warmian-Masurian 1,450,697 3.8 116 4.7
Voivodeship
(Warmińsko-Mazurskie)
13 Świ˛etokrzyskie 1,273,995 3.3 102 4.1
Voivodeship
(Świ˛etokrzyskie)
14 Podlaskie Voivodeship 1,198,690 3.1 118 4.8
(Podlaskie)
15 Lubusz Voivodeship 1,023,317 2.7 83 3.3
(Lubuskie)
16 Opole Voivodeship 1,010,203 2.6 71 2.9
(Opolskie)
The electronic survey was sent to 2459 communes in Poland. A combination of

server and client side scripts has been used for the implementation of this customized
e-survey. Emails containing a link to the questionnaire have been sent in 2 rounds two
weeks apart. The overall response rate was 9.6%, what was satisfactory (because the
sample N = 2459). We have received 136 responses for municipalities after the first
round and 100 after the second attempt, which made a total of 236 valid responses.
With a population of 2459, the minimal sample which should be taken into account for
a survey is 208, assuming a 90% confidence level. Formula (1) was used to calculate
this value. The number of received responses is higher than the minimal sample,
therefore the authors have assumed that the survey results are valid. However all
findings within this paper will be based on the above assumptions. The respondents
have been divided into two groups:
• IMSWM—individual municipal solid waste (MSW) managed by single munici-
pality,
• CMSWM—collaborative MSW management.
P(1 − P)
n= e2 P(1−P)
(1)
Z2
+ N
where:
n—sample size
P—assumed proportion of municipalities which are a member of a CMSWM (P
= 0.3)
e—error margin
Z—the Z value which corresponds to the assumed confidence level (Z = 1645)
N—population size.
3.2 Research Questions
The questionnaire consists of 13 questions, which are structured, as presented in

Fig. 3.
The questions Q2–Q8 are addressed to the CMSWM. They focus on the verifica-
tion of the drivers which have been identified based on the literature review and the
previous in-field studies of the authors. Table 2 presents the scope of questions for
the members of CMSWM’s systems.
The questions Q9–Q13 are addressed to the non-members of the CMSWM’s
system (see Table 3).
Fig. 3 The logic of the research questionnaire (own elaboration)
Table 2 The scope of questions for members of CMSWM’s system (own elaboration)
Questions Scope
Q2 Does cooperation meet the requirements and help to fulfil the goals?
Q3–Q5 The pricing policy
Q6–Q8 Main benefits of being a member of a CMSWM
Table 3 The scope of questions for municipalities which are not member of CMSWM’s system
(own elaboration)
Questions Scope
Q9 Previous experience with being a CMSWM member
Q10 Reasons for withdrawal from the CMSWM
Q11, Q12 Main barriers of being a member of a CMSWM
Q12, Q13 Main expected benefits of being a member of a CMSWM in the future
4 Research Results
The responses have been received from municipalities located in all of the 16 Polish
regions. Table 4 presents the response rates in each region.
The responses in particular regions have been assessed as satisfactory (the mean
value was 9.46% with standard deviation 1.9%), thus the national representation of
the survey’s results has been ensured.
Most of the communes (85%) have not yet established the collaborative struc-
ture in the area of MSW management. Pearson analysis has been undertaken to
measure the correlation between the accession rate to CMSWM’s system and two
most commonly used statistical indicators, namely:
• the municipality area, and
• population.
Table 4 The response rates in each region (own elaboration)

Voivodeship Amount of % of total % of response to total
responses response amount communes in
region
1 Masovian Voivodeship 36 15.3 11.5
2 Silesian Voivodeship 19 8.1 11.4
3 Greater Poland 25 10.6 11.1
Voivodeship
4 Lesser Poland Voivodeship 15 6.4 8.2
5 Lower Silesian 15 6.4 8.9
Voivodeship
6 Łódź Voivodeship 14 5.9 7.9
7 Pomeranian Voivodeship 14 5.9 11.4
8 Lublin Voivodeship 15 6.4 7.0
9 Subcarpathian Voivodeship 13 5.5 8.1
10 Kuyavian-Pomeranian 15 6.4 10.4
Voivodeship
11 West Pomeranian 8 3.4 7.0
Voivodeship
12 Warmian-Masurian 11 4.7 9.5
Voivodeship
13 Świ˛etokrzyskie 8 3.4 7.8
Voivodeship
14 Podlaskie Voivodeship 11 4.7 9.3
15 Lubusz Voivodeship 11 4.7 13.3
16 Opole Voivodeship 6 2.5 8.5
Total 236 100.0
The results Pearson’s analysis has been shown in Table 5. The correlation between
the size (area and population) of the municipality is relatively small. It is worth
mentioning that both, small and large municipalities, are members of the CMSWM’s
system in similar proportion.
The authors have also tested a correlation between the type of a municipality and
their decision to join a CMSWM’s system. In Table 6 is presented the data on the
percentage of the municipalities who have joined a CMSWM (split by type). As can
be observed most municipalities which have joined a CMSWM are classified as rural
Table 5 Pearson analysis

Area in ha Population Question 1
between CMSWM accession
and the municipality’s Area in ha 1
population and area Population 0.319445263 1
Question 1 0.151359432 −0.040921992 1
Table 6 Comparison of the structure of municipality types

Rural (%) Urban (%) Urban-Rural (%) City with district rights
(%)
Municipalities which 47.06 14.71 35.29 2.94
have decided to join a
CMSWM
Municipality’s 63.17 9.64 24.53 2.66
structure in Poland
or urban-rural. However the comparison of the structure of municipality types and

the percentage rates of municipalities which have decided to join a community is
near equality. This means that there is no trend to join CMSWM’s by municipalities
of a given type.
To conclude, there has not been observed a clear correlation between the member-
ship rate in CMSWM’s system and a municipality’s population, its area or its type.
This means that the decision to join a CMSWM is based on other factors than the
ones stated above.
The respondents who are not members of any CMSWM’s system have been asked
whether they would consider joining the inter-municipal cooperation in the future
(e.g. next 2 years). The majority of the respondents 90.6% don’t want to build up the
inter-municipal collaboration. The respondents that don’t belong to any CMSWM’s
system have been asked also about barriers. The summary of their responses is
presented in Fig. 4.
About 45% of the respondents have stated that there is none collaborative structure
in their neighborhood. About 33% of municipalities believe that their own IMSWM’s
system is efficient and there is no need for cooperation in this area. About 13%
Barriers for being a member of collaborative MSW management

(CMSWM)
No CMSWM in the area
Sufficiently developed own WM system
Lack of suitable partners

Difficulties in joint tendering procedure for
the waste collection &processing
Differences in MSWM regulations between
the muncipalities
Other
Higher WM costs
0% 5% 10%15%20%25%30%35%40%45%50%
Fig. 4 Barriers to join a CMSWM’s system

of the respondents have stated that there lack of suitable partners for the creating
of a new CMSWM’s system. The regulations on MSWM which differ between
municipalities have been addressed as barrier by about 8% of respondents. Cost
issues have been mentioned only by 6% of respondents. The respondents have also
mentioned individually identified barriers named as “other” in Fig. 4. They are not
discussed in this paper.
The respondents who consider joining the CMSWM’s system have listed three
potential benefits of such action, as:
• easier access to EU subsidies,
• lower cost of waste collection and processing/recovery,
• lower cost of investment in infrastructure.
The separate set of questions (Q2–Q8) have been directed to the communes who
are members of CMSWM.
Answering to the question Q2 about 86% of communes have declared that
membership in CMSWM meets their requirements and needs. The respondents, who
are members of a CMSWM have been asked (Q3–Q5) also about pricing policy and
potential changes in the existing pricing practices. About 67% of communes use the
same pricing policy within a CMSWM. That means that all the inhabitants within a
particular CMSWM pay the same charge. The typical methods for calculating fees
for MSW management are, as follows:
• based on the number of residents,
• based on the surface of a property,
• based on the amount of water used,
• flat fee for every household.
About 80% of communes are calculating fee based on the number of residents
per household. When determining that fee usually three types of costs are taken into
account: the costs of waste management (collection and disposal of municipal waste,
creation and maintenance of points of selective collection of municipal waste), the
cost of transport and administration costs.
About 78% municipalities don’t plan to change the pricing policy in the next two
years. It can be assumed that the existing pricing policies are well designed.
In questions Q6 and Q7 have been investigated the benefits of participating in
CMSWM. The answers are presented in Fig. 5. These benefits can be also perceived
as “internal and external drivers”, as referred in the previous literature studies.
Only 11% of the communes haven’t identified any benefits of cooperation. Most
of the municipalities have named the cost reduction as the main benefit. The cost
reduction is usually resulting from the economy of scale and it allows lowering
the waste collection fees for the inhabitants. The cost factor has been examined in
detailed, and its elements have been identified as:
• cost of waste processing,
• cost of waste transportation.
Drivers to be a member of a CMSWM
More eficient promotion and PR 67%

Lower cost of waste collection
58%
& processing/recovery
Easier infrastructure investments 44%
Simpler tendering procedures 36%
Lower waste transportation costs 33%
Easier access to EU subsidies 25%
None 11%
Presteige 11%
Other 3%
0% 10% 20% 30% 40% 50% 60% 70%
Fig. 5 Benefits of an inter-municipal cooperation by municipal solid waste management
About 44% of respondents have listed as the benefit also an easier access to
resources and possibilities for new infrastructure investments. About 25% of the
respondents have identified as benefit also new opportunities for external financing
(national and EU funds).
5 Discussion
The scope of collaboration by MSW management is still relatively low in Poland.

The current policy aims to encourage this kind of collaboration. In our empirical
survey we search for the answer for the following research questions:
• RQ1: Which factors (drivers) are encouraging the municipalities to collaborate
by municipal solid waste management?
• RQ2: Which factors (barriers) are making the collaboration by MSW management
difficult?
• RQ3: How can decisions to engage in such cooperation be facilitated?
The empirical results have shown that the cost reduction is the most important
factor for creating inter-municipal collaboration. It is easier to achieve the economy
of scale in a CMSWM. The costs of MSW collection, transportation, processing and
recovery are lower in case of collaborative supply chain especially in case of small
communes. These results are consistent with the previous research of Bel and Mur
(2009), Bel et al. (2013). They have stated that by inter-municipal cooperation a cost
reduction and the economy of scale can be achieved. Creating a CMSWM allows to
lower the costs for the inhabitants in the area. The inhabitants are key stakeholders,
so decreasing charges for them is an important goal of the municipality, whose

governors are dependent on inhabitants (by direct election).
Second important driver is the new possibility for external financing (national and
international funds) in case of infrastructure investments. In our studies this driver
is described by the two elements:
• easier infrastructure investments,
• easier access to EU subsidies.
In Poland the EU development grants are an important source of financing of
investments. So far the collaborative MSW management entities have easier access
to apply for them.
Third important driver, which has been identified in our survey is more effi-
cient promotion and PR. The Polish regulations oblige the municipalities to promote
selective collection of waste. The efficient information campaign also helps munici-
palities to solve problems with illegal dumping or avoiding the MSW sorting by the
inhabitants.
The analysis of the survey results shows that the increased bargaining power is
the less significant factor. The least important factor is “organizational values” which
has been only mentioned by 11.1% respondents.
The second research question aims to examine the barriers for joining CMSWM.
Our findings are consistent with the previous literature statements. The lack of estab-
lished collaborative structure in the neighborhood or lacks of suitable partners are
important barriers for development of a CMSWM in Poland. Furthermore a big group
of communes believe that the already existing IMSWM’s infrastructure is sufficient
and there is no need for cooperation. The non-parametrical statistics tests (chi square
p values were insignificant) have showed that the other barriers are less influential.
The difficulties in joint tendering procedure for the waste collection, processing and
recovery are mentioned by some municipalities. The two least important barriers
are potential higher cost of CMSWM and more complicated tendering procedures.
However these barriers cannot be completely ignored. These results correspond
with the findings of Garrone et al. (2010), who have stated that inter-municipal
collaboration might suffer from higher coordination costs, due to the inefficiency of
management.
In order to find the answer for the last research question the results from our
survey have been grouped into categories assigned to the model and normalized.
The first step is to calculate the relative values of the drivers (ir ). We have assumed
that the drivers should be maximized and we have used the formula from Kosacka
et al. (2015).
i d − i dmin
ir driver = (2)
i dmax − i dmin
where:
id —is the recent value of the driver
idmin —is the minimum value for the driver

idmax —is the maximum value of the driver.
Then we repeat the procedure to calculate the relative states of barriers. We have
made the assumption that barriers should be minimized. We have used the formula
for relativization from Kosacka et al. (2015).
i b − i bmin
irbarrier = 1 − (3)
i bmax − i bmin
where:
ib —is the recent value of the barrier
ibmin —is the minimal value for the barrier
ibmax —is the maximum value of the barrier.
Each driver has different importance level ηid , where 〖0 < η〗_id ≤ 1. The closer
the value of ηid to 1, the more important that driver.
Each barrier has different importance level ηid , where 〖0 < η〗_ib ≤ 1. The
closer the value of ηib to 1, the more important is that barrier.
The importance determinant is calculated separately for drivers and barriers from
Formula 4 and 5.
xid = ηid ∗ ir d (4)
where:
xid —is a importance determinant of ith driver
xib = ηib ∗ ir b (5)
xib —is a importance determinant of ith barrier.
Table 7 presents the calculation of the importance determinants for drivers based
on Formulas (2 and 4).
Table 8 presents the calculation of the importance determinants for barriers based
on Formulas (3 and 5).
Table 7 Calculation of the

Drivers ηid xid
importance determinants for
drivers Cost reduction and economy of scale (d1) 0.33 0.305533
New financing possibilities (d2) 0.27 0.185173
More efficient promotion and PR (d3) 0.20 0.13334
Increased bargaining power (d4) 0.13 0.048147
Organization values (d5) 0.07 0.007407
Total 0.6796
Table 8 Calculation of the importance determinant for barriers

Barriers ηib irbarrier xib
No CMSWM in the area (b1) 0.29 0.565 0.1614
Sufficiently developed IMSWM’s system (b2) 0.24 0.675 0.1607
Lack of suitable partners (b3) 0.19 0.875 0.1667
Difficulties in joint tendering procedure (b4) 0.14 0.91 0.1300
Different regulations (b5) 0.10 0.925 0.0881
Higher cost of cooperation (b6) 0.05 0.945 0.0450
Total 0.7519
In Table 8 it was necessary to add one barrier which has not been initially identi-
fied in the preliminary theoretical framework (Fig. 3). For the respondents the own
existing infrastructure was an important factor to not start the collaboration.
At present the existing barriers are stronger than drivers to start cooperation. The
“lack of existing structure” that the municipality might join, as well as “lack of
suitable partners” to start a CMSWM from scratch are the important barriers.
The empirical results have allowed us for verifying and modifying the preliminary
framework presented in Fig. 3. The final version of the model is presented in Fig. 6.
Fig. 6 Framework to enable decision to join cooperation for MSW

6 Conclusions
The paper presents the issues related to inter-municipal collaboration by MSW

management with focus on municipalities. On the example of Polish communes
we have discussed the barriers and drivers of the inter-municipal collaboration. The
empirical studies in this area are rather limited. Our paper contributes to the sustain-
ability literature by providing empirical evidence from a relatively big group of
municipalities in order to verify theoretical statements. Based on the theoretical
review and the results from our survey the framework to enable decision to join
cooperation for MSW has been created. The main findings of our paper are:
• at present the existing barriers are stronger than drivers to start cooperation,
• lack of existing structure that the municipality might joint is one of the main
barriers,
• lack of suitable partners to start a CMSWM from scratch is one of the main
barriers,
• existence of sufficient own infrastructure to manage MSW is very important
blocking factor to start building up an inter-municipal collaboration,
• the main driver for cooperation is cost reduction and possibility to achieve
economy of scale,
• the driver of high importance with the potential for enabling the future cooperation
(helping to overcome main barriers) is “new financing possibilities” (thus, its
importance determinant is higher than any of the three major barriers).
The policy makers should develop the policy to encourage building up new
inter-municipal cooperation by MSW management guiding the potential partners
and providing networking tools. More external financing which is dedicated to the
CMSWM should enable to establish cooperation.
The analysis is not free of certain limitations which also suggest further areas
for study. First of all, the analysis has included communes only in Poland, so the
presented results might be country-specific biased. It might be also beneficial to go
beyond the communes and to examine also other stakeholders in the CMSWM.
For example public and private companies which perform the waste collection,
transportation and processing tasks on behalf of the communes are very important
stakeholders. Finally, due to the applied survey approach, more detailed analyses of
particular relationships have been impossible to conduct. The in-depth studies should
be conducted in the municipalities which belong to collaborative MSW management
entities.
References
Al-Jarallah R, Aleisa E (2014) A baseline study characterizing the municipal solid waste in the
State of Kuwait. Waste Manag 34(5):952–960
Asase M, Yanful EK, Mensah M, Stanford J, Amponsah S (2009) Comparison of municipal solid
waste management systems in Canada and Ghana: a case study of the cities of London, Ontario,
and Kumasi, Ghana. Waste Manag 29(10):2779–2786
Bel G, Mur M (2009) Inter-municipal cooperation, privatization and waste management costs:
evidence from rural municipalities. Waste Manag 29(10):2772–2778
Bel G, Warner ME (2008) Does privatization of solid waste and water services reduce costs? A
review of empirical studies. Resour Conserv Recycl 52:1337–1348
Bel G, Fageda X, Mur M (2013) Why do municipalities cooperate to provide local public services?
An empirical analysis. Local Gov Stud 39(3):435–454
Blaeschke F (2014) What drives small municipalities to cooperate? Evidence from Hessian
municipalities (No. 14-2014). Joint Discussion Paper Series in Economics
Dijkgraaf E, Gradus RHJM (2007) Collusion in the Dutch waste collection market. Local Gov Stud
33:573–588
EC (2008) Directive 2008/98/EC of the European Parliament and of the Council on Waste
Framework Directive 19 November 2008. http://ec.europa.eu/environment/waste/legislation/a.
htm, Accessed 01.02.2014
Garrone P, Grilli L, Rousseau X (2010) Restructuring public enterprises: evidence from Italian
municipal utilities. Available in SSRN: http://ssrn.com/abstract=1610124
Golinska P (2009) The concept of an agent-based system for planning of closed loop supplies in
manufacturing system. In: Omatu S et al (eds) IWANN 2009 part II. LCNS vol 5518. Springer,
Berlin, pp 346–349
Golinska P, Kawa A (2011) Recovery network arrangements WEEE case. In: Golinska P, Fertsch
M, Marx Gomez J (eds) Information technologies in environmental engineering—trends and
challenges, Springer, Berlin, pp 579–593
Govindan K, Popiuc MN, Diabat A (2013) Overview of coordination contracts within forward and
reverse supply chains. J Clean Prod 47:319–334
Karlsson C (2009) Researching operations management. Taylor and Francis, New York
Kempny D (1998) Co-markership: Zarz˛adzanie dostawami w biznesie przyszłości. In: Zarz˛adzanie
łańcuchami dostaw. Conference proceedings. Katowice
Kosacka M, Golinska-Dawson P, Mierzwiak R (2015) Sustainability classification for SMEs from
the remanufacturing sector. Chiang Mai J Sci (in Press)
Lambert DM, Knemeyer AM, Gardner JT (2004) Supply chain partnerships: model validation and
implementation. J Bus Logist 25(2):21–42
Morrissey AJ, Browne J (2004) Waste management models and their application to sustainable
waste management. Waste Manag 24(3):297–308
Ratajczak-Mrozek M (2012) Global business networks and cooperation within supply chain as a
strategy for high-tech companies’ growth. J Entrepren Manag Innov 8(1):35–51
Sarmah SP, Acharya D, Goyal SK (2006) Buyer vendor coordination models in supply chain
management. Eur J Oper Res 175(1):1–15
Schary PB, Skjøtt-Larsen T (2001) Managing the global supply chain. Handelshøjskolens forlag,
Copenhagen
Sørensen RJ (2007) Does dispersed public ownership impair efficiency? The case of refuse collection
in Norway. Public Adm 85:1045–1058
Tanskanen J-H (2000) Strategic planning of municipal solid waste management. Resour Conserv
Recycl 30:111–133
Visvanathan C, Trankler J (2003) Municipal solid waste management in Asia: a comparative
analysis. In: Workshop on sustainable landfill management, pp 3–5
Walker H, Di Sisto L, McBain D (2008) Drivers and barriers to environmental supply chain
management practices: lessons from the public and private sectors. J Purchas Supply Manag
14(1):69–85
Warner ME, Hefetz A (2003) Rural–urban differences in privatization: limits to the competitive
state. Environ Plan C: Gov Policy 21:703–718
Witkowski J (2010) Zarz˛adzanie łańcuchem dostaw: koncepcje, procedury, doświadczenia. Polskie
Wydawnictwo Ekonomiczne, Warszawa
A Novel Formulation for the Sustainable
Periodic Waste Collection Arc-Routing
Problem: A Hybrid Multi-objective
Optimization Algorithm
Erfan Babaee Tirkolaee, Alireza Goli, Gerhard-Wilhelm Weber,

and Katarzyna Szwedzka
Abstract Municipal solid waste (MSW) management is among the essential tasks
of municipalities that requires large amounts of fixed/variable and investment costs.
In this system, the processes of collection and transportation include the major
part of these costs. On the other hand, greenhouse gas (GHG) emission as envi-
ronmental aspect and citizenship satisfaction as social aspect are also of particular
importance, which are vital requirements for MSW management that need to be
taken into account. In this study, a novel mixed-integer linear programming (MILP)
model is developed to formulate the sustainable periodic capacitated arc routing
problem (PCARP) for MSW management. The objectives are to concurrently mini-
mize the total cost and total environmental emission, and maximize citizenship
satisfaction. To solve the problem efficiently, a hybrid multi-objective optimization
algorithm, namely, MOSA-MOIWOA is designed based on multi-objective simu-
lated annealing algorithm (MOSA) and multi-objective invasive weed optimiza-
tion algorithm (MOIWOA). To increase the algorithm performance, the Taguchi
design method is employed to set the parameters optimally. The validation of the
proposed methodology is evaluated using several problem instances in the literature.
Finally, the obtained results demonstrate the high efficiency of the proposed model
and algorithm to solve the problem.
E. B. Tirkolaee (B)
Department of Industrial Engineering, Istinye University, Istanbul, Turkey
e-mail: e.babaee@ustmb.ac.ir
A. Goli
Department of Industrial Engineering, Yazd University, Yazd, Iran
e-mail: a.goli@stu.yazd.ac.ir
G.-W. Weber · K. Szwedzka
Faculty of Engineering Management, Poznań University of Technology, Poznań, Poland
e-mail: gerhard.weber@put.poznan.pl
K. Szwedzka
e-mail: katarzyna.szwedzka@put.poznan.pl
G.-W. Weber
Institute of Applied Mathematics, Middle East Technical University, Ankara, Turkey

78 E. B. Tirkolaee et al.
Keywords Sustainable waste collection · Municipal solid waste · Periodic

capacitated arc routing problem · Multi-objective invasive weed optimization
algorithm · Multi-objective simulated annealing · Taguchi design method
1 Introduction
Solid wastes generated in urban districts are known as municipal solid waste (MSW).
There are various kinds of MSW that can be generated by residents, schools and
universities, industrial companies, hospitals and clinical centers, public centers such
as streets, markets, bus stops, parks, etc. The indisputable expansion of cities and
the consequent increase of the population, particularly in recent years, has led to
a growing consumption of materials. This has also resulted in an increase of waste
generation in urban districts, such that in 2012, cities worldwide generated 1.3 billion
tons of MSW that is equal to 1.2 kg per capita per day. Concerning these trends and
figures, it is anticipated that the annual rate of waste generation reaches up to 2.2
billion tons by 2025 (Statista 2019). This issue is also highly perceptible in advanced
countries. For example, in Western Europe, despite the successful training system,
the annual generation of MSW per capita is over 600 kg (Gapminder 2019).
As a result, the increased amount of waste generated in recent years requires
an essential higher budget for the processes of collection and transportation. For
example, the budget dedicated to MSW management in Toronto, Canada was $382.2
million in 2018 (Toronto City Council 2019). Furthermore, according to the reports
by local organizations in Malaysia, more than $1 billion is paid for MSW manage-
ment such that this amount constitutes almost 70–80% of the organizations’ incomes
(Budhiarta et al. 2012). In this way, a strict and effective management is much needed
to ensure the essential service level for establishing the requirements of public health
and maintaining policies for the maximum amount of recycling/recovery. Unfor-
tunately, based on the limited available resources, it is hard to establish a waste
management system for municipalities which are in charge of the direct collection,
transportation and disposal.
Inefficient management and disposal of MSW lead to the pollution of water
(groundwater and surface water resources), soil and air. In urban districts, lack of
timely MSW collection results in the output of stagnant leachates in line with the
growth of insects and the infection of diseases.
Moreover, the emission of greenhouse gases (GHGs) associated with MSW trans-
port has attracted much attention in recent decades, including in the European Union
(EU) (Inghels et al. 2016). According to the European Environment Agency (EEA),
waste collection and transportation accounts for up to 5% of direct GHG emissions,
mainly due to short distances that MSW is shipped. However, following the adverse
GHG emission growth associated with different modes of transport, this figure will
rise to about 40% of net greenhouse gas emissions by 2020 (EEA 2008).
Furthermore, job creation opportunity is worth considering in this field. Waste-
collection vehicles need drivers and crew to perform the collection and transportation.
Providing a sustainable waste collection and transportation system requires the study
A Novel Formulation for the Sustainable Periodic … 79
of conflicts and trade-offs between economic, environmental and social objectives

subject to the main operating constraints.
Therefore, this research tries to investigate the sustainability aspect of MSW
collection and transportation problem by designing an efficient methodology
including mathematical model and solution techniques. Accordingly, a novel mixed-
integer linear programming (MILP) model is developed to formulate the sustainable
periodic capacitated arc routing problem (PCARP). The objectives are to minimize
the total cost of the MSW management system, minimize the total pollution emis-
sion and maximize the total job opportunities. To solve the problem, a hybrid multi-
objective optimization algorithm, namely, MOSA-MOIWOA is designed. Moreover,
the Taguchi design method is employed to provide a higher efficiency for the proposed
algorithm.
The remaining sections of the paper are organized as follows. Section 2 provides
a quick review on the background of the study. Section 3 explains the problem and
the main assumptions of the model. Moreover, the proposed mathematical model
is given in this section. Section 4 introduces the proposed solution techniques. The
results and comparative analyses are given in Sect. 5. Finally, Sect. 6 provides the
concluding remarks and suggestions for future studies.
2 Background
The arc routing problem (ARP) is a kind of routing problem that has been specifically
studied to formulate the vehicle routing operations in waste collection, snow removal
of streets, street painting and other street-related services. This type of routing was
firstly introduced by Golden and Wang (1981), which is a specific area of vehicle
routing problem (VRP). Based on the literature, VRP is an extension of the traveling
salesman problem (TSP) which was studied by a wide range of researchers, for
example, see Lin and Kernighan (1973), Paletta and Triki (2004) and Lou et al.
(2020). Furthermore, ARP is known as an extension of the Chinese Postman Problem
(CPP) which can be found in many studies such as Pearn (1994), Filipiak et al. (2009)
and Çodur and Yılmaz (2020).
The most important research works on ARP and in the area of municipal solid
waste management are summarized as follows. Bautista et al. (2008) modified the
ARP by applying changes to the node routing to solve the waste collection problem in
the municipality of a town of Barcelona. These changes were due to road constraints
such as no U-turn to return from the edge. They solved the problem using the ACS
algorithm, which was based on the nearest-neighborhood and closest-placement
methods. Rodrigues and Soeiro Ferreira (2015) developed a mixed capacitated arc
routing problem (CARP) for the collection and transportation of municipal waste.
They considered a heterogeneous fleet of vehicles and multiple disposal sites in the
proposed problem and solved the problem by CPLEX solver using benchmark prob-
lems. Babaee Tirkolaee et al. (2016) applied an SA algorithm to solve the robust
CARP under fuzzy demand for urban waste collection. To improve the performance
of the proposed algorithm, they employed the Taguchi design method to adjust the
algorithm parameters and evaluated the performance of the algorithm compared to
the CPLEX solver. A comprehensive review of ARP was conducted by Mourão and
Pinto (2017) which considers the most important researches from 2010 to 2017.
Tirkoaee et al. (2018) developed a robust CARP to solve the urban waste collection
problem by considering driver and crew shifts. To solve the proposed problem and
validate their mathematical model, they designed random examples and solved them
by an exact method and SA algorithm. A hybrid genetic algorithm (GA) was devel-
oped by Tirkolaee et al. (2018) to solve the multi-trip capacitated green arc routing
problem to collect MSW. To evaluate the performance of the proposed algorithm,
they generated several random examples in different sizes and evaluated the results
compared to the CPLEX solver. Tirkoaee et al. (2019) proposed a MILP formula-
tion for the problem of multi-trip CARP for urban waste collection. To solve the
problem, they developed an improved max-mini ant system (MMAS) algorithm and
evaluated the performance of the algorithm compared to the heuristic algorithms and
the simplified version of the max-min ant colony optimization algorithm by bench-
mark samples. Recently, several studies have been done in the area of uncertainty. A
robust bi-objective MILP model was introduced by Babaee Tirkoaee et al. (2019) for
multi-period multi-trip CARP under fuzzy demand. The model aimed to minimize
the total cost and the longest travelled distance of vehicles simultaneously. They
implemented multi-objective invasive weed optimization (MOIWO) algorithm and
ε-constraint method to solve the problem. Babaee Tirkoaee et al. (2020) developed
an improved hybrid ant colony optimization (ACO) algorithm to solve the multi-
trip CARP under fuzzy demand for urban waste collection. They implemented their
proposed model as a case study in Isfahan, Iran.
By scrutinizing the background, it is concluded that no research has been under-
taken to study the economic, environmental and social aspects of urban waste
management. Accordingly, a novel formulation is developed for the sustainable
PCARP aimed at minimizing the total cost and total pollution emission, and maxi-
mizing the total job opportunity. Moreover, an efficient multi-objective algorithm is
designed to solve the problem in large scales.
3 Problem Description
This section introduces the sustainable PCARP and its specific characteristics. The
operational objective is to determine the optimal number of required waste-collection
vehicles among a fleet of homogeneous vehicles as well as constructing the optimal
planning of routes. The sustainability is addressed by studying its three main aspects
(economic, environmental and social) as separate objective functions in the proposed
model. The first objective function seeks to minimize the total cost including the
traversing costs and usage costs of vehicles. Total pollution emission is regarded as
Fig. 1 Schematic view of the proposed problem
the second objective function to be minimized. The third objective function is to maxi-
mize job opportunities through increasing the number of hired labor. Accordingly,
the sustainability aspect of the problem can be studied.
The proposed network can be displayed by a graph G = (V, E), where V = {1,
2… n} denotes the set of nodes and E represents the set of edges. Each pair of
nodes constitutes an edge of the network and each edge contains two arcs in opposite
directions. Here, each edge in the set E may be a waste edge (required edge) or
non-waste edge (traversing edges).
Thus, ER ⊆ E displays the set of required edges
so that ER = (i, j)|dijt > 0 , where dijt represents the demand of edge (i, j) in tth
period. So, each non-waste edge has a demand equals to 0. Moreover, node number
1 stands for the depot and node number n shows the disposal site.
Figure 1 presents a schematic view of the problem.
As can be seen, the suggested MSW collection network in Fig. 1 is composed of
seven required edges (1–2, 2–3, 3–4, 4–8, 5–6, 6–7, 7–8), one depot node and one
disposal site. As a possible solution for a single planning period, all required edge
are covered exactly by one vehicle and two trips. The required and traversing edges
are specified by trash bins and dashed lines, respectively. The optimal routing for the
first and second trips are 1–2 → 2–3 → 3–4 → 4–8 and 1–8 → 8–7 → 7–6 → 6–5
→ 5–8, respectively. Finally, the vehicle moves back to the depot for completing its
tour.
The main assumptions of the model are given as follows:
I. Each required edge is covered only by one vehicle.
II. Separate locations are considered for the depot and disposal site in the network.
III.Vehicles begin their first trip from the depot and end it at the disposal site.
Then start their possible 2nd, 3rd,…, pth trips from the disposal site and end
at there again.
IV. A fleet of homogeneous vehicles is taken into account.
V. Vehicles have a maximum available service time.
VI. Usage cost of vehicles include drivers’ and crew’s wage cost, fuel cost, hiring
cost, etc.,
VII. A set of planning periods is regarded to serve the required edges.
VIII. Environmental pollution is considered for the transportation of waste in urban

districts.
IX. Social aspect is studied by increasing the number of hired labor.
Now, the main components of the proposed mathematical model including sets
and indices, parameters and variables are listed as follows.
Sets and indices
V Set of nodes; i, j ∈ V ,
E Set of edges; (i, j) ∈ E,

ER Set of required edges; ER = (i, j)|dijt > 0 ,
K Set of homogeneous vehicles; k ∈ K,
P Set of vehicle trips; p ∈ P,
T Set of planning periods; t ∈ T ,
S Each possible subset of edges,
V [S] Set of nodes constituting S,
Parameters
cij Distance of edge (i, j),

W Available capacity for each vehicle,
dijt Demand of edge (i, j) in tth period,
Tmax Maximum available time for vehicles,
M A large number,
tij Traversing time of edge (i, j),
cvk Usage cost of kth vehicle,
θ Conversion factor of distance to cost
Gij Amount of pollution emission released by traversing edge (i, j),
σ Number of required workforce for each vehicle,
Decision variables
p
xijkt Number of traversing the edge (i, j) ∈ E by kth vehicle in pth trip and tth period,
p
yijkt 1 if edge (i, j) ∈ ER is served by kth vehicle in pth trip and tth period, otherwise 0
ukt 1 if kth vehicle is employed in tth period, otherwise 0
p
LTk Total loading time of kth vehicle in pth trip and tth period,
p
UTk Total unloading time of kth vehicle in pth trip and tth period,
Now, the developed MILP formulation of the problem is as follows:

⎛ ⎞
p

minimize Z1 = θ ⎝ cij xijkt ⎠ + cvk ukt (1)
(i,j)∈E p∈P t∈T k∈K t∈T k∈K
p
minimize Z2 = Gij xijkt (2)
(i,j)∈E p∈P t∈T k∈K

maximize Z3 = σ ukt (3)
t∈T k∈K
p
p
xijkt = xjikt ∀i ∈ V [E]; (i, j) ∈ E, ∀k ∈ K, ∀p ∈ P, ∀t ∈ T , (4)
i∈V [E] j∈V [E]
p p
yijkt + yjikt = 1 ∀(i, j)or(j, i) ∈ ER , ∀t ∈ T , (5)
p∈P k∈K
p
dijt yijkt ≤ W ∀k ∈ K, ∀p ∈ P, ∀t ∈ T , (6)
(i,j)∈ER
p p
yijkt ≤ xijkt ∀(i, j) ∈ E, ∀k ∈ K, ∀p ∈ P, ∀t ∈ T , (7)
p
xijk ≤ Mukt ∀k ∈ K, ∀t ∈ T , (8)
p∈P (i,j)∈E
p
p
LTkt = ul dij yijkt ∀k ∈ K, ∀p ∈ P, ∀t ∈ T , (9)
(i,j)∈ER
p
p
UTkt = uu dij yijk ∀k ∈ K, ∀p ∈ P, ∀t ∈ T , (10)
(i,j)∈ER
p
p
p
LTkt + UTkt + tij xijkt ≤ Tmax ∀k ∈ K, ∀t ∈ T , (11)
p∈P p∈P p∈P (i,j)∈E
p
p
xjhkt ≤ M / V [S],
xijkt ∀S ⊆ E; {1, n} ∈
(j,h)∈S / [S],j∈V [S]
i∈V
∀k ∈ K, ∀p ∈ P, ∀t ∈ T , (12)

1 2
x1jkt ≥ xnjkt ∀k ∈ K, ∀t ∈ T , (13)
j∈V [E] j∈V [E]
p
p+1
xnjkt ≥ xnjkt ∀p ∈ {2, 3, . . . , |P| − 1}, ∀k ∈ K, ∀t ∈ T , (14)
j∈V [E] j∈V [E]
p
x1jkt = ukt ∀k ∈ K, ∀p = 1, ∀t ∈ T , (15)
(1, j) ∈ E
j ∈ V [E]\{1, n}
p
xjnkt = ukt ∀k ∈ K, ∀p = 1, ∀t ∈ T , (16)
(j, n) ∈ E
j ∈ V [E]\{1, n}
p
xnjkt ≤ ukt ∀k ∈ K, ∀p ∈ P\{1}, ∀t ∈ T , (17)
(n, j) ∈ E
j ∈ V [E]\{1, n}
p
xjnkt ≤ ukt ∀k ∈ K, ∀p ∈ P\{1}, ∀t ∈ T , (18)
(j, n) ∈ E
j ∈ V [E]\{1, n}
p p
xijkt ∈ Z + , yijkt ∈ {0, 1}, ukt ∈ {0, 1} ∀(i, j) ∈ E, ∀k ∈ K, ∀p ∈ P, ∀t ∈ T , (19)
p p
LTkt , UTkt ≥ 0 ∀(i, j) ∈ E, ∀k ∈ K, ∀p ∈ P∀t ∈ T . (20)
Objective function (1) minimizes the total waste collection and transportation
cost which is composed of the traversing costs and usage costs of vehicles. Objective
function (2) minimizes the total pollution emission released by vehicles. Objective
function (3) maximized the number of hired labor in all periods. Constraint (4)
represents the flow balance in the network for vehicles. Constraint (5) guarantees
that required edges are served only by one vehicle. Constraint (6) shows the capacity
limitation of vehicles. Constraint (7) states that the required edge can be served by
the vehicle traversing it; i.e., vehicles may just traverse an edge without serving it.
Constraint (8) indicates that vehicles can be employed if only their usage costs are
paid. Constraints (9) and (10) calculate the sum of loading and unloading times for
vehicles in each period and trip, respectively. Constraint (11) expresses the maximum
available service time for vehicles in each period. Constraint (12) eliminates the
potential sub-tours for vehicles in each period and trip. Constraints (13) and (14)
form the sequence of vehicles’ trips from 1 to P. Constraints (15) and (16) guarantee
that the first trip starts from the depot and ends at the disposal site, respectively.
Constraints (17) and (18) indicate that if more than one trip is required, the next
trips (second, third, fourth, etc.) start from the disposal site and end at there again.
Constraints (19) and (20) display the domain of the variables.
4 Methodology
This section provides the proposed solution techniques to validate, solve and analyze
the proposed problem. Accordingly, MOIWOA-MOSA is designed as a hybrid multi-
objective optimization algorithm based on MOIWOA and MOSA. Moreover, the ε-
constraint method is applied to solve the model exactly using CPLEX solver/GAMS
software.
4.1 Heuristic for Generating Initial Solutions
In this section, a heuristic algorithm is developed to generate the initial solutions of

the problem. The main implementation steps are given as follows:
Step 1: At the starting point of the algorithm, choose the first planning period.
Otherwise, if there is any planning period remaining, choose the next planning
period among the existing periods and go to Step 2. Otherwise, go to Step 6.
Step 2: Choose a vehicle randomly among the available vehicles to begin the first
trip from the depot and go to Step 3.
Step 3: Among the existing required edges, consider all the required edges that can
be added to the trip based on the capacity and available service time limitations
of the vehicle, then choose one edge with the minimum distance from the depot
and go to Step 5. Otherwise, if there is no eligible required edge satisfying one of
these limitations, move to the disposal site.
Step 4: If the maximum available service time allows the vehicle to construct
another trip, begin the next trip from the disposal site and go to Step 3, otherwise,
go to Step 5.
Step 5: If there is a remaining required edge to be served, go to Step 3. Otherwise,
choose the shortest path to the disposal site and then go to Step 1.
Step 6: Move to the depot and complete the tour.
Step 7: Report the obtained solution.
4.2 MOSA
Simulated annealing (SA) is a local search algorithm with a great ability to prevent
from being trapped in local optima. This algorithm is very effective for solving
nonconvex or discrete problems. Therefore, SA is employed to solve integer program-
ming problems efficiently (Glover and Kochenberger 2006). From its introduction
to the present, SA has demonstrated a high efficiency in solving large combinatorial
optimization problems (Kubotani and Yoshimura 2003). Furthermore, the simplicity
of implementation, convergence and hill-climbing for avoiding local optima are taken
into account as the main superiority factors.
Fig. 2 Pseudo-code of
MOSA (Nam and Park 2000)
Accordingly, MOSA generates non-dominated solutions by using a simple prob-

ability function that tries to provide the solutions constituting the Pareto optimal
front (Tirkolaee et al., 2020). Based on this probability function, the total space
of objective is covered uniformly generating as many possible non-dominated and
well-dispersed solutions. These features have made MOSA a fast reliable algorithm
compared to the other existing multi-objective algorithms with a wide range of appli-
cations. The mechanism of the suggested MOSA is adapted from Nam and Park
(2000). The superiority of MOSA against other EAs is that there is no need to
employ high memory for keeping the population information. The pseudo-code of
the suggested MOSA algorithm is given in Fig. 2.
It should be noted that MOSA is applied to each initial solution generated by
the heuristic algorithm in Sect. 4.1. Finally, we can provide a set of the high-quality
initial solutions.
4.3 MOIWO
The Invasive Weed Optimization algorithm (IWO) is an evolutionary and intelligent

algorithm which was inspired by the processes of propagation, survival and weed
adaptability. This algorithm was introduced by Mehrabian and Lucas (2006). Weed
is a phenomenon that aims to provide the best environment for life and adapt to
environmental conditions rapidly and at the same time, it is resistant to changes.
At the beginning, the weed searches for a huge number of children, which can
accordingly increase the quantity and available coverage of the environment (search
behavior), then it continues to grow in competition with increasing quality to grow
(greedy behavior) due to the capacity constraint.
i. The initial population (a given number of seeds) is generated and dispersed in the
first stage. At the second stage, the dispersed seeds generate seeds themselves
after growing and becoming a plant in terms of fitness and competence. In the
third stage, these seeds (childs) are scattered and grow near its parent. Eventually,
the second and third stages are repeated until the population reaches a given limit;
otherwise, the remaining plants will continue to live better and the remainder
will be ruined.
ii. The implementation of IWO has a suitable performance for all test functions.
To simulate the colonizing behavior of weeds, the main principal features of the
process are regarded as follows (Goli et al. 2019b):
iii. A certain number of seeds is spread over the search space.
iv. Each seed grows to a flowering plant and generates childs (seeds) with respect
to the fitness value.
v. The generated seeds are randomly dispersed over the search space and become
new plants.
vi. This process is continued until the maximum number of plants is reached. There-
after, only the plants with lower fitness values can survive and generate new
seeds and the others are ruined. Now, this process continues until reaching the
maximum number of iterations. Finally, the plant with the best fitness value is
regarded as the nearest solution to the optimum.
Now, MOIWOA is a multi-objective variant of the IWO algorithm that was
first introduced by Nikoofard et al. (2012) and has been extensively employed by
researchers (Goli et al. 2019a, b).
The execution steps of MOIWOA is given as follows:
Step 1: Generating initial seeds
The initial solutions are called initial seeds. These initial seeds are the output of
the MOSA (Sect. 4.2) applied to the solutions generated by the heuristic algorithm
(Sect. 4.1).
Step 2: Seed reproduction
In this phase, a member of the population is allowed to generate seeds regarding
its own and colony’s lowest and highest fitness values according to Eq. (21):
f − fworst
S = Smin + (Smax − Smin ) . (21)
fbest − fworst
where S min and S max denote the minimum and maximum number of seeds, respec-
tively. Furthermore, fworst and fbest represent the worst and best fitness value, respec-
tively. After specifying the number of seeds, reproduction is performed by executing
the following sub-steps:
Sub-step 1: Exchange the routing plan for a given trip of the vehicle by another
vehicle in a period.
Sub-step 2: Select two different trips of two vehicles randomly. If there is a
common edge/edges between these trips, select one of them randomly, then divide
these trips into two parts. The first part of the first trip is combined with the second
part of the second trip and the other two components of the trips are combined.
Consequently, two new trips are generated.
Sub-step 3: Select two trips of a vehicle randomly. If there are two common edges
among them, exchange the sequences of the common edges.
Sub-step 4: Select an edge in a trip of vehicle randomly and reverse its direction.
Sub-step 5: Select a part of a trip randomly and reverse the direction of the included
edges.
Eventually, all the obtained solutions are checked in terms of feasibility and only
the feasible solutions are kept in the population.
Step 3: Competitive exclusion
The number of plants will reach its maximum value by fast reproduction after
several iterations, nevertheless, it is expected that the desired plants are re-generated
more than the undesired ones. Moreover, it is critical to reach the maximum number
of plants (pmax ) to efficiently handle the speed of MOIWOA (Goli et al. 2019a).
In this stage, the solutions are sorted based on the non-dominate sorting technique
that is described in the following. The weaker solutions are removed to keep pmax
solutions for the next step.
Step 4: Non-dominate sorting
To rank a set of solutions, the following parameters are quantified for each solution:
np as the number of solutions and S p as the set of solutions dominated by solution p.
According to this technique, the value of crowding distances is the basic factor
for sorting the solutions, which is calculated by Eq. (22):
m m
Ij+1 Ij−1
fm − fm
dIjm = dIjm + ∀m. (22)
fm − fmmin
max
where fmmax and fmmin stand for the maximum and minimum value of mth objective.
m m
Ij+1 Ij−1
Here, fm and fm show the value of mth objective for the next and previous
solution of jth solution according to the sorted value of mth objective function,
respectively.
Finally, the pseudo-code of the suggested MOIWOA is depicted in Fig. 3.
4.4 MOIWOA-MOSA
The hybrid algorithm of this study works by generating high-quality initial solutions
using MOSA to be incorporated into the MOIWOA. Finally, the best possible solution
will be found by MOIWOA.
Fig. 3 Pseudo-code of proposed MOIWOA (Goli et al. 2019a)
4.5 Taguchi Design Method
As was clarified in the previous sections, MOSA and MOIWOA have several
adjustable parameters that can affect the performance and final results. Hence,
providing a suitable combination of these parameters can highly enhance the
performance and efficiency of the algorithms.
Two approaches for performing the test are known (Taguchi et al. 2005): (i)
Standard analysis of variance (ANOVA) and (ii) Signal to noise (S/N) ratio. The
amount of S/N indicates the amount of dispersion around a certain value, or how the
obtained solutions have altered over different experiments. To reduce the dispersion
of the objective functions as far as possible, S/N ratio plays an important role. This
ratio stands for noise factors along with controllable parameters. By investigating
among different Taguchi tables, the relevant table to L 27 is taken into account, which
is implemented using Minitab statistical software.
After applying the test on the input data in Tables 1 and 2, the obtained optimal
values are given for MOSA and MOIWOA, respectively.
4.6 ε-Constraint Approach
The ε-constraint approach is known as an effective way to deal with multi-objective

optimization problems in the literature, which can generate Pareto solutions. Here,
Table 1 Parameter of the MOSA algorithm

Parameters Initial values Optimal values
1 2 3
Maximum number of iterations (A) 100 150 200 200
Initial temperature (B) 700 800 900 800
Temperature reduction rate (C) 0.9 0.95 0.99 0.99
Boltzman constant (D) 30 50 70 50
Table 2 Parameter of the MOIWOA

Parameters Initial values Optimal values
1 2 3
Initial plant (N) 10 20 30 30
Minimum seed (S min ) 7 9 12 7
Maximum seed (S max ) 100 150 200 200
Maximum plant (Pmax ) 50 100 150 150
Maximum iteration (MI) 100 200 300 300
the Pareto front can be achieved by the ε-constraint approach based on Eq. (23)
(Bérubé et al. 2009):
minimizef1 (X )
subject to
x ∈ X,
f2 (X ) ≤ ε2 ,
...
fn (X ) ≤ εm . (23)
The execution steps of the ε-constraint approach are given in Fig. 4.

According to the offered ε-constraint approach, the first objective function is taken
into account as the main objective function and the second and third objective func-
tions as sub-objective functions. Now, the final formulation relates to the proposed
model of the study is displayed in Eq. (24):
minimizef1 (X )
subject to
x ∈ X,
f2 (X ) ≤ ε2 ,
f3 (X ) ≤ ε3 . (24)
Fig. 4 Execution steps of the proposed ε-constraint
5 Computational Results
This section provides the model validation and comparisons between the proposed
solution techniques through solving 10 problem instances in Babaee Tirkolaee et al.
(2019). The parameters took value using a uniform distribution. For example, the
demand parameter follows a uniform distribution uniform (1, 4). Moreover, the input
information of these problem instances is represented by Table 3.
The obtained results by the proposed solution techniques are shown in Table 4.
With respect to the multi-objective essence of the suggested model, MOSA-
MOIWOA and the ε-constraint approach are compared using the mean of ideal
distance (MID), number of solutions (NOS) and CPU time metrics. MID is employed
to calculate the mean distance of the Pareto solutions from the ideal solution or the
origin of the coordinates. Based on Eq. (25), the lower value of MID indicate more
efficiency for the solution methods.
Table 3 Information of the problem instances

Problem No. of nodes No. of edges No. of required No. of periods No. of vehicles
edges
P1 7 13 8 1 2
P2 9 18 14 2 3
P3 12 28 20 3 4
P4 13 38 26 4 5
P5 14 45 33 5 6
P6 16 55 40 7 7
P7 18 75 54 10 8
P8 20 100 70 12 10
P9 22 150 110 15 12
P10 25 200 150 20 15
Table 4 Comparison between MOIWOA and ε-constraint

Problem ε-constraint (EPC) MOSA-MOIWOA
MID NOS CPU time MID NOS CPU time
P1 1029 7 49.3 1068 13 3.4
P2 1179 14 127.4 1259 24 5.6
P3 1797 25 289.9 1904 53 12.9
P4 2468 29 724.3 2694 79 17.7
P5 3079 30 1423.5 3122 127 23.6
P6 4849 30 2864.2 5091 136 31.5
P7 6168 34 3654.2 6418 150 42.5
P8 7285 35 5843.45 7380 150 53.9
P9 8193 39 10,000 8220 150 64.8
P10 –a – – 9793 150 79.3
Average 4002.22 27 2775.14 4695.49 103.2 33.52
a No solution found
1 N 3
2
MID = fsol,m (25)
NOS s=1 m=1
where fsol,m represents the mth objective value for the sth solution.
Moreover, it should be noted that a run time limitation of 10,000 s is regarded to
evaluate the performance of the solution techniques.
Figures 5, 6 and 7 depict the comparisons for the metrics between solution
techniques in different problems.
As can be seen in Table 4, the ε-constraint approach couldn’t solve P10 within
the run time limitation of 1000 s due to the complexity of the proposed problem in
large sizes and the low efficiency of CPLEX solver to solve it. On the other hand, the
suggested MOSA-MOIWOA solved the problem instances in less than 80 s. In fact,
the comparison of the average CPU time values for these two solution techniques
identifies that the MOSA-MOIWOA just spends about 1.2% of the CPU time required
7000
CPU time value (s)
6000
5000
4000
3000 EPC
2000 MOSA-MOIWOA
1000
0
1 2 3 4 5 6 7 8 9 10
Problems
Fig. 5 Comparisons of CPU time values

12000
10000
MID value
8000
6000 EPC
4000 MOSA-MOIWOA
2000
0
1 2 3 4 5 6 7 8 9 10
Problems
Fig. 6 Comparisons of MID values
160
140
120
NOS value
100
80
EPC
60
40 MOSA-MOIWOA
20
0
1 2 3 4 5 6 7 8 9 10
Problems
Fig. 7 Comparisons of NOS values
for the ε-constraint approach. Figure 5 represents this comparison trends for the CPU
time values.
Moreover, by analyzing the values obtained for the two MID and NOS metrics,
the quality of the final solutions can be evaluated. The lower MID and the more
NOS reflect a more efficient solution technique. Figure 6 displays that these two
solution techniques have a close behavior against each other in terms of MID value.
Furthermore, the comparison of NOS values in Fig. 7 demonstrates that the MOSA-
MOIWOA has a relatively higher efficiency in this index.
5.1 Sensitivity Analysis
This section provides a sensitivity analysis on P10 as the largest-sized problem

that is the closest one to the real-life scale. For this purpose, the parameter of the
maximum available service time (Tmax ) is studied under different change intervals
and the behavior of the objective functions is evaluated using MOSA-MOIWOA.
The obtained results are shown in Table 5 and Figs. 8, 9 and 10.
Table 5 Obtained results for the sensitivity analysis

Objective functions Parameter’s change interval
−20% −10% 0% +10% +20%
Total cost 258,968.28 255,257.37 244,127.56 228,654.59 208,819.67
Total pollution emission 1735.35 1985.39 2115.80 2462.81 2533.73
Total job opportunity 379 368 351 324 311
280000
Total cost value ($)
260000
240000
220000
200000
180000
-20% -10% 0% 10% 20%
Change interval
Fig. 8 Sensitivity analysis of the 1st objective function
3000
Total Pollution (kg)
2500
2000
1500
1000
-20% -10% 0% 10% 20%
Change interval
Fig. 9 Sensitivity analysis of the 2nd objective function

Total Job opportunities
400
350
300
250
-20% -10% 0% 10% 20%
Change interval
Fig. 10 Sensitivity analysis of the 3rd objective function

Table 6 Change percentages of the objective functions against parameter’s change intervals
Objective functions Parameter’s change interval
−20% −10% 0% +10% +20%
Total cost 6.079 4.559 – −6.338 −14.463
Total pollution emission −17.981 −6.164 – 16.401 19.753
Total job opportunity 7.977 4.843 – −7.692 −11.396
As can be seen in Table 5 and Figs. 8, 9, 10, all objective functions reflect remark-
able fluctuations against the changes of Tmax , but in various behavioral directions.
Accordingly, the 1st and 3rd objective functions decreased by the increase of Tmax .
This is due to the generated decrease in the usage costs of vehicles and this reduc-
tion leads to a significant decrease in the number of hired labor. However, the 2nd
objective directly increased by the increase of Tmax as it was expected due to more
usage of vehicles and less movement from the disposal site to the operational zone.
Table 6 represent different change percentages of the objective functions against
the four change intervals considered for Tmax . Accordingly, it is revealed that the
most increase is related to the 2nd objective function; i.e., total pollution emission,
where a 19.753% increase occurs for the 20% increase in Tmax . On the other hand, the
most decrease is again related to this objective function, where a −17.981% decrease
occurs for the 20% decrease in Tmax . Overall, increasing the parameter will improve
the 1st and 3rd objective functions but it leads to a worse value for the 2nd objective
function.
Based on the obtained results, managers can investigate these trade-offs to find
the optimal level of the resources to be provided and allocated in the waste collection
system. In fact, sensitivity analysis is a useful tool to identify the optimal policy in
a real-world situation.
6 Conclusions and Outlook
This study designed an effective methodology to determine the optimal plans for
waste-collection routes and the required number of vehicles. For this purpose, a novel
MILP model was developed to formulate a multi-objective multi-trip sustainable
PCARP. The objectives of the problem are to concurrently minimize the total cost,
total pollution emission and total job opportunities, respectively. To validate the
proposed mathematical model and cope with its tri-objectiveness, the ε-constraint
approach was employed using CPLEX solver/GAMS software. Moreover, to solve
the problem efficiently in large scales, a multi-objective meta-heuristic algorithm,
namely, MOSA-MOIWOA was then developed. To improve the efficiency of the
algorithm, the Taguchi design method was also applied to set the parameters. The
performance of the MOSA-MOIWOA was then evaluated using various problem
instances and three metrics of MID, NOS and CPU time in comparison with the
ε-constraint method. The obtained results proved that the MOSA-MOIWOA can
provide high-quality solutions within a much shorter CPU time. Moreover, the MID
values were close to each other but the proposed algorithm could yield much better
NOS value. Eventually, a sensitivity analysis was implemented on the maximum
available service time of vehicles to assess the behavior of the objective functions
and offer useful managerial insights and decision aids. It was obvious that the total
cost and total job opportunity have indirect relations with this parameter, but the total
pollution emission reflects a direct relation.
According to the main limitations of the study, following suggestions are raised
for future research directions:
(i) Considering multiple depots and disposal sites within the network,
(ii) Taking into account the locational decisions for depots and disposal sites,
(iii) Extending the MSW management network by adding more facilities such as
incinerators and recycling/recovery sites,
(iv) Developing the social sustainability by minimizing the number of garbage
pick-up to reduce social discomfort of odor presence time,
(v) Studying the effect of milk-run strategy in the problem,
(vi) Applying uncertainty techniques such as fuzzy theory, stochastic program-
ming, grey systems and robust optimization,
(vii) Employing other multi-objective algorithms like non-dominated sorting
genetic algorithm III (NSGA-III).
References
Babaee Tirkolaee E, Goli A, Pahlevan M, Malekalipour Kordestanizadeh R (2019) A robust bi-

objective multi-trip periodic capacitated arc routing problem for urban waste collection using a
multi-objective invasive weed optimization. Waste Manage Res 37(11):1089–1101
Babaee Tirkolaee E, Mahdavi I, Seyyed Esfahani MM, Weber GW (2020) A hybrid augmented ant
colony optimization for the multi-trip capacitated arc routing problem under fuzzy demands for
urban solid waste management. Waste Manage Res 38(2):156–172
Bautista J, Fernández E, Pereira J (2008) Solving an urban waste collection problem using ants
heuristics. Comput Oper Res 35:3020–3033
Bérubé JF, Gendreau M, Potvin JY (2009) An exact ǫ-constraint method for bi-objective combina-
torial optimization problems: application to the Traveling Salesman Problem with Profits. Eur J
Oper Res 194(1):39–50
Budhiarta I, Siwar C, Basri H (2012) Current status of municipal solid waste generation in Malaysia.
Int J Adv Sci Eng Inf Technol 2(2):129–134
Çodur MK, Yılmaz M (2020) A time-dependent hierarchical Chinese postman problem. Central
Eur J Oper Res 1–30
EEA E (2008) Better management of municipal waste will reduce greenhouse gas emissions. Support
document to EEA Briefing, 1
Filipiak KA, Abdel-Malek L, Hsieh HN, Meegoda JN (2009) Optimization of municipal solid
waste collection system: case study. Pract Periodical Hazard Toxic, Radioactive Waste Manag
13(3):210–216
Gapminder (2019) United Nations, Department of Economic and Social Affairs, Population
Division. World Population Prospects. https://www.gapminder.org/data/documentation/gd003/.
Accessed 15 Apr 2019
Glover FW, Kochenberger GA (eds) (2006) Handbook of metaheuristics, vol 57. Springer Science
and Business Media
Golden BL, Wong RT (1981) Capacitated arc routing problems. Networks 11(3):305–315
Goli A, Tirkolaee EB, Malmir B, Bian GB, Sangaiah AK (2019a) A multi-objective invasive
weed optimization algorithm for robust aggregate production planning under uncertain seasonal
demand. Computing 101(6):499–529
Goli A, Zare HK, Tavakkoli-Moghaddam R, Sadeghieh A (2019b) Application of robust opti-
mization for a product portfolio problem using an invasive weed optimization algorithm. Numer
Algebra, Control Optim 9(2):187
Inghels D, Dullaert W, Vigo D (2016) A service network design model for multimodal municipal
solid waste transport. Eur J Oper Res 254(1):68–79
Kubotani H, Yoshimura K (2003) Performance evaluation of acceptance probability functions for
multi-objective SA. Comput Oper Res 30(3):427–442
Lin S, Kernighan BW (1973) An effective heuristic algorithm for the traveling-salesman problem.
Oper Res 21(2):498–516
Lou CX, Shuai J, Luo L, Li H (2020) Optimal transportation planning of classified domestic garbage
based on map distance. J Environ Manage 254:109781
Mehrabian AR, Lucas C (2006) A novel numerical optimization algorithm inspired from weed
colonization. Ecol Inf 1(4):355–366
Mourão MC, Pinto LS (2017) An updated annotated bibliography on arc routing problems. Networks
70(3):144–194
Nam D, Park CH (2000) Multiobjective simulated annealing: a comparative study to evolutionary
algorithms. Int J Fuzzy Syst 2(2):87–97
Nikoofard AH, Hajimirsadeghi H, Rahimi-Kian A, Lucas C (2012) Multiobjective invasive weed
optimization: application to analysis of Pareto improvement models in electricity markets. Appl
Soft Comput 12(1):100–112
Paletta G, Triki C (2004) Solving the asymmetric traveling salesman problem with periodic
constraints. Netw Int J 44(1):31–37
Pearn WL (1994) Solvable cases of the k-person Chinese postman problem. Oper Res Lett
16(4):241–244
Rodrigues AM, Soeiro Ferreira J (2015) Waste collection routing—limited multiple landfills and
heterogeneous fleet. Networks 65(2):155–165
Statista (2019) https://www.statista.com/markets/408/topic/435/waste-management/. Accessed 15
Apr 2019
Taguchi G, Chowdhury S, Wu Y (2005) Taguchi’s quality engineering handbook, vol 1736. Wiley,
Hoboken, NJ
Tirkolaee EB, Alinaghian M, Bakhshi Sasi M, Seyyed Esfahani MM (2016) Solving a robust
capacitated arc routing problem using a hybrid simulated annealing algorithm: a waste collection
application. J Ind Eng Manag Stud 3(1):61–76
Tirkolaee EB, Mahdavi I, Esfahani MMS (2018a) A robust periodic capacitated arc routing problem
for urban waste collection considering drivers and crew’s working time. Waste Manag 76:138–146
Tirkolaee EB, Hosseinabadi A, Soltani M, Sangaiah A, Wang J (2018b) A hybrid genetic algorithm
for multi-trip green capacitated arc routing problem in the scope of urban services. Sustainability
10(5):1366
Tirkolaee EB, Alinaghian M, Hosseinabadi AAR, Sasi MB, Sangaiah AK (2019) An improved
ant colony optimization for the multi-trip capacitated arc routing problem. Comput Electr Eng
77:457–470
Toronto City Council (2019) City Budget. Report, City Government Toronto. Available
at: https://www.toronto.ca/city-government/budgetfinances/city-budget/previous-budgets/2018-
citybudget/. Accessed 15 Apr 2019
Tirkolaee EB, Goli A, Faridnia A, Soltani M, Weber GW (2020) Multi-objective optimization for
the reliable pollution-routing problem with cross-dock selection using Pareto-based algorithms. J
Clean Prod 276(10):122927
A Perishable Product Sustainable Supply
Chain Network Design Problem
with Lead Time and Customer
Satisfaction using a Hybrid
Whale-Genetic Algorithm
Alireza Goli, Erfan Babaee Tirkolaee, and Gerhard-Wilhelm Weber
Abstract In this research, designing a sustainable, multi-level, multi-products, and

multi-period closed-loop supply chain network for perishable products is addressed.
For this purpose, an integrated mathematical model is proposed. The main objectives
are to minimize the production, distribution, and customer satisfaction related costs,
minimize total CO2 emissions, and maximize social responsibility. The contributions
of this research include considering lead time for production and delivering perishable
products in the supply chain network design problem, and proposing a novel hybrid
algorithm based on whale optimization algorithm (WOA) and genetic algorithm
(GA). To solve the problem and optimize the mathematical model, the proposed
hybrid algorithm is implemented on several test problems in different sizes. The
obtained results are compared with augmented epsilon constraint in order to evaluate
the performance of the proposed algorithm. The results indicate that the proposed
algorithm provides Pareto solutions with acceptable quality and diversity.
Keywords Closed-loop supply chain · Sustainable supply chain · Whale

optimization algorithm · Genetic algorithm
1 Introduction
Supply chain management consists of all movements as raw material storage, inven-
tory in hand, and transportation of finished products from the source point to the
A. Goli (B)
Department of Industrial Engineering, Yazd University, Yazd, Iran
e-mail: a.goli@stu.yazd.ac.ir
E. B. Tirkolaee
Department of Industrial Engineering, Istinye University, Istanbul, Turkey
e-mail: e.babaee@ustmb.ac.ir
G.-W. Weber
Faculty of Engineering Management, Poznań University of Technology, Poznań, Poland
e-mail: gerhard.weber@put.poznan.pl
Institute of Applied Mathematics, Middle East Technical University, Ankara, Turkey

100 A. Goli et al.
final consumption point. There are three strategic, tactical, operational, and oper-
ational levels in supply chain management. From the aspect of the supply chain
design, determining the optimal location of facilities and the number of materials or
products distributed between each facilitation in order to meet customers’ demands
(Mardan et al. 2019; Burgess et al. 2017).
Concerns have been raised about the effects of human works and human traf-
ficking in recent years, and this has led to greater sustainability across the chain.
Supply chain sustainability is one of the newest and most applicable concepts in
supply chain management. Carter and Rogers (2008) presented a comprehensive
conceptual framework for sustainable supply chain management. The overall dimen-
sions of sustainability include economic, environmental, and social responsibility.
The sustainable supply chain is a developed form of traditional supply chain manage-
ment that tries to provide the economic, social, and environmental requirements of
the supply chain through the flow of material between different levels. Sustainable
supply chain management provides a competitive advantage for partners. In this
regard, designing a sustainable supply chain network is a complex process in which
all economic, social, and environmental aspects are used simultaneously (Mardani
et al. 2020).
Due to the importance of the presented topic, in this research, a sustainable multi-
product closed-loop supply chain network designed is addressed. The levels of the
studied closed-loop chain include suppliers, manufacturers, distributors, customers,
collection centers, and disposal centers. For this supply chain, all three dimensions
of sustainability are optimized simultaneously. In other words, the objectives are
to minimize total costs (economic objective), minimize total environmental pollu-
tion (environmental objective), and maximize total job opportunities (social objec-
tive). Moreover, customer satisfaction and delivery lead-time are considered. In this
regard, a multi-objective mathematical model is proposed to find optimal sustainable
closed-loop supply chain network. In order to optimize this model, a novel hybrid
algorithm based on whale optimization algorithm (WOA) and genetic algorithm
(GA) is proposed which is called HWOAGA.
2 Literature Review
Babazadeh et al. (2017) have designed a biodiesel reproducible supply chain network.
In their proposed model, minimizing the total cost of the biodiesel supply chain and
minimizing its environmental impacts, is modeled under uncertainty in the form
of a multi-objective contingency programming model. In the problem presented in
this paper, strategic and operational decisions such as location decisions, allocation,
facility deployment capacity, inventory turnover, etc. are determined under realistic
assumptions. Zhuo and Wei (2017) presented a nonlinear integer for a location and
inventory problem in a closed-loop supply chain. In this study, the quantity of demand
is presented as an exponential relation of the price determined. To solve this model, an
A Perishable Product Sustainable Supply Chain Network … 101
innovative heuristic method is applied to determine the inventory level and ordering
period.
Burgess et al. (2017), studied the integration of the forward and reverse logistics
in order to design a closed-loop supply chain with considering demand uncertainty.
For this purpose, in addition to minimizing the total costs, the reduction of CO2
emissions was also optimized. An ant colony algorithm has been developed to solve
this model.
Cao et al. (2018) proposed a mathematical model for a sustainable supply chain
network design problem by considering perishable multi-products consumption. In
this research, the authors presented tested their large-scale model using a genetic
algorithm. Rad and Nahavandi (2018) presented an integrated multi-objective model
to design a green closed-loop supply chain network considering the quality of prod-
ucts and discounts in the supply chain. The objectives of this model include reducing
costs, reducing environmental pollution, and increasing customer satisfaction. The
results of this study show that reducing environmental pollution will lead to good
results in terms of customer satisfaction. Ghelichi et al. (2018) designed a green
supply chain for the production of diesel cars. In this regard, a scenario-based multi-
objective mathematical model is proposed and optimized with a stochastic program-
ming approach. The objectives of this model include reducing costs and reducing
environmental pollution. The epsilon constraint method has been developed to solve
it under various scenarios.
Liang and Quesada (2018) studied the fuel supply chain. They have identified and
optimized the reduction of energy costs and greenhouse gas emissions as the most
important goals of this supply chain. The research is being implemented in Japan
and has provided an ideal solution that strikes a balance between energy costs and
environmental pollution. Fard and Hajaghaei (2018) propose a three-level decision
model to formulate the problem of designing a forward/reverse supply chain. They
have used several meta-algorithms including taboo search (TS), variable neighbor-
hood search (VNS), particle swarm optimization (PSO), and water wave algorithm
(WWO) to solve their proposed model. Mardan et al. (2019) designed a multi-period
and multi-product supply chain. In this model, two economic and environmental
goals are examined. Initially, the two goals have become a goal comprehensively.
Then they used the Bender analysis method to optimize it. The results show that the
new model consumes 13% of total chain costs. Niranjan et al. (2019) investigated
and optimized a multi-channel closed-loop supply chain. In this supply chain, the
coverage radius is intended to meet customer demand. This supply chain is opti-
mized under the two objectives of total cost and total supply chain contamination.
The optimized CLPEX solver is used. The results indicate the efficiency of the chain
structure studied.
102 A. Goli et al.
Yavari and Grally (2019) proposed a closed-loop green supply chain optimization
model under uncertainty. In this model demand is uncertain and a robust optimization
approach is used to meet it. In this chain the products are perishable. Two objectives
of total cost and total pollution are optimized in this chain. A new innovative method
called YAG has been used for this optimization. The results show a 1.65% improve-
ment in this method over other methods. Goli et al. (2019) designed a closed-loop
supply chain network with financial flow consideration. In this research, a two-
objective mathematical model is presented. The first is to increase net supply chain
assets and the second is to increase job creation as the supply chain social responsi-
bility. The Epsilon constraint method and gray wolf optimization (GWO) are used
to design this supply chain network. The results show that taking into account the
financial flow can bring more profits to the whole supply chain.
Recently, Rabbani et al. (2020) proposed a new mathematical model for location-
allocation in a sustainable supply chain. In this research, different technology levels
for the transportation fleet are considered. Moreover, Hybrid Robust Possibilistic
Programming, is applied and a case study is solved by the improved augmented
epsilon constraint and some sensitivity analysis is performed.
Finally, the most important contributed researches are reviewed in Table 1.
Based on the articles reviewed in Table 1, it is clear that in previous research,
several methods have been used to optimize the supply chain network design problem.
Among the methods used, the genetic algorithm has been presented as an efficient
method. However, none of them used the WOA which is one of the novel meta-
heuristic algorithms. Therefore, the main innovation of this research can be consid-
ered in the development of a new meta-heuristic algorithm based on hybrid whale
optimization and genetic algorithms to solve the closed-loop supply chain network
design problem.
3 Proposed Mathematical Model
In this research, a sustainable supply chain for perishable products is investigated.

Accordingly, economic, environmental, and social aspects are considered as inde-
pendent objective functions. In the studied supply chain, essential raw materials
are supplied from different suppliers. After production, products are sent to different
distribution centers (DCs) and they store them after inspection and distribute at appro-
priate times between customers. The returned products are shipped to the hybrid
collection and disposing centers. The reusable products are shipped to production
plants and the remaining are disposed. The structure of this supply chain is shown
in Fig. 1.
The distributor has the ability to store goods and deliver in the coming periods.
This condition causes a percentage of products in the distributor’s warehouse to be
corrupted and needs to be returned to the plants. The manufacturing units process
the returned products from distribution centers and hybrid centers, then return them
A Perishable Product Sustainable Supply Chain Network …
Table 1 Brief review and comparison of the literature
Author(s) Year Location Allocation Forward Reverse Multi-commodity Multi Environmental Social Methodology
logistic logistic period aspect aspect
√ √ √ √
Babazadeh 2017 Stochastic
et al. (2017) programing
√ √ √ √ √
Zhuo and 2017 Heuristic
Wei (2017)
√ √ √ √ √
Burgess et al. 2017 Ant colony
(2017) algorithm
√ √ √ √
Cao et al. 2017 Genetic algorithm
(2018)
√ √ √ √
Rad and 2018 Exact solution
Nahavandi
(2018)
√ √ √ √
Ghelichi 2018 Epsilon constraint
et al. (2018)
√ √ √ √
Liang and 2018 Exact solution
Quesada
(2018)
√ √ √
Fard and 2019 VNS-PSO-WWO
Hajiaghaeii
(2018)
√ √ √ √ √
Mardan et al. 2019 Benders
(2019) decomposition
√ √ √ √
Niranjan 2019 Exact solution
et al. (2019)
(continued)
103
104
Table 1 (continued)
Author(s) Year Location Allocation Forward Reverse Multi-commodity Multi Environmental Social Methodology
logistic logistic period aspect aspect
√ √ √ √
Yavari and 2019 YAG
Geraeli
(2019)
√ √ √ √ √
Goli et al. 2019 GWO
(2019)
√ √ √ √
Rabbani 2020 augmented epsilon
et al. (2020) -constraint
√ √ √ √ √ √ √ √
This 2020 Hybrid WOA-GA
research
A. Goli et al.
Fig. 1 Structure of studied supply chain
to the distribution to be sold to customers. Other returned products are considered as

waste products and are disposed. In order to meet the demand of each customer in a
period, the supply chain should provide this amount in specified lead time. Moreover,
customer satisfaction is determined by the rare of receiving products in each period.
Customers need to get their products at a fixed rate and the fluctuation of receiving
products has a distinct cost to the supply chain.
Other assumptions of the mathematical model are as follows:
I. The supply chain has 5 main levels, including supplier, manufacturer, distri-
bution centers, customers, and hybrid collection and disposing centers.
II. Different transport systems are considered each level of the supply chain with
fixed and variable costs.
III. Only one transportation system can be used between each supplier and
distributor, as well as between each distributor and customer.
IV. The supply chain is required to meet the total demand of customers in specified
lead time.
V. There are several potential centers for the establishment of distribution centers,
one or more of which must be established.
VI. The number of facilities to be established is not known beforehand, but the
mathematical model should provide the best possible scenarios, given the
amount of demand and establish costs.
VII. The transportation system selection is used for sending final products from
plants to DCs and also return products from DCs to plants.
Now, the main components of the proposed mathematical model including sets
and indices, parameters and variables are listed as follows.
Indices
S Set of suppliers fixed locations (s = 1, 2, …, S),

I Set of plants fixed locations (i = 1, 2, …, I),
J Set of potential locations for distribution centers (j = 1, 2, …, J),
C Set of customers fixed locations (c = 1, 2, …, C),
(continued)
106 A. Goli et al.
(continued)
D Set of hybrid collection and dispose fixed locations (d = 1, 2, …, D),
P Set of products (p = 1, 2, …, P),
R Set of raw materials (r = 1, 2, …, R),
L Set of transportation systems (l = 1, 2, …, L),
T Set of time period index (t = 1, 2, …, T ).
Parameters
t
Dcp Demand of customer c for product p in period t,
t
RCcp Rate of returning product p from customer c in period t to hybrid collection/dispose
centers,
RHpt Rate of returning product p from hybrid collection/dispose centers to plants in period t,
LTcp Lead time to deliver customer c demand for product p,
t
SCsr Unit purchasing cost of raw material r from supplier s in period t,
MCip t Unit of production cost of product p at plant i in period t,
t
ICjp Unit cost of each inspection and recycling of product p at the distribution center j in
period t,
t
HCjP Unit maintenance cost of product p at distribution center j in period t,
FYjt Fixed cost of establishing the distribution center j in period t,
t
CSsr Capacity of supplier s for material r during period t,
CXit Production capacity at plant i in period t,
CYjt Capacity of the distribution center j in period t,
t
CSIsir Unit transportation cost of raw material r from supplier s to plant i in period t,
t
CIJijpl Unit transportation cost of product p from plant i to distribution center j in period t
with transport system l,
t
CJCjcpl Unit transportation cost of product p from distribution center j to customer c with
transport system l during period t,
t
CJIjipl Unit transportation cost of product p from distribution center j to plant i in period t
with transport system l,
t
CCDcdp Unit transportation cost of product p from customer c to hybrid center d in period t,
t
CDIdip Unit transportation cost of product p from hybrid center d to plant i in period t,
CTRtl Fixed cost of transporting material l during period t,
CUScpt Customer dissatisfaction cost for variable rate of receiving product p to customer c
during period t,
(continued)
(continued)
t
ESIsir Unit CO2 pollution emission resulting from the transfer of raw material r from the
supplier s to the plant i during period t,
t
EIJijpl Unit CO2 pollution emission resulting from the transfer of product p from plant i to
distribution center j in period t with transport system l,
t
EJIjipl Unit CO2 pollution emission resulting from the transfer of product p from distribution
center j to plant i during period t with transport system l,
t
EJCjcpl Unit CO2 pollution emission resulting from the transfer of product p from distribution
center j to customer c with transport system l during period t,
t
ECip Unit CO2 pollution emission resulting from the production of product p in plant i in
period t,
JClt Number of job creation by selecting transportation l during the period t,
nrp Coefficient of consumption of raw material r in product p,
mp Capacity occupation for producing product p,
Rp Return rate of product p from distribution centers to plants,
Rpp Percentage of processing product p from the returned products,
BM An arbitrary big number
Decision variables
t
QSIsir Amount of raw material r sent from the supplier s to plant i in period t,
QIipt Amount of product p produced in plant i in period t,
t
QIJijpl Amount of the product p sent from plant i to the distribution center j with the
transportation system l in period t,
IN Vjpt Inventory of product p in distribution center j at the end of period t,
t
QJCjcpl Amount of product p shipped from distribution center j to the customer c with the
transportation system l during the period t,
t
QJIjipl Amount of the product p sent from the distribution center j to the plant i with the
transport system l during the period t,
t
QCDcdp Amount of product p shipped from the customer c to the hybrid center d during the
period t,
t
QDIdip Amount of product p transferred from hybrid center d to plant i during the period t,
Yjt A binary variable is equal to 1 if at the point j in period t, the distribution center is
established,
Atijl A binary variable is equal to 1 if the transportation system l connects the plant i and
the distribution center j in period t,
t
Bjcl A binary variable is equal to 1 if the transportation system l connects the
distribution center j to client c in period t.
108 A. Goli et al.
Mathematical model equations

minimize Z1 = FYit (Yjt )
t j

t
+ (QSIsir + QS0irt ) SCsrt + QIipt MCipt
t s i r i p t

t t
+ QJCjcpl ICJp
t c j p l

+ IN Vjpt HCpt
t j p

t t
+ QSIsir CSIsir
t s i r

t t
+ QIJijpl CIJijpl
t i j p l

t t
+ QJCjcpl CJCjcpl
t j c p l

t t
+ QJIjipl CJIjipl
t i j p l

t t
+ QCDcdp CCDcdp
t c d p

t t
+ QDIdip CDIdip
t d i p

+ Atijl CTRtl + t
Bjcl CTRtl
t j i l t j c l

t t t−1
+ CUScpl QJCjcpl − QJCjcpl (1)
t j c p l

minimize Z2 = QIipt ECipt
i p t

t t
+ QSIsir ESIsir
t s i r

t t
+ QIJijpl EIJijpl
t i j p l

t t
+ QJCjcpl EJCjcpl
t j c p l

t t
+ QJIjipl EJIjipl (2)
t i j p l

maximizeZ3 = Atijl JClt + t
Bjcl JClt (3)
t j i l t j c l
subject to

t t t
nrp QIJijpl = QSIsir + nrp QJIjipl
j P l s j P l

t
+ nrp QDIdipl ∀i, r, t, (4)
d P l

IN Vjpt−1 + t
QIJijpl = IN Vjpt + t
QJCjcpl
i l c l

t
+ QJIjipl ∀j, p, t, (5)
i l

t t
QJCjcpl = Dcp ∀c, p, t, (6)
j l t ′ ≤t−LTcp

t t t
QCDcdp = Dcp RCcp ∀c, p, t, (7)
d

t t
QDIdip = QCDcdp RHpt ∀d , p, t, (8)
i t c

t
QSIsir ≤ CSsrt ∀s, r, t, (9)
i

t
mp QIJijpl ≤ CXit ∀i, t, (10)
j p l

IN Vjpt + t
QJCjcpl ≤ CYjt Yjt ∀j, t, (11)
p c p l

Atijl ≤ 1 ∀i, j, t, (12)
l

t
Bjcl ≤ 1 ∀j, c, t, (13)
l

Atijl ≤ t
QIJijpl ∀i, j, l, t, (14)
p

t t
Bjcl ≤ QJCjcpl ∀j, c, l, t, (15)
p

t t
QIJijpl + QJIjipl ≤ BMAtijl ∀i, j, l, t, (16)
p
110 A. Goli et al.

t t
QJCjcp ≤ BMBjcl ∀j, c, l, t, (17)
p

t
QJIjipl = Rp IN Vjpt ∀j, p, l, t, (18)
i
⎛ ⎞

t
QIJijpl = QIipt + Rpp ⎝ t ⎠
QJIjipl ∀i, p, l, t, (19)
j j
t
QSIsir , QIipt , QIJijpl
t
, IN Vjpt , QJCjcpl
t t
, QJIjipl t
, QCDcdp t
, QDIdip
≥ 0; Yjt , Atijl , Bjcl
t
∈ {0, 1}
∀i, j, p, l, t, s, c. (20)
Equation (1) presents the total costs of each chain period. These costs include
the fixed costs of establishing distribution centers and purchasing from suppliers,
production costs, operating costs at DCs, inventory costs at distribution centers, trans-
portation costs by various transportation systems in the supply chain, and customer
dissatisfaction cost. Equation (2) also minimizes the total amount of CO2 released
from product production and transportation along with different levels of the supply
chain. Equation (3) maximizes the social responsibility of the supply chain and it
consists of total created jobs in the supply chain.
Equation (4) shows that the amount of raw material imported to each plant in each
period is equal to the output of that plant in the same period. Equation (5) ensures
that for each of the products in each of the periods, the amount entered into each
distribution center as well as the remaining inventory of the previous period is equal
to the amount sent to customers and the rest of the inventory at the end of the period.
This equation is known as the balance of inventory. Equation (6) states that for each
product and in each period, the available amount in each distribution center must be
able to meet the demand for that product. This demand should be delivered in the
specified lead time for each product and each customer.
Equation (7) calculates the total received products in hybrid centers. Equation (8)
calculates the number of reusable products in hybrid centers which are shipped to
plants. Equation (9) ensures that the amount of each raw material sent from suppliers
does not exceed their capacity. Equation (10) states the constraint of the capacity
of materials in the factories as suppliers. Equation (11) states that the remaining
inventory at each distribution center should not exceed their capacity. Equations (12)
and (13) indicate that among all transportation systems, just one of them can be
selected for transformation from a specified orientation and specified destination.
Equations (14) and (15) point out that the transportation system is used when a
shipment is transformed from orientation to a destination. Equations (16) and (17)
state that members of a chain that does not have a connection no goods are also
sent. Equation (18) determines the amount of returned goods as a percentage of the
distributor’s inventory. Equation (19) points out that the amount of goods sent from
the plant to the distribution centers is equal to the amount produced at the plant and
is a percentage of returned products that are being redeveloped. The domain of the
variables is given in Eq. (20)
4 Solution Method (HWOAGA Algorithm)
Using novel meta-heuristic algorithm in optimizing NP-HARD problems are

addressed and suggested in different researches in the field of optimization (Tirko-
laee et al. 2019, 2020; Goli and Davoodi 2018; Davoodi and Goli 2019; Sangaiah
et al. 2020). Conside ring that the designed mathematical model is a multi-objective
model, multi-objective optimization methods should be used to solve this model. In
this regard, HWOAGA and epsilon constraint are used.
In the whales’ lifestyle, it is preferred to hunt small fishes near the water surface by
creating bubbles surrounding the prey. The WOA algorithm is one of the nature-based
and population-based optimization algorithms based on the Whale lifestyle that can
be used in a variety of contexts. This algorithm is consists of sieging, bubble attack,
and hunting. These methods have been designed into a mathematical formulation.
In WOA, in the sieging operator, different solutions are achieved by increasing
the value of a controllable parameter as a. By selecting random values for vector A
between −1 and +1, a search agent can be applied.
In bubble attach, first it calculates the distance between the wall located in the X*
and Y. Then, the position of the whale will be updated to converge walls in the best
possible location.
Hunting operator is applied by using |A| > 1 mode, while the best solution is chosen
when |A| < 1 to update the position of the search agents. The WOA algorithm has
the ability to choose between circular or spiral motion. Finally, the WOA algorithm
ends with satisfying the termination conditions.
The whales can identify and surround the hunting grounds. Since the optimal
design location in the search space is not known, by comparison, the algorithm
assumes that the best candidate for the present is either target hunting or near-optimal.
After the best search agent has been identified, other search agents try to update
their location to the best search agent (Mirjalili and Lewis 2016). This behavior is
expressed through Eqs. (21) and (22):
D = c.X ∗ (t) − X (t) (21)
X (t + 1) = X ∗ (t) − A.D (22)
In the above equations, t represents the iteration of the algorithm, C and A are the
coefficients, and X* is the best position obtained and X(t) is the current wall position.
It should be noted that the value of X* is updated in each iteration. Equations (23)-(24)
are employed to determine the values of A and C.
112 A. Goli et al.
A = 2a.r − a (23)
C = 2.r (24)
As indicated before, a is a controllable parameter that is reduced iteration by

iteration to reach zero at the final iteration (Mirjalili and Lewis 2016). Moreover, r
is a random vector that takes value in [0, 1].
In order to design the HWOAGA, Eqs. (21)–(24) is utilized to apply sieging,
bubble attack operators. Moreover, to have a better search in solution space, crossover
and mutation operators are being used. However, as the proposed mathematical model
is multi-objective, to rank the solutions and report best Pareto solutions, fast non-
dominate sorting which has been proposed by Deb (2000) is applied. The flowchart
of the proposed hybrid algorithm is represented in Fig. 2.
The defined chromosomes in this study follow two structures. In the first structure,
binary variables are defined and vectors with values between 0 and 1 are generated.
The cells with less than 0.5 value mean not-select/not-establish and if more than 0.5
means select/establish.
In the second structure, positive variables are determined. In this structure, each
cell will have a value between 0 and 1. This number specifies the percentage of
sending materials or final products from a source to a destination in the supply chain.
In order to meet the constraints on the flow of materials and products, the values in
this structure will be normalized.
For example, if the number of potential DCs is 3 and the number of customers is
5, an example of a chromosome designed according to Figs. 3 and 4 is shown.
According to Fig. 3, DC2 and DC3 are established among the three distribu-
tion centers. In this regard, only two of these distributors will deliver products to
customers. Therefore, the values of each column in structure 2 between DC2 and
DC3 are normalized. The interpretation of this chromosome is presented in Fig. 4.
According to Fig. 4, 52% of customer 1 demand is met by DC2 and the rest by
DC3. According to the proposed solution representation, all constraints related to
the material flow in the supply chain will be met. However, supply chain capacity
constraints may be violated. To address this, a penalty will be added to all three
objective functions if the facility capacity is violated.
5 Numerical Results
In order to evaluate the performance of HWOAGA, first, the algorithm parameters

are tuned by applying the Taguchi design of experiment method. Accordingly, three
values (three levels) are suggested for each parameter and then, the best value for
each parameter is obtained according to signal to noise index. The suggested and
best value for HWOAGA parameters are presented in Table 2. It should be noted
that the population size and maximum iterations are considered as a two related
Fig. 2 Flowchart of the proposed HWOAGA algorithm
Structure 1 0.41 0.72 0.93
Customer 1 Customer 2 Customer 3 Customer 4 Customer 5
Structure 2 DC1 0.61 0.29 0.43 0.27 0.35

DC2 0.45 0.73 0.28 0.34 0.19
DC3 0.35 0.91 0.73 0.58 0.39
Fig. 3 An example of proposed solution representation

114 A. Goli et al.
Structure 1 0 1 1
Customer 1 Customer 2 Customer 3 Customer 4 Customer 5

DC1 0 0 0 0 0
Structure 2
DC2 0.52 0.44 0.27 0.37 0.32
DC3 0.48 0.56 0.73 0.63 0.68
Fig. 4 Chromosome interpretation
Table 2 Taguchi results for HWOAGA

Parameters Levels Best value
1 2 3
WOA controllable parameter (a) 2 1.5 1 2
Crossover percentage (Pc) 0.7 0.8 0.9 0.8
Mutation percentage (Pm) 0.3 0.2 0.1 0.1
(Population size, Number of iteration) (200, 50) (100, 100) (50, 200) (200, 50)
parameters and the common suggested value are applied. The main reason is that
these parameters affect the solution time directly.
After setting HWOAGA parameters, the performance of this algorithm is assessed.
For this purpose, 10 test problems are generated in different dimensions using (Goli
et al. 2019). The used indices for comparing the solution method are as Eqs. (25)–(30).
n
i=1 (MID − Ci )2
SNS = (25)
n−1

Ci = f21 + f22 + f23 (26)

3

Maxspread = (Minfi − Maxfi )2 (27)
i=1
2 2 2
n f 1i −f 1best f 2i −f 2best f 3i −f 3best
i=1 f 1max min + max min + f 3max min
total −f 1total f 2total −f 2total total −f 3total
MID = (28)
n

n f1i (x)−f1ibest (x) f2i (x)−f2ibest (x) f3i (x)−f3ibest (x)
i=1 f1ibest (x)
+ f2ibest (x)
+ f3ibest (x)
RAS = (29)
n
n
i=1 d̄ − di
SM = (30)
(n − 1)d̄
The information on these test problems is described in Table 3. These test problems
are applied with HWOAGA and also Augmented Epsilon constraint (AEC) which
is proposed by Sangaiah et al. (2019). It should be noted that for each method of
solving time limitation is considered as 3600 s.
After implementing HWOAGA and AEC with different test problems, some
multi-objective indices are calculated and reported in Tables 4 and 5.
Table 3 Information on numerical examples

Test # # # # # # Raw # # Time
problems Suppliers Plants DC Hybrid Products materials Transportation periods
centers systems
1 5 2 4 2 1 1 1 7
2 7 3 6 3 3 2 1 7
3 9 4 8 4 5 3 1 9
4 11 5 10 5 7 4 2 9
5 13 6 12 6 9 5 2 12
6 15 7 14 7 11 6 2 12
7 17 8 16 8 13 7 3 15
8 19 9 18 9 15 8 3 15
9 21 10 20 10 17 9 3 20
10 23 12 22 12 19 10 4 20
Table 4 AEC results for solved test problems

Test problems MID Max_Spread SM NPS RAS Spacing CPU time
1 13.290 11.600 7.91 7 0.270 6.940 6.18
2 14.198 12.581 9.541326 10 0.353 9.558 12.95
3 15.261 14.930 10.06833 11 0.542 12.316 79.36
4 16.697 15.812 15.92292 11 0.723 17.144 425.85
5 17.244 16.501 19.61764 13 0.922 22.102 994.13
6 18.638 22.395 23.40268 15 1.547 22.914 1272.68
7 20.090 31.184 29.90024 17 2.112 33.732 2463.88
8 20.744 40.822 35.35586 19 3.072 35.858 3600.00
9 21.413 42.707 53.01155 19 3.660 48.813 3600.00
10 0.000 0.000 0 0 0.000 0.000 Not solved
Average 17.508 23.170 22.748 13.55 1.46 23.264 1383.892
116 A. Goli et al.
Table 5 HWOAGA results for solved test problems

Test problem MID Max_Spread SNS NPS RAS Spacing CPU time
1 18.869 10.008 7.6622 6 0.345 7.065 13.95
2 18.750 18.431 10.0910 18 0.463 10.935 15.97
3 15.845 16.542 11.1755 26 0.652 14.679 32.97
4 18.627 22.470 22.0571 29 0.957 18.436 57.49
5 25.336 20.822 24.8277 30 1.040 27.163 73.85
6 19.234 30.336 23.6967 30 1.577 32.113 96.28
7 29.121 39.136 38.2629 37 2.427 45.893 127.49
8 30.477 49.408 52.1947 43 4.393 43.912 133.64
9 23.715 47.347 65.1100 50 4.110 53.015 157.96
10 23.690 56.216 60.4938 50 6.326 66.930 169.93
Average 22.366 31.071 31.5572 31.9 2.229 32.014 87.953
5.1 Review the Solution Methods Based on the MID Index
The average of this index for AEC was 17.5 and for the HWOAGA algorithm, it was
22.36. The results of this indicator indicate that HWOAGA amount was higher in all
examples. The reason for this is that the AEC method introduces the best possible
values, so for the innovative method, the small distance with these values is equivalent
to the quality of this method. The results based on MID index are presented in Fig. 5.
35.000
30.000
25.000
20.000
MID
15.000
10.000
5.000
0.000
1 2 3 4 5 6 7 8 9 10
Test problem
AEC HWOAGA
Fig. 5 Comparing HWOAGA and AEC on MID index

60.000
50.000
Max spread
40.000
30.000
20.000
10.000
0.000
1 2 3 4 5 6 7 8 9 10
Test problem
AEC HWOAGA
Fig. 6 Comparing HWOAGA and AEC on max spread index
5.2 Review the Solution Methods Based on the Max Spread

Index
The average of this index for the AEC was 23.1 and for the HWOAGA algorithm
was 31.07. As shown in Fig. 6, the AEC method has only been able to provide a
better value than the HWOAGA in the first test problem. In other ones, HWOAGA
method provides a better value than this index, which indicates the effectiveness of
this method in finding the solutions to the intended problem.
5.3 Review the Solution Methods Based on the SNS Index
The average of this index for AEC was 22.7 and for the HWOAGA algorithm was
31.55. Regarding the SNS index, like the Max spread, the HWOAGA algorithm is
also well-known for finding optimal solutions. By increasing the issue number, the
superiority of the HWOAGA method is seen in the SNS index. The results based on
SNS index are presented in Fig. 7.
5.4 Review the Solution Methods Based on the NPS Index
The average of this index for the Epsilon constraint was 13.55 and for the HWOAGA
algorithm was 31.90. The NPS index is one of the most important indicators that
shows the superiority of the HWOAGA algorithm over AEC. In AEC method the
ability to find various answers decreases due to constraints in the problem. But in the
HWOAGA algorithm, due to its random and intelligent search, it is possible to find
118 A. Goli et al.
70
60
50
40
SNS
30
20
10
0
1 2 3 4 5 6 7 8 9 10
Tet problem
AEC HWOAGA
Fig. 7 Comparing HWOAGA and AEC on SNS index
60
50
40
NPS
30
20
10
0
1 2 3 4 5 6 7 8 9 10
Test problem
AEC HWOAGA
Fig. 8 Comparing HWOAGA and AEC on NPS index
more optimal solutions, which is why in the NPS index a much better performance
of this algorithm is seen. The results based on NPS index are presented in Fig. 8.
5.5 Review the Solution Methods Based on the RAS Index
The average of this index for the AEC was 1.46, and for the HWOAGA algorithm
was 2.22. As it can be seen in Fig. 9, the HWOAGA algorithm offers more values
in compare with AEC. This difference is small in a number of examples. The small
difference between the HWOAGA algorithm and the AEC method, which is on
average about 0.6, indicates the ability of the HWOAGA algorithm.
7.000
6.000
5.000
4.000
RAS
3.000
2.000
1.000
0.000
1 2 3 4 5 6 7 8 9 10
Test problem
AEC HWOAGA
Fig. 9 Comparing HWOAGA and AEC on RAS index
5.6 Review the Solution Methods Based on the Spacing Index
The average of this index for the AEC was 23.26 and for the HWOAGA algorithm
was 32.01. The analysis of this index is similar to the RAS index. The AEC method
provides fewer values, and the HWOAGA algorithm tries to provide a SNS close to
this method. The above results show that this algorithm has been relatively successful
in its attempt to be close to the ideal solution. The results based on Spacing index
are presented in Fig. 10.
80.000
70.000
60.000
Spacing
50.000
40.000
30.000
20.000
10.000
0.000
1 2 3 4 5 6 7 8 9 10
Test problem
AEC HWOAGA
Fig. 10 Comparing HWOAGA and AEC on spacing index

120 A. Goli et al.
Fig. 11 Comparison of solving time of two methods
5.7 Review the Solution Methods Based on CPU Time
The results show that the EAC has an average time of 1383.89 s. While this index
for HWOAGA is 87.95. Figure 11 shows the comparison of the solving time of these
two methods.
As seen in Fig. 11, in the AEC method, the solving time has an ascending trend
with a significant slope. It has solved test problem 7 and 8 as long as possible and also
did not have capabilities to solve problem 10. This is while the HWOAGA algorithm
has increased the solving time with a very small slope. In summary, after analyzing
different indices, it is indicated that the HWOAGA algorithm, with spending much
less time than AEC, can provide very close outputs to optimal answers and even
works better on some other indicators than AEC. This issue properly shows the
ability of the HWOAGA algorithm to solve the problem raised in this research.
5.8 Sensitivity Analysis
In this section, the effect of the demand parameter on each objective function is
investigated. The reason for choosing the demand parameter among all the parameters
is that in real conditions the fluctuation of this parameter is much higher than other
ones and on the other hand, it has a great impact on the supply chain structure.
Accordingly, the fluctuation of −20 to +20% is considered on this parameter
and the value of each objective function is measured. The results are summarized in
Table 6.
The trend of each objective by changing the demand parameter is illustrated as
Figs. 12, 13 and 14.
Figure 12 shows that with an increase in demand, a linear effect on the economic
objective is observed. In other words, increasing demand causes supply chain costs
Table 6 Sensitivity analysis results

Fluctuation rate −20% −10% 0% 10% 20%
Z1 0 0 0 0.019 0.021
Z2 2230.45 2251.29 2307.12 2360.96 2399.81
Z3 862.95 872.98 883.01 893.04 903.07
Fig. 12 Effect of demand fluctuation on the first objective (economic)
Fig. 13 Effect of demand fluctuation on the second objective (environmental)
to be increased linearly. The results of Fig. 13 show that increasing demand leads to
finding a growth in the second objective function with a non-constant slope. At the
level of −20%, the lowest slope, and at the level of +20%, the highest value of this
objective function is observed. Figure 14 shows that negative demand changes do not
affect the social objective function. In other words, reducing the amount of demand
122 A. Goli et al.
Fig. 14 Effect of demand fluctuation on the third objective (social)
cannot change the social responsibility of the supply chain. However, increasing
demand increases non-linear supply chain social responsibility.
6 Conclusion and Future Studies
In this study, the optimization of multi-product and multi-period closed-loop supply

chain design problem was investigated. The supply chain understudy was commen-
surate with the supply of perishable products. In this regard, the sustainable supply
chain is considered as an essential and effective concept. The proposed mathematical
model can provide suitable solutions because of considering all internal and external
stations in the way of meeting customers’ demands. By considering the nature of
dairy products, delivery time, and customer satisfaction in supply chain design were
investigated. In order to optimize this mathematical model, a new meta-heuristic
algorithm called HWOAGA was developed. In this algorithm, the algorithms of
the whale optimization algorithm and Genetic algorithm were used. The results of
comparing this algorithm with AEC show that the proposed algorithm produces much
more Pareto solutions, the quality of Pareto solutions is acceptable and the dispersion
is good. It also has a shorter time to solve test problems in comparison to the AEC
method. The managerial insights of this study suggest that applying the mathematical
model and the proposed solution method can help supply chain managers to improve
financial, environmental, and social performance. In order to develop this research,
it is suggested to develop the mathematical model under uncertainties in demand and
environmental parameters by using possibilistic or stochastic programming. Also,
the robust optimization approach can be an effective way of dealing with uncertainty
in this problem.
References
Babazadeh R, Razmi J, Pishvaee MS, Rabbani M (2017) A sustainable second-generation biodiesel

supply chain network design problem under risk. Omega 66:258–277
Burgess TF, Grimshaw P, Huaccho Huatuco L, Shaw NE (2017) Mapping the operations and supply
chain management field: a journal governance perspective. Int J Oper Prod Manag 37(7):898–926
Cao C, Li C, Yang Q, Liu Y, Qu T (2018) A novel multi-objective programming model of relief
distribution for sustainable disaster supply chain in large-scale natural disasters. J Clean Prod
174:1422–1435
Carter CR, Rogers DS (2008) Sustainable supply chain management: toward new theory in logistics
management. Int J Phys Distrib Logist Manag 38(5):360–387
Davoodi SMR, Goli A (2019) An integrated disaster relief model based on covering tour using hybrid
Benders decomposition and variable neighborhood search: application in the Iranian context.
Comput Ind Eng 130:370–380
Deb K, Agrawal S, Pratap A, Meyarivan T (2000, September) A fast elitist non-dominated sorting
genetic algorithm for multi-objective optimization: NSGA-II. In: International conference on
parallel problem solving from nature. Springer, Berlin, pp 849–858
Fard AMF, Hajaghaei-Keshteli M (2018) A tri-level location-allocation model for forward/reverse
supply chain. Appl Soft Comput 62:328–346
Ghelichi Z, Saidi-Mehrabad M, Pishvaee MS (2018) A stochastic programming approach toward
optimal design and planning of an integrated green biodiesel supply chain network under
uncertainty: a case study. Energy 156:661–687
Goli A, Davoodi SMR (2018) Coordination policy for production and delivery scheduling in the
closed loop supply chain. Prod Eng Res Devel 12(5):621–631
Goli A, Zare HK, Tavakkoli-Moghaddam R, Sadegheih A (2019) Multiobjective fuzzy mathematical
model for a financially constrained closed-loop supply chain with labor employment. Comput
Intel
Liang L, Quesada HJ (2018) Green design of a cellulosic butanol supply chain network: a case
study of sorghum stem bio-butanol in Missouri. BioResources 13(3):5617–5642
Mardan E, Govindan K, Mina H, Gholami-Zanjani SM (2019) An accelerated benders decomposi-
tion algorithm for a bi-objective green closed loop supply chain network design problem. J Clean
Prod 235:1499–1514
Mardani A, Kannan D, Hooker RE, Ozkul S, Alrasheedi M, Tirkolaee EB (2020) Evaluation of
green and sustainable supply chain management using structural equation modelling: a systematic
review of the state of the art literature and recommendations for future research. J Clean Prod
249:119383
Mirjalili S, Lewis A (2016) The whale optimization algorithm. Adv Eng Softw 95:51–67
Niranjan T, Parthiban P, Sundaram K, Jeyaganesan PN (2019) Designing a omnichannel closed
loop green supply chain network adapting preferences of rational customers. Sādhanā 44(3):60
Rabbani M, Hosseini-Mokhallesun SAA, Ordibazar AH, Farrokhi-Asl H (2020) A hybrid robust
possibilistic approach for a sustainable supply chain location-allocation network design. Int J
Syst Sci Oper Logist 7(1):60–75
Rad RS, Nahavandi N (2018) A novel multi-objective optimization model for integrated problem
of green closed loop supply chain network design and quantity discount. J Clean Prod
Sangaiah AK, Tirkolaee EB, Goli A, Dehnavi-Arani S (2019) Robust optimization and mixed-
integer linear programming model for LNG supply chain planning problem. Soft Comput 1–21
Sangaiah AK, Goli A, Tirkolaee EB, Ranjbar-Bourani M, Pandey HM, Zhang, W (2020) Big data-
driven cognitive computing system for optimization of social media analytics. IEEE Access
8:82215-82226
Tirkolaee EB, Goli A, Weber GW (2019) Multi-objective aggregate production planning model
considering overtime and outsourcing options under fuzzy seasonal demand. In: Advances in
manufacturing II. Springer, Cham, pp 81–96
124 A. Goli et al.
Tirkolaee EB, Goli A, Weber GW (2020) Fuzzy mathematical programming and self-adaptive
artificial fish swarm algorithm for just-in-time energy-aware flow shop scheduling problem with
outsourcing option. IEEE Trans Fuzzy Syst. https://doi.org/10.1109/TFUZZ.2020.2998174
Yavari M, Geraeli M (2019) Heuristic method for robust optimization model for green closed-loop
supply chain network design of perishable goods. J Clean Prod 226:282–305
Zhuo H, Wei S (2017) Gaming of green supply chain members under government subsidies—based
on the perspective of demand uncertainty. In: Proceedings of the tenth international conference
on management science and engineering management. Springer, Singapore, pp 1105–1116
Practical Aspects of Application the 3R
(Reduce, Recycle, Reuse)
How to Assess Internal Transport
in Terms of Sustainability
in the Recycling Industry?—Case Study
Izabela Kudelska and Monika Kosacka-Olejnik
Abstract Auxiliary processes are relevant element of the production process, due to
the fact that they affect the technical and economic efficiency. In the modern produc-
tion process there are many material flows between which or during which techno-
logical operations that process the input material into a finished product are carried
out. Incorrect design or inadequate organization of supporting processes can cause
serious disruptions or even financial losses, what may be analyzed from the perspec-
tive of sustainability. The chapter is focused on the process of internal transport, what
is justified by influence of that process on the company’s economic effects, the quality
of employees’ work and the quality of the natural environment. Authors analyze the
internal transport process in a company representing recycling industry which is End-
of-Life Vehicles disassembling enterprise. This sector is affecting People and Planet
during generating Profits (3P concept). In disassembling company internal transport
process includes transport of: ELV’s, vehicles parts, waste, but also it covers handling
of materials including laying, fixing and measuring. Material flow is complex because
of diversity in terms of materials’ size, shape, mass and quantity. In the result the
internal transport process requires precise organization, which should be adopted to
the company to ensure safe shipment of cargo with minimal costs. The basic purpose
of the chapter is to prepare a proposal of method for internal transport assessment in
terms of sustainability. Authors use the following research methods: literature review,
case study, brainstorming. Chapter consists of three parts. The first part shows the
role of the internal transport process as well as the differences between the transport
process in a typical enterprise and disassembling company. In the second part, mate-
rial flows occurring in disassembling facility are described. In the last part problems
were identified and the directions of further research were indicated.
Keywords Internal transport · Disassembling company · Sustainable development
I. Kudelska (B) · M. Kosacka-Olejnik

Faculty of Engineering Management, Poznań University of Technology, Poznań 60-965, Poland
e-mail: izabela.kudelska@put.poznan.pl

128 I. Kudelska and M. Kosacka-Olejnik
1 Introduction
In authors’ opinion there is not attached enough attention to the issue of internal
transport process both in Business and Academia. Internal transport is considered in
the paper as an essential part of inbound logistics, affecting the most a production
process, and as a consequence logistics activities supporting it. In the paper the
internal transport process includes all activities related to material handling in order
to: supply the production process, remove waste and prepare goods for distribution.
All undertaken activities have to occur at the company’s area.
Authors of the paper treat internal transport as a process that mainly supports
production. Considering that, it was stated that internal transport process plays an
important role in a production process owing to the fact that it shortens the production
cycle, gives rhythm, and facilitates work. As a consequence, it was claimed that the
transport process should ensure the movement of a certain amount of cargo on the
shortest possible transport routes, with the maximum use of means of transport and
the minimum use of the machinery. Following previous studies (Nowotyńska et al.
2017a, b; Dima 2013), it was stated that transport includes a set of activities such
as: moving, reloading and manipulations (stacking, fixing, counting). As a result of
effective transport control, output parameters of the production process, including:
the level of manufacturing costs, the volume of stocks and work-in-progress as well
as work efficiency, are satisfactory for managers.
Authors claimed that an economic efficiency of production depends not only on
modern technology, but also on properly designed and organized process of internal
transport, what justifies works on the specified topic.
Considering the high importance level of internal transport process for the
business activity, authors chose a disassembling company as an object of the
conducted research. Disassembling facility was strongly recommended as company
with high impact on people, planet and profit, on the basis of previous research
conducted by authors of the paper (Kudelska and Kosacka 2014, Project of number
503223/11/140/DSMK/4132: Development of a method for the management of parts
resulting from the dismantling of end-of-life vehicles taking into account the prin-
ciples of sustainable development; Kosacka and Kudelska 2015, Project of number
503225/11/140/DSMK/4136: Development of a method for storing parts resulting
from the dismantling of end-of-life vehicles). What is more it has been observed an
increasing trend in vehicles number, what results in high growth of vehicle recy-
cling sector in Poland. Considering the fact, that companies and customers are more
aware of sustainability, it requires to introduce sustainability in business activity and
all processes, including internal transport process. However there are made efforts on
adopt sustainability issues in business activity, there is lack of solutions dedicated for
internal transport. In authors opinion companies should have possibility to improve
their impact on people, planet and profit considering their internal transport activity.
Means of the internal transport often require manual operation. Authors claimed
that, in many Polish companies a large part of the work is carried out by hand or by
simple means of transport, what affects people. Even when there is a cutting-edge
How to Assess Internal Transport in Terms … 129
technology there is always required a man who control it. Moreover, all means of
transport generate environmental burden and costs for the company. Taking that into
consideration it was stated that internal transport should be sustainable.
In order to meet the specified need, authors prepared a procedure for assessment of
the internal transport realization to support sustainability policy introduction. It was
stated that companies increase their competitiveness becoming more sustainable, but
the problem lies in answering the question: how to check it? particularly in the context
of internal transport. Consequently, there was prepared a list of research questions:
• Q1: What requirements should meet an internal transport process in terms of
sustainability?
• Q2: What is the structure of the internal transport process realized in disassembling
company?
• Q3: How to assess the internal transport process in disassembling company in
terms of sustainability?
In order to achieve the specified goals there was used the case study method.
The paper consists of four parts. In the Sect. 2, there was presented a disassembling
company as a specific kind of business from the perspective of material flows and
processes. Section 3 was focused on the theoretical background of the proposed
method. On the basis of this part, the method was verified in the next parts. In the
summary there were included directions for future research.
2 Disassembling Company as a Specific Business Domain
In order to understand the specific features of disassembling company as represen-

tative of the recycling sector, in the presented chapter there were presented material
flows and processes realized in that kind of business.
2.1 Material Flows Organization in a Disassembling

Company
Disassembling facilities operating in Poland have become a research object in the

presented study. It was stated, that disassembling company becomes a key player
in the recycling network, which is responsible for hazardous waste management
(End-of-Life vehicles), affecting all sustainability pillars (Kosacka and Golińska
2014).
It was stated, that the importance level of disassembling company, considering
Environment and People, is relatively high. In the result it is under law supervision
of the Recycling Act of ELVs [polish adaptation of Directive 2000/53/CE (2000)] in
Fig. 1 Material flows in

disassembling and
manufacturing
business—comparison
the aspect of: recycling and recovery rate, technical and organizational conditions,
waste and parts storage, etc.
In order to discover differences and similarities of disassembling company and
typical manufacturer, there was made a material flow comparison, presented in the
Fig. 1
With reference to Fig. 1, it was stated, that in manufacturing company material
flows begin with the supply of raw materials from external transport to the main
warehouse and then by internal transport to the departments. Departmental trans-
port enables the flow of materials through production sections. It ends with inter—
departmental transport directed to the finished goods warehouse, from which the
distribution begins. In contrast, in disassembling facility there are supplied complex
products—ELV’s. Moreover, there may be perceived also differences in output, as
there are two main groups of materials, including: reusable parts (bought by end-
users, car mechanics, shops with car parts) and waste intended for recycling or
utilization by proper companies.
The next differentiating factor is related to types of customers and suppliers, as
key suppliers for disassembling companies are and-users of vehicles, also collection
points, insurance companies and public authority. What is more, supplier becomes
very often a customer, what is very specific situation on the market.
In previous study authors have examined, that the most valuable flow is flow
of reusable parts (Czwajda et al. 2017), that is the most diversified and requires
appropriate warehousing as well as well-fitted infrastructure.
2.2 Processes in a Disassembling Company
It was stated, that internal transport is always related to the primary production
process, as it is a part of it, so firstly there should be recognized the disassembling
process in order to analyze transport process.
Fig. 2 Architecture of processes in disassembling facility
The disassembling process is controlled by demand on parts which is uncertain.

Customer may order part directly in place, where company is located or via the
Internet or phone.
In the disassembling company, there are made following processes: ELV’s adop-
tion,1 preparation for storage, vehicles storage, disassembling of vehicle or partic-
ular part (e.g. engine), waste/parts’ storage, waste/parts’ distribution. Processes are
connected with material and information flows, however they would not occur
without transport operations (internal and external), what was presented in the Fig. 2.
With reference to Fig. 2, ELV’s adoption process is proceeded by external trans-
port which may be realized by supplier or by receiver (disassembling facility). During
adoption process ELV is weighted, what requires movement to the weight station.
After completing the formalities, vehicles is moved to the station where preparation
is made. There are removed some unique parts as well as waste (e.g. catalyst, fluids),
which require transport. Prepared ELV is moved to the storage yard for vehicles,
where it is kept until order for parts appears or disassembling request is made, in
accordance to disassembling schedule. During storage, there are removed waste,
which require transport. There are two options for disassembling. Firstly, fragmen-
tary disassembling, where part is removed according to the order (sell at place). Part
is transported to the sales department, where the payment is realized or it is stored
until the shipment is made. Secondly, there is a full disassembling, if ordered part is
hard to reach and it requires full disassembling or if vehicle is planned to be disas-
sembled as a whole according to schedule. It was stated, that there is no procedure
for disassembling process realization because scope of work is dependent on many
1 This process was presented in details in previous study in work: (Kosacka et al. 2015).
Fig. 3 Disassembled parts
factors including: the technical condition of parts, quality of parts’ connections, tech-
nical solutions made by car manufacturer. All disassembled parts are presented in
the Fig. 3.
With reference to Fig. 3, it was stated that the internal transport requires a lot of
work in company under study. Disassembling process has the greatest demand on
transport operations due to the fact that it results in generation of the large amount of
waste and parts, which should be moved to appropriate storage container/area. Some
of parts like engine, car seats are disassembled separately at appropriate stations,
what also generates demand on transport.
Considering the fact that, Polish recycling system is manual system, where there
are simple machines and tools supporting manual operations made by people, it was
stated that all processes in disassembling company are affecting people substantially.
What is more, considering the specific kind of business it may be perceived a high
influence of disassembling business on sustainable development policy realization.
3 Assessment of the Internal Transport Process in Terms

of Sustainability—Methodology
In order to answer the question: “How to assess the internal transport process in
terms of sustainability”, there was developed a procedure presented in the Fig. 4:
Firstly, there should be prepared a list of guidelines on internal transport
consideration in the aspect of sustainability.
However the term “Sustainable development” is widely known from 1987, the
definition proposed by WCED (Brundtland 1987), is unclear and without any prac-
tical connotations. In the result it was claimed, that the term sustainability is not
Fig. 4 Procedure for the internal transport assessment considering sustainability issues
understood by people. Authors claimed that it is not possible to assess the sustain-
ability of each process in such of conditions, so there was made an attempt to define
requirements for internal transport process that would be realised in terms of sustain-
able development policy introduction. In order to solve that problem, there was used
a literature review method.
The basis for sustainable development are three pillars, including social, ecolog-
ical and economic aspects. As it was mentioned in the Introduction, internal trans-
port process affects people, planet and economy. In the result it was stated that
internal transport process which support sustainable development policy realiza-
tion, should ensure transport of various materials (raw materials, semi-finished prod-
ucts, materials and the collection of finished products and waste) between stations
and/or departments at the company, that would meet the following requirements
simultaneously:
• R1: cost reduction;
• R2: comfort for the Employee improvement;
• R3: negative influence on the Environment reduction through machines and orga-
nization that will ensure future generations access to the Environment at least at
the same level that we have.
With reference to the Fig. 4 according to the guidelines on internal transport
consideration in the aspect of sustainability, there should be made an analysis of
transport realization in details with the use of case study method. Authors recommend
to perform some observations in the company under study.
Firstly, there should be identified operations and parties engaged into their real-
ization in order to map the process realization. Secondly, there should be identified
resources, including human resources and material resources (means of transport and
material). It is relevant to define the labour intensity of activities (influence on people)
and the volume of work that should be done. Lastly, there should be made an analysis
of factors affecting transport in the particular company. Authors recommended to use
an Ishikawa diagram for that purpose.
After the deep analysis of the internal transport in the company, there should
be defined a list of achievements for the internal transport process in terms of
sustainability, considering requirements listed beforehand (R1–R3).
There was recommended the brainstorming method use with experts from the
industry and academia. In authors opinion, in order to ensure simplicity of the assess-
ment, there should be determined a questionnaire with a list of achievements that
allow the assessment, according to the objective of the paper. Achievements should
be divided into three groups, in order to assess separately each pillar of sustainability,
what corresponds with the requirements R1–R3. Each achievement should be formed
as a question with possible answer: YES, NO.
In authors opinion each of achievement should be equally important.
The level of accomplished achievements allow to evaluate the internal transport
process in each of considered aspects. The obtained result will be in a range of 0–
100%. The biggest advantage is that all achievement that were not accomplished
show direction for future development for the company. Authors recommended to
use an obtained score to define future actions, not to point out what should be done but
what is time pressure for changes, following the guidelines presented in the Table 1.
Table 1 Guidelines for future actions

Level Score of assessment (%) Required actions Time pressure
1 <0–33> Required Immediately
2 (33–66> Corrective When organizational and financial
state allow for that
3 (66–100> Improvement Expected but not required
With reference to data shown in the Table 1, there were prepared guidelines
which should be followed during assessment. Authors suggested to use simple scale
with three ranges of percentage value of accomplished achievements, specified in
the previous step. There were noticed information about time pressure of required
actions, however there were not included guidelines what to do. Owing to the fact
that, decisions should be made by Employees of the specified company, there is only
obtained result of total score and the required minimum to achieve better level of
assessment.
The procedure described theoretically in presented chapter was verified in the
next chapter for a disassembling company, which represents vehicle recycling sector
in Poland.
4 Assessment of the Internal Transport Process in Terms

of Sustainability in the Recycling Sector
The method proposed in the previous section was used in order to assess the internal
transport process in the one of the biggest polish disassembling facilities. The
company employs 10 people, including the owner, processing around 1200 ELV’s
annually.
With reference to Fig. 4, following requirements R1–R3, there was made an
analysis of internal transport realization in the company under study.
Internal transport in disassembling company includes operations presented in the
Fig. 2, where there were described operations, engaged parties and material stream.
Considering that, authors stated that those operations may be divided into following
types:
• Production transport—transport operation related to the primary process (disas-
sembling);
• Warehouse-production transport—related to waste and parts reception and
shipping, including their storage in warehouse area;
• Inter-departmental transport, as a part of production transport, realized inside
Disassembling Department, divided into working-stand type (at disassembling
position) and transport between stations (e.g. between engine disassembling
station and car seats disassembling station).
Considering fact that Polish recycling companies have limited resources, which
are engaged into many manual operations, there have been made analysis of
resources, including means of transport used in a typical disassembling company,
people engagement and material flows, what was described in the Table 2.
Table 2 Means of transport used in disassembling facility

Load unit Transport route Packaging Means of transport
Start End unit
Forklift Platform Worker
trolley
ELV ELV’s’ storage Disassembling – x
yard station (car lift)
Engine Disassembling Disassembling – x
station (car lift) station
Cooler Disassembling Disassembling – x
station (car lift) station
Body Disassembling Shredder – x
station (car lift)
Wheel Disassembling Vulcanization – x
station (car lift) position
LPG Disassembling Waste – x
installation/fuel station (car lift) container
tank
suspension Disassembling Parts’ – x x
system station (car lift) warehouse
Liquids Disassembling Waste Barrels for x
station (car lift) containers liquids
Headlights, Disassembling Parts’ – x
bumpers station (car lift) warehouse
Car seat Disassembling Seats’ – x x
station (car lift) container
Glass Disassembling Parts’ – x
station (car lift) warehouse
Dashboard Disassembling Container for – x
station (car lift) plastics
Other waste Disassembling Containers x
station (car lift)
With reference to information in the Table 2, it was stated that however there
are used forklifts and platform trolleys, vast of transport operations in disassem-
bling facility is performed by people. Consequently, it was stated, that internal trans-
port in disassembling company is very labour-intensive, as workers are very often
responsible for transport.
In the infrastructure of internal transport besides: car lift, forklift, supermarket
trolley, platform trolley, there should be also distinguished transport routes, including
transport routes inside disassembling hall and warehouse, and outside transport
routes, between particular departments (ELVs’ storage yard, vulcanization station,
etc.).
Transport is made between stations and departments, what is a result of transport

relations, responsible for the movement of the material. The following groups of
relationships were distinguished:
• transport relations;
• delivery relations;
• reception relations.
With reference to Table 1, it was stated that transport analysis is much easier and
its result is more transparent if it can be presented in a simple and compact way. It
was recommended to examine the volume of material flows with the use of Sankey
diagram as a specific type of flow diagram, in which the width of the arrows is shown
proportionally to the flow quantity (Fig. 5).
In the Fig. 5, there was presented Sankey diagram which represents all mate-
rial flows into company under study, including ELVs (blue), waste (green) and
parts (orange). The widths of the bands are directly proportional to the volume of
items. The scheme puts a visual emphasis on the major transfers or flows within a
system. It was claimed, that material flows are complex. The material volume that
flows between individual objects is hard to identify because ELVs represent various
completeness levels. In addition, parts are in a different technical condition, which
may cause that they may not be redistributable. Authors stated that parts create the
most diversified stream, which includes: gearbox, alternator, clutch, axle, wheel hub,
starter, drive shaft, flywheel, bridge, engine, radiator, shock absorber, spring, rocker,
beam, McPherson column, lamps, door, bonnet, rear hatch, bumper, fender, mirror,
windows, seat, rear seat, navigation, gas installation, roof rack, trailer hook, tire, rim,
hubcap, unique elements. It was stressed, that an essential part of transport process
description is its correct identification, what is equal to decomposition at various
hierarchical levels: system, environment, process and related attributes. Decompo-
sition was made on processes occurring in disassembling company. In each of these
processes, the relationships that occur between the various stages of disassembling
process, have been isolated and analysed. In the result, there were identified factors
influencing the increase in labour intensity of the means of transport.
What is more, research and observation in a chosen disassembling facility, allowed
to define and systematize the factors occurring and affecting the growth of lead time
of transport operations, what (Table 3).
Factors presented in the Table 3 show that, with the increase of labour consump-
tion there is related growth of transport costs resulting even from poor surfaces,
insufficiently wide passageways or insufficient service. Considering that, authors
stated that identification of those factors and determination their impact, is essential
for disassembling facility assessed in the context of sustainability issues.
Considering the disassembling process in cybernetic terms, it is a series of consec-
utive operations from which the output of the previous operation is part of the next
operation’s input, what was shown in Fig. 6.
With reference to Fig. 6, it was noticed that there is a series of material displace-
ments during which technological operations are carried out. Therefore, each process
Fig. 5 Sankey diagram for material flows in disassembling company (adopted from Report of
research on Sustainability in Remanufacturing operations, SIRO 2012–2014)
Table 3 Factors affecting the increase of working time of the means of transport in disassembling
facility (adopted from Fijałkowski 2003)
Group of factors Factor type Description
Human factors Experience Possibilities and effort of a given operator in
relation to the experienced operator (model)
Tiredness Requirement of rest time
Inattention Lack of proper supervision may result in longer
transport time as well as increase the risk of
transported part’s damage
Working parameters Distance Lack of use of maximum speed at short distance
Height Setting in layers or on higher shelves requires
additional time
Driving direction Varied time of forward and backward movement
Technical condition The technical condition of the means of transport
has an impact on time losses (e.g. frequent failures)
as well as on the environment
Utilization factors Pedestrian traffic Delays caused by pedestrians and other vehicles
Obstacles Low and narrow routes, obstacles at the routes
Surface The type and condition of the surface on which the
means of transport is moving
Doors/gates At some of the doors/gates, the transport must be
released because of its size
Cargo type and size The value of the movement times depends on the
volume of the load, weight and type
Lighting The lighting quality affects transport as it requires
more caution
Order Disorder on the transport and storage surface delays
the work of the means of transport
Weather Poor weather (rain/snow) reduces work efficiency
outside the disassembly hall
Elevation Elevations limit the speed of the means of transport
Fig. 6 The cybernetic approach to the transport process in disassembling company

Fig. 7 Factors affecting transport in disassembling company
should correspond to the proper transport process, because only then can you get the
required economic and technical results.
Consequently, it was stated that, poor transport organization in many cases cause
serious disruptions of the disassembling process, including:
• Unforeseen breaks, resulting from unpunctual delivery of ELV or waste/parts
collection;
• Growth of demand on storage area;
• Changes in employees’ responsibilities e.g. disassembling workers deal with
transport activities;
• Deterioration of health and safety at work.
It was claimed that many factors have an influence on transport in disassembling
company, however they affect each other, e.g. lack of parts/waste removal from the
dismantling station will require storage at the station, which will: reduce the storage
space, make work more difficult for the employee considering aspect of health and
safety and increase the risk of environmental pollution. There was made an analysis
of all factors affecting transport, presented in the Fig. 7.
According to Fig. 7, there were specified five groups of factors, including charac-
teristic of input (ELVs), characteristic of output (parts, waste), employees, technology
and market factors.
On the basis of observations and interviews with employees of the company under
study, as well as literature review, there were specified basic transport methods.
The preliminary issue is the analysis of transport methods, which depends on the
following factors:
• What to transport (material type, shape, weight, dimensions)?
• In what it should be transported (container)?
• Where it should be transported (route, start and end)?
• How often to transport?
• What are the transport distances?
• At what speed transport should take place?
• What are the transport routes (pavement hardness, slope)?
• What should be the degree of mechanization or automation.

To sum up, the above characteristic of internal transport process in the disas-
sembling facility was used in order to assess it according to sustainability policy
realization.
With reference to the procedure presented in the Fig. 4, in the next step there was
prepared a brainstorming session, where there were three researchers from area of
production familiar with sustainability issues and two representatives of the disassem-
bling facility (owner, production manager). One author of the paper was a chairman
of brainstorming session, the other noted all ideas. In the result of conducted research,
there were prepared three lists of achievements for each sustainability requirement
determined for internal transport process in disassembling facility, presented in the
Tables 4, 5 and 6.
The first list of achievements was prepared for the economic issues, considering
requirement R1, assessed for the company under study was presented in the Table 4.
According to the prepared list of achievements (Table 4), the company where the
research was carried out, obtained a result of economic aspect of development of
internal transport at the level of 50%.
In that case, it was claimed that company should improve their result, consid-
ering min. two changes in internal transport process to achieve the level of min.
66% (Table 1, level 2), which should be introduced when it is possible considering
organizational and financial conditions.
The second list of achievements was prepared for the social issues, considering
requirement R2, assessed for the company under study. With regard to engagement
of workers in transport operations in disassembling facility, authors recommended to
analyse actions towards employees, considering the list of achievements (Table 5).
Table 4 List of achievements to assess economic aspect of internal transport

Achievement YES NO
There many accidents in transport (min. one per day) X
There empty running during the transport X
There are often breaks in the dismantling process caused by the bad ELVs’ delivery X
organization and the parts’ receipt/supply to the station (min. three per day)
The depreciation costs of transport infrastructure are high X
A lot of materials damaged during internal transport (min. 20% of total) X
Each means of transport is empty for more than 20% of working time X
Transport operations are not scheduled X
There are many reloading points on the ELV’s disassembly route X
There are many unnecessary operations of handing and transporting materials X
There are some changes of means of transport in transporting each part/waste from X
start to end?
Total number of answers 5 5
Total score of economic aspect (%) 50
Table 5 List of achievements to assess social aspect of internal transport

Achievement YES NO
There are accidents occurring at work during internal transport (min. once per X
week)?
Employee responsible for internal transport have a wide range of duties also not X
related to transport
There is no employee’s satisfaction research in the company X
There are many manual activities during transport X
Manual transport may be replaced by mechanical transport X
There is no close relationships with customers and suppliers X
Most of means of transport pose a threat to the life and health of the employee X
Most of means of transport are not ergonomic X
Worker has to move manually a load with a weight exceeding 25 kg? X
There is no customers’ satisfaction research in the company X
Total score of social aspect (%) 40
Table 6 List of achievements to assess ecological aspect of internal transport

Question YES NO
Cargo units with waste are not secured during transport and storage X
Containers/canisters are damaged X
Waste/parts are temporarily stored on the floor X
There are used even broken means of internal transport X
Means of transport were not considered according to ecological parameters X
Road transport surface is destroyed X
Transport units are transported incorrectly X
Liquid and gaseous materials are not transported in separate containers X
Forklifts are not adapted to the working conditions X
Forklifts have invalid current technical inspection documents X
Total score of ecological aspect (%) 50
Considering disassembling business there should be noticed, its internal and

external stakeholders. Internal activity is related to employees, while external activity
is directed at customers, suppliers and the local community. It was assumed, that
activities in the social area can bring many benefits. These benefits can be considered
in the external dimension: improving the image and reputation, increasing customers’
loyalty, as well as in the internal dimension: maintaining the best employees,
increasing the motivation of managers and employees, improving relationships with
employees, that is based on trust, integration and group cohesion.
research was carried out, obtained a result of social aspect of development of internal
transport at the level of 40%, below the score of economic aspect. In that case, it was
claimed that company should improve their result, considering min. three changes in
internal transport process to achieve the level of min. 66% (Table 1, level 2), which
should be introduced when it is possible considering organizational and financial
conditions.
The last list of achievements was prepared for the ecological issues, considering
requirement R3, assessed for the company under study. As it was stated before,
disassembling facility affects not the Environment due to the fact that it manages
hazardous waste. A list of achievements for the particular aspect was presented in
the Table 6.
research was carried out, obtained a result of ecological aspect of development of
internal transport at the level of 50%, below the score of economic aspect. In that
case, it was claimed that company should improve their result, considering min. two
changes in internal transport process to achieve the level of min. 66% (Table 1, level
2), which should be introduced when it is possible considering organizational and
financial conditions.
To sum up, disassembling facility has limited human, financial and material
resources. For the proper implementation of tasks from the perspective of imple-
menting the principles of sustainable development policy, this business require a
support for the decision-making process in the area of internal transport. However,
internal transport is often treated superficially, it was stated that it is essential for
all processes undertaken in the disassembling facility, what justifies research on that
topic.
5 Summary
Sustainability is a modern and relevant issue which should be practically imple-

mented at the operational level of business, however the translation is not so easy
and require experts knowledge.
In the paper authors have chosen disassembling company considering high poten-
tial in the context of sustainability in order to prepare the assessment method for
internal transport process evaluation. It was claimed, that internal transport is essen-
tial and it influenced other business activities. According to the research the main
factors, that have an impact on transport and the choice of a means of transport are:
the shape, size, construction of the good. When the transport is designed, there should
be answered the following questions:
• Why and what kind of material have to be transported?
• Where is start and stop for material transport and why?
• When and what is the amount of material to be transported?
Considering those questions, there may be eliminated the randomness in the

internal transport design solution. Moreover, during the analysis, it is not enough
to examine only technical aspects, but also the economic, social and environmental
criteria should be noted.
In the paper there was determined a procedure that allow assessment of the internal
transport in terms of sustainability, what is the biggest advantage in authors opinion.
The presented method requires some improvements, however the most relevant is
that the objective was to consider meaningful of the internal transport from the
perspective of the sustainability issues. Authors are aware of the limitation of the
proposed method as the list of achievement is opened and it was specified for partic-
ular company, however it should be considered as an advantage. The list may be
improved and fit for a company.
In future research authors plan to create more precise performance measurement
system to assess not only each of sustainability aspects, but also sustainability state
of the whole internal transport as well as other processes.
Acknowledgements This paper refers to the research conducted under Statutory activity, financed
by MNiSW/Poznań University of Technology, Project ID: 11/140/DSMK/4136.
References
Brundtland GH (1987) Report of the World Commission on environment and development: our
common future. United Nations
Czwajda L, Kosacka M, Kudelska I (2017) Prediction markets as a decision support tool in disassem-
bling companies. DEStech Transactions on Engineering and Technology Research, (ICPR2017),
pp 238–243
Dima IC (2013) Industrial production management in flexible manufacturing systems. Published
in the United States of America by Engineering Science Reference, IGI Global
Directive 2000/53/EC (2000) Directive 2000/53/EC of the European Parliament and the Council
on end-of-life vehicles (2000). Off J Eur Commun
Fijałkowski J (2003) Transport wewn˛etrzny w systemach logistycznych. Warsaw University of
Technology Publishing House, Warsaw
Kosacka M, Golińska P (2014) Assesment of sustainability in dismantling station-case study. Res
Logist Prod 4(2):135–145
Kosacka M, Kudelska I (2015) Project of number 503225/11/140/DSMK/4136: development of a
method for storing parts resulting from the dismantling of end-of-life vehicles
Kosacka M, Kudelska I, Golińska-Dawson P (2015) How properly value ELVs? Concept of the tool
of ELVs assessment for dismantling station. Case study. In: Proceedings of the 25th international
conference on flexible automation and intelligent manufacturing (FAIM 2015), pp 208–215
Kudelska I, Kosacka M (2014) Project of number 503223/11/140/DSMK/4132: development of a
method for the management of parts resulting from the dismantling of end-of-life vehicles taking
into account the principles of sustainable development
Nowotyńska I, St Kut, Krauz M (2017a) Internal transport as integrated logistics parts in production
part 1. BUSES Technol Oper Transp Syst R 18(12):1548–1551
Nowotyńska I, St Kut, Krauz M (2017b) Internal transport as integrated logistics parts in production
part 2. BUSES Technol Oper Transp Syst R 18(12):1552–1555
Report of research on Sustainability in Remanufacturing operations. SIRO, 2012–2014
Principle of the Cognitive Grinding
of Reuse Materials
Adam Mroziński, Józef Flizikowski, Kazimierz Bieliński, and Marek Macko
Abstract Cognitive grinding or active recycled and reuse machinery monitoring—

continuous supervising, diagnosing, managing, controlling, compensating, docu-
menting; a process of acquiring and transferring streams of information (usually
source information) about the analyzed, developed object, process and relations
between the same and the environment that can be used to realize the postulated state:
knowledge creation (theory and innovation), environment melioration (harmfulness)
and technical system optimization (design)—depending on technology needs and
engineer imaginations. The objective of this paper is to provide a mathematical
description, optimization of the states and changes in the grinding of recycled mate-
rials and machine space, their surface and volume during movement (idle and working
movement) of the components and design assemblies in the multi-hole grinding
process.
Keywords Cognitive grinding · Recycling · Machinery monitoring
1 Introduction
Knowledge creation, coming as result of the system development, optimization,

modernization and innovation—creative action (creating). Melioration—intentional
activities of a technical system and boundary zone; activities that enhance, improve
A. Mroziński (B) · J. Flizikowski · K. Bieliński

Faculty of Mechanical Engineering, University of Technology and Life Science in Bydgoszcz,
Kaliskiego 7 Street, 85-789 Bydgoszcz, Poland
e-mail: adammroz@utp.edu.pl
J. Flizikowski
e-mail: fliz@utp.edu.pl
K. Bieliński
e-mail: kbiel@utp.edu.pl
M. Macko
Kazimierz Wielki University in Bydgoszcz, Chodkiewicza 30 Street, 85-064 Bydgoszcz, Poland
e-mail: mackomar@ukw.edu.pl

146 A. Mroziński et al.
and restore properties of the environment and not only limit technological harm-
fulness. Optimum—coming in property of the machinery design (construction) or
system state with respect to the criteria that enable rational evaluation of the state.
Active, cognitive grinding and monitoring, investigations into multi-disc grinders
demonstrate that it is possible to acquire knowledge of, describe and utilize, for
design and structural purposes, the characteristics that indicate the relations between
speeds, idle movement, loads and the indicators of motion variables in the grinding
space.
The objective of this paper is to provide a mathematical description, optimization
of the states and changes in the grinding recycled materials and machine space,
their surface and volume during movement (idle and working movement) of the
components and design assemblies in the multi-hole grinding process.
2 Process Engineering
The development of the process, devices, system and machinery construction in the
innovation of the plastics engineering depends on a motivation, knowledge, capital
and ready markets. The innovation is carried on methodically, on the basis of the
system mathematical model.
The versatile equation including all the novelty beings in the operating systems,
from idea till elimination, has the form:

L H̄ , Ē, R̄, Θ, t = P(s̄, z̄, Θ, t − t0 ), (1)
where:
H̄ —performance characteristics as output quantities (efficiency),
Ē—inner elements ((nS) construction) and outer elements (ready markets),
R̄—connections of elements (relations, reactions, correlations of elements),
Θ, (t1 − t0 )—time,
s̄—intentional control,
z̄—disturbances.
According to designation, the functional plastics recycling for energy engineering
spheres as technical system—is the whole of its external operating possibility:
• human potential PL (t),
• technical potential PT (t),
• energy—material (plastic) potential PE (t),
• controlling potential PS (t).
Function of operating potential:
Pd (t) = Φ P L (t), P T (t), P E (t), P S (t) ,

(2)
Principle of the Cognitive Grinding of Reuse Materials 147
The following ones belong to indicators describing the operating potential (the
description is limited to controlling potential exclusively, as the basic concept tool
of designer’s activity):
• temporary course of real executive possibilities, π d (t)
• volume of operation used actively, usefully M d (t)
• theoretical possibilities and operations needs, ε,
and especially:
Pd (t) = πd (t) · Md (t) · ε. (3)
Operating (energy) potential equation in the period [t0, T]:
T T T
Pd (T ) = Pd (to ) − pdE (t)dt − pds (t)dt + pdo (t)dt (4)
to to to
where:
Pd (t o )—initial operating potential,
pdE E(t)—density of effectively used stream of potential,
pds (t)—density of lost stream of potential,
pdo (t)—density of recovered (or obtained from the environment) stream of
potential.
Taking energetically plastics grinding for energy engineering aims into account
we obtain:
T T T
E s o
Pem (T ) = Pem (to ) − pem (t)dt − pem (t)dt + pem (t)dt (4a)
to to to
where:
Pem (t o )—initial energy-material potential (e-m) of plastics grinding system,
pEem (t)—flux density of effectively used e-m raw, plastics potential,
psem (t)—flux density of wasted and lost e-m plastics potential,
poem (t)—flux density of e-m recreated plastics potential, (or only retrieved from
environment).
The principle of the cognitive control support in the direction of getting the
extreme solution can be defined:
X∗ ∈ ϕ : ∧ Z (x) ≥ Z X ∗ ,

(5)
x∈ϕ
in the case of minimization of energy consumption (Z = eR or Pem (T ))

X∗ ∈ ϕ : ∧ Z (x) ≤ Z X ∗ ,

(6)
x∈ϕ
in the case of maximization of energetic milling indicator (Z = eR).

If the target point is known in the target space (e.g. eR < 75 kJ kg−1 and/or eR >
60), it is possible to conduct the procedure aiming at approaching the given solution.
The procedure means searching for such δS, δER which are expressed by the
following formula:
Smax − Smin E Rmax − E Rmin

δS = ⇒ 0, δ E R = ⇒ 0, (7)
Ssr E sr
This way a new objective function is obtained. It is in the form of the distance
between the target condition and the countess condition in the target space:
Z d (x) = Z min − Z (x), (8)
where:
Zd —the distance between the solution quality vector Z(x) and the target solution
Zmin .
In the case of Euclidean norm, the distance is expressed by the following formula:
1
2
Z dl (x) = [Z imin − Z i (x)]2 , (9)

i=1
where:
Zimin —the value of unitary energy consumption for the target solution,
Zi (x)—the value of the unitary energy consumption for the designed solution.
The grinding process is described by a number of indicators such as: efficiency,
reliability, outcome, product quality. Each indicator requires different sources of
information.
In the grinding engineering, there are three main groups of objects (Bieliński and
Flizikowski 2013; Ministry of Economy 2009):
• material that is considered in three states:
– at input,
– during mechanical processing,
– at output,
• machine, as a functional system that may include:
– drive motor (motors),
– mechanical gear,
– grinding unit,
• process considered with respect to categories related to:
– energy,
– cost-efficiency,
– environmental aspects,
– organisational aspects,
– time,
– social aspects.
The above shows that grinding is a multi-stage process that requires detailed
knowledge on properties of specific groups of objects: the material, the functional
unit and the grinding process itself.
Grinding process indicators. In the case of biomaterials recycling, the
energy-related indicators that significantly affect the grinding process include
(Flizikowski 2002, 2008; Flizikowski, Bielinski 2013; Macko et al. 2011; Powierża
1997; Tomporowski 2012a):
• total energy of fracture propagation,
• crack stress,
• crack resistance,
• load during collision and cutting,
• collision duration,
• digestibility of organic material, performance ratio of ground product incineration,
• relation of dimensions before and after the grinding process,
• increase of specific surface area.
Many theoretical works known as theories have been published on the subject
including theories by: Griffith, Behrens, Rumpf, Schonerta, Kerlin, Flizikowski.
Figure 1 shows a schematic diagram showing variables, constants and disruptions
taken into account during active monitoring of the grinding process.
Choosing a strategy for active, cognitive monitoring and grinding of technical
objects and must take into account models and objectives, for example, the highest
efficiency of the grinding process and final product quality (e.g. optimal) (Bieliński
and Flizikowski 2013):
e E = f (Po , Pe , Ps , Pod , E j , Onq ) (10)
where:
Po —initial potential,
Pe —effectively used potential,
Ps —ineffectively lost potential,
Pod —potential recovered from technology or the surroundings,
E j —unit energy consumption,
Onq —low quality product, waste, loss, defect operations etc.
THE SURROUNDINGS (ENVIRONMENT)
TECHNOLOGICAL
AND ENERGY- SECONDARY
RELATED NON- MONITORING ZONE SM GOODS
RENEWABLE SERVICES
GOODS
RAW SAFETY OF
MATERIALS GRINDER ST SUPPLIES
Czynniki stałe
Cs
NEEDS
MATERIALS Cg TOP gran_C AQ ?P COMPENSATION
Zmienne niezależne Cm ER
Yot = Xst Y s t = X ot
Zmienne zależne
Cr eR
TECHNOLOGICAL ? max ?S
RELATED GOODS
RENEWABLE PRODUCTS
GOODS
Cz
SUN Czynniki zakłócające POWER
WASTE ENERGY
Fig. 1 General diagram of elements and relations between the grinding system (ST), the
environment (OT), the monitoring cognitive zone (SM = SG) (boundary zone)
As a result of the development of modern IT technologies, active monitors, opti-

mization numerical methods (statistical and deterministic) have displaced previously
used analytical methods. The principle of operation of deterministic methods is to aim
at the optimum value by using appropriate algorithms. The best results are obtained
by an effective combination of both methods.
The technical system (ST) includes the grinding unit (Fig. 2) connected directly
by a drive shaft or through the gear to an electric motor (motors).
Each drive motor is equipped with a dedicated inverter that controls its opera-
tion. The technical system also includes physical quantity measurement units and
necessary systems:
• for material transport and feeding,
• wiring systems (power supply, measurement, control, data transmission, safety,
visualization and alarm),
• ground material collection, transport and packing system.
The boundary zone (SG) is made up of programmes, algorithms, executive units,
modules of active monitoring system (SM) used to identify, acquire, analyse, verify
and exchange data and to configure the system, control inverters of motors that
drive the grinding unit. This zone also consists of modules and tools used to analyse
data that supplements the monitoring system, such as image identification, Genetic
Algorithms, MS EXCEL, STATISTICA.
THE SURROUNDINGS (ENVIRONMENT)
TECHNOLOGICAL
AND ENERGY- MONITORING ZONE SM SECONDARY
RELATED NON- GOODS
RENEWABLE SERVICES
GOODS REPORTS AND CONFIGURA-
DRIVERS DATABASE EXPLORER
GRAPHS TION
RAW SAFETY OF
MATERIALS SUPPLIES
GRINDER ST
NEEDS
MATERIALS
COMPENSATION
On-line, Off-line
On-line, Off-line
Yot = Xst Yst = Xot
TECHNOLOGICAL
RELATED GOODS
RENEWABLE PRODUCTS
GOODS
SUN POWER
OTHER
GENETIC MS EXCEL, DATA
ANALYTICAL CONTROL
ALGORITHMS STATISTICA EXCHANGE
MODULES
WASTE ENERGY
Fig. 2 Detailed diagram of elements and relations between the grinding system (ST), the
environment (OT), the extended monitoring zone (SM = SG) (boundary zone)
When implementing a monitoring concept adopted to increase knowledge about

the recycled and reuse materials grinding process, the following must be chosen as
part of the technical infrastructure selection procedure: information sources, types of
measured quantities, interval values for measurement and measured data recording,
implementation of simple and complex algorithms for aggregation of data. Reference
values must also be defined for further analysis and standard and custom reports on
test results must be pre-defined, for those interested in achieving the highest efficiency
of the grinding process.
3 Cognitive Motional Characteristics
Cognitive, motional, usable characteristics and multi-disc and multi-hole grinding

outcome variables: power demand (PR = f (n)), degree of fineness (λ = f (n)) and
mass target efficiency (Qm = f (n), Qc ≤ Qm ) depend on the common area of the edges
of two holes (S c , S T ), density and volume of grain in the working space (ρ m , V g ),
rotational, angular and linear speed of a component and time (n, ω, v, Θ, t i ) − L(PR ,
′
λ, Qm , Qc ) = P(Sc , ST , ρ̃nm+1 , Vḡ , n, ω, v, Θ, ti ); they also depend on the volumetric
dosing of mass feed q(0;1). The usable characteristics of grinding, dependent on
the movement of grain and grinder components were named the motional character-
istics of grain (Flizikowski and Co-authors 2005; Flizikowski 2011a; Knosala and
(a) (b)
Fig. 3 RWT-5KZ multi-hole five-disc grinder working unit and space (Flizikowski 2011b;
Tomporowski 2012b): a grain filling in two adjoining working holes of the quasi-cutting unit:
Tn−1 do Tn+2 —subsequent grinding discs, hG —height of material column before the cutting plane,
hD —height of material column behind the cutting plane, VG g —calculated volume of material before
the cutting plane, VD
g —calculated volume of material behind the cutting plane, SC —common area of
the quasi-cutting pair of holes; b cross-section of the multi-disc unit: 1—bearing, 2—grinding disc
(so-called “preceding” disc), 3—grinding disc (so-called “subsequent” disc), 4—body, 5—shaft,
6—pulley
Co-authors 2002; Niederliński 1987; Sidor 2006; Tomporowski and Opielak 2012;
Zawada and Co-authors 2005; Ziemba et al. 1980).
To the active, cognitive grinding monitoring, determine motional characteris-
tics of reuse grain and grinder working unit, two states were assumed which are
dependent on the linear speed of the grinding holes edges (Fig. 3) (Flizikowski
2011b; Tomporowski 2012b): the first one—idle state, involving only movement and
mixing, exclusive of grinding (linear speed of points on edges—below 0.7 m s−1 ),
second—working state, with significant grinding initiators (above 0.7 m s−1 ).
The positioning of grains being ground in the working space of a multi-hole
grinder (Fig. 3a) is described by statistical distribution of its length. Because the
material present in the holes of this same disc is characterised by the same particle
size ρ and is subject to the same grinding and cutting process in each hole, its state
for the purpose of this analysis is indexed with the cutting number (m) and the disc
number (n) (Tomporowski, 2011a, b):
lmax
ρnm : [0, lmax ] → [0, 1], ρnm dl = 1. (11)
0
The thickness of the nth disc, for the purpose of this analysis, is marked with the
symbol yn , and the height up to which the material fills the hole in the nth disc prior
to the kth cut by the symbol ỹn(k) .
In efficient grinding and cutting, length distribution of grains which filled the
empty space in the (n + 1)th disc changes as per the following dependency (Fig. 3a):
lmax
m x 1
ρ̃n+1 (x) = An,m ρnm = 1− m
m
ρn (x) + m ρnm (l)dl,
yn+1 − ỹn+1 yn+1 − ỹn+1
x
(12)
while of those left in the nth disc as per:
lmax
x 1
ρ̃nm+1 (x) = B̃n,m ρnm m
= 1 − m ρn (x) + m ρn (l)dl, (13)
ỹn ỹn
x
where:
A, B—stochastic operators for mth cut, nth disc.
It was assumed for simplification purposes that, subsequent to grinding, distribu-
tion of granulated product in the hole spaces of the (n + 1)th disc will be uniform
(cut fraction and that present in the hole before cutting will mix) and it will therefore
k
be the weighted average from ρn+1 i ρnk :
m m
m ỹn+1 m−1 yn − ỹn+1
ρn+1 (x) = ρn+1 + An,m ρnm (x) (14)
yn+1 yn+1
During the modelling of the common part surface, integration of grinding momen-
tary cross-section was employed (Flizikowski and Co-authors 2005; Flizikowski
2011b; Tomporowski 2011b):
x2 x2
2
2 1/2
1/2
b1 − R12 − (x − a1 )2

SC = b2 + R2 − (x − a2 ) dx − dx
x1 x1
(15)
where:
a1 , a2 , b1 , b2 —C1 and C2 hole centres coordinates,
R1 , R2 —holes radius vector.
Based on what has been said, distribution of grain length in ground material which
filled the empty space of the (n + 1)th disc changes as follows:
lmax
m x 1
ρ̃n+1 (x) = An,m ρnm = (1 − D )ρnm (x) + D ρnm (l)dl (16)
h h
x
whereas in the material left within nth disc in the following way (analogical
reasoning):
lmax
x 1
ρ̃nm+1 (x) = B̃n,m ρnm = (1 − D )ρnm (x) + G ρn (l)dl (17)
h h
x
It must be remembered that the column of the material being cut is not the entire
material that has been moved to the lower hole. Its volume is:
S c · hD whereas that of the entire material moved from the preceding hole to the
subsequent hole is VgD (αc ) − Vnm .
It means that the second and third component in (16) must be multiplied by the
relation between these volumes:
Sc x
ρ̃nm+1 (x) = B̃n,m ρnm = (1 − m
m )ρn (x)
VgD (αc ) − Vn+1
lmax
Sc
+ D m ρn (l)dl, (18)
Vg (αc ) − Vn+1
x
It was assumed for simplification purposes that, subsequent to cutting, the grain
length distribution in the (n + 1)th disc will be uniform (cut fraction and that present
in the hole prior to cutting will mix), and therefore it will be the weighted average
m−1 m−1
from ρn+1 and ρ̃n+1 .
m m
m Vn+1 m−1 V D − Vn+1
ρn+1 (x) = ρn+1 + Bn,m ρnm (x) (19)
VD VD
The filling level of the quasi-cutting unit and thus the efficiency of the cutting
process depend on the value of the function V D , V G and S c which in turn depend on
the direction of the effective gravitation and on the total volume of material in both
m
holes before cutting (Vn+1 + Vnm ).
Other theory, models and indicators of variables are provided based on optimiza-
tion tests.
With the stabilised motion of grinding components ((n, ω, v) = const), for the
analysed scope of polymer waste granulate movement and with the use of functional
models (unit energy consumptions and modified Froude number), it is possible to
acquire knowledge of and design mutual relations within the multi-hole and multi-
disc system of the grinder working unit: filling of the transport and grinding area,
power, process target and mass efficiency as well as linear (circumferential) and
rotational speeds of operating discs (Fig. 4).
Fig. 4 New structure of grinding conditions W: PP-start and products material features Wm ,
grinding process features Wp , and design of machine elements Wd (Openings, holes = 5,
= 10)
4 Cognitive Compensation of the Technical System
Figure 5 shows a block diagram representing the operational algorithm of the cogni-
tive measurement system of the environmental optimization of the recycled and
reuse grinding process (compensatory monitoring of the energy/material parame-
ters of the grinding process). On the basis of this algorithm, the system design
Fig. 5 The algorithm of cognitive compensatory monitoring of energy/material parameters of

special grinding systems
is presented: the process structural system (KP)—technical and control structural

system (KC)—that handles the system input. The system of compensatory moni-
toring of the energy/material parameters of the grinding process was created on the
basis of an IT concept that consists in the centralization of all measurement data and
their calculation in a single database, with the application that manages the system
works in client-server arrangement.
Monitoring the energy/material parameters of the environmental recycled and
reuse materials grinding (Fig. 5) is a precondition for the understanding the dynamics
of the key phenomena occurring in special disc grinders (KP).
At the same time, measurement data recorded in the database of the system may
be used as input data for advanced analytic (genetic) algorithms, whose purpose is
to generate signals that automatically control (KJ) the operation of the grinder. A
concept was developed and a system built for monitoring energy/material parameters
of grinding.
5 Solutions of Cognitive Grinding
It was assumed in the experimental verification that the trajectory of grain moving in
the area between the radial lines of the external and internal cone (holes positioning
envelope) is a continuous line—in terms of movement and mixing (first state) and
continuous and disrupted (for the duration of grinding)—in the case of the second
state (efficient edge speed, for e.g. corn: vR > 0.7 m s−1 ). As the speed of grains during
movement and mixing is low and because grains are stopped for the duration of quasi-
cutting and there is a potential increase in the flying movement of grain specks as
a result of bouncing subsequent to grinding, it can be assumed for simplification
purposes that the second state has the motion trajectory and time equal to t p+m+r =
t p+m .
Figure 5 illustrates the operation of the solution for cognitive grinding, active
monitoring energy/material characteristics and parameters of the grinding process
within the target structure. The basic design of each monitoring system includes
an object layer, a data concentration and registration layer, a computer and data
transmission layer, and application layer including a database and an application
managing the system.
An object layer consists of motors (KC) that drive grinder discs (KP) and a system
for power electronic control and measurement of physical parameters (KC = KJ)
(Flizikowski 2011a; Zawada and Co-authors 2005; Ziemba et al. 1980).
The machine—5-disc recycled and reuse materials grinder (KP) is cognitive
controlled, regulated and compensated by a system of pDrive inverters (KC = KJ)
installed in one control cabinet—SR 230/400 V 63A IP44 (KL) with the controller
(KC) of the DSK material feeder (KL).
6 Solutions and Results Analysis
With the cognitive stabilised motion of grinding components ((n, ω, v) = const),

for the analysed scope of plastics grain movement and with the use of functional
models, it is possible to acquire knowledge of and design mutual relations within
the multi-hole and multi-disc system of the grinder working unit: filling of the trans-
port and grinding area, power, process target and mass efficiency as well as linear
(circumferential) and rotational speeds of operating discs.
The systematised environment and technical system characteristics of idle running
and load, based on specialist calculations and investigations into recycled and reuse
materials grinders, indicate that the filling of the quasi-cutting unit, and therefore effi-
cient, power demand and energy consumption in the granulate/re-granulate cutting
process, depend on the values of operating speeds of both quasi-cutting and feeding.
These, in turn, depend on motional and drive parameters of individual grinding units
and the sum of momentary cross-sections, material volume in adjacent holes, making
up a grinding unit. The grain and granulate motion characteristics, at various stages
of disintegration and movement, depend on the common areas of the preceding and
subsequent holes and their filling level with material both before and behind the
cutting plane.
Applying a practical approach related to the structure and operation of recycling
machines–assuming a environment and design solutions (as a logical conjunction of
criteria and structural features of a quasi-cutting unit) within the conceptual space,
providing an optimal solution from the point of view of the selected criteria including
objective, minimum power, auto-adjustment and multi-level structure–it is possible
to propose a new pro-developmental solution with regard to further analyses of the
integrated grinding system in the field of permissible variability of structural features
and processing parameters.
7 Conclusions
The methodology of calculations and examination of the characteristics of recy-

cled and reuse granulate, grain motion, for idle and loaded grinding, may lead to
improvement and development of processing machines.
The active, cognitive grinding, monitoring selected characteristics of grain motion
point to the need for reaching a compromise between the two basic functions: move-
ment and grinding within the intra-hole working space. Proposed and partly verified
models will facilitate selection of optimal structural features and multi-disc grinding
process parameters. It is a useful and desired course, resulting ultimately in obtaining
a high energy product with a defined form, structure and repeatable dimensions.
The analysis of the current studies and structural basics of PVC-, PP- and PE-waste
grain grinders, as well as detailed mathematical descriptions of the grinding process in
relation to the structure of disintegrating units confirm the possibility of development
and experimental verification of mathematical models useful for cognitive control,

optimisation (modernisation and advancement) of multi-disc grinding structures.
Models and corresponding mathematical dependencies facilitate efficient designing
and planning of multi-hole grinding systems utilisation.
For the purpose of the design of special cognitive systems machines, devices
and lines for recycled and reuse materials (biological and recycles corn) grinding
and the compilation of statistical data, procedures for active, environmental and
compensatory monitoring of grinding parameters were proposed. The algorithm that
enables identification of: product quality in technology; control based on product
quality indicators, the actual state of effectiveness of mechanical processing and
harmlessness of the energy system effects (in CO2 emission equivalent units: 960.2
MgCO2 /GWh of saved electric energy). Procedures, based on the active, cognitive
monitoring algorithm, were used to support innovation of creative activities, i.e.
environmentally friendly actions, that is those actions that produce specific positive
environmental results in the energy-related, clean, recycling mechanical processing
of recycled and reuse materials and other energy corn materials.
Cognitive controlled technologies and activities lead to standards described as:
clean processes, high quality materials and products, effective use of energy in oper-
ations, and energy saving. The analysis concerned a multi-disc grinder of environ-
mentally enhanced design at various stages of development including: industrial
testing stage, prototype testing stage, certification stage, implementation and mass
production planning (developmental implementation).
For the purpose of rough evaluation and compensation of structural system opera-
tional parameters, the basic criteria of eco-innovation were applied: energy consump-
tion, material consumption, operational efficiency and effectiveness, the overall waste
balance (waste balance, waste produced).
References
Flizikowski JB (2002) Dissertation on construction, [in Polish]. ITEE, Radom, Poland

Flizikowski JB (2008) Plastic devices for energy engineering. NANOENERGY, Tapis Rouge, Paris,
France
Flizikowski JB (2011a) Intelligent grinding system. In: Chemical engineering and equipment no
3/2011. SIGMA-NOT Sp. z o.o., Warsaw, Poland, pp 22–23
Flizikowski JB (2011b) Levels of intelligent grinding system. In: Chemical engineering and
equipment no 3/2011. SIGMA-NOT Sp. z o.o., Warsaw, Poland, pp 24–26
Flizikowski J, Bielinski K (2013) Technology and energy sources monitoring: control, efficiency,
and optimization. IGI GLOBAL USA 2013; ISBN13: 9781466626645, ISBN10: 146662664X,
EISBN13: 9781466626959, pp 248
Flizikowski JB, co-authors (2005) Implementation project of intelligent system in special mills
construction. MNiI, ATR-AGH, Warsaw, Bydgoszcz, Krakow, Poland
Knosala R, Co-authors (2002) Application of artificial intelligence in production engineering [in
Polish]. Warsaw, Poland
Niederliński S (1987) System and control, [in Polish]. PWN, Warsaw, Poland
Ministry of Economy (2009) Polish energy policy until 2030. Warsaw, Poland
Macko M, Boniecka M, Drop A (2011) Life cycle assessment of grinders rising SoliWorks sustain-
ability application, [in Polish]. Chemical engineering and equipment no 3/2011. SIGMA-NOT
Sp. z o.o., Warsaw, Poland, pp 49–50
Powierża L (1997) Outline of bioagriculture engineering system, [in Polish]. Part I. ITEE, Radom,
Poland
Sidor J (2006) Studies, models and methods of designing vibratory mills, [in Polish], vol 150.
Publisher AGH ROZPRAWY MONOGRAFIE, Krakow, Poland
Tomporowski A (2011a) Structure development of biological material shredders. Part I and II [in
Polish]. Chemical Engineering and Equipment no 3/2011. SIGMA-NOT Sp. z o.o., Warsaw,
Poland, pp 75–78
Tomporowski A (2011b) Effectiveness of drive and innovative construction solutions of multidisc
grinders for biomass grain—case study, [In Polish]. LTN, Lublin, Poland
Tomporowski A (2012a) Filling model for the working multi-disc biomass grain grinding unit.
In: The archive of mechanical engineering, vol LIX, no 2. Poland: Polish Academy of Science,
Committe of Machine Design, Versita, Warsaw, pp 155–174
Tomporowski A (2012b) Stream of efficiency of rice grains multi-disc grinding. Maintenance Reliab
14(2):150–153
Tomporowski A, Opielak M (2012) Structural features versus multi-hole grinding efficiency.
Maintenance Reliab 14(3):223–228
Zawada J, Co-authors (2005) Introduction to mechanics of machine crumbing process, [in Polish].
ITE, Radom-Warsaw, Poland
Ziemba S, Jarominek Wł, Staniszewski R (1980) Problems of systems theory, [in Polish].
Ossolineum, PAN, Wrocław, Poland
Smart-Tracking Systems Development
with QR-Code and 4D-BIM for Progress
Monitoring of a Steel-Plant
Blast-Furnace Revamping Project
in Korea
H. Y. Roh, E. B. Lee , I. H. Jung, and C. Y. Kim
Abstract Blast furnace revamping in the steel industry is one of the most important
works to complete the complicated equipment within a short period of time-based
on the interfaces of various types of work. P company has planned to build a Smart
Tracking System based on the wireless tag system to comply with the construction
period and reduce costs, ahead of the revamping of blast furnace scheduled for
construction in February next year. It combines detailed design data with wireless
detection technology to grasp the stage status of design, storage, and installation.
Then, it graphically displays the location information of each member to the plan
and the actual status in connection with Building Information Modeling (BIM) 4D
Simulation. QR Code is used as a wireless tag to check the receiving status of
core equipment considering the characteristics of each item. Then, DB in the server
system is built, status information is input. By implementing BIM 4D Simulation
data using DELMIA, the information on location and status is provided. In terms of
logistics digitization, the system’s features allow suppliers to monitor the real-time
status of the 4D system provided on a Web basis, enabling suppliers to accurately
identify and supply the delivery times for the facilities they need to manufacture and
supply. Besides, the project progress control can be managed quickly and accurately
by identifying the location of major delivered facilities in real-time to maximize
the efficiency of construction, and by identifying the status information of major
facilities, i.e. installation status in real-time. This digitalization ultimately results in
savings in the manpower involved in the project and contributes to lower investment
costs as a whole.
Keywords Wireless tracking system · Blast furnace revamping · BIM 4D ·

Real-time detection of facilities
H. Y. Roh · E. B. Lee (B) · I. H. Jung · C. Y. Kim

Pohang University of Science and Technology, 77 Cheongam-Ro, Nam-Gu, Pohang, Korea
e-mail: dreblee@postech.ac.kr

162 H. Y. Roh et al.
1 Introduction
An active wireless recognition system (based on QR-code) in a 4D-BIM environment

has been proposed. In connection with the new aging of the population, the construc-
tion period of the mid-term facility replacement business is increasing due to the aging
of construction personnel and the lack of skilled workers. As a result, productivity
is expected to decrease naturally. This deterioration is expected to be more severe in
blast furnace repair. Steel plant features consist of various facilities and construction
processes (instrumentation, machinery, electricity, civil engineering, and construc-
tion). In other words, it is composed of various equipment processes, so it takes a
lot of time and money to rework when errors and corrections occur. Therefore, it
is essential to use primary data for IT-based system data accumulation and repro-
ducibility analysis to prevent design and construction errors and to reduce costs based
on technology accumulation. In this chapter, we propose a design and construction
system that integrates wireless tracking technology in a BIM environment for a new
concept of technical grafting. The main goal of the system is to improve consistency
through maintaining data consistency at the design-purchase-construction stage in
the EPC engineering stage and eliminate errors by linking with the BIM technology.
The proposed system can input step data for equipment in real-time after the produc-
tion of the product by the manufacturer. In the case of the yard, it has been input
the data for four areas for the location and checks the detailed material information
through tagging in each step to prevent errors due to crosstalk. Furthermore, the 4D
viewer can check the status and location of the facility in real-time, and can check
the preparation status of the target facility in advance in connection with the 4D
Schedule, and has the advantage of easy real-time tracking. With these advantages,
problems of scheduled work can be identified in advance to minimize errors.
2 Related Work
2.1 Conventional Wireless System Technology
For the efficient and automatic inventory management system of logistics and distri-
bution, we mentioned the need for wireless tagging technology. Several examples
of wireless tagging-based technology are introduced, and methods using RFID or
Bluetooth are introduced. First, for inventory automation of outdoor warehouses,
we proposed a system that tracks and manages products with unmanned drones by
attaching Bluetooth Low Energy (BLE) and beacon technology to products. However,
the results of this study needed to derive actual results for the problem of beacon
distance recognition according to the drone’s route (Bae et al. 2018). Similar tech-
nologies have been introduced mainly for logistics management and indoor location-
based system research of smart factories based on Bluetooth beacon communication
(Am-Suk 2015; Park et al. 2015). Also, the business model components are organized
Smart-Tracking Systems Development with QR-Code … 163
by classifying the cases of presenting the To-Be model for companies participating
in the international logistics process using RFID wireless tagging technology (Choi
et al. 2010; Jang et al. 2007) and examples for the activation of the RFID tech-
nology industry. There are techniques presented (Joung and Kim 2013). Besides,
a technology that automatically updates product information to a database using a
mobile-robot and an unmanned mobile robot based on RFID wireless communication
was introduced (Son and Do 2011).
2.2 The Advantages of Proposed System
In this paper, it has been introducing the technology to visualize products in the 4D
BIM environment of the process based on the QR code wireless tagging technology.
The advantages of QR codes can be widely used in manufacturing, distribution,
logistics, and marketing. On-site product information and ordering information are
automatically converted into digital systems by QR code recognition, which is very
easy to configure as a database. The converted database can be used to order and
inspect products, and it is easy to collect and track information about products sold or
received. This advantage of QR code helps you to search for distribution and distribu-
tion systems effectively and quickly. In other words, it can accurately provide users
with information about products sold or received (Park 2013; Gu and Zhang 2011;
Yoon et al. 2011; Dey 2013; Goyal et al. 2016; Bhargava et al. 2014; Ramsden 2008).
The QR code can also be used for construction management and informs workers,
including construction and construction procedures so that they can quickly search
for construction-related information. Additionally, worker information may include
the company the worker belongs to, whether safety training has been completed, and
the extent of the approved work. To this end, a QR code is attached to the worker’s
helmet so that the manager’s work information can be checked immediately. Due to
the attachment of the QR code, the construction manager does not have to visit the
construction site directly with the construction documents and has the convenience
to check the current status of the work (Lorenzo et al. 2014).
Where, the smart tracked product information is stored as a central server that
manages the BIM platform and database, and the visualization result is derived
based on the data stored in the server. The derived result can be accessed in the form
of a separately developed UI (User Interface) for the end-user. Central server access
through the designed UI can be accessed using a smartphone or PC. That is, based
on the IP allowed access to the central server, you can access anytime, anywhere,
through a device that can use the Internet. This enables process managers to have
very convenient and efficient management in process control and supervision.
3 Smart Tracking System
3.1 The Wireless System Modeling
Product QR code information recognition of the proposed system captures product

QR code information in the form of a smartphone device or mobile-robot in the field.
The captured data is transmitted to the central server and stored through the Web,
and can be linked to the 4D BIM environment at the process site through the CM
(Construction Management) Platform. Where, the QR code is scanned based on the
QR code data collected and taken. If the target QR code is treated as unconfigured
data, it is switched to the main screen with a search failure notification. On the other
hand, once the QR code is recognized, it is divided into an input-capable case and an
input-incapable case according to the accepted QR code classification. An example
of a classification system for identified data is a process-related data classification
system such as delivery, receipt, storage, equipment, and inspection. And the overall
system configuration of the proposed system is shown in Fig. 1.
The applicable range of the designed system is the EP equipment, the main unit,
the furnace, and the gas cleaning class supply EP equipment for the 1st runner of
furnace facility. QR code is attached to the recognition object, and the QR code
information is DB. Afterward, a UI-type QR-code operation screen is provided for
end-users. The following figure is an example of the main screen of the developed
UI (Fig. 2).
Where, the definition of each function is as follows:
• A: Project information
• B: Facility information
• C: Supplier information
• D: Function button for each status
• E: Search and Exit button.
Fig. 1 The proposed system

Fig. 2 The main screen of

user interface
And it can check the delivery status of the product by date through the next screen
in more detail (Fig. 3).
• A: Unstoring date and status
• B: Unstoring button
• C: Warehousing date and status
• D: Warehousing button.
Fig. 3 The main screen of delivery status of the product

Table 1 The required data

Division Property Data
list
Project Project period Contact number of P
information company
Project name Contact number of
contractor
Contractor NULL
Equipment Item name Count
information Picture of item Weight
Packing tracking Supply flag
number
Tag code Supplier name
Class number Packing number
Drawing number Contact number of
supplier
Status Un-storing date Install status
information Un-storing status Installation location
Warehousing date Measurement date
Warehousing status Measurement status
Open storage date Installation
completion date
Storage location Installation
completion status
Install date Install status
In other words, by using the provided interface, you can conveniently check the
progress and schedule of the entire process, and track the status of the operator.
Where, the contact number is used when contacting the P company’s project
management (PM), construction management (CM), or business personnel. Project
information indicates information about construction information. The construction
manager can quickly identify documents by construction by checking the document
number referring to database information. Information such as product quantity,
weight, and drawing drawings of the product are useful when checking whether
the product information matches the order. Besides, the status information includes
the progress status of each construction job and the date showing the history. The
main entities derived from the inputs listed in Table 1 are project equipment and
contractors. The main objects are project equipment and contractors. For security
reasons, if the user scope is restricted, a user entity is added, and the figure below
shows the relationship structure according to the substance.
In Fig. 4, the red attribute is primary, and the information marked with green
sequel means the foreign key. And the interface table is implemented based on the
diagram in Fig. 4.
Fig. 4 The diagram of entities relationship
3.2 The 4D BIM Simulation
When the database configuration work is completed in the interface table through
the wireless system modeling work, the collected items are periodically uploaded
to the existing BIM platform DB. In order to link the useful tracking function with
the 4D BIM environment, 3D modeling work must be preceded. The basic format of
3D modeling work is derived as a 3D XML file and is additionally compatible with
data such as.dxf and.dwg (Dassault system based modeling). As mentioned earlier,
in this study, the smart tracking system is integrated into 3D and 4D environments to
provide users with information on construction items quickly. Therefore, the drawing
status information uploaded to the BIM platform database must be changed to color
information. For this, the essential functions of B. I (Business Intelligence) supported
by the 3D experience application are used. B. I function helps construction managers
easily identify the necessary information (Bernérus and Karlsson 2016). By utilizing
the B. I. function, 3D models can be displayed in a predetermined color on the open
storage at each stage of installation completion. To link the status information of each
item to the 3D model, you need to connect the 3D model from the BIM database with
the smart tracking system and tag data used to specify the product. The recommended
process is to provide the generated code in advance to the 3D model according to
pre-specified rules when creating a 3D model. That way, you can assign tag codes to
many 3D model objects without having to find something to manage. The 3D model
list is preferentially extracted and displayed as color information, and experts link
tag data to the list. And the final list connected by experts consists of a database
inside the interface table.
4 Experiment
4.1 Wireless System and User Interface
When the proposed system was applied to an actual construction site, an experiment
was conducted to check the efficiency. The experimental method goes through the
procedure where the construction site user utilizes a smartphone to recognize the
QR code and configure the known QR code as a database in the interface table.
As mentioned above, any device that can use the web environment does not matter.
When a database recognized as a wireless system is configured, the 3D modeling
database is also set on the same server in the manner mentioned in Sect. 3. When
the databases required for the experiment are configured, the engineer can check
whether the smart tracking function and data configuration procedure are correctly
performed in the server environment and the reliability of the set database. For the
experiment, there are 3D experience licenses and some necessary provisions, and the
table below lists them.
After the necessary provisions in Table 2 were all composed of data in the interface
table, the results of the reliability verification experiment were confirmed, and the
figures below show the experimental results.
Figure 5 shows the databases organized in the interface table (job information,
product information, construction schedule, wireless system tagging information,
etc.). And Fig. 6 shows the progress of a specific construction schedule (visualized
in the form of a bar at the top of the UI) and connection attribute information with
construction manager information. Looking at Fig. 5 first, the QR code and product
information are visualized when the active Downloads button on the right tab of the
Table 2 Check list

Division Description
Wireless system Connection via QR code
Direct connection not through QR code
Check of Item search result
Entering status information according to scenario and check of result
Functionality of main page
Equipment information Display of status information of items in 3D model
Check of location information of item in 3D model
Display of status information of items in 4D model
Fig. 5 The result of interface table data check
screen is pressed, based on the QR code captured by the smartphone in the field. If it
is the job information, the status information window for the job is visualized, and if
it is the job worker information, the necessary knowledge of the worker and whether
or not the job safety training is completed are visualized. By checking the derivation
results in Fig. 5, the job manager can quickly and accurately review and correct the
site conditions and progress of construction work. Next, the results of Fig. 6 are first
visualized the progress of the construction work in the form of a bar at the top, and
it can be confirmed that the call connection function is activated by recognizing the
connection information with practitioners for each operation. In other words, the
general manager in charge of construction reduces the task of separately managing
contact information of each worker, so that it is possible to control the work process
schedule quickly and accurately.
When the proposed system is applied to the site through the results of experiments,
hundreds, thousands of items are targeted, unlike experiments involving 13 issues.
The proposed smart tracking system is a concept of a pilot-test performed before
application to the construction site. And the site application procedure is as follows:
• (First): The items to be managed are selected from all items in the construction
project.
• (Second): Set the product details for the product to be handled. Here, if issues are
managed in too small or large units, it is difficult for the user to grasp the current
status of each item, so it is necessary to select an appropriate group.
Fig. 6 The result of system progress and task manager call connection information check
• (Third): QR codes are distributed to suppliers of things to be managed, and tags

with QR codes are attached to each item. In this case, the attachment method is
attached in such a way that it can be maintained during the maintenance period,
taking into account characteristics such as how to use, size, or environment of
each product.
• (Fourth): Typically, tags printed on paper are used due to price advantages, but
for items placed in extreme situations, use a QR code lasered on an aluminum
plate to prevent the tag from falling or being damaged.
• (Fifth): Suppliers scan tickets and update item status on delivery.
• (Sixth): Experts with sufficient knowledge of each item connect the 3D model to
the tag data of the target item to construct the final interface table database.
Fig. 7 The result of the mechanical drawing applied to the experiment
4.2 Mechanical Drawing
To visualize in a 4D BIM environment, 3D modeling work is required in advance.

In this paper, 3D modeling was performed based on the blast furnace drawing of the
Korean steel company P, and Fig. 7 shows the modeling results.
Where, the left side of Fig. 7 shows the attributes for the mechanical drawing
elements, and the upper right tab indicates the status for each color. The parts
marked in dark pink for each color represent securely separate elements. White
is the current process. Yellow and dark blue show the shipping and loading status
of the process product, respectively. Orange means the yard, and purple color is the
assembly completion process. Dark green indicates the installation status; light pink
indicates the completion status, and finally, bright green indicates the installation
status. The drawing results in Fig. 7 represent each color based on the process data
collected by the wireless system and stored in the interface table database. In other
words, by checking the process status with drawings, data add-on is possible in the
4D BIM environment.
5 Conclusion
In this paper, we proposed a smart tracking system using QR code-based wireless

tagging technology. The purpose of the proposed method is to reduce the cost and
reduce working hours due to the decrease in worker productivity in construction or
various industrial sites by introducing smart construction technology. Furthermore,
by using this system, project managers were able to manage projects effectively. As an
example, it was confirmed that the final administrator could access the project through
the user interface environment by constructing the interface table inside the central
server and providing the user interface environment. Besides, this approach was able
to confirm the construction project’s progress by-product, construction work, worker,
product release, product receipt, and product shape using the Web. The utilized results
were verified through experiments and summarized the necessary processes until
the results of the verified pilot-tests were applied to industrial sites. The proposed
smart tracking system will be gradually introduced into the blast furnace process,
which has been tested, until next year. Also, the proposed intelligent tracking system
architecture will be applied universally beyond the blast furnace process involved in
the experiment. Due to this, in the future, the smart process system can be expected to
decrease worker productivity due to the aging population and ensure efficient work
for managers.
Funding Acknowledgements The authors acknowledge that this research was sponsored by (1)
the Korea Ministry of Trade Industry and Energy (MOTIE/KEIT) through the Technology Inno-
vation Program funding for (1) “Artificial Intelligence Big-data (AI-BD) Platform for Engineering
Decision-support Systems (grant number = 20002806)”, and (2) POSCO for “Smart-Tracking of
Construction Progress for Blast-Furnace Retrofit based on 4D BIM”.
References
Am-Suk Oh (2015) Smart factory logistics management system using house interior position
tracking technology based on Bluetooth Beacon. Korea Inst Inf Commun Eng 19(11):2677–2682
Bae SM, Han KH, Lee HY, Hong KS (2018) Automation of inventory checking system for outdoor
warehouse. J Soc Korea Ind Syst Eng 41(3):138–144
Bernérus E, Karlsson M (2016) Simulation with 3DEXPERIENCE-evaluation of software for
production flow simulation in manufacturing industry
Bhargava N, Kumawat A, Bhargava R (2014) Demonstration of barcodes to QR codes through text
using document software. Int J Innov Res Sci Eng Technol 3(9):16243–16249
Choi H-R, Kim H-S, Hong S-G, Park Y-S, Kim H-Y, Shin J-J, Ha J-S (2010) Improvement of port
entrance system using RFID. J Korea Acad Ind Cooperation Soc 11(2):709–719
Dey S (2013) New generation of digital academic-transcripts using encrypted QR code™: use of
encrypted QR code™ in mark-sheets (academic transcripts). Paper presented at the 2013 inter-
national mutli-conference on automation, computing, communication, control and compressed
sensing (iMac4s)
Goyal S, Yadav S, Mathuria M (2016) Exploring concept of QR code and its benefits in digital
education system. Paper presented at the 2016 international conference on advances in computing,
communications and informatics (ICACCI)
Gu Y, Zhang W (2011) QR code recognition based on image processing. Paper presented at the
international conference on information science and technology
Jang K-Y, Lee C-H, Kim J-G, Lim S-K, Yoo W-S (2007) An empirical study on RFID application
to the container terminal gate management system. IE Interfaces 20(1):69–78
Joung S-I, Kim DS (2013) Case study for RFID applications from business model perspective. J
Inf Technol Appl Manag 20(1):197–216
Lorenzo TM, Benedetta B, Manuele C, Davide T (2014) BIM and QR-code. A synergic application
in construction site management. Procedia Eng 85:520–528
Park E (2013) A study on the application of qr code through case study-focused on logistics and
distribution sector. J Korea Port Econ Assoc 29(4):247–264
Park Y, Back J, Proksa G, Bae K, Do K, Kim J (2015) Attendance check system based on Bluetooth
beacons. Paper presented at the proceedings KICS of international Conference on communication
Ramsden A (2008) The use of QR codes in education: a getting started guide for academics.
Accessible desde 28(10). http://zabdulkhaaliqcapstone.yolasite.com/resources/getting_started_
with_QR_Codes.pfg
Son M-H, Do Y-T (2011) Mobile robot system design for RFID-based inventory checking. J Korea
Rob Soc 6(1):49–57
Yoon H, Park N, Lee W, Jang Y, Woo W (2011) QR code data representation for mobile augmented
reality. Paper presented at the international AR standards meeting

Xem Trang 35 Của Cái Này Có Bảng Các Định Nghĩa Nhé

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Xem Trang 35 Của Cái Này Có Bảng Các Định Nghĩa Nhé

Uploaded by

Copyright:

Available Formats

Accelerat ing t he world's research.

Principle of the Cognitive Grinding

Cite this paper Downloaded from Academia.edu 

Get the citation in MLA, APA, or Chicago styles

Related papers Download a PDF Pack of t he best relat ed papers 

CIRCULAR ECONOMY APPLICAT ION OF REVERSE LOGIST ICS IN T HE RECYCLING OF URBAN T R…

Paulina Golinska-Dawson Editor

Environmental Issues in Logistics and Manufacturing

More information about this series at http://www.springer.com/series/10152

ISSN 2193-4614 ISSN 2193-4622 (electronic)

© Springer-Verlag GmbH Germany, part of Springer Nature 2020

This Springer imprint is published by the registered company Springer-Verlag GmbH,

Poznań, Poland Paulina Golinska-Dawson

Reverse Logistics Flows and Network’s Configuration Problems

Practical Aspects of Application the 3R (Reduce, Recycle, Reuse)

Principle of the Cognitive Grinding of Reuse Materials . . . . . . . . . . . . . . . 145

Abstract The circular economy concept encourages the redesign of industrial

Keywords Logistics service providers · Logistics infrastructure · Circular

© Springer-Verlag GmbH Germany, part of Springer Nature 2020 3

• cradle-to-cradle (McDonough and Braungart 2002),

incremental efficiency gains to deliver transformative change’ (Preston 2012, p. 1).

Fig. 1 The elements of the

In Fig. 1 is presented the summary of the elements of the bottom-up CE transition

2.1 Innovative Practices

The circular economy pushes the frontiers of environmental sustainability at a

transition towards the CE is “predominantly issues of innovation, technical systems,

• reconfiguration, when selected niche-innovations are developed (as regimes face

2.2 Circular Business Models

Summarizing, the business model is implemented:

focuses on the e-business support. It implements the sustainable practices by

approach e.g. (Allwood 2014), the CE practices in material flow management in a

4 Conclusions and Further Research

Usama Awan, Narmeen Kanwal, and Mohammed Khurrum S. Bhutta

Keywords Circular economy · Resource regenerative · Product design ·

© Springer-Verlag GmbH Germany, part of Springer Nature 2020 19

There is an ongoing discussion about the significance and application of CE in a

1.1 Systematic Literature Review Methodology

1.2 The Emergence of the Circular Economy Concept

2 The Definition of Circular Economy Debate

The exponential growth of the CE concept is evident in recent literature, however,

However, how CE is to be understood by the business and consumer marketing

give further insights into our understanding of exo-system, Chronosystem in addi-

Péter Egri, Balázs Dávid, Tamás Kis, and Miklós Krész

Abstract Recycling waste materials has become increasingly important recently

Keywords Facility location · Robust optimization · Economies of scale · Reverse

© Springer-Verlag GmbH Germany, part of Springer Nature 2020 37

1 Introduction and Motivation

In the following subsections we formulate the uncapacitated facility location problem

2.1 Uncapacitated Facility Location Problem

2.2 Economies of Scale Problem

2.3 Robust Optimization Problem

Robust optimization can be modeled as a multi-objective optimization problem,

3.1 Determining the Worst Case Cost Effectively

3.2 Local Search Heuristic

3.3 Tabu Search Heuristic

3.4 Exact Solution of the Robust Problem

3.5 Bilevel Integer Program Formulation

X iπi1 ≥ Yπ′ i1 ∀i ∈ I (13)

Then, the inner maximization problem of (12) takes the form

subject to (13)–(16) and the constraints

Constraints (22)–(24) are the constraints of the inner optimization problem.

As a conclusion it can be observed that including robustness is the most important

Table 1 Average performance using 50 locations

of a larger number of opened facilities (which usually happened with unrealistically