You are on page 1of 22

Automation in Construction 134 (2022) 104064

Contents lists available at ScienceDirect

Automation in Construction
journal homepage: www.elsevier.com/locate/autcon

Applications of fuzzy hybrid techniques in construction engineering and


management research
Phuong H.D. Nguyen a, Aminah Robinson Fayek b, *
a
Department of Civil and Environmental Engineering, University of Alberta, Edmonton, AB, Canada
b
Tier 1 Canada Research Chair in Fuzzy Hybrid Decision Support Systems for Construction, NSERC Industrial Research Chair in Strategic Construction Modeling and
Delivery, Professor, Hole School of Construction Engineering, Department of Civil and Environmental Engineering, University of Alberta, Edmonton, AB, Canada

A R T I C L E I N F O A B S T R A C T

Keywords: Construction engineering and management (CEM) becomes more complex with increasing project size and
Fuzzy hybrid techniques complexity. Various fuzzy hybrid techniques have been implemented to handle subjective uncertainty and
Systematic literature review vagueness in CEM problems. The objective of this paper is to investigate applications of fuzzy hybrid techniques
Content analysis
across different CEM applications (e.g., prediction, decision making, optimization) through a systematic review
Construction engineering and management
of 255 journal articles published from 2004 to 2021. A checklist of selection criteria for choosing an appropriate
fuzzy hybrid technique to solve a specific CEM problem was also established. This study contributes to the body
of knowledge by providing a state-of-the-art review of existing fuzzy hybrid techniques utilized in CEM to (1)
demonstrate their capabilities to overcome limitations of some standard techniques in solving complex con­
struction problems, (2) determine selection criteria for their applications, and (3) examine the applicability of
each fuzzy hybrid technique’s category to given practical construction problems.

1. Introduction demonstrating abilities of the standard techniques and also overcome


their limitations.
Increased project size and complexity as well as project delivery Current CEM literature shows that utilizing fuzzy hybrid techniques
risks, conflicting criteria, frequently inconsistent requirements, and the improves modeling capacity and effectiveness in addressing CEM
dynamic nature of construction lead to highly complex construction problems involving incomplete data, subjective uncertainties, and am­
engineering and management (CEM) problems involving vagueness, biguity [5,8]. For example, the hybridization of fuzzy logic and opti­
subjectivity, and uncertainty. To address these challenges, researchers mization methods can enhance a model’s ability to capture both
have implemented many modeling and computing approaches, such as subjective uncertainty and probabilistic uncertainty in the optimization
optimization, machine learning, multi-criteria decision making process [9]. Integration of fuzzy logic with machine learning methods,
(MCDM), and simulation [1,2]. However, most current applied methods such as artificial neural networks (ANNs), can overcome the latter’s
encounter difficulties in handling incomplete data, subjective judge­ limitation in handling linguistically expressed information from input
ments, and linguistically expressed information in practical construction data [10]. The literature also reveals that fuzzy logic has been hybrid­
problems. Construction researchers have used fuzzy logic – a human-like ized with diverse modeling, computing, and decision-making techniques
reasoning approach – to tackle these challenges and improve modeling in construction research over the past decades [4,7,10]. Therefore,
efficiency [3,4]. Using fuzzy logic alone, however, is limited with commonly used fuzzy hybrid techniques in construction research topics
respect to solving all facets of a CEM problem. Thus, construction re­ can be classified based on particular types and purposes of the applied
searchers use hybridized fuzzy logic with standard techniques, such as computing techniques in construction problems, including optimization,
MCDM, simulation, optimization, and machine learning, to enhance simulation, machine learning, and MCDM. Accordingly, major applica­
their capabilities in performing dynamic modeling and computing pro­ tions of fuzzy hybrid techniques in CEM research can be grouped into
cesses [5,6]. Hybridization is defined as the process of integrating two or four categories: fuzzy hybrid machine learning, fuzzy hybrid MCDM,
more techniques to capitalize on strengths and overcome restrictions of fuzzy hybrid optimization, and fuzzy hybrid simulation. Fuzzy hybrid
the techniques [7]. Therefore, fuzzy hybrid techniques are capable of machine learning techniques are capable of learning from data and are

* Corresponding author.
E-mail address: aminah.robinson@ualberta.ca (A. Robinson Fayek).

https://doi.org/10.1016/j.autcon.2021.104064
Received 29 June 2021; Received in revised form 29 October 2021; Accepted 21 November 2021
Available online 30 November 2021
0926-5805/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

often utilized in CEM modeling problems related to classification and presents conclusions and possible future research directions regarding
prediction [11,12,13,14]. Fuzzy hybrid MCDM techniques are often potential improvements of each fuzzy hybrid category and possible
used to develop decision-aid models or systems that can accommodate applications of fuzzy hybrid techniques with advanced technologies in
various subjective judgements and linguistically expressed information construction.
of decision makers [15,16,17,18]. Fuzzy hybrid optimization techniques
are often implemented to solve multi-objective optimization problems, 2. Literature review
such as time-cost trade-offs, resource usage, and optimal solutions in
CEM research [19,20,21,22]. Fuzzy hybrid simulation techniques are A review of previous studies shows a variety of applications of fuzzy
capable of simulating construction systems (projects or activities), pro­ logic in construction research since Zadeh [36] first introduced it. Many
cesses, and agents, and they are applied to forecast the behavior of these studies have since reviewed and summarized practices of fuzzy set
systems under different conditions [23,24,25,26]. theory, fuzzy logic, and fuzzy hybrid techniques in construction, as
Fuzzy hybrid techniques have been increasingly used to select and shown in Table 1.
support applications of advanced technologies, such as building infor­ Chan et al. [40] provided the first comprehensive review of 52 fuzzy-
mation modeling (BIM), geographic information system (GIS), sensors, related studies from 1996 to 2005, which focused on categorizing fuzzy-
modular construction, alternative reality technologies (e.g., augmented related applications into two broad fields: fuzzy set / fuzzy logic and
reality / AR, virtual reality / VR), and emerging innovations (e.g., big fuzzy hybrid techniques in some common construction domains,
data analytics) in CEM research [27,28,29,30]. For example, Hoseini including decision making, performance, evaluation/assessment, and
et al. [31] utilized a fuzzy hybrid MCDM approach to investigate in­ modeling. Simić et al. [39] reviewed 54 fuzzy-related studies from 1966
fluences of BIM on CEM applications. As a result, the top three BIM to 2016, with a focus on supporting decision making in selection and
features that contribute to project success include parametric capability assessment of suppliers. Their study found that only 12% of the collected
of the 3D model, clash detection, and integrated design and develop­ studies dealt with fuzzy set theory, while 88% of them presented the use
ment of collaboration. Implementing advanced technologies also gen­ of fuzzy hybrid MCDM techniques in various models for selection of
erates a large volume of data and information that can be used with suppliers. As a result, they suggested implementing fuzzy hybrid tech­
modeling and computing techniques, such as decision making, optimi­ niques to solve complex decision-making problems in construction,
zation, prediction, and simulation, to solve practical construction especially in ranking, assessing, and selecting appropriate suppliers. To
problems. However, vast amounts of data collected from advanced improve risk management in construction projects, Islam et al. [38]
technologies often include incompleteness, vagueness, and subjective conducted a review of 83 papers concentrating on fuzzy logic and fuzzy
uncertainty due to the dynamic nature of construction projects [32]. To hybrid techniques published from 2005 to 2017. Differently from Chan
overcome these challenges, researchers use fuzzy hybrid techniques (e. et al. [40], they categorized fuzzy-related construction applications into
g., fuzzy optimization, fuzzy machine learning, fuzzy MCDM, fuzzy three groups: basic fuzzy-related applications with fuzzy set theory and
simulation) to perform data analysis in CEM research that utilizes
advanced technologies. CEM studies related to fuzzy hybrid techniques Table 1
and advanced technologies have been conducted for various processes of Previous fuzzy-related construction literature reviews.
a construction project, such as pre-design and planning, pre-
Reference Journal Paper Title Review Number of
construction, construction, controlling, and closing [27,28,29,30]. Periods Reviewed
Applications of fuzzy hybrid techniques are quickly evolving and Papers
widely implemented in a variety of practical construction problems Review fuzzy
[33,34,35]. However, no detailed content analysis has been done to multi-criteria
demonstrate integrated categorization and itemized applications of Chen and decision making in
Applied Soft
fuzzy hybrid techniques for solving specific CEM problems. Further­ Pan
Computing
construction 2007–2017 165
[37] management using
more, the current CEM literature also lacks guidelines for determining
a network
the applicability of specific fuzzy hybrid techniques to problems within a approach
particular CEM application area. Another research gap exists regarding a Neuro-fuzzy
systematic review of applications of fuzzy hybrid techniques in sup­ Tiruneh
systems in
porting the selection of advanced technologies (e.g., BIM, GIS, sensors) Automation in construction
et al. 2000–2020 116
Construction engineering and
and handling subjective uncertainty and incompleteness of data in [10]
management
different processes of construction projects. To address these existing research
research gaps, the objectives of this study are to (1) investigate existing Current research
and emerging applications of fuzzy hybrid techniques in CEM research, trends and
application areas
(2) determine selection criteria for applications of fuzzy hybrid tech­ Islam Advanced
of fuzzy and hybrid
niques to solve CEM problems, and (3) recommend potential improve­ et al. Engineering
methods to the risk
2005–2017 83
ments for each fuzzy hybrid category by utilizing the hybridization [38] Informatics
assessment of
between fuzzy logic and multiple standard techniques and their appli­ construction
cability to CEM studies that involve advanced technologies in con­ projects.
50 years of fuzzy
struction, as directions for future research. set theory and
The rest of this paper is structured as follows. First, an overview of Simić
Journal of models for supplier
et al. 1966–2016 54
recent applications of fuzzy logic in construction research is given. Then, Applied Logic assessment and
[39]
a systematic literature review methodology is presented, which uses selection: a
literature review
content analysis to highlight applications of fuzzy hybrid techniques in
Overview of the
CEM research. This is followed by the integrated categorization of fuzzy Journal of application of
hybrid techniques across different CEM application areas along with Chan et al. Construction “fuzzy techniques”
1996–2005 52
detailed discussions of selected papers in each category that demon­ [40] Engineering and in construction
strate the capabilities of fuzzy hybrid techniques to solve complex CEM Management management
research
problems. Next, the discussion section provides a checklist for identifi­
cation and recommendation of selection criteria for applications of fuzzy
hybrid techniques to solve CEM problems. Finally, the last section

2
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

fuzzy logic; extended fuzzy-related applications with fuzzy arithmetic, (PRISMA) approach presented by Moher et al. [41] and guidelines
fuzzy expert system, fuzzy synthesis evaluation, fuzzy inference system, from Booth et al. [42] to avoid biased selections and establish a
and fuzzy consensus qualitative analysis; and fuzzy hybrid techniques comprehensive body of knowledge on the reviewed subject. The meth­
with fuzzy probability (e.g., fault tree, Monte-Carlo simulation, Bayesian odology includes four steps, as shown in Fig. 1: background, systematic
networks), fuzzy MCDM, fuzzy cloud model, and fuzzy integral process. literature review methodology, results, and discussion.
Their results indicated that fuzzy hybrid MCDM, such as fuzzy analytical The goals of the literature review were to (1) investigate recent ap­
network process (FANP), is suitable to handling the complex interde­ plications of fuzzy hybrid techniques in CEM, (2) support selection of
pendence among construction risk factors. They found that FANP was appropriate applications of fuzzy hybrid techniques to solve specific
commonly used in infrastructure, tunnel, and power plant projects. In construction problems, and (3) explore promising future research trends
addition, their study proposed using fuzzy Bayesian belief network in regarding applications of fuzzy hybrid techniques in CEM. The
risk assessment and management because of its ability to incorporate comprehensive review of applications of fuzzy hybrid techniques
new information into risk analysis. Tiruneh et al. [10] investigated concentrated on dynamic modeling and computing approaches (e.g.,
construction applications of neuro-fuzzy hybrid systems (NFS), a subset machine learning, MCDM, optimization, simulation). CEM researchers
of fuzzy hybrid machine learning techniques, by reviewing 116 fuzzy- and practitioners can use the results to understand the capabilities of
related studies from 2000 to 2020. Their study reviewed and catego­ these techniques in solving construction problems, including the six
rized NFS into five core groups: gradient-based algorithms, hybrid application areas of prediction, evaluation/assessment, planning/man­
learning, population-based algorithms, extreme learning machine, and agement, process modeling, system modeling, and performance. Spe­
support vector machine. They also found that most NFS were used in cifically, the results identify what fuzzy hybrid techniques are
predictive modeling of construction problems followed by some other implemented and which problems they solve. Second, a list of recom­
applications in decision making, evaluation/assessment, and system mended selection criteria was developed for choosing suitable fuzzy
modeling. Tiruneh et al. [10] produced a list of five common perfor­ hybrid techniques to solve problems within these six CEM application
mance evaluation criteria– accuracy, interpretability, convergence areas. This study identified problems and/or challenges that CEM re­
speed, computational complexity, and local minima trapping – that are searchers face when choosing a fuzzy hybrid technique. Finally, a review
used to select an appropriate NFS to solve construction problems. Ac­ of recent applications of fuzzy hybrid techniques and advanced tech­
cording to their review, NFS applications in construction concentrate nologies (BIM, GIS, sensors, modular construction, alternative reality
more on accuracy and overlook interpretability of model outcomes. technologies such as AR and VR) and emerging innovations (e.g., big
More recently, Chen and Pan [37] collected 165 papers related to ap­ data analytics) is discussed. Study results will inform future construction
plications of fuzzy MCDM in construction management and analyzed research in coordinating fuzzy hybrid techniques with modern
them using a network approach in order to determine relationships technologies.
between fuzzy logic / fuzzy set and MCDM applications. Their collected
documents involved 37 single-hybrid fuzzy MCDM (i.e., hybridization of 3.1. Selection of journals and papers
fuzzy logic with a standard MCDM technique) and 17 multiple-hybrid
fuzzy MCDM (i.e., hybridization of fuzzy logic with two or more stan­ Highly ranked journals in the field of construction research were
dard MCDM techniques). They found that the majority of fuzzy MCDM selected from multiple online libraries, including Google Scholar, Sci­
techniques were used in the operation-and-maintenance process (44%), ence Direct, Emerald Insight, Web of Science, IEEE Xplore, Wiley Online,
followed by pre-design and planning (39%), and that the most-used and SpringerLink, and different databases from the top publishers, such
fuzzy hybrid MCDM technique was fuzzy analytic hierarchy process as American Society of Civil Engineers (ASCE), Elsevier, IEEE Trans­
(AHP). actions, and Taylor & Francis Group. The three databases from which
The previous literature review studies provide different applications the most collected articles were collected include Science Direct (75
of fuzzy set theory, fuzzy logic, and fuzzy hybrid techniques in various papers), ASCE (63 papers), and Emerald Insight (46 papers). Selecting
CEM application areas. The current literature indicates that fuzzy set multiple databases instead of a single generalized search engine maxi­
theory and fuzzy logic can be constantly applied to tackle complex mized the coverage of selected articles related to applications of fuzzy
construction problems that involve linguistically expressed information hybrid techniques in the CEM research community. Previous fuzzy-
and subjective uncertainty by using appropriate fuzzy membership related review and content analysis studies in construction used a sin­
functions and linguistic variables. In other cases, where fuzzy logic and gle generalized search engine for their literature search. For instance,
fuzzy set theory may not be the best fit because of random uncertainty Chen and Pan [37] utilized Web of Science to collect journal papers
and high dynamism in construction projects, the hybridization of fuzzy related to applications of fuzzy MCDM in construction management.
logic with other modeling and computing techniques is recommended. However, using a single generalized search engine can result in fewer
Although previous studies provided detailed investigations into fuzzy collected fuzzy-hybrid-related articles, which impacts the subsequent
hybrid MCDM and NFS in construction management, other fuzzy hybrid content analysis and the ability for researchers to generalize adequate
techniques, such as fuzzy optimization and fuzzy simulation, have not findings [10]. This study also attempted to maximize the coverage of the
been investigated in detail. Additionally, previous studies concentrated total fuzzy-related articles by performing a specific search into indi­
on solving problems within a few CEM application areas, including vidual target reputable journals listed in the Scimago Journal & Country
prediction and decision making. This study aims to extend the review of Rank list. The following criteria were used to select papers, journals, and
current construction literature with a detailed and integrated categori­ conference proceedings to include in the study. First, a paper or journal
zation of fuzzy hybrid techniques in CEM research and recommenda­ had to have at least one of the following terms in the title, abstract, or
tions for selecting an appropriate fuzzy hybrid technique to solve keywords: assessment, building, construction, decision, engineer, engineer­
problems within a specific CEM application area. ing, fuzzy logic, fuzzy sets, fuzzy systems, infrastructure, management,
modeling/modelling, performance, productivity, project, simulation, or soft
3. Systematic literature review methodology computing. Journals had to be published in English, peer reviewed, and
published regularly (e.g., weekly, bi-weekly, monthly) and have pub­
This study included a systematic literature review methodology with lished at least one paper related to the application of fuzzy hybrid
content analysis for collecting, organizing, and analyzing articles pub­ techniques in construction research. Since applications of fuzzy hybrid
lished in high-quality journals presenting applications of fuzzy hybrid techniques in construction from 1996 to 2005 were investigated and
techniques in CEM research. The applied methodology is based on the summarized in a thorough literature review study by Chan et al. [40], a
Preferred Reporting Items for Systematic Reviews and Meta-Analyses criterion for paper selection in this study was that papers must have been

3
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Fig. 1. Systematic literature review methodology

published from 2004 to 2021, presented hybridization of fuzzy logic collected information [43]. The leading benefit of content analysis is its
with another technique, and been specifically related to CEM. ability to accommodate a large volume of information and handle un­
The search process for this study considered title, abstract, keywords, structured data that might consist of meaningful conceptions of data
concluding remarks, and the content of relevant papers. After excluding sources [44]. Content analysis can be qualitative or quantitative,
duplicate and irrelevant papers, 255 papers from 80 well-known jour­ depending on the nature of the study [43]. Qualitative content analysis
nals in the CEM domain were selected for content analysis. Journals with emphasizes determining meaning by grouping data into categories
the most selected papers in this study include Automation in Construction based on interpretation of the contents (e.g., themes, trends, patterns).
(29 papers), American Society of Civil Engineers (ASCE) Journal of Con­ Quantitative content analysis extends the qualitative approach by
struction Engineering and Management (27 papers), Canadian Journal of generating numerical values for the categorized data (e.g., frequencies,
Civil Engineering (12 papers), Construction and Building Materials (11 ratings, rankings) and calculating overall frequency of a topic
papers), Engineering, Construction and Architectural Management (11 pa­ mentioned throughout the entire collection. This study used both
pers), ASCE Journal of Management in Engineering (9 papers), Journal of qualitative and quantitative analysis. First, the contents of selected pa­
Civil Engineering and Management (8 papers), International Journal of pers, including (1) application domain, (2) research problem, (3)
Project Management (7 papers), Construction Management and Economics research objective, (4) fuzzy hybrid technique(s) used, (5) reason to
(5 papers), and ASCE Journal of Computing in Civil Engineering (5 papers). select the fuzzy hybrid technique(s) if applicable, and (6) dataset used to
train and evaluate the fuzzy hybrid technique if applicable, were
3.2. Content analysis extracted. Then, fuzzy hybrid techniques were grouped into relevant
categories based on type of modeling and computing techniques and
Formal content analysis was used to investigate possible trends and year of publication. Next, fuzzy hybrid techniques were matched with
patterns, dominant aspects, and relevant conclusions from applications specific construction problems they were applied to. Finally, commonly
of fuzzy hybrid techniques in construction problems in the 255 selected used fuzzy hybrid techniques in CEM research were identified and
papers. Content analysis is a robust observational research method for reviewed in detail.
systematically organizing and evaluating the symbolic contents of CEM applications can be classified using different approaches, such
recorded documents and/or communications (either written or visual) as knowledge areas, process groups, and application areas [45]. In
to determine major facets and develop valid inferences from the addition to the ten originally classified project management knowledge

4
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

areas of integration, scope, cost, quality, schedule, risk, resource, planning, execution, and controlling. Decision making is a broad
communication, procurement, and stakeholder management, the Proj­ application category that encompasses a wide range of knowledge areas
ect Management Institute (PMI) also developed a construction-focused and/or process groups. Specifically, in CEM research, decision-making
extension with four new knowledge areas: safety, environmental, problems include investigation, reasoning, prioritization, selection,
financial, and claim management [10]. Process groups involve typical and optimization of multiple knowledge areas (e.g., cost, safety, quality,
stages of a construction project, such as initiation, planning, execution, risk, and stakeholder management) and process groups (e.g., initiation,
controlling, and closing. CEM consists of multidisciplinary application planning, execution, and controlling). Accordingly, decision making
categories, such as prediction, evaluation/assessment, performance appears to be necessary part of some categorized CEM application areas
management, process/system modeling, and optimization, that crosscut in this study, such as evaluation/assessment, planning/management,
knowledge areas and process groups [10,40] as shown in Fig. 2. and performance. Therefore, the six categorized application areas
CEM application areas can be categorized using content analysis to concisely produce comprehensive results covering state-of-the-art CEM
classify applications under related headings (e.g., title, abstract, and applications across knowledge areas and process groups.
keywords) [46,47,10]. The content analysis results show that CEM
problems can be grouped into the six application areas: prediction, 4. Categorization of fuzzy hybrid techniques in CEM
evaluation/assessment, planning/management, process modeling, sys­ applications
tem modeling, and performance. Prediction problems include fore­
casting construction project performance (e.g., project duration, cost Based on the detailed content analysis, applications of fuzzy hybrid
overrun, cash flow, client satisfaction), behaviors of construction com­ techniques from 255 selected papers were categorized into the six CEM
ponents (e.g., structural beams and columns, construction materials), application areas of prediction, evaluation/assessment, planning/man­
and construction productivity (e.g., activities, labor) [12,48,49,50]. agement, process modeling, system modeling, and performance. Table 2
Evaluation/assessment problems consist of decision making and pre­ shows that prediction problems represent 19% of construction prob­
diction that support selection of project delivery methods, contractors, lems, and about 96% of fuzzy machine learning techniques used in
sub-contractors/suppliers, construction means and methods, construc­ prediction problems were fuzzy hybrid applications while about 4%
tion technologies, and productivity improvement strategies [15,17,51]. were fuzzy optimization techniques. Evaluation/assessment problems
Planning/management problems include analysis, decision making, and represent 48% of construction problems, for which fuzzy hybrid MCDM
optimization of project resources, scheduling, time-cost tradeoffs, (64%) represents the majority of fuzzy hybrid techniques used, followed
structural design, site layout planning, and supply chain management by fuzzy hybrid machine learning (26%), fuzzy hybrid simulation (7%),
[19,21,52]. Process modeling problems include analysis, simulation, and fuzzy optimization (3%). Planning/management problems repre­
and prediction of construction processes, such as road traffic controls, sent 20% of construction problems, for which fuzzy hybrid optimization
earthwork operations, bidding process, and pipeline maintenance techniques consist of 58% of total fuzzy hybrid techniques used, fol­
[33,53]. System modeling problems include analysis and simulation of lowed by fuzzy hybrid machine learning (21%) and fuzzy hybrid MCDM
construction systems, such as heating, ventilation, and air conditioning (21%). Process modeling problems represent 7% of construction prob­
(HVAC) systems, supply chain network design, construction structural lems, and fuzzy hybrid simulation (82%) represents the majority of
analysis, and infrastructure management [25,54]. Performance prob­ fuzzy hybrid techniques applied to construction process modeling
lems include optimization and prediction of time performance (e.g., problems, followed by fuzzy hybrid machine learning (18%). System
project durations), cost performance (e.g., cost controls), productivity modeling problems represent 2% of construction problems, for which
performance (e.g., estimation of productivity), risk management, dis­ the majority of fuzzy hybrid applications are fuzzy hybrid simulation
putes and claims, and client satisfaction [55,56,57]. Implementing the (67%), followed by fuzzy hybrid machine learning (33%). Performance
multidisciplinary categorization approach shown in Fig. 2 allows a problems represent 4% of construction problems, with a fairly equal
systematic classification of CEM application areas regardless of knowl­ implementation of all four categories of fuzzy hybrid techniques: fuzzy
edge area or process group. For example, prediction is an essential CEM hybrid machine learning (30%), fuzzy hybrid MCDM (26%), fuzzy
application domain that cuts across multiple knowledge areas and pro­ hybrid optimization (26%), and fuzzy hybrid simulation (18%).
cess groups. Evaluation/assessment problems consist of analysis, deci­ The following sections give detailed results and discussions of each
sion making, and selection of various knowledge areas such as resource, fuzzy hybrid category. To illustrate essential capabilities of each fuzzy
cost, quality, and schedule, in multiple process groups including hybrid technique in practical construction problems, selected related

Fig. 2. Multidisciplinary categorization of CEM application areas.

5
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Table 2
Summary of fuzzy hybrid techniques in CEM applications.
CEM Application Category Fuzzy Hybrid Technique Category

Optimization Machine Learning MCDM Simulation Total Percent

Prediction 2 47 0 0 19%
Evaluation/ Assessment 3 31 80 7 48%
Planning/ Management 30 10 10 0 20%
Process Modeling 0 4 0 18 7%
System Modeling 0 2 0 4 2%
Performance 3 4 3 2 4%
Total Percent 14% 38% 36% 12% 100%

papers are also discussed. Paper selection was based on three criteria: (1) used to develop the fuzzy inference system [5,40]. Current literature
papers that developed and/or implemented different fuzzy hybrid identifies a few fuzzy hybridizations with ANNs categorized as cooper­
techniques in a given category were selected to demonstrate diverse ative neuro-fuzzy systems, such as Mirhadi and Zayed’s [79] predictive
variants of fuzzy hybrid applications in CEM; (2) papers that utilized model for construction productivity. Fuzzy neural networks are another
fuzzy hybrid techniques to solve different construction problems were type of fuzzy hybridization with ANNs in CEM problems [80]. These
selected to illustrate adequate capabilities of the technique in solving techniques use the same architecture and learning algorithms as a
various practical problems; and (3) the most recent papers were selected standard ANN model, with differences in handling input variables and
if a large number of papers satisfied the first two criteria for a given logical operations [5]. Fuzzy neural networks can handle both crisp and
fuzzy hybrid category. fuzzy numbers, while only crisp inputs can be entered into ANNs.
Additionally, a fuzzy neural network utilizes logical fuzzy operations in
its hidden layer(s), while an ANN’s hidden layers deal with classical
4.1. Fuzzy hybrid machine learning arithmetic operations. Using logical fuzzy operations helps fuzzy neural
networks obtain a more transparent and interpretable reasoning process
The use of computer science and artificial intelligence (AI) ap­ [79].
proaches has been significantly increased in order to support modeling Fuzzy clustering techniques, a discipline of unsupervised machine
and computing efforts in solving practical construction problems [58]. learning, are used to classify data into meaningful classes based on
Researchers have combined machine learning, a discipline of AI appli­ similarities and/or dissimilarities between data points [59,81]. Appli­
cations that concentrates on developing computational methods and cations of fuzzy clustering techniques (e.g., fuzzy c-means, context-
models that have the ability to learn from data and solve different specific fuzzy inference systems) have been used to develop data-
problems combined with fuzzy logic to capitalize on the ability of driven inference systems for predictive modeling [82], while crisp
resulting models to accommodate subjectivity, incomplete information, clustering applications are only used for data classification [81,83]. In
and ambiguity [5,10]. The content analysis results of this study show some cases, fuzzy clustering techniques are used to classify data for
that fuzzy hybrid machine learning techniques represent the majority structural damage detection [84], development of control systems for
(38%) of fuzzy hybrid applications in construction problems. Fuzzy pavement deflection tests [85], monitoring project schedules [56], and
hybrid machine learning techniques are often utilized in construction pavement design [59]. In other cases, fuzzy clustering techniques are
problems such as data classification [11,59], predictive modeling used in developing fuzzy inference systems for prediction, such as fuzzy
[12,60,61], and pattern recognition [62]. Common fuzzy hybrid ma­ rule-based inference systems for selection of project delivery methods
chine learning techniques include fuzzy hybridizations with ANNs [86] and predictive models for construction labor productivity [87].
[48,63,64,65] and fuzzy clustering techniques [14,29,66]. Fuzzy clustering techniques can also be combined with other computing
CEM researchers have utilized various fuzzy hybridizations with techniques, such as pattern recognition and differential evolution algo­
ANNs to overcome some limitations of a standalone ANN model, rithm, to produce a multi-hybrid model. Table 3 presents selected papers
including (1) incapability to accommodate subjective uncertainty, (2) for fuzzy hybrid machine learning techniques in construction.
requiring a higher level of interpretability for outcomes from the
reasoning and prediction processes, and (3) necessity of large datasets
for training, validation, and testing procedures [46,67]. Three common 4.2. Fuzzy hybrid multi-criteria decision making (MCDM)
fuzzy hybridizations with ANN consist of NFS, cooperative neuro-fuzzy
systems, and fuzzy neural networks [5,68]. NFS techniques represent the Construction problems often consist of a variety of decision-making
majority of fuzzy hybridizations with ANN in CEM research, where the scenarios with multiple selection criteria involving complexity and un­
learning algorithm of ANN is combined with a fuzzy inference system as certainties [40]. MCDM techniques have been implemented in con­
a single hybrid fuzzy ANN model [10,69,70]. In addition, fuzzy ANN can struction problems to help decision makers identify, classify, select,
also be integrated with other standard techniques, such as artificial bee and/or rank a set of decision alternatives while balancing conflicting
colony, particle swarm optimization, and system dynamics, as multi- criteria to derive optimal decisions. CEM researchers have combined
hybrid fuzzy ANN models. Classification of NFS techniques can be MCDM techniques with applications of fuzzy logic to better accommo­
based on the structure of the fuzzy component (Mamdani or Takagi date subjectivity and ambiguity in modeling processes [37]. Fuzzy
Sugeno), learning algorithm (backpropagation, random, supervised, or MCDM techniques have been widely used in construction, including
unsupervised learning), and fuzzy approach (type-1 and type-2) [68]. fuzzy AHP [101,102], the Technique for Ordering Preference by Simi­
Two common NFS types are adaptive network-based fuzzy inference larity to Ideal Solution (TOPSIS) [103,104], VIsekriterijumska opti­
system (ANFIS) developed by Jang [71] and the evolutionary fuzzy mizacija i KOmpromisno Resenje (VIKOR) [105,106], the Preference
neural inference model developed by Cheng and Ko [72]. Applications Ranking Organization METHod for Enrichment Evaluations (PROM­
of NFS techniques in construction are found in prediction/forecasting of ETHEE) [107], and Elimination et Choix Traduisant la Realite (ELEC­
construction materials’ strengths [73,74,75], project cost estimation TRE) [108].
[76], and evaluation of subcontractor and project manager [77,78]. Fuzzy AHP is an analytical approach that provides measurements
Cooperative neuro-fuzzy systems include two independent components, and assessments using pairwise comparisons between alternatives and is
an ANN model and a fuzzy inference system, where the ANN model is capable of capturing subjective judgements and ambiguity from experts’

6
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Table 3
Papers selected for fuzzy hybrid machine learning techniques in CEM research.
CEM Application Year Reference Type of Hybrid Technique CEM Application Journal
Category Hybridization

Fuzzy Hybridizations with ANNs


Dastgheib et al. Project cost
Prediction 2021 Single hybrid ANFIS Canadian Journal of Civil Engineering
[48] management
Construction material
Prediction 2021 Qi et al. [11] Multi-hybrid ANFIS, artificial bee colony Construction and Building Materials
strength
Golafshani et al. Construction material
Prediction 2020 Multi-hybrid ANFIS, Grey Wolf Optimizer Construction and Building Materials
[12] strength
Construction material- Practice Periodical on Structural Design and
Prediction 2020 Madani et al. [13] Multi-hybrid ANFIS, ANNs
cement Construction
Khotbehsara et al. Construction material-
Prediction 2018 Single hybrid ANFIS Construction and Building Materials
[89] mortar
Khademi and Jamal Construction material- International Journal of Structural
Prediction 2017 Single hybrid ANFIS
[90] concrete Engineering
Rezaie and Sadighi Construction material- International Journal of Civil Engineering
Prediction 2017 Single hybrid ANFIS
[49] concrete and Technology
Project cost Journal of Civil Engineering and
Prediction 2017 Wang et al. [70] Single hybrid ANFIS
management Management
Construction material-
Prediction 2016 Zhou et al. [92] Multi-hybrid NFS Construction and Building Materials
concrete
Sadrossadat et al. Construction material-
Prediction 2016 Single hybrid ANFIS Construction and Building Materials
[93] soil
Al-zharani et al. Construction material-
Prediction 2016 Single hybrid ANFIS Construction and Building Materials
[94] concrete
Construction material- International Journal of Structural and Civil
Prediction 2016 Siraj et al. [73] Single hybrid ANFIS
concrete Engineering Research
Evaluation/ Naderpour and Structural design Practice Periodical on Structural Design and
2019 Single hybrid ANFIS
assessment Mirrashid [88] analysis Construction
Evaluation/ Construction project Engineering, Construction and Architectural
2019 Utama et al. [64] Single hybrid ANFIS
assessment selection Management
Evaluation/ NFS, particle swarm Infrastructure
2017 Liu et al. [21] Multi-hybrid Structure and Infrastructure Engineering
assessment optimization maintenance
Evaluation/ Construction material-
2017 Khademi et al. [69] Single hybrid Fuzzy ANNs Frontiers of Structural and Civil Engineering
assessment concrete
Evaluation/
2016 Kabir et al. [91] Single hybrid ANFIS Pipeline maintenance Structure and Infrastructure Engineering
assessment
Evaluation/ Omar and Fayek
2016 Multi-hybrid Fuzzy ANNs Project performance Automation in Construction
assessment [80]
Planning/ Construction International Journal of Construction
2021 Sarihi et al. [60] Multi-hybrid ANFIS
management productivity Management
Planning/ Structural design
2020 Soares et al. [63] Multi-hybrid ANFIS Engineering Structures
management analysis
Project risk
Process modeling 2018 Peña et al. [65] Multi-hybrid ANFIS, Monte Carlo simulation Expert Systems with Applications
management
Mirahadi and Zayed Construction
Process modeling 2016 Single hybrid Fuzzy ANNs Automation in Construction
[79] productivity
Gerami Seresht and Construction
System modeling 2020 Multi-hybrid NFS, system dynamics Applied Soft Computing
Fayek [66] productivity
Fuzzy Clustering Techniques
Gerami Seresht Construction
Prediction 2020 Single hybrid Fuzzy clustering Applied Soft Computing
et al. [66] productivity
Ouma and Hahn Construction
Prediction 2017 Single hybrid Fuzzy c-means Automation in Construction
[29] management
Evaluation/ Project delivery Journal of Construction Engineering and
2020 Nguyen et al. [14] Single hybrid Fuzzy clustering
assessment selection Management
Evaluation/ Fuzzy c-mean, pattern Project delivery Journal of Construction Engineering and
2020 Nguyen et al. [86] Multi-hybrid
assessment recognition selection Management
ASCE-ASME Journal of Risk and Uncertainty
Planning/ Pavement design-load
2017 Li et al. [59] Single hybrid Fuzzy clustering in Engineering Systems, Part A: Civil
management determination
Engineering
Planning/ Tsehayae and Fayek Construction Journal of Construction Engineering and
2016 Single hybrid Fuzzy inference system
management [87] productivity Management
Tsehayae and Fayek Construction
System modeling 2016 Single hybrid Fuzzy inference system Construction Innovation
[67] productivity
Fuzzy clustering, differential Project resource
Performance 2016 Tran et al. [55] Multi-hybrid Alexandria Engineering Journal
evolution management
Other Fuzzy Hybrid Machine Learning Techniques
Evaluation/ Mohamad and Tran Construction Journal of Construction Engineering and
2021 Single hybrid Fuzzy Bayesian network
assessment [95] inspection Management
Evaluation/ Abad and Naeni Fuzzy fault tree, event tree Project risk International Journal of Construction
2020 Multi-hybrid
assessment [96] analysis management Management
Evaluation/
2019 Lin and Huang [32] Single hybrid Fuzzy rule-based reasoning Design management Applied Sciences
assessment
2019 Shoar et al. [99] Single hybrid Fuzzy fault tree analysis Construction Innovation
(continued on next page)

7
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Table 3 (continued )
CEM Application Year Reference Type of Hybrid Technique CEM Application Journal
Category Hybridization

Evaluation/ Project risk


assessment management
Planning/ Shoar and Banaitis Journal of Civil Engineering and
2019 Single hybrid Fuzzy fault tree analysis Labor productivity
management [98] Management
Fuzzy extreme learning
Planning/ Phoemphon et al.
2018 Multi-hybrid machine, particle swarm Site location Applied Soft Computing
management [100]
optimization
International Journal of Construction
Process modeling 2020 Rezakhani [97] Single hybrid Fuzzy Bayesian network Project scheduling
Management

evaluations [17]. This type of fuzzy hybrid technique has been widely analysis for infrastructure management [25], water network design
used in various decision-making processes in construction, including [154], and pipeline maintenance [155]. FMCS can also be incorporated
construction material selections [15], project cost management with FABM in modeling construction crew performance [24]. FDES is
[109,110], project risk management [16,111,112], supplier selection often applied to modeling construction processes consisting of various
[51,102], sustainability analysis [113,114,115], and project team activities performed in a sequence to predict duration of the process or
development [116]. Fuzzy TOPSIS can be used to assess and rank al­ usage of necessary resources. Recent FDES applications in CEM research
ternatives based on a set of selection criteria associated with pre­ include modeling project scheduling [26,156] and managing project
determined weights. It can handle subjective uncertainty and vagueness resources [157]. FSD is a hybrid integration of fuzzy logic with system
created by the participation of experts in assessing selection criteria for dynamics method, developed by Levary [158]. Hybrid integration en­
identifying optimal alternatives [37], and it is capable of analyzing fuzzy hances capabilities of (1) handling the dynamic nature and in­
input numbers, which allows experts to provide their evaluation using terrelationships among different components of the construction system,
linguistic terms rather than a precise quantitative assessment for deci­ (2) using fuzzy membership functions to represent system variables that
sion alternatives and selection criteria. Fuzzy TOPSIS is often used in involve subjective uncertainty, and (3) using fuzzy rule-based inference
construction safety [117], construction productivity [118], road pave­ system in FSD to model uncertainty of the interactions among system
ment selection [119], construction material selection [120,121,122], variables. Common applications of FSD in construction problems
project risk management [35,123,124], off-site construction manage­ include construction risk assessments [23,159], construction produc­
ment [125], site layout planning [3,20], and supplier selection tivity [160,161], quality management in construction projects [162],
[103,126]. Fuzzy VIKOR method concentrates on assessing decision quantification of the impacts of construction claims [163], modeling
alternatives, based on a set of conflicting criteria and a statistical metric social sustainability in construction [33], and modeling project contin­
to calculate the distance of the alternatives from the best alternative. gency [164]. FABM has been increasingly used in construction problems
This hybrid technique is used to solve decision-making problems in over the past decade to model complex construction systems with
project risk management [18,127] and in selection of advanced tech­ interacting elements, which is defined as individual agents in agent-
nologies (e.g., BIM) [31] in construction. While fuzzy AHP can be used based modeling (ABM). The hybridization of fuzzy logic with ABM to
with other techniques, such as fuzzy Delphi and FANP as a multi-hybrid develop FABM can augment model capability by using fuzzy member­
model, the majority of fuzzy TOPSIS, fuzzy VIKOR, and fuzzy ELECTRE ship functions instead of probabilistic distributions to represent sub­
were found as single hybrid models. Table 4 presents selected papers for jective characteristics of agents. FABM techniques are often used in
fuzzy hybrid MCDM techniques in construction. modeling construction crew performance [24,53,165]. Although some
multi-hybrid models were found between fuzzy Monte Carlo and agent-
based simulation, fuzzy cognitive maps and structural equation
4.3. Fuzzy hybrid simulation
modeling, or FABM and MCDM, most fuzzy simulation techniques have
been used as a single hybrid model. Table 5 presents selected papers for
Fuzzy hybrid simulation techniques are capable of handling the dy­
fuzzy hybrid simulation techniques in construction.
namic nature of construction problems (systems and processes) and
capturing the interrelationships among variables that affect the behav­
iors of the systems and/or processes, such as project size, duration, 4.4. Fuzzy hybrid optimization
project complexity, and project risks [152]. Fuzzy hybrid simulation has
been implemented for many complex construction problems to provide CEM research anticipates many complex nonlinear optimization
sufficient solutions under different scenarios. This technique represents problems, such as monitoring project schedule to simultaneously ach­
12% of applications of fuzzy hybrid techniques in construction. The ieve multiple project objectives, planning project site layouts, managing
content analysis results for this study show that common fuzzy simula­ construction supply chain, and assessing working conditions of con­
tion techniques have been implemented in CEM problems, including struction structures [170,171]. For example, Bakry et al. [172] devel­
fuzzy Monte Carlo simulation (FMCS), fuzzy discrete event simulation oped a fuzzy optimization and buffering algorithm for determining the
(FDES), fuzzy system dynamics (FSD), and fuzzy agent-based modeling optimum scheduling solution of repetitive construction projects with
(FABM). different objectives including time, cost, and users’ desired confidence
FMCS has been implemented mostly in construction risk manage­ in the final schedule. Their results indicated that fuzzy optimization
ment because of its ability to simultaneously model and process various algorithm is capable of modeling subjective uncertainties associated
uncertainties, including probabilistic and subjective uncertainty of with input variables. Fuzzy-related optimization applications can be
project risk factors [5]. For example, Sadeghi et al. [153] utilized FMCS used to handle fuzzy variables and any fuzzy restrictions involved in the
in simulating project cost estimation problems and found that it provides optimization process [173,174] and/or enhance the strengths and
decision makers with more detailed project cost results compared to the overcome shortcomings of a standard optimization algorithm
standard Monte Carlo simulation. In addition, their study concluded that [175,176]. The content analysis results of this study show that fuzzy
an increase in the number of variables that require representation using hybrid optimization techniques are formulated by combining fuzzy logic
fuzzy membership functions (i.e., subjective variables) escalates fuzzi­ with numerous standard optimization approaches including evolu­
ness in simulation outcomes. FMCS techniques are often used in risk tionary algorithms [19], particle swarm optimization [20], ant colony

8
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Table 4
Papers selected for fuzzy multi-criteria decision-making (MCDM) techniques in CEM research.
CEM Application Year Reference Type of Hybrid Technique CEM Application Journal
Category Hybridization

Fuzzy AHP
Evaluation/ Construction material
2021 Figueiredo et al. [15] Single hybrid Fuzzy-AHP Building and Environment
assessment selection
Evaluation/
2020 Mirhosseini et al. [128] Multi-hybrid Fuzzy DEMATEL, ANP Project management Sustainability
assessment
Evaluation/ Project risk Journal of Construction Engineering
2019 Beltrão and Carvalho [16] Single hybrid Fuzzy AHP
assessment management and Management
Evaluation/ Project risk Journal of Construction Engineering
2019 Monzer et al. [112] Single hybrid Fuzzy AHP
assessment management and Management
Evaluation/ Structural design
2019 Yang et al. [17] Multi-hybrid Fuzzy AHP Journal of Bridge Engineering
assessment analysis
Evaluation/ Benchmarking: An International
2018 Kumar et al. [51] Multi-hybrid Fuzzy Delphi, AHP-DEMATEL Supplier selection
assessment Journal
Evaluation/ Tomczak and Jaśkowski Project team
2018 Single hybrid Fuzzy AHP KSCE Journal of Civil Engineering
assessment [116] development
Evaluation/ Sustainability
2018 Jang et al. [113] Single hybrid Fuzzy AHP Journal of Management in Engineering
assessment analysis
Evaluation/ Sustainability Journal of Construction Engineering
2017 Zhao et al. [114] Multi-hybrid Fuzzy AHP, ANP
assessment analysis and Management
Evaluation/ Sustainability
2017 Inti and Tandon [115] Multi-hybrid Fuzzy AHP, MCDM Journal of Infrastructure Systems
assessment analysis
Evaluation/ Prascevic and Prascevic Journal of Civil Engineering and
2017 Single hybrid Fuzzy AHP Project plan selection
assessment [130] Management
Evaluation/ Plebankiewicz and Kubek Journal of Construction Engineering
2016 Single hybrid Fuzzy AHP Supplier selection
assessment [102] and Management
Evaluation/ Abdul-Rahman et al. Sustainability Proceedings of the Institution of Civil
2016 Single hybrid Fuzzy AHP
assessment [131] analysis Engineers - Engineering Sustainability
Evaluation/ Structure and Infrastructure
2016 El-Abbasy et al. [132] Single hybrid Fuzzy AHP Pipeline maintenance
assessment Engineering
Planning/ Project risk Engineering, Construction and
2019 Aladağ and Işık [111] Single hybrid Fuzzy AHP
management management Architectural Management
Planning/
2018 Mavi and Standing [110] Multi-hybrid Fuzzy DEMATEL, ANP Project management Journal of Cleaner Production
management
Project cost International Journal of Managing
Performance 2020 Afzal et al. [109] Multi-hybrid Fuzzy AHP, simulation
management Projects in Business
Fuzzy AHP, fuzzy preference Infrastructure Journal of Performance of Constructed
Performance 2017 Omar et al. [129] Multi-hybrid
program maintenance Facilities
Fuzzy TOPSIS
Evaluation/ Road pavement
2020 Pasha et al. [119] Single hybrid Fuzzy TOPSIS Soft Computing
assessment selection
Evaluation/ Journal of Civil Engineering and
2020 Mohandes et al. [117] Single hybrid Fuzzy TOPSIS Construction safety
assessment Management
Evaluation/ Construction material International Journal of Construction
2019 Reddy et al. [120] Single hybrid Fuzzy TOPSIS
assessment selection Management
Evaluation/ Fuzzy comprehensive Project risk
2019 Gebrehiwet and Luo [35] Multi-hybrid Symmetry
assessment evaluation, TOPSIS management
Evaluation/ Project risk
2019 Norouzi and Namin [123] Single hybrid Fuzzy TOPSIS Civil Engineering Journal
assessment management
Evaluation/ Mathiyazhagan et al. Construction material Journal of Advances in Management
2019 Single hybrid Fuzzy TOPSIS
assessment [122] selection Research
Evaluation/ Interior design and Engineering, Construction and
2019 Juan et al. [28] Single hybrid Fuzzy TOPSIS
assessment decoration Architectural Management
Evaluation/ Project risk
2018 Yazdi [124] Single hybrid Fuzzy TOPSIS Safety Science
assessment management
Evaluation/ Fuzzy TOPSIS, fuzzy positive/
2017 Liang et al. [133] Multi-hybrid Project plan selection Journal of Infrastructure Systems
assessment negative-ideal solution
Evaluation/
2017 Jang et al. [103] Single hybrid Fuzzy TOPSIS Supplier selection Journal of Management in Engineering
assessment
Evaluation/
2017 Maghsoodi et al. [134] Single hybrid Fuzzy TOPSIS Project performance KSCE Journal of Civil Engineering
assessment
Evaluation/ Journal of Civil Engineering and
2017 Polat et al. [126] Single hybrid Fuzzy TOPSIS Supplier selection
assessment Management
Evaluation/ Contract
2016 Al-Humaidi [135] Single hybrid Fuzzy TOPSIS Journal of Management in Engineering
assessment management
Evaluation/ Construction Journal of Multi-Criteria Decision
2016 Fayek and Omar [118] Single hybrid Fuzzy TOPSIS
assessment productivity Analysis
Evaluation/
2016 Ning et al. [3] Single hybrid Fuzzy TOPSIS Site layout planning Automation in Construction
assessment
Planning/ Construction material International Journal of Construction
2019 Fazeli et al. [121] Single hybrid Fuzzy TOPSIS
management selection Management
Planning/ Off-site construction Architectural Engineering and Design
2018 Arashpour et al. [125] Single hybrid Fuzzy TOPSIS
management management Management
2018 Song et al. [20] Single hybrid Fuzzy TOPSIS Site layout planning
(continued on next page)

9
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Table 4 (continued )
CEM Application Year Reference Type of Hybrid Technique CEM Application Journal
Category Hybridization

Planning/ Journal of Computing in Civil


management Engineering
Fuzzy VIKOR
Evaluation/ Project risk Journal of Natural Gas Science and
2019 Mete et al. [18] Single hybrid Fuzzy VIKOR
assessment management Engineering
Evaluation/ International Journal of Industrial
2019 Hoseini et al. [31] Multi-hybrid Fuzzy VIKOR, FANP Project management
assessment Engineering and Production Research
Evaluation/ Project risk
2018 Wang et al. [127] Single hybrid Fuzzy VIKOR Applied Soft Computing
assessment management
Fuzzy ELECTRE
Evaluation/
2016 Wu et al. [108] Single hybrid Fuzzy ELECTRE Project plan selection Energy Conversion and Management
assessment
Evaluation/ Project team International Journal of Strategic
2014 Chaghooshi et al. [136] Single hybrid Fuzzy ELECTRE
assessment development Decision Sciences
Evaluation/ Yazdani-Chamzini et al. Project risk International Journal of Management
2013 Single hybrid Fuzzy ELECTRE
assessment [137] management Science and Engineering Management
Other Fuzzy MCDM Techniques
Evaluation/ Engineering, Construction and
2021 Eghbali-Zarch et al. [138] Multi-hybrid Fuzzy IDOCRIW, WASPAS Project management
assessment Architectural Management
Evaluation/ Karamoozian and Wu Fuzzy failure mode and Project risk Engineering, Construction and
2020 Single hybrid
assessment [139] effective analysis management Architectural Management
Evaluation/ Gharanfoli and Project risk Journal of Multi Criteria Decision
2019 Single hybrid Fuzzy DEMATEL, ANP
assessment Valmohammadi [140] management Analysis
Evaluation/ Hatefi and Tamošaitienė Project risk Journal of Civil Engineering and
2019 Multi-hybrid Fuzzy DEMATEL, ANP
assessment [141] management Management
Evaluation/ Fuzzy failure mode and effect Project risk Journal of Construction Engineering
2019 Yazdani et al. [142] Multi-hybrid
assessment analysis management and Management
Evaluation/ Fuzzy qualitative flexible Engineering, Construction and
2019 Liang and Chong [143] Single hybrid Supplier selection
assessment multiple attributes Architectural Management
Evaluation/ Sedady and Beheshtinia Construction project
2019 Single hybrid Fuzzy TOPKOR Management of Environmental Quality
assessment [144] selection
Evaluation/ International Journal of Construction
2019 Chen et al. [27] Multi-hybrid Fuzzy MAUT Project management
assessment Management
Evaluation/ Mohande and Zhang
2019 Single hybrid Fuzzy best worst method Construction safety Safety Science
assessment [145]
Evaluation/
2018 Gunduz et al. [146] Single hybrid fuzzy structural equation model Construction safety Automation in Construction
assessment
Evaluation/ Ardeshir and Mohajeri International Journal of Injury Control
2018 Multi-hybrid Fuzzy DEMATEL, ANP Construction safety
assessment [147] and Safety Promotion
Evaluation/ Journal of Urban Planning and
2018 Boostani et al. [148] Multi-hybrid Fuzzy DEMATEL, ANP Site layout planning
assessment Development
Evaluation/ Fuzzy evaluation based on Project resource Archives of Civil and Mechanical
2018 Ghorabaee et al. [149] Single hybrid
assessment distance from average solution management Engineering
Evaluation/ Project risk
2016 Ameyaw and Chan [150] Single hybrid Fuzzy risk-allocation criteria Journal of Infrastructure Systems
assessment management
Evaluation/ Singhaputtangkul and Journal of Construction Engineering
2016 Single hybrid Fuzzy consensus scheme Project plan selection
assessment Zhao [151] and Management
Planning/ Project risk
2018 Chatterjee et al. [2] Multi-hybrid Fuzzy preference relation, FANP Symmetry
management management

systems [22], optimum system hierarchy analysis [177], artificial bee involving scheduling, classification, and vehicle routing [22,183,184].
colony algorithms [178], adaptive fuzzy Pareto set [179], and linear The CEM literature shows that the majority of fuzzy optimization
programming [180]. techniques were used in terms of a single hybrid approach. Table 6
Fuzzy evolutionary algorithms are population-based optimization presents selected papers for fuzzy hybrid optimization techniques in
algorithms with two common applications of genetic algorithm and construction.
differential evolution that have been used to accommodate subjectivity
and vagueness in construction optimization problems [5]. These hybrid 5. Discussions
techniques are commonly used in project scheduling [176], structural
design [19,173], site layout planning [52], supply chain management This section presents (1) the identification and development of se­
[174], and supplier selection [171]. Fuzzy particle swarm optimization lection criteria for the implementation of fuzzy hybrid techniques to
approach, another population-based algorithm deriving optimum solu­ solve CEM problems and (2) suitability of fuzzy hybrid techniques to
tions in terms of particles, has been used to handle uncertainties, such as specific CEM application areas.
subjectivity, imprecision, and ambiguity, in various optimization prob­
lems [6]. This hybrid technique is often used in construction problems
5.1. Selection criteria of fuzzy hybrid techniques for construction
involving project resource management [20], project scheduling [181],
problems
and site layout planning [182]. Fuzzy ant colony optimization, a prob­
abilistic multi-agent technique, has been widely implemented in CEM
A variety of applications of fuzzy hybrid techniques have been used
research to capture subjective uncertainty and vagueness in multi-
to solve problems within a specific CEM application area. The diversity
objective optimization problems. Applications of fuzzy ant colony
of techniques applied in various CEM problems leads to difficulties in
optimization are used in many combinatorial optimization problems
evaluating and comparing the performance and accuracy of fuzzy hybrid

10
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Table 5
Papers selected for fuzzy simulation techniques in CEM research.
CEM Application Year Reference Type of Hybrid Technique CEM Application Journal
Category Hybridization

Fuzzy Monte Carlo Simulation


Evaluation/ Parvizsedghy et al.
2015 Single hybrid FMCS Pipeline maintenance Structure and Infrastructure Engineering
assessment [155]
Gouri and Srinivas Stochastic Environmental Research and Risk
Process modeling 2017 Single hybrid FMCS Water network design
[154] Assessment
Infrastructure ASCE-ASME Journal of Risk and Uncertainty in
System modeling 2017 Attarzadeh et al. [25] Single hybrid FMCS
management Engineering Systems, Part A: Civil Engineering
Fuzzy Discrete Event Simulation
Journal of Construction Engineering and
Process modeling 2016 Sadeghi et al. [26] Single hybrid FDES Project scheduling
Management
Project resource
Performance 2015 Sadeghi et al. [157] Single hybrid FDES Automation in Construction
management
Fuzzy System Dynamics
Evaluation/ Project risk
2020 Siraj and Fayek [164] Single hybrid FSD Canadian Journal of Civil Engineering
assessment management
Evaluation/ Etemadinia and Project risk
2018 Single hybrid FSD International Journal of Construction Management
assessment Tavakolan [23] management
Rostamnezhad et al. Sustainable Engineering, Construction and Architectural
Process modeling 2020 Multi-hybrid FSD, DEMATEL
[33] development Management
Engineering, Construction and Architectural
Process modeling 2019 Nasirzadeh et al. [163] Single hybrid FSD Project management
Management
Gerami Seresht and Construction
Process modeling 2018 Single hybrid FSD Manuscript submitted for publication
Fayek [160] productivity
Nojedehi and Construction
Process modeling 2017 Single hybrid FSD KSCE Journal of Civil Engineering
Nasirzadeh [161] productivity
Fuzzy Agent-Based Modeling
Construction crew Journal of Construction Engineering and
Process modeling 2020 Raoufi and Fayek [24] Multi-hybrid FMCS, FABM
performance Management
Construction crew
Process modeling 2020 Kedir et al. [165] Multi-hybrid FABM, MCDM Journal of Management in Engineering
performance
Construction crew
Process modeling 2018 Raoufi and Fayek [53] Single hybrid FABM Journal of Computing in Civil Engineering
performance
Other Fuzzy Simulation Techniques
Evaluation/ Engineering, Construction and Architectural
2020 Luo et al. [166] Multi-hybrid FCM, SEM Project management
assessment Management
Evaluation/
2020 Naji et al. [167] Single hybrid Fuzzy SEM Project management IEEE Transactions on Engineering Management
assessment
Evaluation/ FIS, Latin hypercube Project risk
2019 Haqiqat et al. [168] Multi-hybrid Journal of Project Management
assessment simulation management
Journal of Construction Engineering and
Process modeling 2018 Khanzadi et al. [169] Single hybrid Fuzzy cognitive map Change management
Management

techniques. A guideline is needed regarding applicability, accuracy, and accommodate uncertainty and imprecision in modeling and computing
reliability of implementing fuzzy hybrid techniques in specific CEM processes [5,81]. Subjectivity, vagueness, and lack of data in modeling
research areas. The current literature shows that multiple performance increase the complexity of CEM problems, leading to concerns regarding
evaluation criteria, such as accuracy, validity, applicability, and intel­ prediction accuracy, decision-making reliability, interpretability of
ligibility, need to be considered when selecting a fuzzy hybrid technique outcomes, and implementation complexity [189]. In addition, every
to solve construction problems [10,37], and these criteria reflect the CEM problem requires a different set of criteria for selecting fuzzy
dynamic nature of the construction industry and variability in categories hybrid techniques depending on the problem’s characteristics and
of CEM application areas. CEM researchers and practitioners can select a application domains. For example, CEM prediction problems require
fuzzy hybrid technique based on their expertise, research interests, and modeling approaches with the ability to perform trend analysis and
specific CEM application areas (e.g., prediction, decision making, opti­ pattern recognition for predicting and high prediction accuracy [10],
mization). For example, the majority of previous studies have selected while evaluation/assessment problems need modeling approaches that
fuzzy machine learning techniques for CEM problems related to pre­ can prioritize available alternatives and determine the optimal option
diction [68] while fuzzy optimization techniques have often been [37]. Results of the detailed content analysis of 255 publications show
selected for planning/management problems [5]. Fuzzy MCDM tech­ that decisions in selecting an appropriate fuzzy hybrid technique can be
niques have been used to investigate and solve most evaluation/ made based on six categories of performance evaluation criteria: (1)
assessment problems [37]. Most system and process modeling problems data availability, (2) processing ability, (3) computational complexity,
in CEM research have implemented fuzzy simulation techniques [8,38]. (4) interpretability, (5) accuracy, and (6) implementation complexity, as
Researchers selected a variety of fuzzy hybrid techniques in perfor­ shown in Table 7. Each criterion is described in a following section.
mance problems where fuzzy optimization, fuzzy machine learning, and
fuzzy MCDM were applied [39,157,188]. To support subjective judge­ 5.1.1. Accuracy
ments in current selection of fuzzy hybrid techniques in CEM research, a Accuracy is the most important criterion in selecting fuzzy hybrid
comprehensive list of selection criteria in terms of fuzzy hybrid tech­ techniques for CEM research and solving practical construction prob­
niques and their applications across six CEM research areas is needed. lems [4,10]. Accuracy requires techniques with the ability to capture
Concerns in selecting fuzzy hybrid techniques over single modeling uncertainty and vagueness of model outputs, obtain high validity, pro­
and computing methods relate to the ability of models to deal with data duce low training and testing errors, produce the least root-mean-square
availability (e.g., linguistically expressed information) and error and/or mean absolute error between the target values and the

11
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Table 6
Papers selected for fuzzy hybrid optimization models in CEM research.
CEM Application Year Reference Type of Hybrid Technique CEM Application Journal
Category Hybridization

Fuzzy Hybrid Evolutionary Models


Planning/ Marzouk and Sustainable construction
2020 Single hybrid Fuzzy genetic algorithm Decision Science Letters
management Abdelakder [185] materials
Planning/ Building deterioration
2020 Ahmed et al. [170] Single hybrid Fuzzy genetic algorithm Algorithms
management prediction
Planning/ IEEE Transactions on Engineering
2017 Liu et al. [174] Single hybrid Fuzzy genetic algorithm Supply chain management
management Management
Planning/
2017 Song et al. [52] Single hybrid Fuzzy genetic algorithm Site layout planning International Journal of Civil Engineering
management
Planning/
2017 Tu et al. [171] Single hybrid Fuzzy genetic algorithm Supplier selection Journal of Infrastructure Systems
management
Planning/ Haghighi and Ayati
2016 Single hybrid Fuzzy genetic algorithm Structural design-dam Journal of Intelligent and Fuzzy Systems
management [173]
Planning/ Fuzzy differential
2016 Cheng and Tran [176] Single hybrid Project scheduling KSCE Journal of Civil Engineering
management evolution
ASCE-ASME Journal of Risk and
Performance 2020 Bagheri et al. [19] Single hybrid Fuzzy genetic algorithm Structural design Uncertainty in Engineering Systems, Part A:
Civil Engineering
Fuzzy Hybrid Particle Swarm Optimization Models
Site layout planning/
Planning/ Fuzzy particle swarm
2018 Song et al. [199] Single hybrid Project resource Automation in Construction
management optimization
management
Planning/ Fuzzy particle swarm
2014 Ma and Xu [181] Single hybrid Project scheduling Canadian Journal of Civil Engineering
management optimization
Planning/ Fuzzy particle swarm
2012 Xu and Li [182] Single hybrid Site layout planning Automation in Construction
management optimization
Fuzzy Hybrid Ant Colony Systems Models
Evaluation/ Abbasianjahromi et al.
2014 Single hybrid Fuzzy ant colony systems Project team development Project Management Journal
assessment [22]
Planning/
2014 Sambhoo et al. [183] Single hybrid Fuzzy ant colony systems Project plan selection Applied Soft Computing
management
Planning/
2011 Kalhor et al. [186] Single hybrid Fuzzy ant colony systems Time-cost tradeoff Automation in Construction
management
Others Fuzzy Hybrid Optimization Models
Evaluation/
2018 Zhang and Guan [187] Single hybrid Fuzzy bow-tie analysis Project risk management Journal of Management in Engineering
assessment
Planning/ Cheng and Prayogo fuzzy teaching–learning
2017 Single hybrid Structural design-beam Engineering with Computers
management [175] optimization
Planning/ ShangGuan et al. Transportation Research Record: Journal of
2016 Single hybrid Fuzzy pareto set Infrastructure planning
management [179] the Transportation Research Board
Fuzzy linear Engineering, Construction and Architectural
Performance 2016 Bakry et al. [172] Single hybrid Project scheduling
programming Management

values predicted by fuzzy hybrid model, and achieve high optimization information from input variables. Prediction problems require fuzzy
effectiveness and efficiency [5,39,40]. Specifically, fuzzy hybrid ma­ hybrid techniques to model highly dimensional and complex data,
chine learning techniques applied to prediction problems are required to accommodate a mix of quantitative and qualitative inputs, and capture
have the ability to achieve high accuracy in classification and predic­ uncertainty and vagueness of model outputs. Evaluation/assessment
tion, capture uncertainty and vagueness of model outputs, produce low problems require applied fuzzy hybrid techniques that can accommo­
training and testing errors in classification outcomes, and produce the date subjective judgment and preferences of multiple decision makers
least root-mean-square error and/or mean absolute error in prediction and model a large number of parameters. Planning/management prob­
outcomes. In evaluation/assessment problems, fuzzy hybrid techniques lems require fuzzy hybrid techniques that can model a large number of
need the ability to obtain high validity. In planning/management, pro­ parameters and capture uncertainty and vagueness of model outputs. In
cess modeling, system modeling, and performance problems, the process modeling and system modeling problems, fuzzy hybrid tech­
selected technique must have the ability to achieve high optimization niques need the capability to capture uncertainty and vagueness of
effectiveness and efficiency and capture uncertainty and vagueness of model outputs and model a large number of parameters. Finally, per­
model outputs. formance problems require techniques with the ability to accommodate
subjective judgment and preferences of multiple decision makers.
5.1.2. Data availability
Data availability requires techniques with the ability to model highly 5.1.3. Computational complexity
dimensional and complex data, accommodate a mix of quantitative and Computational complexity refers to the variety of algorithm mathe­
qualitative inputs, capture subjectivity and vagueness, and be applicable matics and number of iterations required to perform data training and
to CEM research with advanced technologies (e.g., BIM, GIS, VR, sen­ different types of analyses [10,68]. The complexity of different algo­
sors, modular construction) [10,40,81]. This criterion is essential in rithms in fuzzy hybrid techniques requires techniques with the ability to
every fuzzy hybrid application in construction (fuzzy machine learning, model a large number of parameters (high dimensionality), perform
fuzzy MCDM, fuzzy optimization, and fuzzy simulation), because the scenario and sensitivity analyses, capture dynamic systems and re­
underlying purpose of integrating fuzzy logic with standard modeling lationships, avoid local minima trapping, perform trend analysis and
and computing methods is to capitalize on the strength of fuzzy hybrid pattern recognition in predictive models (e.g., time series), and perform
techniques to handle subjective uncertainty and linguistically expressed need analysis [37,40]. Similar to data availability, computational

12
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

Table 7 the ability to perform trend analysis, pattern recognition in predictive


Selection criteria for applications of fuzzy hybrid techniques in CEM. models, and scenario and sensitivity analyses.
Selection Criteria Specific Selection Criteria Reference
Category 5.1.4. Interpretability
• Ability to capture uncertainty Interpretability is the degree to which outcomes of fuzzy hybrid
and vagueness of model applications are understandable [10,40]. This subjective property in­
outputs cludes degrees of transparency and the ability to prioritize available
• Ability to obtain high validity alternatives for determining the optimal option. Most applications of
• Ability to produce low
training and testing errors
fuzzy hybrid techniques in construction have limitations in their ability
(classification accuracy) to obtain a high interpretability due to the completeness in modeling
• Ability to produce the least [4] [5] [7] [10] [28] [37] and computing formulation. Interpretability is critical to some con­
Accuracy root-mean-square error and/ [38] [39] [40] [58] [68] struction problems, including evaluation/assessment, planning/man­
or mean absolute error be­ [97] [150] [190] [193]
agement, process modeling, and performance problems. Fuzzy hybrid
tween the target values and
the values predicted by fuzzy techniques applied to evaluation/assessment problems needs to also be
hybrid model (prediction capable of prioritizing available alternatives for determining the optimal
accuracy) solution [37]. On the other hand, CEM researchers implementing fuzzy
• Ability to achieve high hybrid techniques in prediction problems usually seek techniques that
optimization effectiveness
and efficiency
produce high interpretability without compromising accuracy of the
• Ability to model highly modeling outcomes.
dimensional and complex
data 5.1.5. Processing ability
• Ability to accommodate a mix
Processing ability of fuzzy hybrid techniques is defined based on
of quantitative and qualitative [5] [8] [9] [32] [35] [37]
inputs [38] [39] [40] [53] [66] their convergence and computing speeds in modeling and producing
Data availability outcomes in different analyses [10,12]. For instance, this criterion can
• Ability to capture subjectivity [81] [158] [189] [191]
and vagueness [192] represent how fast the convergence speeds of fuzzy hybrid machine
• Applicable to CEM research learning techniques are in training and testing data [40,68]. Processing
with advanced technologies
ability can be justified based on the amount of time required for training
(e.g., BIM, GIS, VR, sensors,
modular construction) and testing data with fuzzy hybrid machine learning [10] and the
• Ability to model a large memory requirements for processing an iteration to optimize the pa­
number of parameters (high rameters with fuzzy hybrid optimization [19]. This criterion is common
dimensionality)
for fuzzy hybrid applications in processing modeling and performance.
• Ability to perform scenario
and sensitivity analyses Specifically, CEM researchers have applied fuzzy hybrid simulation
• Ability to capture dynamic (FMCS, FDES, FSD, and FABM) techniques in modeling construction
[6] [7] [10] [30] [33]
Computational
systems and relationships
[37] [38] [40] [48] [50]
processes [5]. In other cases, some fuzzy hybridizations with ANNs,
• Ability to avoid local minima including NFS techniques and fuzzy neural networks, also prioritize the
complexity [68] [71] [81] [115]
trapping
[189] [191] [193] computational speeds of fuzzification, inference, and defuzzification
• Ability to perform trend
analysis and pattern processes [12]. Processing ability is highly recommended in selecting
recognition in predictive automated fuzzy hybrid techniques to quickly produce modeling
models (e.g., time series) outcomes.
• Ability to perform need
analysis
• Transparency
5.1.6. Implementation complexity
• Ability to prioritize available [8] [10] [37] [38] [40] Implementation complexity, or feasibility of model development,
Interpretability
alternatives for determining [69] [123] [149] [192] refers to the ease of applying the fuzzy hybrid techniques to solve spe­
the optimal option cific CEM problems. For example, fuzzy clustering techniques can be
• Fast convergence and [10] [24] [26] [40] [68]
Processing ability easily implemented using various pre-defined functions/packages in R
analyzing speeds [71] [158] [190]
• Availability of commercial programming language [14,86]. However, other fuzzy hybrid machine
Implementation
software packages, open- [14] [68] [86] [189] learning techniques, such as population-based NFS, require the use of
complexity
source coding, and self-coding comprehensive coding skills or commercial software packages [10,68].
Implementation complexity involves availability of commercial soft­
ware packages, open-source coding, and self-coding. The decision to
complexity is required in all applications of fuzzy hybrid techniques in apply a fuzzy hybrid technique can depend on how complicated it is to
CEM research. In prediction problems, fuzzy hybrid machine learning develop and implement that technique [189]. Implementation
and optimization techniques are required to be capable of performing complexity occurs in all CEM application areas, including prediction,
trend analysis and pattern recognition in predictive models (e.g., time evaluation/assessment, planning/management, process modeling, sys­
series) and modeling a large number of parameters (high dimension­ tem modeling, and performance. Applications of fuzzy machine learning
ality). Evaluation/assessment problems require fuzzy hybrid techniques techniques, such as fuzzy clustering, NFS, fuzzy neural networks, and
with the ability to perform need analysis, capture complex interdepen­ fuzzy Bayesian networks, have low implementation complexity since a
dence amongst factors, and perform a mix of qualitative and quantita­ variety of computerized functions and commercial software packages
tive data analysis. Planning/management problems need techniques exist and can be used in their development. Similarly, fuzzy simulation
with the ability to model a large number of parameters. Process techniques, including FMCS, FDES, FSD, and FABM, are developed with
modeling and system modeling problems require techniques having the various computer-based applications and software packages. On the
ability to avoid local minima trapping, capture dynamic systems and other hand, most current fuzzy MCDM and optimization techniques rely
relationships among variables, model a large number of parameters, and on sophisticated formulations and development processes, so they often
perform scenario and sensitivity analyses. Fuzzy hybrid techniques obtain a high level of implementation complexity.
applied to performance problems require fuzzy hybrid techniques with

13
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

5.2. Suitability of fuzzy hybrid techniques to solve construction problems qualitative inputs, model highly dimensional and complex data, model a
large number of parameters, and capture uncertainty and vagueness of
Based on the detailed content analysis of 255 fuzzy hybrid studies model outputs. In this study, the appropriateness of a given fuzzy hybrid
across six CEM application areas, this study determined the suitability of technique to solve problems within a particular CEM application area
fuzzy hybrid techniques to solve problems within specific CEM appli­ was assessed based on the degree to which it satisfies the specific se­
cation areas. Each CEM application area consists of a set of specific se­ lection criteria within that area. A thorough analysis of selected fuzzy
lection criteria derived from the six categories shown in Table 7. For hybrid studies was performed to determine the degree of suitability of
example, fuzzy hybrid techniques selected to solve prediction problems fuzzy hybrid techniques for CEM problems, which was described in
need to satisfy a set of specific selection criteria including classification terms of the linguistic assessments expressed as “High,” “Moderate,” and
accuracy, prediction accuracy, implementation complexity, and the “Low.” For instance, in evaluation/assessment problems, the majority of
ability to perform trend analysis and pattern recognition in predictive fuzzy MCDM techniques (e.g., fuzzy AHP, fuzzy TOPSIS, fuzzy VIKOR,
models (e.g., time series), accommodate a mix of quantitative and fuzzy PROMETHEE, fuzzy ELECTRE) have the ability to prioritize

Table 8
Selection checklist for fuzzy hybrid techniques in CEM research.
CEM Application Specific Selection Criteria Fuzzy Hybrid Techniques* Reference
Category
Optimi- Machine MCDM Simu-
zation Learning lation

Ability to perform trend analysis and pattern recognition in predictive models [81] [191]
High
(e.g., time series) [194]
Classification accuracy High [10] [14] [40]
Prediction accuracy High [10] [40]
Ability to accommodate a mix of quantitative and qualitative inputs Mod. High [38] [194]
Prediction Ability to model highly dimensional and complex data High High [7] [10]
Ability to model a large number of parameters High High [4] [5] [37]
Ability to capture uncertainty and vagueness of model outputs Mod. High [7] [10]
Applicable to CEM research with advanced technologies (e.g., BIM, GIS, VR,
High [1] [29]
sensors, modular construction)
Implementation complexity High Low [189]
Ability to perform need analysis High [10] [40]
Ability to prioritize available alternatives for determining the optimal option High [5] [37]
Ability to capture complex interdependence amongst factors High [37]
Mix of qualitative and quantitative data analysis Low Low High Low [37] [194]
Ability to accommodate subjective judgment and preferences of multiple
Mod. High High Mod. [7] [39] [193]
decision makers
Evaluation/
Ability to achieve high transparency (i.e., interpretability) without lessening
Assessment Mod. Mod. High Mod. [10] [191]
accuracy
Ability to obtain high validity Low Mod. High Mod. [37] [193]
Ability to model a large number of parameters Mod. High High Mod. [37]
Applicable to CEM research with advanced technologies (e.g., BIM, GIS, VR,
Low Low High [1] [28]
sensors, modular construction)
Implementation complexity High Low Mod. Low [189]
Ability to achieve high optimization effectiveness and efficiency High [5]
Ability to achieve high transparency (i.e., interpretability) without lessening
High High High [10] [191]
accuracy
Planning/ Ability to model a large number of parameters High High High [7] [37]
Management Ability to capture uncertainty and vagueness of model outputs Mod. High High [7] [10]
Applicable to CEM research with advanced technologies (e.g., BIM, GIS, VR,
Mod. Mod. [1] [30] [195]
sensors, and modular construction)
Implementation complexity High Low Mod. [189]
Ability to avoid local minima trapping High [68]
Ability to capture dynamic systems and relationships among variables High [5] [190]
Processing and computational speeds High High [68] [192]
Ability to capture uncertainty and vagueness of model outputs High High [7] [10]
Process modeling
Ability to achieve high transparency (i.e., interpretability) without lessening
High Mod. [10] [191]
accuracy
Ability to model a large number of parameters High Mod. [7] [37]
Implementation complexity Low Low [189]
Ability to achieve high optimization effectiveness and efficiency High [5]
Ability to avoid local minima trapping High [10] [68]
Ability to capture dynamic systems and relationships among variables High [7] [39]
System modeling
Ability to perform scenario and sensitivity analyses Low High [6] [38]
Ability to model a large number of parameters High Mod. [7] [37]
Implementation complexity Low Low [189]
Ability to perform trend analysis and pattern recognition in predictive models
High [81] [194]
(e.g., time series)
Ability to achieve high transparency (i.e., interpretability) without lessening
High High [7] [10] [40]
accuracy
Performance Processing and computational speeds High High High [10] [68]
Ability to accommodate subjective judgment and preferences of multiple
Mod. High High Mod. [7] [39]
decision makers
Ability to perform scenario and sensitivity analyses Low High [6] [38]
Implementation complexity High Low Mod. Low [189]
*
Mod. = Moderate

14
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

available alternatives for determining the optimal option [5,37,40]; including radio-frequency identification (RFID) transponders and
therefore, a “High” linguistic expression was given for fuzzy MCDM for readers, RFID with GPS chips, ultrasound, GPS receivers, and infrared,
that specific selection criterion. In planning/management problems, can be used to produce data for models utilizing fuzzy inference system.
although most fuzzy machine learning techniques, such as ANFIS, fuzzy Fuzzy MCDM and fuzzy hybrid simulation techniques are not found to
ANNs, fuzzy Bayesian belief network, and fuzzy pattern recognition, have been implemented in this application area.
have the ability to model a large number of parameters and capture In evaluation/assessment problems (48% of construction problems),
uncertainty and vagueness of model outputs, represented by a “High” in fuzzy hybrid MCDM techniques are highly recommended because of
each criterion, only fuzzy c-means clustering was found with uncom­ their ability to perform need analysis, prioritize available alternatives
plicated algorithm and ease of use [5,10,189]. As a result, a “Low” lin­ for determining the optimal option, and capture complexity and inter­
guistic expression was given to fuzzy machine learning in the dependence amongst selection criteria. Compared to fuzzy hybrid ma­
implementation complexity criterion. While evaluation/assessment chine learning, fuzzy hybrid simulation, and fuzzy hybrid optimization,
problems often require the ability to achieve high transparency (i.e., fuzzy MCDM techniques achieve higher transparency (i.e., interpret­
interpretability) without lessening accuracy, fuzzy hybrid optimization ability) and validity without lessening accuracy. Although their devel­
techniques, such as fuzzy particle swarm optimization, fuzzy ant colony opment and implementations are more complicated than fuzzy hybrid
systems, fuzzy optimum system hierarchy analysis, and adaptive fuzzy machine learning and fuzzy simulation, fuzzy MCDM techniques are
Pareto set, can only solve the problem with various assumptions made to recommended for evaluation/assessment problems. Fuzzy MCDM are
establish the constraints and objective functions [20,22,177,178,179]. also applicable in CEM research on advanced technologies (e.g., BIM,
However, other fuzzy hybrid optimization techniques, such as fuzzy GIS, VR, modular construction). For instance, Chen et al. [27] developed
linear programming and fuzzy evolutionary algorithms, do not require a a semi-automatic, image-based BIM model using fuzzy multi-attribute
comprehensive set of assumptions in evaluation/assessment problems in utility theory (MAUT) for evaluating sustainable development
the current CEM literature. Therefore, a “Moderate” linguistic expres­ throughout construction project lifecycles. Fuzzy MAUT algorithms help
sion was given to fuzzy hybrid optimization in evaluation/assessment the developed model recognize structure object types and the corre­
problems. Similarly, the suitability of given fuzzy hybrid techniques to sponding materials used to support later assessments in the operation
CEM problems was determined according to specific selection criteria in and maintenance phase. Fazeli et al. [121] incorporated BIM and fuzzy
each CEM application category. Table 8 summarizes a selection checklist TOPSIS to optimize selection of sustainable building components at
for four categories of fuzzy hybrid techniques (optimization, machine early conceptual design stages. BIM-based fuzzy TOPSIS derives out­
learning, MCDM, and simulation) across six CEM applications areas comes for decision making to improve sustainability of building con­
(prediction, evaluation/assessment, planning/management, process struction based on design, quality, and economic factors. Figueiredo
modeling, system modeling, and performance) with relevant references et al. [15] also implemented a BIM-based fuzzy AHP approach to support
supporting the linguistic terms. selection of sustainable construction materials for ensuring the devel­
Because of their ability to perform trend analysis and pattern opment of buildings’ lifecycles. Investigating the use of BIM in CEM,
recognition in predictive models (e.g., time series) and obtain high ac­ Mirhosseini et al. [128] used fuzzy decision-making trial and evaluation
curacy in classification and prediction modeling outcomes [10,40], laboratory-ANP (DEMATEL-ANP) to depict the cause-and-effect re­
fuzzy machine learning techniques, including NFS techniques, fuzzy lationships among leadership competencies required for BIM imple­
clustering, fuzzy pattern recognition, and fuzzy Bayesian networks, are mentations and weight them. Turskis et al. [30] developed a GIS-based
highly recommended for solving practical prediction problems, which model using fuzzy AHP and weighted aggregated sum product assess­
comprise 19% of construction problems. Fuzzy machine learning tech­ ment (WASPAS) to support selection of shopping center construction
niques are also highly capable of modeling highly dimensional and sites. Fuzzy MCDM techniques (e.g., fuzzy Delphi, fuzzy TOPSIS) are
complex data and handling a large number of parameters. Compared also utilized with VR-based applications for supporting decision making
with fuzzy hybrid optimization, fuzzy machine learning techniques in interior design for office buildings [28]. Fuzzy AHP, another fuzzy
perform better in accommodating a mix of quantitative and qualitative MCDM technique, is also suggested for assessing project risks in modular
inputs and capturing uncertainty and vagueness of model outputs. Fuzzy construction. Fuzzy hybrid machine learning techniques are being
hybrid machine learning techniques are also applicable in CEM research considered more in modular construction because they can accommo­
with advanced technologies (e.g., BIM, sensors). For example, Lin and date subjective judgment and preferences of multiple decision makers
Huang [32] utilized common fuzzy classification machine learning and model a large number of parameters with a low implementation
techniques, including decision tree (DT), support vector machine (SVM), complexity [198].
and k-nearest neighbor (k-NN), to filter and assess irrelevant design For planning/management problems (20% of construction prob­
clashes detected by BIM at an early stage and increase accuracy of final lems), fuzzy hybrid optimization techniques are recommended, because
construction documents. The incorporation of multiple supervised fuzzy they can achieve high optimization effectiveness and efficiency
machine learning techniques results in a BIM-based automatic filtering compared to fuzzy hybrid machine learning and fuzzy MCDM tech­
system that identifies irrelevant design clashes in order to avoid po­ niques. In addition, fuzzy hybrid optimization techniques are applicable
tential delays and incurred costs. Ouma and Hahn [29] integrated fuzzy to CEM research with BIM applications. For example, Marzouk and
c-means clustering techniques and imaging accelerometer sensors to Abdelakder [185] implemented a hybrid fuzzy multi-objective non-
support pothole detection on asphalt pavement in highway construction dominated sorting genetic algorithm to compute several types of con­
projects. Such integration can provide highway agencies with an struction emissions (e.g., greenhouse gas footprint, sulfur dioxide) and
optimal compromise between cost, accuracy, and applicability in fuzzy TOPSIS to select the most sustainable construction material al­
detecting potholes on asphalt pavement. Razavi and Haas [196] ternatives based on overall project emissions and total used primary
implemented wireless sensors with a fuzzy machine learning technique energy. Moon et al. [195] developed a BIM-based computer system
to improve automated materials tracking and locating on construction using fuzzy generic algorithms for construction scheduling simulation to
sites. Their results showed that applications of sensors-based fuzzy minimize overlaps in construction activities and improve projects’
inference system can provide accurate location estimation and move­ operational performance. Their results can help project managers
ment detection to assist project managers with efficiently monitoring actively monitor and evaluate project activities with optimal schedule
and planning construction activities. In follow-up research, Razavi and alternatives visualized in a BIM-based four-dimensional computer-aided
Haas [197] continuously promoted the integration of tracking sensors design environment. Where CEM researchers cannot overcome the high
with fuzzy machine learning hybrid techniques to improve construction implementation complexity of fuzzy hybrid optimization techniques,
productivity. Their study showed that a wide range of wireless sensors, both fuzzy hybrid machine learning and fuzzy MCDM are also suitable to

15
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

planning/management problems because of their ability to achieve high increasingly used in CEM research because of their ability to simulta­
transparency (i.e., interpretability) without lessening accuracy, model a neously capture multiple uncertainties in construction problems, such as
large number of parameters, and capture uncertainty and vagueness of subjectivity, vagueness, and incompleteness, to augment existing
model outputs with low or moderate implementation complexity. Fuzzy modeling and computing efforts. Although they have been widely used
hybrid simulation techniques are not found to have been implemented in many CEM application areas to solve complex problems involving
in this application area. incompleteness of data, vagueness, and subjective uncertainty, no
In process modeling problems (7% of construction problems), fuzzy detailed content analysis has previously been conducted regarding an
hybrid simulation techniques are recommended because of their ability integrated categorization and itemization of fuzzy hybrid technique
to capture dynamic systems and relationships among input variables and applications in solving specific CEM problems. Therefore, this paper
to avoid local minima trapping. However, fuzzy hybrid machine provides a systematic literature review of fuzzy hybrid techniques used
learning techniques perform better than fuzzy hybrid simulation tech­ in CEM research publications from highly reputed CEM journals from
niques in such problems, having higher transparency (i.e., interpret­ 2004 to 2021. The content analysis results show that fuzzy hybrid
ability) without lessening accuracy and having the ability to model a techniques applied in CEM can be classified into four categories: fuzzy
large number of parameters (high dimensionality). If CEM researchers hybrid machine learning, fuzzy hybrid MCDM, fuzzy hybrid simulation,
and practitioners seek the ability to capture uncertainty and vagueness and fuzzy hybrid optimization. This paper also matches individual fuzzy
of model outputs and to have fast processing and computational speeds, hybrid techniques to specific CEM application areas, comprising pre­
both fuzzy machine learning and fuzzy simulation techniques are suit­ diction, evaluation/assessment, planning/management, process
able. These fuzzy hybrid techniques also obtain a low implementation modeling, system modeling, and performance management. The results
complexity. Fuzzy MCDM and fuzzy optimization techniques have not of this study show that fuzzy hybrid machine learning and fuzzy hybrid
been found in the application area of process modeling. MCDM are widely implemented in CEM research. Specifically, fuzzy
Similarly, in system modeling problems (2% of construction prob­ hybrid machine learning techniques are mostly used to develop pre­
lems), fuzzy hybrid simulation techniques are highly recommended dictive models and deal with forecasting problems in CEM research.
because of their ability to achieve high optimization effectiveness and Fuzzy hybrid MCDM techniques are often selected for supporting
efficiency, avoid local minima trapping, capture dynamic systems and decision-making processes in evaluation/assessment applications. Fuzzy
relationships among variables, and perform scenario and sensitivity hybrid optimization techniques are commonly used to handle optimi­
analyses. If CEM researchers in this application area seek to handle a zation problems and improve planning and management of construction
large number of parameters, fuzzy hybrid machine learning may be a projects. Fuzzy hybrid simulation techniques are essentially used to
better choice than fuzzy hybrid simulation techniques. Fuzzy MCDM and enhance process and system modeling, providing CEM researchers and
fuzzy optimization techniques are not been found to have been imple­ industry practitioners with different decision-making scenarios.
mented in this application area. Results from the detailed content analysis show that each integrated
In performance problems (4% of construction problems), fuzzy ma­ category of fuzzy hybrid techniques in CEM research involves various
chine learning techniques are recommended because of their ability to hybridizations of fuzzy logic with other standard modeling and
perform trend analysis and pattern recognition in predictive models (e. computing techniques. Single-hybrid fuzzy techniques, which are hy­
g., time series), achieve high transparency (i.e., interpretability) without bridization of fuzzy logic with a standard technique, include fuzzy
lessening accuracy, have fast processing and computational speeds, and particle swarm optimization, FANP, FSD, and FMCS. Multi-hybrid fuzzy
achieve low implementation complexity. Fuzzy MCDM techniques are techniques, which are hybridization of fuzzy logic with two or more
also recommended for accommodating subjective judgment and pref­ standard techniques, include fuzzy AHP-TOPSIS, fuzzy Monte-Carlo
erences of multiple decision makers and to achieve high transparency (i. ABM, and FANP-DEMATEL. The majority of fuzzy hybrid machine
e., interpretability) without lessening accuracy. Fuzzy hybrid simulation learning techniques consist of NFS (hybridization of fuzzy logic and
techniques are suitable for prioritizing a model’s ability to perform ANNs) for prediction/forecasting problems and fuzzy clustering for
scenario and sensitivity analyses, obtain fast processing and computa­ classification problems. Researchers have also used some fuzzy hybrid
tional speeds, and achieve low implementation complexity. machine learning techniques, such as fuzzy c-means clustering and fuzzy
In summary, fuzzy hybrid machine learning techniques are recom­ pattern recognition, to analyze large datasets collected via sensors and
mended for solving problems in the prediction and performance appli­ wearable devices in construction projects. Fuzzy MCDM techniques,
cation areas, because these techniques have the ability to perform trend including fuzzy AHP, fuzzy TOPSIS, and fuzzy VIKOR, have been widely
analysis and pattern recognition in predictive models (e.g., time series) applied in evaluation/assessment problems that involve a large amount
and obtain high classification and prediction accuracy. Additionally, of subjective uncertainty, ambiguity, and incomplete data from various
CEM applications with sensors and BIM can use fuzzy hybrid machine decision makers and/or stakeholders. Fuzzy MCDM techniques have
learning to handle a large volume of data. Fuzzy hybrid MCDM tech­ also been used in studies involving advanced technologies, such as BIM,
niques are recommended in the evaluation/assessment application area GIS, VR, and modular construction, to support different decision-making
because they are capable of performing need analysis, prioritizing scenarios under uncertainty. Fuzzy hybrid optimization techniques,
available alternatives for determining the optimal option, and capturing including fuzzy evolutionary algorithms, fuzzy particle swarm optimi­
complex interdependence amongst criteria. Fuzzy MCDM techniques are zation, and fuzzy ant colony optimization, are mostly applied in plan­
also found in CEM research with BIM, GIS, VR, and modular construc­ ning/management problems because of their ability to provide efficient
tion. Fuzzy hybrid optimization techniques are recommended in the optimization outcomes of time-cost trade-offs and resource usage. Fuzzy
planning/management application area, because they can achieve high hybrid simulation techniques, such as FMCS, FDES, FSD, and FABM,
optimization effectiveness and efficiency. Fuzzy hybrid simulation have been used in modeling and simulating construction systems and
techniques are recommended for solving problems within the process processes.
and system modeling application areas, because they can avoid local This study also identified and determined common categories of
minima trapping as well as capture dynamic systems and relationships selection criteria, including accuracy, data availability, computational
among input variables. complexity, interpretability, processing ability, and implementation
complexity, for fuzzy hybrid applications in CEM research. Each of these
6. Conclusions and future research categories comprises a list of specific selection criteria required for
particular CEM application areas. From the specific selection criteria
This paper presents a state-of-the-art review of applications of fuzzy within each CEM application area, this study examined suitable fuzzy
hybrid techniques in CEM research. Fuzzy hybrid techniques have been hybrid techniques to solve particular CEM problems. Based on the study

16
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

results, fuzzy hybrid machine learning techniques (e.g., NFS, fuzzy other fuzzy hybrid applications in other civil engineering domains. As a
clustering) are recommended for developing predictive modeling to result, although the selection criteria (Table 7) and the selection
forecast and monitor construction productivity, project cost manage­ checklist for fuzzy hybrid techniques in CEM research (Table 8) can
ment, and construction structural and materials testing. Fuzzy hybrid assist researchers in selecting an appropriate fuzzy hybrid technique
MCDM techniques are recommended for tackling decision-making according to their experience and the purpose of the model, the selection
problems that involve multiple decision makers and/or stakeholders in criteria should be considered as recommendations only and require
project delivery method selection, risk assessment and management, further validations in future research.
construction material selection, project plan selection, supplier selec­ Future research directions may be made based on potential im­
tion, project team development, sustainability analysis, and site layout provements in each category of fuzzy hybrid techniques and their
planning. Fuzzy hybrid simulation techniques are recommended for applicability to CEM studies that involve advanced technologies in
capturing the high dynamism and random uncertainty in process and construction. Performance of fuzzy hybrid machine learning techniques
system modeling problems, such as project scheduling, project resource (e.g., fuzzy clustering, NSF) can be improved by incorporating other AI
management, risk management, and construction productivity. Fuzzy (e.g., pattern recognition, data mining) and population-based optimi­
hybrid optimization techniques (e.g., fuzzy evolutionary algorithms, zation algorithms (e.g., evolutionary algorithms, particle swarm opti­
fuzzy particle swarm optimization, fuzzy ant colony optimization) are mization, firefly algorithm). Specifically, fuzzy clustering can be
recommended for solving complex multi-objective optimization prob­ incorporated with pattern recognition to improve a model’s ability to
lems, such as project time-cost trade-offs, site layout planning, supply provide valid inferences and support decision-making problems in CEM
chain management, and project scheduling. The results of this study can research. The integrated fuzzy Bayesian inference system can be a robust
assist CEM researchers and industry practitioners in selecting appro­ decision-aid model with the ability to handle linguistically expressed
priate fuzzy hybrid techniques for solving problems that involve information and subjective uncertainty along with updating new infor­
advanced technologies, including BIM, GIS, VR, sensors, and modular mation in model development. In addition, fuzzy clustering can be
construction. combined with other data mining techniques, such as feature selection
This study makes several contributions to the body of knowledge by and/or hybrid feature selection, to reduce the dimensionality of pre­
providing a state-of-the-art review of existing fuzzy hybrid techniques dictive models. NFS models can be enhanced by incorporating evolu­
utilized in CEM research. First, the systematic review demonstrates the tionary algorithms to optimize and tune model parameters. Results of
capabilities of fuzzy hybrid techniques to overcome limitations of some this study show that most current NFS models are tied with type-1 fuzzy
standard techniques in solving complex construction problems. sets; therefore, future research can attempt to improve performance of
Accordingly, the results of this study determined limitations of standard NFS models by incorporating type-2 fuzzy sets (a more generalized form
modeling and computing methods, such as machine learning, MCDM, of type-1 fuzzy sets) that can handle more uncertainty in the modeling
simulation, and optimization, in solving practical construction prob­ process. NFS models can also be integrated with simulation techniques,
lems, and thus hybridization of fuzzy logic with standard methods is such as Monte Carlo simulation and system dynamics, to enhance the
recommended. Second, the content analysis results provide detailed and performance of predictive models for estimating construction duration
integrated categorization of fuzzy hybrid techniques across different in earthwork operations and tunneling projects as well as forecasting
CEM application areas. Additionally, the categorization of fuzzy hybrid time series trends in construction projects (e.g., renewable energy, oil
techniques matching with specific construction application areas – price). Moreover, using machine learning techniques can also help fuzzy
prediction, evaluation/assessment, planning/management, process hybrid simulation techniques, such as FSD and FABM, overcome the
modeling, system modeling, and project performance – helps illustrate limitation of data availability. Although fuzzy hybrid MCDM techniques
the capabilities of fuzzy hybrid techniques to efficiently and effectively are capable of simultaneously capturing subjectivity, multiple objec­
solve practical construction problems. Finally, this study identifies and tives, and randomness of construction problems, they are unable to
develops selection criteria that can help guide CEM researchers and accommodate the interrelationships among model variables and the
industry practitioners in choosing and examining the applicability of high dynamism of construction processes. Therefore, some fuzzy hybrid
each fuzzy hybrid technique category to given practical construction simulation techniques, such as FSD, FDES, and FMCS, can be incorpo­
problems. rated with fuzzy MCDM in future studies. For example, FSD can be used
This study has several limitations. First, the number of articles ob­ with fuzzy AHP in supporting maintenance decision making in highway
tained from construction journals for each fuzzy hybrid category was road projects by assessing the evolution of road conditions and per­
uneven because of different usages and the commonness of fuzzy hybrid forming MCDM. Based on the aforementioned suggestions from this
techniques in CEM research. For example, fewer studies have concen­ paper, future research can develop more robust models with greater
trated on fuzzy hybrid simulation than other fuzzy hybrid techniques (e. capacity to tackle many facets of construction problems, including
g., machine learning, MCDM, and optimization) since implementations incompleteness of data, vagueness, often-conflicting requirements,
of FDES and FABM are still in their infancy in CEM. In addition, some subjective uncertainty, and random uncertainty.
articles related to fuzzy hybrid techniques published in journals that Fuzzy hybrid techniques have been increasingly used to support se­
were not included the reviewed databases might not be collected and lection of advanced technologies (e.g., BIM, modular construction) in
included in the content analysis. Therefore, some studies focusing on construction as well as modeling and computing the data collected using
fuzzy hybrid techniques in other discipline areas, such as storm water cutting-edge technologies (e.g., sensors, GIS, VR). Most collected studies
design, infrastructure planning and management, pavement design, related to advanced technologies in construction involve the majority of
structural design, structural analysis, structural deterioration and dam­ fuzzy MCDM techniques implemented in evaluation/assessment and
age detection, sustainability, and project management, were also planning/management application areas. Future development and
incorporated in the analysis to maximize the coverage of collected ar­ deployment of fuzzy hybrid applications in CEM can be made based on
ticles and secure the reliability and validity of outcomes from the con­ the promising integration of fuzzy hybrid techniques with advanced
tent analysis. Discussions regarding the use of fuzzy hybrid techniques in technologies, alternative reality technologies, and emerging innovations
the two CEM application categories of performance and system such as big data analytics. For example, it is possible that applications of
modeling are limited because few studies are available in the current BIM-based fuzzy optimization models can be used in construction to
construction literature. Second, the proposed six categories of perfor­ support high-level visualization and assessment of planning and man­
mance evaluation criteria – data availability, processing ability, agement issues. Although there is a lack of GIS-based studies using fuzzy
computational complexity, interpretability, accuracy, and implementa­ hybrid techniques in CEM, the use of fuzzy MCDM and optimization
tion complexity – were obtained from a content analysis that included techniques in a GIS environment to support construction site layout and

17
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

planning in promising. GIS-based fuzzy hybrid approaches (e.g., fuzzy grey wolf optimizer, Constr. Build. Mater. 232 (2020), 117266, https://doi.org/
10.1016/j.conbuildmat.2019.117266.
MCDM, fuzzy hybrid optimization) are promising tools for handling
[13] H. Madani, M. Kooshafar, M. Emadi, Compressive strength prediction of
incomplete, imprecise, and subjective uncertainties in decision-making nanosilica-incorporated cement mixtures using adaptive neuro-fuzzy inference
and optimization processes for construction site planning and manage­ system and artificial neural network models, Pract. Period. Struct. Des. Constr. 25
ment. The current literature also indicates future opportunities in up­ (3) (2020) 04020021, https://doi.org/10.1061/(asce)sc.1943-5576.0000499.
[14] P. Nguyen, D. Tran, B. Lines, Fuzzy set theory approach to classify highway
trend integrations of fuzzy hybrid techniques with big data analytics. To project characteristics for delivery selection, J. Constr. Eng. Manag. 146 (5)
accommodate large datasets, fuzzy hybrid machine learning techniques (2020) 04020044, https://doi.org/10.1061/(asce)co.1943-7862.0001829.
are promising tools because of their ability to develop fast and sufficient [15] K. Figueiredo, R. Pierott, A. Hammad, A. Haddad, Sustainable material choice for
construction projects: a life cycle sustainability assessment framework based on
modeling in predicting future outcomes. As a subset of AI, fuzzy machine BIM and fuzzy-AHP, Build. Environ. 196 (2021), 107805, https://doi.org/
learning can quickly train a machine to learn and automatically produce 10.1016/j.buildenv.2021.107805.
models that can analyze large and complex datasets. This future direc­ [16] L. Beltrão, M. Carvalho, Prioritizing construction risks using fuzzy AHP in
Brazilian public enterprises, J. Constr. Eng. Manag. 145 (2) (2019) 05018018,
tion can help improve modeling and analysis efforts in labor produc­ https://doi.org/10.1061/(asce)co.1943-7862.0001606.
tivity, performance, health and safety behaviors, and planning practices [17] Y. Yang, J. Peng, C. Cai, J. Zhang, Improved interval evidence theory-based fuzzy
with big data technologies, such as the internet of things (IoT) and AHP approach for comprehensive condition assessment of long-span PSC
continuous box-girder bridges, J. Bridg. Eng. 24 (12) (2019) 04019113, https://
sensors. Fuzzy hybrid techniques can help researchers investigate hid­ doi.org/10.1061/(ASCE)BE.1943-5592.0001494.
den patterns, correlations, and other insights related to workers’ ergo­ [18] S. Mete, F. Serin, N. Oz, M. Gul, A decision-support system based on Pythagorean
nomics, labor health conditions, productivity, performance, and safety fuzzy VIKOR for occupational risk assessment of a natural gas pipeline
construction, J. Natural Gas Sci. Eng. 71 (2019), 102979, https://doi.org/
risks. Future research trends in construction may include hybridization
10.1016/j.jngse.2019.102979.
of fuzzy logic with other AI methods, such as natural language pro­ [19] M. Bagheri, B. Keshtegar, S. Zhu, D. Meng, J. Correia, A. De Jesus, Fuzzy
cessing and Markov decision process for complex CEM problems that reliability analysis using genetic optimization algorithm combined with adaptive
require heavy computational intelligence. descent chaos control, ASCE-ASME J. Risk Uncertainty Eng. 6 (2) (2020)
04020022, https://doi.org/10.1061/ajrua6.0001064.
[20] X. Song, L. Zhong, Z. Zhang, J. Xu, C. Shen, F. Peña-Mora, Multistakeholder
Declaration of Competing Interest conflict minimization-based layout planning of construction temporary facilities,
J. Comput. Civ. Eng. 32 (2) (2018) 04017080, https://doi.org/10.1061/(ASCE)
CP.1943-5487.0000725.
The authors declare that they have no known competing financial [21] H. Liu, X. Wang, Y. Jiao, X. He, B. Wang, Condition evaluation for existing
interests or personal relationships that could have appeared to influence reinforced concrete bridge superstructure using fuzzy clustering improved by
the work reported in this paper. particle swarm optimization, Struct. Infrastruct. Eng. 13 (7) (2017) 955–965,
https://doi.org/10.1080/15732479.2016.1227854.
[22] H. Abbasianjahromi, H. Rajaie, E. Shakeri, F. Chokan, A new decision-making
Acknowledgments model for subcontractor selection and its order allocation, Proj. Manag. J. 45 (1)
(2014) 55–66, https://doi.org/10.1002/pmj.21394.
[23] H. Etemadinia, M. Tavakolan, Using a hybrid system dynamics and interpretive
This research is funded by the Natural Sciences and Engineering structural modeling for risk analysis of design phase of the construction projects,
Research Council of Canada Industrial Research Chair in Strategic Int. J. Constr. Manag. 21 (1) (2018) 93–112, https://doi.org/10.1080/
Construction Modeling and Delivery (NSERC IRCPJ 428226–15), which 15623599.2018.1511235.
[24] M. Raoufi, A.R. Fayek, Fuzzy Monte Carlo agent-based simulation of construction
is held by Dr. Aminah Robinson Fayek. crew performance, J. Constr. Eng. Manag. 146 (5) (2020) 04020041, https://doi.
org/10.1061/(asce)co.1943-7862.0001826.
References [25] M. Attarzadeh, D.K. Chua, M. Beer, E.L. Abbott, Fuzzy randomness simulation of
long-term infrastructure projects, ASCE-ASME J. Risk Uncertainty Eng. 3 (3)
(2017) 04017002, https://doi.org/10.1061/AJRUA6.0000902.
[1] T. Tan, G. Mills, E. Papadonikolaki, Z. Liu, Combining multi-criteria decision
[26] N. Sadeghi, A.R. Fayek, N. Gerami Seresht, A fuzzy discrete event simulation
making (MCDM) methods with building information modelling (BIM): a review,
framework for construction applications: improving the simulation time
Autom. Constr. 121 (2021), 103451, https://doi.org/10.1016/j.
advancement, J. Constr. Eng. Manag. 142 (12) (2016), 04016071, https://doi.
autcon.2020.103451.
org/10.1061/(ASCE)CO.1943-7862.0001195.
[2] K. Chatterjee, E. Zavadskas, J. Tamošaitienė, K. Adhikary, S. Kar, A hybrid MCDM
[27] L. Chen, Q. Lu, X. Zhao, A semi-automatic image-based object recognition system
technique for risk management in construction projects, Symmetry 10 (2) (2018)
for constructing as-is IFC BIM objects based on fuzzy-MAUT, Int. J. Constr.
46, https://doi.org/10.3390/sym10020046.
Manag. (2019) 1–15, https://doi.org/10.1080/15623599.2019.1615754.
[3] X. Ning, L.Y. Ding, H.B. Luo, S.J. Qi, A multi-attribute model for construction site
[28] Y. Juan, H. Chi, H. Chen, Virtual reality-based decision support model for interior
layout using intuitionistic fuzzy logic, Autom. Constr. 72 (2016) 380–387,
design and decoration of an office building, Eng. Constr. Archit. Manag. 28 (1)
https://doi.org/10.1016/j.autcon.2016.09.008.
(2019) 229–245, https://doi.org/10.1108/ecam-03-2019-0138.
[4] S. Kar, S. Das, P. Ghosh, Applications of neuro fuzzy systems: a brief review and
[29] Y.O. Ouma, M. Hahn, Pothole detection on asphalt pavements from 2D-colour
future outline, Appl. Soft Comput. 15 (2014) 243–259, https://doi.org/10.1016/
pothole images using fuzzy c-means clustering and morphological reconstruction,
j.asoc.2013.10.014.
Autom. Constr. 83 (2017) 196–211, https://doi.org/10.1016/j.
[5] N. Gerami Seresht, R. Lourenzutti, A. Salah, A.R. Fayek, Overview of fuzzy hybrid
autcon.2017.08.017.
techniques in construction engineering and management, in: A.R. Fayek (Ed.),
[30] Z. Turskis, E. Zavadskas, J. Antucheviciene, N. Kosareva, A hybrid model based
Fuzzy Hybrid Computing in Construction Engineering and Management: Theory
on fuzzy AHP and fuzzy WASPAS for construction site selection, International J.
and Applications, Emerald Publishing, Bingley, UK, 2018, pp. 37–107. ISBN 978-
Comput. Commun. Control 10 (6) (2015) 113, https://doi.org/10.15837/
1-7874-3869-9.
ijccc.2015.6.2078.
[6] J.N. Hooker, Hybrid modeling, in: Hybrid Optimization, Springer, New York, NY,
[31] A. Hoseini, S. Ghannadpour, S. Noori, M. Yazdani, Analysis of the influence of
USA, 2011, pp. 11–62. ISBN 978-1-4419-1644-0.
building information modeling (BIM) on construction project management areas
[7] A.R. Fayek (Ed.), Fuzzy Hybrid Computing in Construction Engineering and
of knowledge: a hybrid FANP-FVIKOR approach, Int. J. Indust. Eng. Product. Res.
Management, 1st ed., Emerald Publishing Limited, Bingley, UK, 2018,
30 (1) (2019) 57–92, https://doi.org/10.22068/ijiepr.30.1.57.
pp. 37–107. ISBN 978-1-7874-3869-9.
[32] W. Lin, Y. Huang, Filtering of irrelevant clashes detected by BIM software using a
[8] L. Suganthi, S. Iniyan, A. Samuel, Applications of fuzzy logic in renewable energy
hybrid method of rule-based reasoning and supervised machine learning, Appl.
systems: A review, Renew. Sustain. Energy Rev. 48 (2015) 585–607, https://doi.
Sci. 9 (24) (2019) 5324, https://doi.org/10.3390/app9245324.
org/10.1016/j.rser.2015.04.037.
[33] M. Rostamnezhad, F. Nasirzadeh, M. Khanzadi, M. Jarban, M. Ghayoumian,
[9] M. Islam, M. Nepal, A fuzzy-Bayesian model for risk assessment in power plant
Modeling social sustainability in construction projects by integrating system
projects, Proc. Comput. Sci. 100 (2016) 963–970, https://doi.org/10.1016/j.
dynamics and fuzzy-DEMATEL method: a case study of highway project, Eng.
procs.2016.09.259.
Constr. Archit. Manag. 27 (7) (2020) 1595–1618, https://doi.org/10.1108/ecam-
[10] G. Tiruneh, A.R. Fayek, V. Sumati, Neuro-fuzzy systems in construction
01-2018-0031.
engineering and management research, Autom. Constr. 119 (2020), 103348,
[34] E. Zarei, N. Khakzad, V. Cozzani, G. Reniers, Safety analysis of process systems
https://doi.org/10.1016/j.autcon.2020.103348.
using fuzzy Bayesian network (FBN), J. Loss Prev. Process Ind. 57 (2019) 7–16,
[11] C. Qi, H. Ly, L. Le, X. Yang, L. Guo, B. Pham, Improved strength prediction of
https://doi.org/10.1016/j.jlp.2018.10.011.
cemented paste backfill using a novel model based on adaptive neuro fuzzy
[35] T. Gebrehiwet, H. Luo, Risk level evaluation on construction project lifecycle
inference system and artificial bee colony, Constr. Build. Mater. 284 (17) (2021),
using fuzzy comprehensive evaluation and TOPSIS, Symmetry 11 (1) (2018) 12,
122857, https://doi.org/10.1016/j.conbuildmat.2021.122857.
https://doi.org/10.3390/sym11010012.
[12] E. Golafshani, A. Behnood, M. Arashpour, Predicting the compressive strength of
normal and high-performance concretes using ANN and ANFIS hybridized with

18
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

[36] L.A. Zadeh, Fuzzy sets, Inf. Control. 8 (3) (1965) 338–353, https://doi.org/ [62] L. Yawei, C. Shouyu, N. Xiangtian, Fuzzy pattern recognition approach to
10.1016/S0019-9958(65)90241-X. construction contractor selection, Fuzzy Optim. Decis. Making 4 (2) (2005)
[37] L. Chen, W. Pan, Review fuzzy multi-criteria decision-making in construction 103–118, https://doi.org/10.1007/s10700-004-5867-4.
management using a network approach, Appl. Soft Comput. 102 (2021), 107103, [63] R. Soares, L. Barroso, O. Al-Fahdawi, Response attenuation of cable-stayed bridge
https://doi.org/10.1016/j.asoc.2021.107103. subjected to central US earthquakes using neuro-fuzzy and simple adaptive
[38] M. Islam, M. Nepal, M. Skitmore, M. Attarzadeh, Current research trends and control, Eng. Struct. 203 (2020), 109874, https://doi.org/10.1016/j.
application areas of fuzzy and hybrid methods to the risk assessment of engstruct.2019.109874.
construction projects, Adv. Eng. Inform. 33 (2017) 112–131, https://doi.org/ [64] W. Utama, A. Chan, H. Zahoor, R. Gao, D. Jumas, Making decision toward
10.1016/j.aei.2017.06.001. overseas construction projects, Eng. Constr. Archit. Manag. 26 (2) (2019)
[39] D. Simić, I. Kovačević, V. Svirčević, S. Simić, 50 years of fuzzy set theory and 285–302, https://doi.org/10.1108/ecam-01-2018-0016.
models for supplier assessment and selection: a literature review, J. Appl. Log. 24 [65] A. Peña, I. Bonet, C. Lochmuller, F. Chiclana, M. Góngora, An integrated inverse
(2017) 85–96, https://doi.org/10.1016/j.jal.2016.11.016. adaptive neural fuzzy system with Monte-Carlo sampling method for operational
[40] A.P. Chan, D.W. Chan, J.F. Yeung, Overview of the application of “fuzzy risk management, Expert Syst. Appl. 98 (2018) 11–26, https://doi.org/10.1016/j.
techniques” in construction management research, J. Constr. Eng. Manag. 135 eswa.2018.01.001.
(11) (2009) 1241–1252, https://doi.org/10.1061/(ASCE)CO.1943- [66] N. Gerami Seresht, A.R. Fayek, Neuro-fuzzy system dynamics technique for
7862.0000099. modeling construction systems, Appl. Soft Comput. 93 (2020) 106400, https://
[41] D. Moher, A. Liberati, J. Tetzlaff, D.G. Altman, Preferred reporting items for doi.org/10.1016/j.asoc.2020.106400.
systematic reviews and meta-analyses: the PRISMA statement, Ann. Intern. Med. [67] A. Tsehayae, A.R. Fayek, System model for analysing construction labour
151 (4) (2009) 264–269, https://doi.org/10.1136/bmj.b2535. productivity, Constr. Innov. 16 (2) (2016) 203–228, https://doi.org/10.1108/CI-
[42] A. Booth, A. Sutton, D. Papaioannou, Systematic Approaches to a Successful 07-2015-0040.
Literature Review, Sage, Thousand Oaks, CA, USA, 2016, pp. 36–55. ISBN 978-1- [68] K. Shihabudheen, G. Pillai, Recent advances in neuro-fuzzy system: a survey,
4739-1245-8. Knowl.-Based Syst. 152 (2018) 136–162, https://doi.org/10.1016/j.
[43] R. Fellows, A. Liu, Research Methods for Construction, 3rd ed, Blackwell Science, knosys.2018.04.014.
Oxford, UK, 2008, pp. 81–121. ISBN 140517790X. [69] F. Khademi, M. Akbari, S.M. Jamal, M. Nikoo, Multiple linear regression, artificial
[44] K. Krippendorff, Content Analysis: An Introduction to Its Methodology, 2nd ed., neural network, and fuzzy logic prediction of 28 days compressive strength of
Sage, Thousand Oaks, CA, USA, 2004, pp. 44–95. ISBN 9780761915454. concrete, Front. Struct. Civ. Eng. 11 (1) (2017) 90–99, https://doi.org/10.1007/
[45] PMI, A Guide to the Project Management Body of Knowledge (PMBOK® Guide), s11709-016-0363-9.
6th ed. Project Management Institute Inc., Newtown Square, PA, USA (2017), [70] W. Wang, T. Bilozerov, R. Dzeng, F. Hsiao, K. Wang, Conceptual cost estimations
ISBN 9781628251845, pp. 1–29. using neuro-fuzzy and multi-factor evaluation methods for building projects,
[46] W. Yi, A.P. Chan, Critical review of labor productivity research in construction J. Civ. Eng. Manag. 23 (1) (2017) 1–14, https://doi.org/10.3846/
journals, J. Manag. Eng. 30 (2) (2014) 214–225, https://doi.org/10.1061/(ASCE) 13923730.2014.948908.
ME.1943-5479.0000194. [71] J. Jang, ANFIS: adaptive-network-based fuzzy inference system, IEEE Trans. Eng.
[47] A. Darko, A.P. Chan, Critical analysis of green building research trend in Manag. 23 (3) (1993) 665–685, https://doi.org/10.1109/21.256541.
construction journals, Habitat Int. 57 (2016) 53–63, https://doi.org/10.1016/j. [72] M. Cheng, C. Ko, Object-oriented evolutionary fuzzy neural inference system for
habitatint.2016.07.001. construction management, J. Constr. Eng. Manag. 129 (4) (2003) 461–469,
[48] S. Dastgheib, M. Feylizadeh, M. Bagherpour, A. Mahmoudi, Improving estimate at https://doi.org/10.1061/(ASCE)0733-9364(2003)129:4(461).
completion (EAC) cost of construction projects using adaptive neuro-fuzzy [73] N.B. Siraj, A.R. Fayek, A.A. Tsehayae, Development and optimization of artificial
inference system (ANFIS), Can. J. Civ. Eng. (2021), https://doi.org/10.1139/cjce- intelligence-based concrete compressive strength predictive models, Int. J. Struct.
2020-0399. Civil Eng. Res. 5 (3) (2016) 156–167, https://doi.org/10.18178/ijscer.5.3.156-
[49] M. Rezaie, N. Sadighi, Prediction of slump and density of lightweight concretes 167.
using ANFIS and linear regression, Int. J. Civil Eng. Technol. 8 (10) (2017) [74] I. Mansouri, O. Kisi, Prediction of debonding strength for masonry elements
1635–1648, http://paper.researchbib.com/view/paper/159716. retrofitted with FRP composites using neuro fuzzy and neural network
[50] M. Cheng, N. Hoang, A. Roy, Y. Wu, A novel time-depended evolutionary fuzzy approaches, Compos. Part B 70 (2015) 247–255, https://doi.org/10.1016/j.
SVM inference model for estimating construction project at completion, Eng. compositesb.2014.11.023.
Appl. Artif. Intell. 25 (4) (2012) 744–752, https://doi.org/10.1016/j. [75] Z. Yuan, L. Wang, X. Ji, Prediction of concrete compressive strength: Research on
engappai.2011.09.022. hybrid models genetic based algorithms and ANFIS, Adv. Eng. Softw. 67 (2014)
[51] A. Kumar, A. Pal, A. Vohra, S. Gupta, S. Manchanda, M. Dash, Construction of 156–163, https://doi.org/10.1016/j.advengsoft.2013.09.004.
capital procurement decision making model to optimize supplier selection using [76] M. Cheng, H. Tsai, W. Hsieh, Web-based conceptual cost estimates for
Fuzzy Delphi and AHP-DEMATEL, Benchmarking 25 (5) (2018) 1528–1547, construction projects using evolutionary fuzzy neural inference model, Autom.
https://doi.org/10.1108/bij-01-2017-0005. Constr. 18 (2) (2009) 164–172, https://doi.org/10.1016/j.autcon.2008.07.001.
[52] X. Song, J. Xu, C. Shen, F. Peña-Mora, Z. Zeng, A decision making system for [77] C. Ko, M. Cheng, T. Wu, Evaluating sub-contractors performance using EFNIM,
construction temporary facilities layout planning in large-scale construction Autom. Constr. 16 (4) (2007) 525–530, https://doi.org/10.1016/j.
projects, International, J. Civ. Eng. 15 (2017) 333–353, https://doi.org/10.1007/ autcon.2006.09.005.
s40999-016-0107-1. [78] A. Rashidi, F. Jazebi, I. Brilakis, Neurofuzzy genetic system for selection of
[53] M. Raoufi, A.R. Fayek, Fuzzy agent-based modeling of construction crew construction project managers, J. Constr. Eng. Manag. 137 (1) (2011) 17–29,
motivation and performance, J. Comput. Civ. Eng. 32 (5) (2018) 04018035, https://doi.org/10.1061/(ASCE)CO.1943-7862.0000200.
https://doi.org/10.1061/(ASCE)CP.1943-5487.0000777. [79] F. Mirahadi, T. Zayed, Simulation-based construction productivity forecast using
[54] T. Kazanasmaz, Fuzzy logic model to classify effectiveness of daylighting in an neural-network–driven fuzzy reasoning, Autom. Constr. 65 (2016) 102–115,
office with a movable blind system, Build. Environ. 69 (2013) 22–34, https://doi. https://doi.org/10.1016/j.autcon.2015.12.021.
org/10.1016/j.buildenv.2013.07.011. [80] M.N. Omar, A.R. Fayek, Modeling and evaluating construction project
[55] D. Tran, M. Cheng, A. Pham, Using fuzzy clustering chaotic-based differential competencies and their relationship to project performance, Autom. Constr. 69
evolution to solve multiple resources leveling in the multiple projects scheduling (2016) 115–130, https://doi.org/10.1016/j.autcon.2016.05.021.
problem, Alexandria Eng. J. 55 (2) (2016) 1541–1552, https://doi.org/10.1016/ [81] T.J. Ross, Fuzzy Logic with Engineering Applications, 3rd ed., Wiley, West Sussex,
j.aej.2016.03.038. UK, 2010, pp. 117–400. ISBN: 978-0-470-74376-8.
[56] M. Cheng, D. Tran, Y. Wu, Using a fuzzy clustering chaotic-based differential [82] K. Aydin, O. Kisi, Applicability of a fuzzy genetic system for crack diagnosis in
evolution with serial method to solve resource-constrained project scheduling Timoshenko beams, J. Comput. Civ. Eng. 29 (5) (2015) 04014073, https://doi.
problems, Autom. Constr. 37 (2014) 88–97, https://doi.org/10.1016/j. org/10.1061/(ASCE)CP.1943-5487.0000385.
autcon.2013.10.002. [83] M. Marzouk, M. Alaraby, Predicting telecommunication tower costs using fuzzy
[57] H. Khamesi, S.R. Torabi, H. Mirzaei-Nasirabad, Z. Ghadiri, Improving the subtractive clustering, J. Civ. Eng. Manag. 21 (1) (2015) 67–74, https://doi.org/
performance of intelligent back analysis for tunneling using optimized fuzzy 10.3846/13923730.2013.802736.
systems: Case study of the Karaj Subway Line 2 in Iran, J. Comput. Civ. Eng. 29 [84] L. Yu, J. Zhu, L. Yu, Structural damage detection in a truss bridge model using
(6) (2015) 05014010, https://doi.org/10.1061/(ASCE)CP.1943-5487.0000421. fuzzy clustering and measured FRF data reduced by principal component
[58] Y. Pan, L. Zhang, Roles of artificial intelligence in construction engineering and projection, Adv. Struct. Eng. 16 (1) (2013) 207–217, https://doi.org/10.1260/
management: a critical review and future trends, Autom. Constr. 122 (2021), 1369-4332.16.1.207.
103517, https://doi.org/10.1016/j.autcon.2020.103517. [85] A. Amadore, G. Bosurgi, O. Pellegrino, Classification of measures from deflection
[59] Q. Li, K. Wang, M. Eacker, Z. Zhang, Clustering methods for truck traffic tests by means of fuzzy clustering techniques, Constr. Build. Mater. 53 (2014)
characterization in pavement ME design, ASCE-ASME J. Risk Uncertainty Eng. 3 173–181, https://doi.org/10.1016/j.conbuildmat.2013.11.094.
(2) (2017) F4016003, https://doi.org/10.1061/AJRUA6.0000881. [86] P. Nguyen, D. Tran, B. Lines, Empirical inference system for highway project
[60] M. Sarihi, V. Shahhosseini, M. Banki, Development and comparative analysis of delivery selection using fuzzy pattern recognition, J. Constr. Eng. Manag. 146
the fuzzy inference system-based construction labor productivity models, Int. J. (12) (2020) 04020141, https://doi.org/10.1061/(asce)co.1943-7862.0001950.
Constr. Manag. (2021), https://doi.org/10.1080/15623599.2021.1885117. [87] A. Tsehayae, A.R. Fayek, Developing and optimizing context-specific fuzzy
[61] R. Zuo, Y. Xiong, Big data analytics of identifying geochemical anomalies inference system-based construction labor productivity models, J. Constr. Eng.
supported by machine learning methods, Nat. Resour. Res. 27 (1) (2018) 5–13, Manag. 142 (7) (2016) 04016017, https://doi.org/10.1061/(ASCE)CO.1943-
https://doi.org/10.1007/s11053-017-9357-0. 7862.0001127.

19
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

[88] H. Naderpour, M. Mirrashid, Shear failure capacity prediction of concrete beam- Manag. 145 (3) (2019) 04019003, https://doi.org/10.1061/(asce)co.1943-
column joints in terms of ANFIS and GMDH, Pract. Period. Struct. Des. Constr. 24 7862.0001614.
(2) (2019) 04019006, https://doi.org/10.1061/(asce)sc.1943-5576.0000417. [113] W. Jang, S.K. Lee, S.H. Han, Sustainable performance index for assessing the
[89] M.M. Khotbehsara, B.M. Miyandehi, F. Naseri, T. Ozbakkaloglu, F. Jafari, green technologies in urban infrastructure projects, J. Manag. Eng. 34 (2) (2018)
E. Mohseni, Effect of SnO2, ZrO2, and CaCO3 nanoparticles on water transport and 04017056, https://doi.org/10.1061/(ASCE)ME.1943-5479.0000582.
durability properties of self-compacting mortar containing fly ash: experimental [114] X. Zhao, L. Chen, W. Pan, Q. Lu, AHP-ANP–fuzzy integral integrated network for
observations and ANFIS predictions, Constr. Build. Mater. 158 (2018) 823–834, evaluating performance of innovative business models for sustainable building,
https://doi.org/10.1016/j.conbuildmat.2017.10.067. J. Constr. Eng. Manag. 143 (8) (2017) 04017054, https://doi.org/10.1061/
[90] F. Khademi, S.M. Jamal, Estimating the compressive strength of concrete using (ASCE)CO.1943-7862.0001348.
multiple linear regression and adaptive neuro-fuzzy inference system, Int. J. [115] S. Inti, V. Tandon, Application of fuzzy preference-analytic hierarchy process
Struct. Eng. 8 (1) (2017) 20–31, https://doi.org/10.1504/ logic in evaluating sustainability of transportation infrastructure requiring
IJSTRUCTE.2017.081669. multicriteria decision making, J. Infrastruct. Syst. 23 (4) (2017) 04017014,
[91] G. Kabir, R. Sadiq, S. Tesfamariam, A fuzzy Bayesian belief network for safety https://doi.org/10.1061/(ASCE)IS.1943-555X.0000373.
assessment of oil and gas pipelines, Struct. Infrastruct. Eng. 12 (8) (2016) [116] M. Tomczak, P. Jaśkowski, Application of type-2 interval fuzzy sets to contractor
874–889, https://doi.org/10.1080/15732479.2015.1053093. qualification process, KSCE J. Civ. Eng. (2018) 1–12, https://doi.org/10.1007/
[92] Q. Zhou, F. Wang, F. Zhu, Estimation of compressive strength of hollow concrete s12205-017-0431-2.
masonry prisms using artificial neural networks and adaptive neuro-fuzzy [117] S. Mohandes, H. Sadeghi, A. Mahdiyar, S. Durdyev, A. Banaitis, K. Yahya,
inference systems, Constr. Build. Mater. 125 (2016) 417–426, https://doi.org/ S. Ismail, Assessing construction labours’ safety level: A fuzzy MCDM approach,
10.1016/j.conbuildmat.2016.08.064. J. Civ. Eng. Manag. 26 (2) (2020) 175–188, https://doi.org/10.3846/
[93] E. Sadrossadat, A. Heidaripanah, S. Osouli, Prediction of the resilient modulus of jcem.2020.11926.
flexible pavement subgrade soils using adaptive neuro-fuzzy inference systems, [118] A.R. Fayek, M.N. Omar, A fuzzy TOPSIS method for prioritized aggregation in
Constr. Build. Mater. 123 (2016) 235–247, https://doi.org/10.1016/j. multi-criteria decision making problems, J. Multi-Criteria Decis. Anal. 23 (5-6)
conbuildmat.2016.07.008. (2016) 242–256, https://doi.org/10.1002/mcda.1573.
[94] T.M. Al-zharani, R. Demirboga, W.H. Khushefati, O. Taylan, Measurement and [119] A. Pasha, A. Mansourian, M. Ravanshadnia, A hybrid fuzzy multi-attribute
prediction of correction factors for very high core compressive strength by using decision making model to select road pavement type, Soft. Comput. 24 (21)
the adaptive neuro-fuzzy techniques, Constr. Build. Mater. 122 (2016) 320–331, (2020) 16135–16148, https://doi.org/10.1007/s00500-020-04928-2.
https://doi.org/10.1016/j.conbuildmat.2016.06.019. [120] A. Reddy, P. Kumar, P. Raj, Entropy-based fuzzy TOPSIS framework for selection
[95] M. Mohamad, D. Tran, Risk-based prioritization approach to construction of a sustainable building material, Int. J. Constr. Manag. (2019) 1–12, https://doi.
inspections for transportation projects, J. Constr. Eng. Manag. 147 (1) (2021) org/10.1080/15623599.2019.1683695.
04020150, https://doi.org/10.1061/(asce)co.1943-7862.0001962. [121] A. Fazeli, F. Jalaei, M. Khanzadi, S. Banihashemi, BIM-integrated TOPSIS-Fuzzy
[96] F. Abad, L. Naeni, A hybrid framework to assess the risk of change in construction framework to optimize selection of sustainable building components, Int. J.
projects using fuzzy fault tree and fuzzy event tree analysis, Int. J. Constr. Manag. Constr. Manag. (2019) 1–20, https://doi.org/10.1080/15623599.2019.1686836.
(2020) 1–13, https://doi.org/10.1080/15623599.2020.1790474. [122] K. Mathiyazhagan, A. Gnanavelbabu, B. Lokesh Prabhuraj, A sustainable
[97] P. Rezakhani, Hybrid fuzzy-Bayesian decision support tool for dynamic project assessment model for material selection in construction industries perspective
scheduling and control under uncertainty, Int. J. Constr. Manag. (2020) 1–13, using hybrid MCDM approaches, J. Adv. Manag. Res. 16 (2) (2019) 234–259,
https://doi.org/10.1080/15623599.2020.1828539. https://doi.org/10.1108/jamr-09-2018-0085.
[98] S. Shoar, A. Banaitis, Application of fuzzy fault tree analysis to identify factors [123] A. Norouzi, H. Ghayur Namin, A hybrid fuzzy TOPSIS–best worst method for risk
influencing construction labor productivity: a high-rise building case study, prioritization in megaprojects, Civil Eng. J. 5 (6) (2019) 1257–1272, https://doi.
J. Civ. Eng. Manag. 25 (1) (2019) 41–52, https://doi.org/10.3846/ org/10.28991/cej-2019-03091330.
jcem.2019.7785. [124] M. Yazdi, Risk assessment based on novel intuitionistic fuzzy-hybrid-modified
[99] S. Shoar, F. Nasirzadeh, H. Zarandi, Quantitative assessment of risks on TOPSIS approach, Saf. Sci. 110 (2018) 438–448, https://doi.org/10.1016/j.
construction projects using fault tree analysis with hybrid uncertainties, Constr. ssci.2018.03.005.
Innov. 19 (1) (2019) 48–70, https://doi.org/10.1108/ci-07-2018-0057. [125] M. Arashpour, R. Wakefield, B. Abbasi, M. Arashpour, R. Hosseini, Optimal
[100] S. Phoemphon, C. So-In, D. Niyato, A hybrid model using fuzzy logic and an process integration architectures in off-site construction: theorizing the use of
extreme learning machine with vector particle swarm optimization for wireless multi-skilled resources, Architect. Eng. Design Manag. 14 (1–2) (2018) 46–59,
sensor network localization, Appl. Soft Comput. 65 (2018) 101–120, https://doi. https://doi.org/10.1080/17452007.2017.1302406.
org/10.1016/j.asoc.2018.01.004. [126] G. Polat, E. Eray, B.N. Bingol, An integrated fuzzy MCGDM approach for supplier
[101] J.S. Chou, A.D. Pham, H. Wang, Bidding strategy to support decision-making by selection problem, J. Civ. Eng. Manag. 23 (7) (2017) 926–942, https://doi.org/
integrating fuzzy AHP and regression-based simulation, Autom. Constr. 35 (2013) 10.3846/13923730.2017.1343201.
517–527, https://doi.org/10.1016/j.autcon.2013.06.007. [127] L. Wang, H.-Y. Zhang, J.-Q. Wang, L. Li, Picture fuzzy normalized projection-
[102] E. Plebankiewicz, D. Kubek, Multicriteria selection of the building material based VIKOR method for the risk evaluation of construction project, Appl. Soft
supplier using AHP and fuzzy AHP, J. Constr. Eng. Manag. 142 (1) (2016) Comput. 64 (2018) 216–226, https://doi.org/10.1016/j.asoc.2017.12.014.
04015057, https://doi.org/10.1061/(ASCE)CO.1943-7862.0001033. [128] S. Mirhosseini, R. Kiani Mavi, N. Kiani Mavi, B. Abbasnejad, F. Rayani,
[103] W. Jang, H.U. Hong, S.H. Han, S.W. Baek, Optimal supply vendor selection model Interrelations among leadership competencies of BIM leaders: A fuzzy DEMATEL-
for LNG plant projects using fuzzy-TOPSIS theory, J. Manag. Eng. 33 (2) (2017) ANP approach, Sustainability 12 (18) (2020) 7830, https://doi.org/10.3390/
04016035, https://doi.org/10.1061/(ASCE)ME.1943-5479.0000474. su12187830.
[104] O. Taylan, A.O. Bafail, R.M. Abdulaal, M.R. Kabli, Construction projects selection [129] T. Omar, M.L. Nehdi, T. Zayed, Performance of NDT techniques in appraising
and risk assessment by fuzzy AHP and fuzzy TOPSIS methodologies, Appl. Soft condition of reinforced concrete bridge decks, J. Perform. Constr. Facil. 31 (6)
Comput. 17 (2014) 105–116, https://doi.org/10.1016/j.asoc.2014.01.003. (2017) 04017104, https://doi.org/10.1061/(ASCE)CF.1943-5509.0001098.
[105] S. Ebrahimnejad, S.M. Mousavi, R. Tavakkoli-Moghaddam, M. Heydar, Risk [130] N. Prascevic, Z. Prascevic, Application of fuzzy AHP for ranking and selection of
ranking in mega projects by fuzzy compromise approach: a comparative analysis, alternatives in construction project management, J. Civ. Eng. Manag. 23 (8)
J. Intell. Fuzzy Syst. 26 (2) (2014) 949–959, https://doi.org/10.3233/IFS- (2017) 1123–1135, https://doi.org/10.3846/13923730.2017.1388278.
130785. [131] H. Abdul-Rahman, C. Wang, L.C. Wood, M. Ebrahimi, Integrating and ranking
[106] S.M. Mousavi, F. Jolai, R. Tavakkoli-Moghaddam, A fuzzy stochastic multi- sustainability criteria for housing, Eng. Sustain. 169 (1) (2016) 3–30, https://doi.
attribute group decision-making approach for selection problems, Group Decis. org/10.1680/ensu.15.00008.
Negot. (2013) 1–27, https://doi.org/10.1007/s10726-011-9259-1. [132] M.S. El-Abbasy, H. El Chanati, F. Mosleh, A. Senouci, T. Zayed, H. Al-Derham,
[107] J.R. San Cristobal, Critical path definition using multicriteria decision making: Integrated performance assessment model for water distribution networks, Struct.
PROMETHEE method, J. Manag. Eng. 29 (2) (2013) 158–163, https://doi.org/ Infrastruct. Eng. 12 (11) (2016) 1505–1524, https://doi.org/10.1080/
10.1061/(ASCE)ME.1943-5479.0000135. 15732479.2016.1144620.
[108] Y. Wu, J. Zhang, J. Yuan, S. Geng, H. Zhang, Study of decision framework of [133] R. Liang, Z. Dong, Z. Sheng, X. Wang, C. Wu, Case study of selecting decision-
offshore wind power station site selection based on ELECTRE-III under making schemes in large-scale infrastructure projects, J. Infrastruct. Syst. 23 (4)
intuitionistic fuzzy environment: a case of China, Energy Convers. Manag. 113 (2017) 06017001, https://doi.org/10.1061/(ASCE)IS.1943-555X.0000364.
(2016) 66–81, https://doi.org/10.1016/j.enconman.2016.01.020. [134] A.I. Maghsoodi, M. Khalilzadeh, Identification and evaluation of construction
[109] F. Afzal, S. Yunfei, M. Nazir, S. Bhatti, A review of artificial intelligence based risk projects’ critical success factors employing fuzzy-TOPSIS approach, KSCE J. Civ.
assessment methods for capturing complexity-risk interdependencies: Cost Eng. (2017) 1–13, https://doi.org/10.1007/s12205-017-1970-2.
overrun in construction projects, Int. J. Manag. Proj. Bus. 14 (2) (2021) 300–328, [135] H.M. Al-Humaidi, Construction projects bid or not bid approach using the fuzzy
https://doi.org/10.1108/ijmpb-02-2019-0047. technique for order preference by similarity (FTOPSIS) method, J. Constr. Eng.
[110] R. Mavi, C. Standing, Critical success factors of sustainable project management Manag. 142 (12) (2016) 04016068, https://doi.org/10.1061/(ASCE)CO.1943-
in construction: a fuzzy DEMATEL-ANP approach, J. Clean. Prod. 194 (2018) 7862.0001180.
751–765, https://doi.org/10.1016/j.jclepro.2018.05.120. [136] A.J. Chaghooshi, E. Khanmohammadi, M. Faghei, A. Karimi, Contractor selection
[111] H. Aladağ, Z. Işık, Design and construction risks in BOT type mega transportation using integrated goal programming and fuzzy ELECTRE, Int. J. Strategic Decision
projects, Eng. Constr. Archit. Manag. 26 (10) (2019) 2223–2242, https://doi.org/ Sci. 5 (3) (2014) 65–86, https://doi.org/10.4018/ijsds.2014070104.
10.1108/ecam-08-2018-0351. [137] A. Yazdani-Chamzini, S.H. Yakhchali, M. Mahmoodian, Risk ranking of tunnel
[112] N. Monzer, A.R. Fayek, R. Lourenzutti, N. Siraj, Aggregation-based framework for construction projects by using the ELECTRE technique under a fuzzy
construction risk assessment with heterogeneous groups of experts, J. Constr. Eng. environment, Int. J. Manag. Sci. Eng Manag 8 (1) (2013) 1–14, https://doi.org/
10.1080/17509653.2013.783185.

20
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

[138] M. Eghbali-Zarch, R. Tavakkoli-Moghaddam, K. Dehghan-Sanej, A. Kaboli, [163] F. Nasirzadeh, D. Carmichael, M. Jarban, M. Rostamnezhad, Hybrid fuzzy-system
Prioritizing the effective strategies for construction and demolition waste dynamics approach for quantification of the impacts of construction claims, Eng.
management using fuzzy IDOCRIW and WASPAS methods, Engineering, Constr. Archit. Manag. 26 (7) (2019) 1261–1276, https://doi.org/10.1108/ecam-
Construct. Architect. Manag. (2021), https://doi.org/10.1108/ecam-08-2020- 08-2017-0150.
0617. [164] N. Siraj, A.R. Fayek, Hybrid fuzzy system dynamics model for analyzing the
[139] A. Karamoozian, D. Wu, A hybrid risk prioritization approach in construction impacts of interrelated risk and opportunity events on project contingency, Can.
projects using failure mode and effective analysis, Eng. Constr. Archit. Manag. 27 J. Civ. Eng. (2020), https://doi.org/10.1139/cjce-2020-0032.
(9) (2020) 2661–2686, https://doi.org/10.1108/ecam-10-2019-0535. [165] N. Kedir, M. Raoufi, A.R. Fayek, Fuzzy agent-based multicriteria decision-making
[140] B. Gharanfoli, C. Valmohammadi, Identification and prioritization of construction model for analyzing construction crew performance, J. Constr. Eng. Manag. 36 (5)
projects investment risks using a hybrid fuzzy approach, J. Multi-Criteria Decis. (2020) 04020053, https://doi.org/10.1061/(asce)me.1943-5479.0000815.
Anal. 26 (3-4) (2019) 113–127, https://doi.org/10.1002/mcda.1661. [166] L. Luo, L. Zhang, Q. He, Linking project complexity to project success: a hybrid
[141] S. Hatefi, J. Tamošaitienė, An integrated fuzzy DEMATEL-fuzzy ANP model for SEM–FCM method, Eng. Constr. Archit. Manag. 27 (9) (2020) 2591–2614,
evaluating construction projects by considering interrelationships among risk https://doi.org/10.1108/ecam-05-2019-0241.
factors, J. Civ. Eng. Manag. 25 (2) (2019) 114–131, https://doi.org/10.3846/ [167] K. Naji, M. Mansour, M. Gunduz, Methods for modeling and evaluating
jcem.2019.8280. construction disputes: A critical review, IEEE Access 8 (2020) 45641–45652,
[142] M. Yazdani, M. Abdi, N. Kumar, M. Keshavarz-Ghorabaee, F. Chan, Improved https://doi.org/10.1109/access.2020.2976109.
decision model for evaluating risks in construction projects, J. Constr. Eng. [168] E. Haqiqat, Y. Zare Mehrjerdi, A. Zare Bidaki, Fuzzy inference system-Latin
Manag. 145 (5) (2019) 04019024, https://doi.org/10.1061/(asce)co.1943- hypercube simulation: an integrated hybrid model for OHS risks management,
7862.0001640. Int. J. Project Manag. 4 (2019) 127–140, https://doi.org/10.5267/j.
[143] R. Liang, H. Chong, A hybrid group decision model for green supplier selection: a jpm.2018.11.001.
case study of megaprojects, Eng. Constr. Archit. Manag. 26 (8) (2019) [169] M. Khanzadi, F. Nasirzadeh, M. Dashti, Fuzzy cognitive map approach to analyze
1712–1734, https://doi.org/10.1108/ecam-10-2018-0462. causes of change orders in construction projects, J. Constr. Eng. Manag. 144 (2)
[144] F. Sedady, M. Beheshtinia, A novel MCDM model for prioritizing the renewable (2018) 04017111, https://doi.org/10.1061/(asce)co.1943-7862.0001430.
power plants’ construction, Manag. Environ. Qual. 30 (2) (2019) 383–399, [170] R. Ahmed, T. Zayed, F. Nasiri, A hybrid genetic algorithm-based fuzzy Markovian
https://doi.org/10.1108/meq-05-2018-0102. model for the deterioration modeling of healthcare facilities, Algorithms 13 (9)
[145] S. Mohandes, X. Zhang, Towards the development of a comprehensive hybrid (2020) 210, https://doi.org/10.3390/a13090210.
fuzzy-based occupational risk assessment model for construction workers, Saf. [171] Y. Tu, X. Zhou, J. Gang, J. Xu, W. Shen, B. Lev, Hierarchical supplier selection
Sci. 115 (2019) 294–309, https://doi.org/10.1016/j.ssci.2019.02.018. optimization with multiple items in large-scale construction projects,
[146] M. Gunduz, M. Talat Birgonul, M. Ozdemir, Development of a safety performance J. Infrastruct. Syst. 23 (3) (2017) 04017003, https://doi.org/10.1061/(ASCE)
index assessment tool by using a fuzzy structural equation model for construction IS.1943-555X.0000356.
sites, Autom. Constr. 85 (2018) 124–134, https://doi.org/10.1016/j. [172] I. Bakry, O. Moselhi, T. Zayed, Optimized scheduling and buffering of repetitive
autcon.2017.10.012. construction projects under uncertainty, Eng. Constr. Archit. Manag. 23 (6)
[147] A. Ardeshir, M. Mohajeri, Assessment of safety culture among job positions in (2016) 782–800, https://doi.org/10.1108/ECAM-05-2014-0069.
high-rise construction: a hybrid fuzzy multi criteria decision-making (FMCDM) [173] A. Haghighi, A.H. Ayati, Stability analysis of gravity dams under uncertainty
approach, Int. J. Inj. Control Saf. Promot. 25 (2) (2018) 195–206, https://doi.org/ using the fuzzy sets theory and a many-objective GA, J. Intell. Fuzzy Syst. 30 (3)
10.1080/17457300.2017.1416483. (2016) 1857–1868, https://doi.org/10.3233/IFS-151897.
[148] A. Boostani, F. Jolai, A. Bozorgi-Amiri, Optimal location selection of temporary [174] Q. Liu, J. Xu, F. Qin, Optimization for the integrated operations in an uncertain
accommodation sites in Iran via a hybrid fuzzy multiple-criteria decision making construction supply chain, IEEE Trans. Eng. Manag. 64 (3) (2017) 400–414,
approach, J. Urban Plann. Dev. 144 (4) (2018) 04018039, https://doi.org/ https://doi.org/10.1109/TEM.2017.2686489.
10.1061/(asce)up.1943-5444.0000479. [175] M.-Y. Cheng, D. Prayogo, A novel fuzzy adaptive teaching-learning-based
[149] M.K. Ghorabaee, M. Amiri, J.S. Sadaghiani, E.K. Zavadskas, J. Antucheviciene, optimization (FATLBO) for solving structural optimization problems, Eng.
A new hybrid fuzzy MCDM approach for evaluation of construction equipment Comput. 33 (2017) 55–69, https://doi.org/10.1007/s00366-016-0456-z.
with sustainability considerations, Arch. Civil Mech. Eng. 18 (2018) 32–49, [176] M.-Y. Cheng, D.-H. Tran, An efficient hybrid differential evolution based serial
https://doi.org/10.1016/j.acme.2017.04.011. method for multimode resource-constrained project scheduling, KSCE J. Civ. Eng.
[150] E.E. Ameyaw, A.P.C. Chan, A fuzzy approach for the allocation of risks in public- 20 (1) (2016) 90–100, https://doi.org/10.1007/s12205-015-0414-0.
private partnership water-infrastructure projects in developing countries, [177] S. Kikuchi, N. Kronprasert, S.M. Easa, Aggregate blending using fuzzy
J. Infrastruct. Syst. 22 (3) (2016) 04016016, https://doi.org/10.1061/(ASCE) optimization, J. Constr. Eng. Manag. 138 (12) (2012) 1411–1420, https://doi.
IS.1943-555X.0000297. org/10.1061/(ASCE)CO.1943-7862.0000557.
[151] N. Singhaputtangkul, Z. Zhao, Applying a fuzzy consensus scheme to enhance the [178] J. Xu, P. Wei, Production-distribution planning of construction supply chain
group decision making of a building envelope design team, J. Constr. Eng. Manag. management under fuzzy random environment for large-scale construction
142 (8) (2016) 04016025, https://doi.org/10.1061/(ASCE)CO.1943- projects, J. Industrial Manag. Optimiz. 9 (1) (2013) 31–56, https://doi.org/
7862.0001142. 10.3934/jimo.2013.9.31.
[152] M. Raoufi, N.G. Seresht, A.R. Fayek, Overview of fuzzy simulation techniques in [179] W. ShangGuan, J. Wang, Z. Sheng, X. Yu, B. Cai, J. Wang, Adaptive fuzzy
construction engineering and management, in: Proceedings of the 2016 Annual planning of optimal speed profiles for high-speed train operation on the basis of a
Conference of the North American Fuzzy Information Processing Society Pareto set, Transp. Res. Rec. 2546 (1) (2016) 103–111, https://doi.org/10.3141/
(NAFIPS), IEEE, 2016, pp. 1–6, https://doi.org/10.1109/NAFIPS.2016.7851610. 2546-13.
[153] N. Sadeghi, A.R. Fayek, W. Pedrycz, Fuzzy Monte Carlo simulation and risk [180] A. Afshar, H. Amiri, Risk-based approach to unbalanced bidding in construction
assessment in construction, Comput. Aid. Civil Infrastruct. Eng. 25 (4) (2010) projects, Eng. Optim. 42 (4) (2010) 369–385, https://doi.org/10.1080/
238–252, https://doi.org/10.1111/j.1467-8667.2009.00632.x. 03052150903220964.
[154] R. Gouri, V. Srinivas, A fuzzy approach to reliability based design of storm water [181] Y. Ma, J. Xu, A novel multiple decision-maker model for resource-constrained
drain network, Stoch. Env. Res. Risk A. 31 (5) (2017) 1091–1106, https://doi. project scheduling problems, Can. J. Civ. Eng. 41 (6) (2014) 500–511, https://
org/10.1007/s00477-016-1299-2. doi.org/10.1139/cjce-2013-0232.
[155] L. Parvizsedghy, A. Senouci, T. Zayed, S.F. Mirahadi, M.S. El-Abbasy, Condition- [182] J. Xu, Z. Li, Multi-objective dynamic construction site layout planning in fuzzy
based maintenance decision support system for oil and gas pipelines, Struct. random environment, Autom. Constr. 27 (2012) 155–169, https://doi.org/
Infrastruct. Eng. 11 (10) (2015) 1323–1337, https://doi.org/10.1080/ 10.1016/j.autcon.2012.05.017.
15732479.2014.964266. [183] K. Sambhoo, S. Kadam, A. Deshpande, Ranking of sites for power plant
[156] A.A. Shaheen, A.R. Fayek, S.M. AbouRizk, Methodology for integrating fuzzy installation using soft computing techniques: a thought beyond EIA, Appl. Soft
expert systems and discrete event simulation in construction engineering, Can. J. Comput. 23 (2014) 556–566, https://doi.org/10.1016/j.asoc.2014.05.016.
Civ. Eng. 36 (9) (2009) 1478–1490, https://doi.org/10.1139/L09-091. [184] K.C. Lam, X. Ning, T. Ng, The application of the ant colony optimization
[157] N. Sadeghi, A.R. Fayek, N. Gerami Seresht, Queue performance measures in algorithm to the construction site layout planning problem, Constr. Manag. Econ.
construction simulation models containing subjective uncertainty, Autom. Constr. 25 (4) (2007) 359–374, https://doi.org/10.1080/01446190600972870.
60 (2015) 1–11, https://doi.org/10.1016/j.autcon.2015.07.023. [185] M. Marzouk, E. Abdelakder, A hybrid fuzzy-optimization method for modeling
[158] R.R. Levary, Systems dynamics with fuzzy logic, Int. J. Syst. Sci. 21 (8) (1990) construction emissions, Decision Sci. Lett (2020) 1–20, https://doi.org/10.5267/
1701–1707, https://doi.org/10.1080/00207729008910486. j.dsl.2019.9.002.
[159] F. Nasirzadeh, M. Khanzadi, M. Rezaie, Dynamic modeling of the quantitative risk [186] E. Kalhor, M. Khanzadi, E. Eshtehardian, A. Afshar, Stochastic time–cost
allocation in construction projects, Int. J. Proj. Manag. 32 (3) (2014) 442–451, optimization using non-dominated archiving ant colony approach, Autom. Constr.
https://doi.org/10.1016/j.ijproman.2013.06.002. 20 (8) (2011) 1193–1203, https://doi.org/10.1016/j.autcon.2011.05.003.
[160] N. Gerami Seresht, A.R. Fayek, Dynamic modeling of multifactor construction [187] Y. Zhang, Z. Guan, Selecting project risk preventive and protective strategies
productivity for equipment-intensive activities, J. Constr. Eng. Manag. 144 (9) based on bow-tie analysis, J. Constr. Eng. Manag. 34 (3) (2018) 04018009,
(2018), 04018091, https://doi.org/10.1061/(ASCE)CO.1943-7862.0001549. https://doi.org/10.1061/(ASCE)ME.1943-5479.0000603.
[161] P. Nojedehi, F. Nasirzadeh, A hybrid simulation approach to model and improve [188] H. Ke, W. Ma, X. Gao, W. Xu, New fuzzy models for time-cost trade-off problem,
construction labor productivity, KSCE J. Civ. Eng. 21 (5) (2017) 1516–1524, Fuzzy Optim. Decis. Making 9 (2) (2010) 219–231, https://doi.org/10.1007/
https://doi.org/10.1007/s12205-016-0278-y. s10700-010-9076-z.
[162] F. Nasirzadeh, M. Khanzadi, A. Afshar, S. Howick, Modeling quality management [189] V. Novák, I. Perfilieva, A. Dvoøák, Insight into Fuzzy Modeling, John Wiley and
in construction projects, Int. J. Civil Eng. 11 (1) (2013) 14–22, http://ijce.iust.ac. Sons, Inc, Hoboken, NJ, USA, 2016, pp. 151–188. ISBN: 978-1-119-19318-0.
ir/article-1-646-en.html.

21
P.H.D. Nguyen and A. Robinson Fayek Automation in Construction 134 (2022) 104064

[190] E. Çano, M. Morisio, Hybrid recommender systems: a systematic literature review, [195] H. Moon, H. Kim, V. Kamat, L. Kang, BIM-based construction scheduling method
Intelligent Data Analysis 21 (6) (2017) 1487–1524, https://doi.org/10.3233/ using optimization theory for reducing activity overlaps, J. Comput. Civ. Eng. 29
IDA-163209. (3) (2015) 04014048, https://doi.org/10.1061/(asce)cp.1943-5487.0000342.
[191] P.V. de Campos Souza, Fuzzy neural networks and neuro-fuzzy networks: a [196] S. Razavi, C. Haas, Multisensor data fusion for on-site materials tracking in
review the main techniques and applications used in the literature, Appl. Soft construction, Autom. Constr. 19 (8) (2010) 1037–1046, https://doi.org/10.1016/
Comput. 92 (2020) 106275, https://doi.org/10.1016/j.asoc.2020.106275. j.autcon.2010.07.017.
[192] M. Mohd Adnan, A. Sarkheyli, A. Mohd Zain, H. Haron, Fuzzy logic for modeling [197] S. Razavi, C. Haas, Reliability-based hybrid data fusion method for adaptive
machining process: a review, Artif. Intell. Rev. 43 (3) (2013) 345–379, https:// location estimation in construction, J. Comput. Civ. Eng. 26 (1) (2012) 1–10,
doi.org/10.1007/s10462-012-9381-8. https://doi.org/10.1061/(asce)cp.1943-5487.0000101.
[193] C. Kahraman, S. Onar, B. Oztaysi, Fuzzy multicriteria decision-making: A [198] H. Li, W. Wang, L. Fan, Q. Li, X. Chen, A novel hybrid MCDM model for machine
literature review, Int. J. Comput. Intel. Syst. 8 (4) (2015) 637–666, https://doi. tool selection using fuzzy DEMATEL, entropy weighting and later defuzzification
org/10.1080/18756891.2015.1046325. VIKOR, Appl. Soft Comput. 91 (2020), 106207, https://doi.org/10.1016/j.
[194] A.R. Fayek, Fuzzy logic and fuzzy hybrid techniques for construction engineering asoc.2020.106207.
and management, J. Constr. Eng. Manag. 146 (7) (2020) 04020064, https://doi. [199] X. Song, J. Xu, C. Shen, F. Peña-Mora, F., Conflict resolution-motivated strategy
org/10.1061/(asce)co.1943-7862.0001854. towards integrated construction site layout and material logistics planning: A bi-
stakeholder perspective, Autom. Constr. 87 (2018) 138–157, https://doi.org/
10.1016/j.autcon.2017.12.018.

22

You might also like