Fachrurrazi Jurnal 9428

12/5/2019 The Subcontractor Selection Practice using ANN-Multilayer
Vol 8, No 4 (2017) > Civil Engineering
The Subcontractor Selection Practice using ANN-Multilayer

Authors Authors and Affiliations
Fachrurrazi , Saiful Husin, Munirwansyah , Husaini
Corresponding email:  fachrurrazi@unsyiah.ac.id (mailto:fachrurrazi@unsyiah.ac.id)
Published at : 31 Jul 2017

IJtech : IJtech (http://ijtech.eng.ui.ac.id/) Vol 8, No 4 (2017) (http://ijtech.eng.ui.ac.id/issue/33)
DOI : https://doi.org/10.14716/ijtech.v8i4.9490 (https://doi.org/10.14716/ijtech.v8i4.9490)
Cite this article as:

Fachrurrazi, Husin, .S., Munirwansyah, Husaini, 2017. The Subcontractor Selection Practice using ANN-Multilayer. International Journal of
Technology. Volume 8(4), pp. 761-772
422
Downloads
Abstract
The practice of subcontracting selection emphasizes two important goals: the company's
strategic goal to maximize profits by partnering with subcontractors and the project's
operational goal for obtaining qualified subcontractors. Both goals are achieved by
formulating the best multi-criteria weights. This is not easy to implement due to differences in
subjectivity, viewpoint, and other consideration of assessors, but prioritizing the criterion
weights can reduce these differences. This study presents an ANN (Artificial Neural Network)
with the ability to generalize data. The purpose of the study is to develop an ANN model for
subcontracting selection and to identify significant criteria related to the company's strategic
goal. The initial training of the proposed ANN model utilized 40 subcontractor selection
datasets containing data in the form of a subcontractor selection scheme consisting of 20 criteria and 5 major
groups. Training of ANN model was successful with MSE learning at 1.37269e-7, MSE validation at 0.07985, and
epoch 600 to 800. The quotation price is the significant criterion of the selection, and it has a great outcome for
the contractor strategic goal. The interaction between the subcontractor selection practice and the ANN model
shows that the ANN has an important role in the subcontractor selection practice.
Keywords
ANN model; Company goal; Multi-criteria; Multilayer architecture; Project goal; Subcontractor selection; Weight
DOWNLOAD PDF  (HTTP://IJTECH.ENG.UI.AC.ID/DOWNLOAD/ARTICLE/591)
HOW TO CITE 
WHO CITE THIS PAPER  (HTTPS://SCHOLAR.GOOGLE.COM/SCHOLAR?Q=THE+SUBCONTRACTOR+SELECTION+PRACTICE+USING+ANN-MULTILAYER)
SHARE ARTICLE 
ijtech.eng.ui.ac.id/article/view/591 1/2
12/5/2019 The Subcontractor Selection Practice using ANN-Multilayer
Table of Contents
 Article
 Abstract
Copyright © 2017 Faculty of Engineering

International Journal of Technology
IJTech secretariat, Engineering Center Bld., 2nd Fl.
Faculty of Engineering, Universitas Indonesia
Depok 16424, Indonesia.
ijtech.eng.ui.ac.id/article/view/591 2/2
12/5/2019 Archives
ARCHIVES
Vol 10, No 7 (2019) SE
(http://ijtech.eng.ui.ac.id/issue/59)
ijtech.eng.ui.ac.id/archives 1/41
12/5/2019 Archives
Vol 10, No 6 (2019) SE
12/5/2019 Archives
Vol 10, No 5 (2019)
Vol 10, No 4 (2019)
12/5/2019 Archives
Vol 10, No 3 (2019) SE
12/5/2019 Archives
Vol 10, No 2 (2019)
Vol 10, No 1 (2019)
12/5/2019 Archives
Vol 9, No 8 (2018) SE
12/5/2019 Archives
Vol 9, No 7 (2018) SE
Vol 9, No 6 (2018) SE
12/5/2019 Archives
Vol 9, No 5 (2018)
12/5/2019 Archives
Vol 9, No 4 (2018)
12/5/2019 Archives
Vol 9, No 3 (2018)
12/5/2019 Archives
Vol 9, No 2 (2018) SE
12/5/2019 Archives
Vol 9, No 1 (2018)
12/5/2019 Archives
Vol 8, No 8 (2017) SE
12/5/2019 Archives
Vol 8, No 7 (2017) SE
12/5/2019 Archives
Vol 8, No 6 (2017) SE
12/5/2019 Archives
Vol 8, No 5 (2017)
12/5/2019 Archives
Vol 8, No 4 (2017)
12/5/2019 Archives
Vol 8, No 3 (2017)
12/5/2019 Archives
Vol 8, No 2 (2017) SE
12/5/2019 Archives
Vol 8, No 1 (2017)
12/5/2019 Archives
Vol 7, No 8 (2016) SE
12/5/2019 Archives
Vol 7, No 7 (2016) SE
12/5/2019 Archives
Vol 7, No 6 (2016)
12/5/2019 Archives
Vol 7, No 5 (2016)
12/5/2019 Archives
Vol 7, No 4 (2016)
Vol 7, No 3 (2016)
12/5/2019 Archives
Vol 7, No 2 (2016)
12/5/2019 Archives
Vol 7, No 1 (2016)
Vol 6, No 7 (2015)
12/5/2019 Archives
Vol 6, No 6 (2015)
12/5/2019 Archives
Vol 6, No 5 (2015)
Vol 6, No 4 (2015)
12/5/2019 Archives
Vol 6, No 3 (2015)
12/5/2019 Archives
Vol 6, No 2 (2015)
Vol 6, No 1 (2015)
12/5/2019 Archives
Vol 5, No 3 (2014)
12/5/2019 Archives
Vol 5, No 2 (2014)
Vol 5, No 1 (2014)
12/5/2019 Archives
Vol 4, No 3 (2013)
12/5/2019 Archives
Vol 4, No 2 (2013)
Vol 4, No 1 (2013)
12/5/2019 Archives
Vol 3, No 2 (2012)
12/5/2019 Archives
Vol 3, No 1 (2012)
Vol 2, No 3 (2011)
12/5/2019 Archives
Vol 2, No 2 (2011)
12/5/2019 Archives
Vol 2, No 1 (2011)
Vol 1, No 1 (2010)
 About the Journal
Editorial Board (http://ijtech.eng.ui.ac.id/people)

Focus and Scope (http://ijtech.eng.ui.ac.id/about/5/focus-and-scope)
Online Submissions (http://ijtech.eng.ui.ac.id/about/3/online-submission)
Publication Policy (http://ijtech.eng.ui.ac.id/about/11/publication-policy)
Publication Ethics and Policy (http://ijtech.eng.ui.ac.id/about/10/policy)
Author Guidelines (http://ijtech.eng.ui.ac.id/about/4/author-guidelines)
(http://ijtech.eng.ui.ac.id/about/12)List of Reviewers
 Login
Username
Username or email
Password
12/5/2019 Archives
Password
Not as user? LOGIN

Register (http://ijtech.eng.ui.ac.id/register)
 IJTech
p-ISSN : 2086-9614
e-ISSN : 2087-2100
 Journal Metrics
Metrics by SCOPUS 2018

CiteScore 2018: 1.06
SCImago Journal Rank (SJR): 0.345
Source Normalized Impact per Paper
(SNIP): 1.395
Acceptance rate: 28.32 %

Average time to publish: 156 days.
 IJTech is indexed in:
(http://www.scopus.com)
(http://www.ebsco.com/index.asp)
(http://www.doaj.org)
(http://www.scimagojr.com)
(http://www.indexcopernicus.com)
12/5/2019 Archives
(http://www.crossref.org)
(http://ip-science.thomsonreuters.com/cgi-
bin/jrnlst/jlresults.cgi?PC=EX&Full=international%20journal%20of%20technology)
(http://info.flagcounter.com/AoS7)

12/5/2019 Journal Issue
JOURNAL ISSUE
ijtech.eng.ui.ac.id/issue/33 1/13
31 Jul 2017
Volume 8, Number 4
Fostering Partnerships and Strategic Alliances in Sustainable Infrastructure

Development (http://ijtech.eng.ui.ac.id/article/view/570)
Mohammed Ali Berawi
Publication Date (Online):

Jul 31, 2017
DOI: https://doi.org/10.14716/ijtech.v8i4.9468 (https://doi.org/10.14716/ijtech.v8i4.9468)
Pages : 568-571
Electrolytic Refining of Lead in Molten Chloride Electrolytes

(http://ijtech.eng.ui.ac.id/article/view/571)
Pavel Arkhipov, Yuriy Zaykov, Yulia Khalimullina, Anna Kholkina

Jul 31, 2017
DOI: https://dx.doi.org/10.14716/ijtech.v8i4.9473 (https://dx.doi.org/10.14716/ijtech.v8i4.9473)
Pages : 572-581
Synthesis and Characterization of Hematite Nanoparticles using Ultrasonic

Sonochemistry Method (http://ijtech.eng.ui.ac.id/article/view/575)
Munawar Khalil, Ning Liu, Robert L. Lee

Jul 31, 2017
Pages : 582-590
User Verifiable Multiple Keyword Search Scheme using the Merkle Tree for
Outsourced Data in the Cloud (http://ijtech.eng.ui.ac.id/article/view/577)
Devi Thiyagarajan, R. Ganesan

Jul 31, 2017
Pages : 591-600
Sin Activation Structural Tolerance of Online Sequential Circular Extreme Learning

Machine (http://ijtech.eng.ui.ac.id/article/view/579)
Sarutte Atsawaraungsuk, Tatpong Katanyukul

Jul 31, 2017
Pages : 601-610
A Preliminary Study on Shifting from Virtual Machine to Docker Container for

Insilico Drug Discovery in the Cloud (http://ijtech.eng.ui.ac.id/article/view/580)
Agung Putra Pasaribu, Muhammad Fajar Siddiq, Muhammad Irfan Fadhila, Muhammad H. Hilman,
Arry Yanuar, Heru Suhartanto

Jul 31, 2017
Pages : 611-621
A Stochastic Method based on the Markov Model of Unit Jump for Analyzing Crack
Jump in a Material (http://ijtech.eng.ui.ac.id/article/view/582)
Karima Selmani Bouayoune, El Mostapha Boudi, Aziz Bachir

Jul 31, 2017
Pages : 622-633
Calorific Value Analysis of Azadirachta Excelsa and Endospermum Malaccense as

Potential Solid Fuels Feedstock (http://ijtech.eng.ui.ac.id/article/view/584)
Nazia Hossain, Rafidah Jalil, T.M.I. Mahlia, Juliana Zaini

Jul 31, 2017
Pages : 634-643
Synthesis and Magnetic Characterization of Mn-Ti Substituted SrO.6Fe2-

xMnx/2Tix/2O3 (x = 0.0–1.0) Nanoparticles by Combined Destruction Process
Karina Nur Fitriana, Mas Ayu Elita Hafizah, Azwar Manaf

Jul 31, 2017
Pages : 644-650
Effect of Post-Hydrothermal Treatments on the Physical Properties of ZnO Layer

Derived from Chemical Bath Deposition (http://ijtech.eng.ui.ac.id/article/view/589)
Amalia Sholehah, Akhmad Herman Yuwono, Nofrijon Sofyan, Chairul Hudaya, Muhammad Ikhlasul
Amal

Jul 31, 2017
Pages : 651-661
Experimental Study on Mechanical Properties of Aluminum Alloys under Uniaxial

Tensile Tests (http://ijtech.eng.ui.ac.id/article/view/590)
Olfa Daghfas, Amna Znaidi, Ahmed Ben Mohamed, Rachid Nasri

Jul 31, 2017
Pages : 662-672
Kinetic Model Development for Biogas Production from Lignocellulosic Biomass

Manjula Das Ghatak , Pinakeswar Mahanta

Jul 31, 2017
Pages : 673-680
Numerical Study About the Change in Flow Separation and Velocity Distribution in a
90° Pipe Bend with/without Guide Vane Conditions
Sumit Saha, Nityananda Nandi

Jul 31, 2017
Pages : 681-689
Vapor Chamber Utilization for Rapid Cooling in the Conventional Plastic Injection
Molding Process (http://ijtech.eng.ui.ac.id/article/view/576)
Ahmad Hasnan, Nandy Putra, Wayan Nata Septiadi, Bambang Ariantara, Nasruddin A Abdullah

Jul 31, 2017
DOI: https://dx.doi.org/10.14716/ijtech.v8i4.9493 (https://dx.doi.org/10.14716/ijtech.v8i4.9493)
Pages : 690-697
TRIGA-2000 Research Reactor Thermal-hydraulic Analysis using

RELAP/SCDAPSIM/MOD3.4 (http://ijtech.eng.ui.ac.id/article/view/578)
Anhar Riza Antariksawan, Efrizon Umar, Surip Widodo, Mulya Juarsa, Mukhsinun Hadi Kusuma

Jul 31, 2017
Pages : 698-708
Controlling Unmanned Surface Vehicle Rocket using GPS Tracking Method

Onny Sutresman, Rafiuddin Syam, Sapta Asmal

Jul 31, 2017
Pages : 709-718
Concept Application for Pipelines using a Submerged Floating Tunnel for Use in the
Oil and Gas Industry (http://ijtech.eng.ui.ac.id/article/view/583)
Ery Budiman, I.G.P. Raka, Endah Wahyuni

Jul 31, 2017
Pages : 719-727
Strength Development of High-Performance Concrete using Nanosilica

Jonbi Eddhie

Jul 31, 2017
Pages : 728-736
Measuring the Effect of Strengthened Concrete on the Fracture Characteristics of

Notched Concrete Beams Through a Three-Point Beam Test
Atur P.N. Siregar

Jul 31, 2017
Pages : 737-746
Shaking Table Test of Soil Liquefaction in Southern Yogyakarta

Lindung Zalbuin Mase

Jul 31, 2017
Pages : 747-760
The Subcontractor Selection Practice using ANN-Multilayer

Fachrurrazi , Saiful Husin, Munirwansyah , Husaini

Jul 31, 2017
Pages : 761-772
LIST OF REVIEWERS CONTRIBUTING IN THIS ISSUE (http://ijtech.eng.ui.ac.id/reviewer/33)

 Login
Username
Username or email
Password
Password
Not as user? LOGIN

 IJTech
p-ISSN : 2086-9614
e-ISSN : 2087-2100
 Journal Metrics

(SNIP): 1.395


12/5/2019 International Journal Technology FTUI
About the Journal
International Journal of Technology (IJTech) is quarterly international referred journal with the
objectives to explore, develop, and elucidate the knowledge of engineering design and technology, to
keep practitioners and researchers informed on current issues and best practices, as well as serving as
a platform for the exchange of ideas, knowledge, and expertise among technology researchers and
practitioners.
IJTech is published by Universitas Indonesia, the best university in Indonesia (Times Higher Education-
QS World University Rankings since 2006). Published papers in IJTech strongly represent the use of
sustainable technology in product and service designs for tropical environment. Sustainable
technology has been developed in various Asian countries that comply with the principles of social,
economic, and ecological sustainability and is categorized into the following themes: conventional and
renewable energy, sustainable architecture, construction and environment, advanced material and
manufacturing technology, efficiency and productivity improvement. Technology application and
development for tropical environment that encourage innovation, improve economic benefits,
stimulate business growth, and increasing the quality of human life while still trying to maintain the
balance between the technology itself and the environment indicates the character and the unique
features of IJTech.
IJTech is currently indexed in Google scholar, DOAJ, EBSCO, GISI, Index Copernicus, SCOPUS, SCImago,
Emerging Sources Citation Index (ESCI) Thomson Reuters, Directory of Research Journal Indexing
(DRJI), has been accepted for coverage in selected Elsevier product(s), and a member of CrossRef.
Since 2017, IJTech journal rank raised with h-index 6 and classified into Q2 journal (2nd quartile) for
Engineering field in Scimago Journal database.
ijtech.eng.ui.ac.id/about/1/about-the-journal 1/4
(http://www.scimagojr.com/journalsearch.php?
q=21100235612&tip=sid&exact=no)

 Login
Username
Username or email
Password
Password
Not as user? LOGIN

 IJTech
p-ISSN : 2086-9614
e-ISSN : 2087-2100
 Journal Metrics

(SNIP): 1.395


12/5/2019 Editorial Team
EDITORIAL TEAM
Editor in Chief
1. Dr. Mohammed Ali Berawi, Universitas Indonesia, Indonesia
Managing Editor
1. Dr. Nyoman Suwartha, Universitas Indonesia, Indonesia
Members
1. Dr. Agus Sunjarianto Pamitran, Universitas Indonesia, Indonesia
2. Dr. Anwar Usman, Universiti Brunei Darussalam, Brunei Darussalam
3. Dr. Cecilia Vale, University of Porto, Portugal
4. Dr. Eko Adhi Setiawan, Universitas Indonesia, Indonesia
5. Dr. Eny Kusrini, Universitas Indonesia, Indonesia
6. Dr. Giuseppe Lo Papa, Teagasc Rural Economy Research Centre, Ireland
7. Dr. Hendri Dwi Saptioratri, Universitas Indonesia, Indonesia
8. Dr. Hng Huey Hoon, Nanyang Technological University, Singapore , Singapore
9. Dr. Johannes Widodo, National University of Singapore, Singapore
10. Dr. Lee Wilson, University of Saskatchewan, Canada
11. Dr. Muhamad Asvial, Universitas Indonesia, Indonesia
12. Dr. Muhammad Suryanegara, Universitas Indonesia, Indonesia
13. Dr. Nofrijon Sofyan, Universitas Indonesia, Indonesia
14. Dr. Reza Kia, Islamic Azad University, Iran (Islamic Republic of)
15. Dr. Roy Woodhead, Digital Innovation, Sheffield Business School, Sheffield Hallam University,
United Kingdom
16. Dr. Ruki Harwahyu, Universitas Indonesia, Indonesia
17. Dr. Sam P. Sinha, Scientific Research & Development, United States
18. Dr. Sri Harjanto, Universitas Indonesia, Indonesia
19. Dr. Yung-Jung Hsu, National Chiao Tung University, Taiwan
20. Prof. Dr. Akhmad Herman Yuwono , Universitas Indonesia, Indonesia
21. Prof. Dr. Bambang Sugiarto, Universitas Indonesia, Indonesia
22. Prof. Dr. Dedi Priadi, Universitas Indonesia, Indonesia
23. Prof. Dr. Esah Hamzah, Universiti Teknologi Malaysia, Malaysia
24. Prof. Dr. Hamzah Abdul Rahman, Universiti Malaya, Malaysia
25. Prof. Dr. Hideaki Ohgaki, Kyoto University, Japan
26. Prof. Dr. Isti Surjandari, Universitas Indonesia, Indonesia
27. Prof. Dr. Jong-Taek Oh, Chonnam National University, Korea, Republic of
28. Prof. Dr. Muhammad Idiris Saleh, Universiti Sains Malaysia, Malaysia
29. Prof. Dr. Nandy Putra, Universitas Indonesia, Indonesia
30. Prof. Dr. Raimundo Delgado, University of Porto, Portugal
31. Prof. Dr. Simon P. Ringer, University of Sydney, Australia
32. Prof. Dr. Sutrasno Kartohardjono, Universitas Indonesia, Indonesia
33. Prof. Dr. T. Yuri M. Zagloel, Universitas Indonesia, Indonesia
34. Prof. Dr. Toshio Shudo, Tokyo Metropolitan University, Japan
35. Prof. Dr. Yandi Andri Yatmo, Universitas Indonesia, Indonesia
36. Prof. Dr. Yung- Hui Lee, National Taiwan University, Taiwan
37. Prof. Paramita Atmodiwiryo, Universitas Indonesia, Indonesia
ijtech.eng.ui.ac.id/people 1/4
38. Prof. Rui Calcada, University of Porto, Portugal

 Login
Username
Username or email
Password
Password
Not as user? LOGIN

 IJTech
p-ISSN : 2086-9614
e-ISSN : 2087-2100
 Journal Metrics

(SNIP): 1.395


International Journal of Technology (2017) 4: 761-772
ISSN 2086-9614 © IJTech 2017
THE SUBCONTRACTOR SELECTION PRACTICE USING ANN-MULTILAYER
Fachrurrazi1*, Saiful Husin1, Munirwansyah1, Husaini2,

1
Department of Civil Engineering, Syiah Kuala University Jl. Teuku Nyak Arief, Darussalam, Kota
Banda Aceh, Aceh 23111, Indonesia
2
Department of Mechanical Engineering, Syiah Kuala University, Jl. Teuku Nyak Arief,
Darussalam, Kota Banda Aceh, Aceh 23111, Indonesia
(Received: May 2016 / Revised: March 2017 / Accepted: July 2017)
ABSTRACT
The practice of subcontracting selection emphasizes two important goals: the company's
strategic goal to maximize profits by partnering with subcontractors and the project's
operational goal for obtaining qualified subcontractors. Both goals are achieved by formulating
the best multi-criteria weights. This is not easy to implement due to differences in subjectivity,
viewpoint, and other consideration of assessors, but prioritizing the criterion weights can reduce
these differences. This study presents an ANN (Artificial Neural Network) with the ability to
generalize data. The purpose of the study is to develop an ANN model for subcontracting
selection and to identify significant criteria related to the company's strategic goal. The initial
training of the proposed ANN model utilized 40 subcontractor selection datasets containing
data in the form of a subcontractor selection scheme consisting of 20 criteria and 5 major
groups. Training of ANN model was successful with MSE learning at 1.37269e-7, MSE
validation at 0.07985, and epoch 600 to 800. The quotation price is the significant criterion of
the selection, and it has a great outcome for the contractor strategic goal. The interaction
between the subcontractor selection practice and the ANN model shows that the ANN has an
important role in the subcontractor selection practice.
Keywords: ANN model; Company goal; Multi-criteria; Multilayer architecture; Project goal;
Subcontractor selection; Weight
1. INTRODUCTION
Procurement management in a project requires subcontractor selection. The procurement
specialist, in turn, needs the support of advanced tools to speed up the right decision. This
decision-making process in project management can take many directions, depending on the
problem characteristics. The problems affecting subcontractor selection are quite complex and
can be caused by a large number of criteria. The subcontractors, who are a part of the strategic
decision, are often selected in any construction project implementation. Some researchers, such
as Moselhi et al. (1992), have successfully applied Artificial Neural Network (ANN) to the case
of subcontractor selection. The decisions based on the pattern of the weight criteria usually will
vary and relate to the situation of the project, conditions, and the purpose of selecting the
subcontractors (Azadnia et al., 2012). These changes must be accommodated without ruling out
the experience on the decisions made by evaluator. The criterion weights determined in the
subcontractor selection process should be adapted to the selection purpose, but the process for
adapting the criterion weights is not easy because of variations in the subjectivity, viewpoints,
*Corresponding author’s email: fachrurrazi@unsyiah.ac.id, Tel. +62-651-7552222, Fax. +62-651-7552222

Permalink/DOI: https://doi.org/10.14716/ijtech.v8i4.9490
762 The Subcontractor Selection Practice using ANN-Multilayer
and other considerations of assessors. One approach to minimize these variations is to use an
ANN, a learning algorithm for generalization based on data input and output (Oliveira et al.,
2015). The ANN has the ability to store information from the past and combine it with further
experience, and it is updated more generally using the latest data (Oliveira et al., 2015).
The aim of the present study is to develop an ANN model for subcontractor selection and to
identify the significant criteria that relate to the strategic goal of the main contractor. This aim is
approached by developing the ANN architecture in five groups corresponding to five main
criteria, with each group consisting of input layers that relate to a number of sub-criteria, as
shown in Figure 2. This need is accommodated by determining the applicability of the ANN
object by implementing ANN learning, which is meant to generalize data from any decisions
made (Moselhi et al., 1992; Chiarazzo et al., 2014; Zhang et al., 2016) and could be visualized
in the learning curve. This will improve the generalization of the criteria weights for the
subcontractor selection. The performance improvement will provide continuous and
simultaneous updating of the synaptic information of ANN for each process of the
subcontractor selection. This goal will be achieved by conducting the initial training using 25
datasets of subcontractor selection (as shown in Table 1 in the appendix).
This paper presents a proposed development of the ANN model and its implementation for the
case of subcontractor selection. Some of the sub-methods of ANN, the objects, and the
attributes, have been referenced in several other studies. Nevertheless, this paper presents
specificity for the architecture of ANN model, and the methods for supervised learning, which
are placed on two levels. The characteristics of the architecture of the ANN model are tailored
to the characteristics of subcontractor selection in the current case. A correlation test for
validation is conducted to identify the potential criteria that most influence the decision based
on expert judgment. This test is important for understanding the ANN model pattern that relates
to the correlation of the strategic goal and the performance goal. One of research aim was to
increase the knowledge of the ANN model by proposing an integrated interaction model
between assessor assessment, expert judgment, and the ANN model. The interaction model is
needed to provide an increase in the generating weights and acceleration of the subcontractor
selection process. Testing the performance of the interaction model of ANN will be future
research by other researchers in this field.
2. METHODOLOGY
The current subcontractor selection formally uses the Multi-Criteria Decision Analysis
(MCDA) model, formulated using the Weighted Sum Model (WSM):
, for i ϵ {1, 2, 3, …, n}, and j ϵ {1,2,3, …, m} (1)
where AiWSM denotes the total score for each alternative- ; is the relative weight of the
importance level for each criterion- ; relates the performance value of alternative- and
criteria- ; is the number of alternatives; is the number of criteria; is the maximum number
of alternatives, and is the maximum number of criteria. In this study, ANN is used to
pattern the by adjusting the weights, as shown in Figure 1. However, the weights could not
formally describe the ANN pattern due to the black box.
2.1. Framework of the ANN Model
The ANN model uses a supervised learning mechanism, as shown in Figure 1, that is placed on
the intermediate output (level 2) and on the final output (level 4) of the ANN architecture, as
shown in Figure 2. The performance measurement of the ANN model is analyzed using MSE
(Mean Square Error). The aim is to achieve a condition where the ANN outputs and targets are
equal or the MSE learning is stable, as shown in Figure 3. The other sub-methods of the ANN
refer to other research, such as ANN training, adjustment of the criterion weights, the activation
Fachrurrazi et al. 763
function, the flow of information in a network, and reduction in error. The attributes are,
respectively, supervised learning (Fachrurrazi et al., 2017a), the back propagation algorithm
(Taghavifar et al., 2014; Zuna et al., 2016), sigmoid (Kusumoputro et al., 2016), feed forward
(Euler-Rolle et al., 2016), and gradient descent (Kim et al., 2004).
Figure 1 The supervised learning of ANN (Fachrurrazi et al., 2017a)
The architecture of the ANN has been built using a series of inputs, output layers, hidden
layers, and number of nodes. The nodes are used as the processing signal by implementing the
mathematical definition for a single node (Meruelo et al., 2016):
) (2)
where x and y are the inputs and outputs, respectively, of the th node, is the weight for each
input, is the bias, and is referred to as the activation function. The activation
functions transfer the y output in the hidden layers process as a sigmoid function, in a
mathematical function (Kusumoputro et al., 2016):
(3)
2.2. Dataset
The data, totaling 40 sets used in this research, are from previous research by Fachrurrazi et al.
(2017b). They consist of expert’s judgment data, to be used as the target of ANN, and the
director judgment of the main contractor. The dataset will be divided into two groups,
consisting of 25 sets for ANN learning (as shown in Table 1 in the appendix) and 15 sets for
ANN validation (as shown in Table 2 in the appendix). The splitting of the dataset between
training and validation is based on the principle of independence of data, where 60% is used for
training and 40% for validation of the model.
3. RESULTS
3.1. The Architecture of ANN Model
During the preparation stage of the ANN model, its objects, methods, attributes, and the
architecture are the first considerations. Its characteristics are associated with the problem to be
solved, as shown in Figure 2. This phase is conducted in the numerical experimental of ANN to
find the effective model.
Figure 2 The ANN model architecture
The architecture of the ANN model is built based on the criteria hierarchy structure of actual
subcontractor selection and object definition of the architecture using MATLAB. The objects of
the ANN model consist of 5 groups of input layers (the numbers of input sizes in each group are
successively 13, 2, 3, 1, and 1), 12 hidden layers, 12 bias connects, 5 middle output layers, 1
final output layer, 5 middle targets, and 1 final target. The sizes of the nodes in each hidden
layer are 20, 8, 8, 3, 3, 1, 1, 1, 1, 1, 10, and 1, respectively. The details of objects, attributes, and
architecture are shown in Figure 2.
3.2. Learning of the ANN Model
In the learning phase of the ANN model, the initial randomly generated values of weights are
assigned to the ANN model (Nayak et al., 2016). The progress of the ANN learning can be
followed by observing the learning curve in Figure 3. The learning showed a significant
decrease in the initial until it reached the epochs of 600. After passing those epochs, the
learning became stable. The epochs are iterated for both a forward pass and a backwards pass,
for all the training examples. The ANN model is effective in patterning the subcontractor
selection data with a minimal epoch achievement. Comparisons to other research about this
learning include Creese et al. (1995) with the epochs of 15,000 to 50,000, Loyola et al. (2015)
with the epochs of 200,000, Ko et al. (2007) with the epochs of 4000, and (Albino et al., 1998)
with the epochs of 5000 to 50.000.
The learning performance of the ANN model is achieved at MSE of 1.37269E-07, as shown in
Figure 3. This indicates that the learning of the proposed ANN model is achieved well. The
learning performance is more accurate than that of Loyola et al. (2015) with MSE of 2.8e-07 or
Ko et al. (2007) with RMSE 0.0082, but does still not achieve the result of Albino et al. (1998)
with an MSE learning of 0.0 (zero).
Figure 3 Learning curve of ANN
3.3. The Performance of the ANN Validation

The plotting of the output data, target data, and the error of ANN shows that outputs of the
ANN model are also able to track the pattern of the target data, as shown in Figure 4. The ANN
has therefore been able to act as a model application to solve the problems of multi-criteria for
subcontractor selection.
Figure 4 Output, target, and error of ANN learning
The error of the ANN model, analyzed using the MSE validation of 0.07985, is close to the
validation results of Albino et al. (1998) with an MSE of 0.006 and of Ko et al. (2007), with an
RMSE validation of 0.0141. This study shows three alternatives where the error exceeds 15%;
namely, Q7, Q13, and Q14. However, on average, the overall error generated on the ANN
model validation is still below 15%. This is possible because the data pattern of the alternative
on learning of ANN differs from the data patterns Q7, Q13, and Q14. This difference will
decrease with an increasing amount of data for learning ANN. This explains the adequacy of
the data amount and meets all the situations/conditions in the subcontractor selection process;
therefore, the ANN model can be made more accurate. However, the output pattern of the ANN
model for the subcontractor selection shows a uniform shape toward the target output. Based on
the validation results, we can conclude that the proposed ANN model is able to show the initial
learning to adopt the knowledge of expert judgments, as shown in Figure 5. This agrees with
the opinions of other researchers, who have determined a small bias for a generalized model
using ANFIS (Adaptive Network-based Fuzzy Inference System) (Shahraiyni et al., 2015).
Figure 5 Output, target, error of ANN validation
4. DISCUSSION
4.1. Review the ANN Model for Subcontractor Selection
MSE validation results of the ANN model shows more than its MSE learning. It indicates that
the ANN model still needs training to improve its accuracy. Further training processes will be
carried out with a learning-by-doing mechanism, as shown in Figure 6. The advantage of these
mechanisms, at least for the proposed ANN model, is that they will contribute to harmonize the
strategic decisions by expert judgment and the performance appraisal by assessors. This will
provide a decision for subcontractor selection that has a more general weight, or the pattern of
the criteria weight of the previous assessors will also be considered for the formation of a new
pattern of the criteria weight.
Both the indicators of MSE of learning and validation indicate that the ANN model is able to
recognize the relationship pattern for both assessor's assessments and expert judgment, as
previously reviewed. The correlation test indicates a strong positive between the ANN result
and B1 criteria (quotation price) in both learning data (as shown in Table 3) and validation data
(as shown in Table 4), at 0.82 and 0.74, respectively. The proximity value that exists between
B1 (Quotation Price) and the output from ANN is explained by the R-square of the linear
regression; i.e., 0.67 (0.822) and 0.55 (0.742), respectively. For learning, this means that 67% of
B1 criteria can explain the expert judgment decisions, and 33% (100–67%) is explained by
other criteria. For validation, this means that 55% of B1 criteria can explain the expert judgment
decisions, and 45% (100–55%) is explained by other criteria.
The expert judgment decisions represent the strategic goal of the main contractor, one of which
is to get more profit for the company. This strategic goal is directly related to the B1 criteria.
Nevertheless, the final decision of the expert judgment does not release any other criteria to
support the expert judgment decision. This is in line with Haksever et al. (1995) and Latham
(1994), who illustrate that the use of price criteria as the sole basis for determining competent
subcontractors will not guarantee the performance of the subcontractor. This is possibly
because, in the practice of subcontractor selection, the cost parameter is one of the reasons the
contractor has partnered with the subcontractor. It is also in line with Černá et al. (2016), who
concluded that the crucial factors to the company for a services provider will be providing an
optimal level of services at minimal cost. Elazouni et al. (2000) also asserted that
subcontractors help contractors to overcome problems, including the need for special expertise,
resource shortage, and financial limitations. This is a primary goal of the main contractor and
will be achieved through the strategic decision.
Table 3 Correlation test of criteria to the ANN learning

A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A12 A12 A13 B1 B2 C1 C2 C3 D1 E1
0.03 0.57 -0.06 -0.07 0.07 -0.06 -0.23 -0.09 0.07 -0.34 0.10 -0.13 0.10 0.82 0.17 0.27 -0.04 0.32 -0.48 0.04
Table 4 Correlation test of criteria to the ANN validation

A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A12 A12 A13 B1 B2 C1 C2 C3 D1 E1
0.38 -0.26 0.34 -0.48 -0.29 0.23 0.24 0.12 -0.18 0.25 0.03 0.13 -0.17 0.74 0.48 -0.28 -0.12 0.38 0.27 -0.06
In the subcontractor selection practice, several criteria for the minimum requirements must be
met by the subcontractor’s performance, such as quality assurance criteria (A2). This is
important, but the assessor assessment should be above the performance of the requirement, so
the weight is not too significant. A difference in score from A2 is found in both alternatives of
P5 (learning data of 0.9, as shown in Table 1) and Q13 (validation data of 0.3, as shown in
Table 2). This difference of A2 (between learning and validation data) does not give
significance to the final score of the expert judgment.
Some other criteria are not requirements, but they will get additional scoring from the assessor
if they provide additional performance; one example is compression of schedule / D1. A
difference in score from A2 is found in alternative P5 (in learning data of 0.0) and Q13 (in
validation data of 0.2). Thus, the difference in score from A2 (between learning data and
validation data) does not give significance to the final score of expert judgment (successfully,
0.72 and 0.64), Even the Q13 final score is smaller, which may be influenced by other larger
criteria on alternative P5.
In the case of P4 and P13 alternatives, the B1 score of the assessor has the same value (0.5 and
0.5), and it is not in line with the final score of the expert judgment, which is the difference
(0.53 and 0.46). This indicates that the expert judgment has a consideration for other criteria
beyond the B1 criteria. Based on these conditions, we conclude that the decision of expert
judgment is a strategic decision that is correlated with the quotation price (B1), and it does not
neglect the importance of other criteria as decision support. This is the best practice for
subcontractor selection, where the quotation price is part of the overall evaluation criteria that
have significantly higher weight than other criteria. In this section, we conclude that the ANN
model has been able to perform the initial learning to manage expert judgment decisions.
Subsequent learning improvements will be made with the implementation of the ANN model in
the subcontractor selection, as shown in Figure 6.
4.2. Implementation of the ANN Model for Subcontractor Selection

In the subcontractor selection practice, two things need to be achieved (Bailey, 2016): the
company's strategic goal, which is maximized profits by the partnership, and the operational
goal, which is to get a qualified subcontractor. Strategic goals are achieved through expert
judgment decisions, while the operational goals are achieved through the performance
measurement of the assessors. Both goals are implemented in the practice of subcontracting
selection, as shown in Figure 6. This is in line with the opinion of Ko et al. (2017), who stated
that the performance evaluation of the subcontractor by general contractors consists of two
stages: primary and final scores. The interrelation between subcontractor performance and the
strategic goal needs a connection using the ANN model. Finally, the ANN model will be able to
substitute in full for the expert judgment or the directors in the decision-making of the
subcontractor selection.
The role of the assessors to conduct the performance assessment will require the criteria and
their weights. The performance criteria based on Fachrurrazi et al. (2017b), show that the
weight criteria vary between each assessor, as indicated by the standard deviation of the
criterion weights. This is in line with the subcontractor selection practice, where the main
contractor usually gives a freedom to determine the criterion weight within a specified range to
the assessor. Diversity in the weight criteria needs to be generated to achieve the ideal weights
of all assessors, which will represent the entire process of the subcontractor selection. This is an
important role for the ANN model for generating accuracy. The generalization capability of
ANN, which will form the patterns of criterion weight in each evaluation process of the
subcontractor, will improve the knowledge of ANN models. Furthermore, the general pattern of
ANN models will improve the performance of the decision results. The interaction between the
subcontractor selection process and the application of the ANN model (as shown in Figure 6)
shows that the ANN model will have an important role in the subcontractor selection process.
Figure 6 Interaction model of the subcontractor selection process and ANN Model
In the implementation phase, it will be interesting to change the ANN models into Simulink
models (Salmi et al., 2012), as this is intended to enhance the excellent performance of the
model as an application to continue the auto training in every real case. This implementation
phase will provide decision of the subcontractor selection with better accuracy and more
general weighting criteria (Oliveira et al., 2015).
5. CONCLUSION
A proposed ANN model, where its architecture is prepared by following the hierarchy structure
of decision criteria, has been successfully built for solving the problems of the selection of
subcontractors. The proposed ANN model is effective in patterning the data of the
subcontractor selection with minimal learning epochs of 600 to 800. The proposed ANN model
is able to identify patterns in the data quickly. The performance of learning has been achieved
in MSE-learning of 1.37269e-7 and MSE-validation of 0.07985. The findings show that the
ANN model has good validity, even though it is still lacking when compared to its MSE
learning. This indicates that the proposed ANN model needs increases in the raw dataset for
training. This stage will be conducted in the implementation stage of the ANN model.
Our use of the ANN model has identified the Quotation Price criterion (B1) as a significant
influence over the other criteria using the correlation method. This criterion (B1) is the critical
criterion that will give the greatest opportunity in partnering for the subcontractor and for
nomination in the selection.
The practice of selecting subcontractors in Indonesia, particularly in the reviewed case, has two
important stages for conducting the subcontractor evaluation; namely: (1) a performance
appraisal that relates to project needs and is conducted by assessors; and (2) a strategic
assessment that relates to the needs of the main contractor firms and is conducted by expert
judgment. The ANN role in the subcontractor selection process, which is to collaborate on the
assessor assessments and the expert judgment decision, has also been described. The
implementation phase of the ANN model is a crucial stage for generalizing the criterion weights
to improve the performance of the subcontractor selection.
6. ACKNOWLEDGEMENT
The Authors appreciate to the parties in their contribution; we would like thank Director of PT.
Waskita Karya and staff, and the CEM (Civil Engineering Management) Laboratory staff of
Syiah Kuala University. Special thanks go to the reviewers for their great comments for
improving this paper.
7. REFERENCES
Albino, V., Garavelli, A.C., 1998. A Neural Network Application to Subcontractor Rating in
Construction Firms. International Journal of Project Management, Volume 16(1), pp. 9–14
Azadnia, A.H., Saman, M.Z.M., Wong, K.Y., Ghadimi, P., Zakuan, N., 2012. Sustainable
Supplier Selection based on Self-organizing Map Neural Network and Multi-criteria
Decision Making Approaches. Procedia - Social and Behavioral Sciences, Volume 65, pp.
879–884
Bailey, J.M., 2016. Subcontractor Selection in the Construction Industry-Development of A
Research Approach to Investigate Selection Criteria, Methods, Value Creation Initiatives
And Supply Chain Management, Master's thesis, Norwegian University of Science and
Technology (NTNU), Trondheim, Norway
Černá, L., Buková, B., 2016. Supplier Evaluation Methodology in the Logistics Company.
Procedia Engineering, Volume 134, pp. 377–385
Chiarazzo, V., Caggiani, L., Marinelli, M., Ottomanelli, M., 2014. A Neural Network Based
Model for Real Estate Price Estimation Considering Environmental Quality of Property
Location. Transportation Research Procedia, Volume 3, pp. 810–817
Creese, R.C., Li, L., 1995. Cost Estimation of Timber Bridges using Neural Networks. Cost
Engineering, Volume 37(5), pp. 17–22
Elazouni, A.M., Metwally, F.G., 2000. D-SUB: Decision Support System for Subcontracting
Construction Works. Journal of Construction Engineering and Management, Volume
126(3), pp. 191–200
Euler-Rolle, N., Skrjanc, I., Hametner, C., Jakubek, S., 2016. Automated Generation of
Feedforward Control using Feedback Linearization of Local Model Networks. Engineering
Applications of Artificial Intelligence, Volume 50, pp. 320–330
Fachrurrazi, F., Away, Y., Husin, S., 2017b. The Weights Detection of Multi-criteria by using
Solver. International Journal of Electrical and Computer Engineering (IJECE), Volume
7(2), pp. 858–868
Fachrurrazi, Husin, S., Tripoli, Mubarak, 2017a. Neural Network for the Standard Unit Price of
the Building Area. Procedia Engineering, Volume 171, pp. 282–293
Haksever, M.A., Demir, I.H., Giran, O., 2001. Assessing the Benefits of Long-Term
Relationships Between Contractors and Subcontractors in the UK. International Journal
for Construction Marketing, Volume 3(1), pp. 1–10
Kim, G., An, SH., Kang, K-I., 2004. Comparison of Construction Cost Estimating Models
based on Regression Analysis, Neural Networks, and Case-based Reasoning. Building and
Environment, Volume 39, pp. 1235–1242
Ko, C. H., Cheng, M. Y., Wu, T. K., 2007. Evaluating Sub-contractors Performance using
EFNIM. Automation in Construction, Volume 16(4), pp. 525–530
Kusumoputro, B., Sutarya, D., Faqih, A., 2016. Performance Analysis of an Automatic Green
Pellet Nuclear Fuel Quality Classification using Modified Radial Basis Function Neural
Networks. International Journal of Technology, Volume 7(4), pp. 709-719
Latham, M., 1994. Constructing the Team. HMSO London Department of the Environment,
ISBN: 0-11-752994-X
Loyola, R.D.G., Pedergnana, M. Gimeno García, S., 2016. Smart Sampling and Incremental
Function Learning for Very Large High Dimensional Data. Neural Networks, Volume 78,
pp. 75–78
Meruelo, A.C., Simpson, D.M., Veres, S.M., Newland, P.L., 2016. Improved System
Identification using Artificial Neural Networks and Analysis of Individual Differences in
Responses of an Identified Neuron. Neural Networks, Volume 75, pp. 56–65
Moselhi, O., Hegazy, T., Fazio, P., 1992. Potential Applications of Neural Networks in
Construction. Canadian Journal of Civil Engineering, Volume 19(3), pp. 521–529
Nayak, J., Naik, B., Behera, H.S., 2016. A Novel Nature-Inspired Firefly Algorithm with
Higher Order Neural Network: Performance Analysis. Engineering Science and
Technology, Volume 19(1), pp. 197–211
Oliveira, G.G., Pedrollo, O.C., Castro, N.M.R., 2015. Simplifying Artificial Neural Network
Models of River Basin Behavior by an Automated Procedure for Input Variable Selection.
Engineering Applications of Artificial Intelligence, Volume 40, pp. 47–61
Salmi, T., Bouzguenda, M., Gastli, A., Masmoudi A., 2012. MATLAB/Simulink Based
Modelling of Solar Photovoltaic Cell. International Journal of Renewable Energy
Research, Volume 2(2), pp. 1–6
Shahraiyni, H.T., Sodoudi, S., Kerschbaumer, A., Cubasch, U., 2015. A New Structure
Identification Scheme for ANFIS and its Application for the Simulation of Virtual Air
Pollution Monitoring Stations in Urban Areas. Engineering Applications of Artificial
Intelligence, Volume 41, pp. 175-182
Taghavifar, H., Mardani, A., 2014. Application of Artificial Neural Networks for the Prediction
of Traction Performance Parameters. Journal of the Saudi Society of Agricultural Sciences,
Volume 13(1), pp. 35–43
Zhang, Y., Zhang, E., Chen, W., 2016. Deep Neural Network for Halftone Image Classification
based on Sparse Autoencoder. Engineering Applications of Artificial Intelligence, Volume
50, pp. 245–255
Zuna, H., Hadiwardoyo, S., Rahadian, H., 2016. Developing A Model of Toll Road Service
Quality using an Artificial Neural Network Approach. International Journal of
Technology, Volume 7(4), pp. 562–570
APPENDIX
Table 1 Dataset of input and target for training of the ANN model (Fachrurrazi et al., 2017b)
Multi-Criteria of The Decision Alternative (Number of Subcontractors)
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19 P20 P21 P22 P23 P24 P25
A. Subcontractor Credibility
1. Company profile
a. Management Capabilities
• Quality system
- ISO certification, similar A1 0.60 1.00 0.80 0.50 0.60 0.60 0.60 0.90 0.60 0.60 0.50 1.00 0.00 1.00 0.50 0.50 0.20 1.00 0.80 0.90 0.50 0.50 0.80 0.65 1.00
- Quality assurance A2 0.30 0.50 0.60 1.00 0.90 0.50 0.65 0.50 0.50 0.40 0.50 0.30 0.90 0.00 0.00 0.50 0.30 1.00 0.60 1.00 0.60 0.50 0.60 0.90 1.00
- Company profile A3 0.20 1.00 0.60 1.00 0.40 0.40 0.80 0.55 0.50 0.70 0.60 0.50 0.60 0.50 0.50 0.80 0.50 1.00 0.20 1.00 0.70 1.00 0.80 0.65 1.00
• Financial Stability
- Balance Sheet A4 0.80 0.60 0.30 0.90 0.60 0.60 0.55 0.60 0.60 0.30 0.55 0.65 0.10 0.10 0.60 0.90 0.50 1.00 0.20 0.50 0.25 0.20 0.80 0.80 1.00
- Bank guarantee A5 0.80 0.90 0.30 1.00 0.50 0.70 0.65 0.20 0.80 0.30 0.30 0.70 0.90 0.50 0.20 1.00 0.30 1.00 0.30 0.60 0.90 0.85 0.25 0.90 1.00
b. Technology capability
• Facilities A6 0.80 0.70 0.50 0.60 0.80 0.30 0.60 0.80 0.50 0.20 0.70 0.85 0.90 0.60 0.70 0.50 0.80 0.75 0.10 0.80 0.65 0.45 0.45 0.60 0.50
• Transport A7 0.80 0.60 0.60 0.70 0.50 0.10 0.80 0.20 0.40 0.50 0.60 0.20 0.50 0.30 0.80 1.00 0.60 0.65 0.70 0.30 0.90 0.55 0.80 0.60 0.60
• Equipment A8 0.80 0.70 0.60 0.80 0.50 0.10 0.50 0.80 0.40 0.50 1.00 0.40 0.70 0.50 0.10 1.00 0.30 0.50 0.90 0.30 0.80 0.90 0.60 0.45 0.70
2. Contract Trustworthy
a. Project Experience A9 0.60 1.00 0.50 0.90 0.30 0.90 0.25 0.30 0.60 0.30 0.50 0.30 0.80 0.50 0.50 0.30 0.40 0.90 0.30 0.50 0.60 0.65 0.85 0.60 1.00
b. Project achievement A10 0.40 1.00 0.90 0.20 0.30 0.30 0.65 0.70 0.30 0.50 1.00 0.80 0.70 0.60 0.80 0.50 0.50 0.80 0.30 0.40 0.40 0.80 0.50 0.30 1.00
c. Type, amount of insurance A11 0.90 0.60 0.90 0.30 0.50 0.80 0.50 0.60 0.50 0.40 1.00 1.00 0.90 0.70 0.30 1.00 0.20 1.00 0.90 0.90 0.90 0.35 0.60 0.50 1.00
d. Registered in associations A12 1.00 0.20 0.90 0.50 0.50 0.60 0.40 0.50 0.50 0.20 0.50 1.00 0.80 0.50 0.10 1.00 0.90 1.00 0.10 0.30 0.80 0.60 0.45 0.80 0.90
e. Company legitimate A13 0.50 0.90 1.00 0.50 0.80 1.00 0.60 0.20 0.50 0.80 0.40 1.00 0.50 0.40 0.80 0.80 0.50 1.00 0.80 0.20 0.40 0.85 0.30 0.60 1.00
B. Quotation
1. Quotation Price B1 0.35 0.25 0.50 0.60 0.80 0.70 0.65 0.40 0.30 0.10 0.15 0.05 0.60 0.50 0.25 0.30 0.40 0.60 0.70 0.30 0.20 0.50 0.30 0.40 0.20
2 Methods of Payment B2 0.00 0.25 0.20 0.30 0.80 0.40 0.10 0.50 1.00 1.00 1.00 0.20 0.50 0.00 0.00 0.30 0.00 0.50 0.60 0.40 0.20 0.20 0.20 0.40 0.40
C. Technical Capabilities
1. Expertise of personnel C1 0.80 0.10 0.50 0.30 0.90 1.00 0.10 0.20 0.70 0.80 0.90 0.30 0.90 0.80 0.50 0.60 0.30 0.20 0.80 0.90 1.00 0.50 0.60 0.80 0.60
2. Specializes in working methods C2 0.50 0.60 0.40 0.10 0.55 0.80 0.35 0.65 0.45 0.20 0.30 0.80 0.70 0.55 0.60 0.40 0.50 0.80 0.30 0.20 0.60 0.70 0.50 0.95 0.40
3. Material specification C3 0.70 0.60 0.90 0.90 0.90 0.90 0.90 0.80 0.60 0.90 0.70 0.90 0.60 0.80 0.70 0.50 0.60 0.90 0.80 0.50 0.70 0.90 0.60 0.70 0.80
D. Execution Time
1. Compression of schedule D1 0.20 0.30 0.40 0.10 0.00 0.20 0.60 0.30 0.10 0.20 0.50 0.60 0.50 0.40 0.60 0.50 0.40 0.30 0.10 0.15 0.25 0.35 0.60 0.45 0.55
E. Type of Project References
1. Number of similar work E1 0.80 0.60 0.90 0.80 0.70 0.30 0.10 0.20 0.50 0.55 0.45 0.65 0.00 0.35 0.60 0.10 0.00 0.00 0.40 0.30 0.20 0.20 0.30 0.50 0.60
Data for Target
A. Subcontractor Credibility 0.69 0.76 0.64 0.67 0.57 0.53 0.59 0.55 0.54 0.43 0.62 0.71 0.58 0.52 0.48 0.76 0.44 0.90 0.50 0.66 0.65 0.58 0.62 0.65 0.89
B. Quotation 0.29 0.25 0.45 0.55 0.80 0.65 0.56 0.42 0.42 0.25 0.29 0.08 0.58 0.42 0.21 0.30 0.33 0.58 0.68 0.73 0.20 0.45 0.28 0.40 0.23
C. Technical Capabilities 0.67 0.43 0.60 0.43 0.78 0.90 0.45 0.55 0.58 0.63 0.63 0.67 0.73 0.72 0.60 0.50 0.47 0.63 0.63 0.53 0.77 0.70 0.75 0.82 0.60
D. Execution Time 0.20 0.30 0.40 0.10 0.00 0.20 0.60 0.30 0.10 0.20 0.50 0.60 0.50 0.40 0.60 0.50 0.40 0.30 0.10 0.15 0.25 0.35 0.60 0.45 0.55
E. Type of Project References 0.80 0.60 0.90 0.80 0.70 0.30 0.10 0.20 0.50 0.55 0.45 0.65 0.00 0.35 0.60 0.10 0.00 0.00 0.40 0.30 0.20 0.20 0.30 0.50 0.60
Final Score of The Expert Judgment 0.43 0.37 0.53 0.55 0.72 0.62 0.50 0.42 0.35 0.35 0.40 0.31 0.54 0.46 0.35 0.37 0.33 0.55 0.60 0.62 0.33 0.47 0.38 0.50 0.41
Output of ANN Learning 0.43 0.37 0.53 0.55 0.72 0.62 0.50 0.42 0.35 0.35 0.40 0.31 0.54 0.46 0.35 0.37 0.33 0.55 0.60 0.62 0.33 0.47 0.38 0.50 0.41
772 The Subcontractor Selection Practice using Ann-Multilayer
Table 2 Dataset of input and target for validation of the ANN model (Fachrurrazi et al., 2017b)
Data for Validation
Multi-Criteria of The Decision Alternative (Number of Subcontractors)
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15
(1) (2) (3)
A. Subcontractor Credibility
1. Company profile
a. Management Capabilities
• Quality system
- ISO certification, similar A1 1.00 1.00 0.00 0.80 0.80 0.90 0.80 0.50 0.20 0.60 0.50 0.10 0.50 0.90 0.40
- Quality assurance A2 1.00 1.00 1.00 0.60 1.00 0.50 0.50 0.10 0.30 0.30 0.80 0.60 0.30 0.40 0.50
- Company profile A3 0.60 0.70 0.30 1.00 0.50 0.80 0.20 0.50 0.20 0.80 0.60 0.20 0.50 0.80 0.80
• Financial Stability
- Balance Sheet A4 0.70 0.50 0.80 0.70 0.85 0.50 0.20 0.50 0.80 0.25 0.50 0.80 0.20 0.50 0.60
- Bank guarantee A5 1.00 1.00 1.00 0.50 0.40 0.80 0.80 0.70 0.60 0.80 0.90 0.40 0.30 0.30 0.80
b. Technology capability
• Facilities A6 0.20 0.50 0.60 0.50 1.00 0.90 0.60 0.40 0.60 0.50 0.30 0.30 0.40 0.80 0.50
• Transport A7 1.00 1.00 0.30 1.00 0.50 0.45 0.50 0.90 0.70 0.35 0.40 0.20 0.80 0.60 0.50
• Equipment A8 0.75 0.50 0.30 0.60 0.50 0.70 0.90 0.70 0.20 0.80 0.20 0.50 0.90 0.30 0.50
2. Contract Trustworthy
a. Project Experience A9 1.00 0.50 0.70 1.00 0.40 0.35 1.00 0.80 0.60 0.75 0.10 0.60 1.00 0.50 0.70
b. Project achievement A10 0.50 0.10 0.40 0.20 0.50 0.15 1.00 0.50 0.40 0.50 0.50 0.20 0.50 0.70 0.50
c. Type, amount of insurance A11 0.50 1.00 0.60 1.00 1.00 0.50 1.00 0.30 0.50 1.00 0.60 0.70 0.90 0.60 0.60
d. Registered in associations A12 1.00 1.00 0.30 1.00 0.80 0.20 1.00 0.70 0.60 0.75 0.80 0.20 0.40 0.80 0.60
e. Company legitimate A13 0.70 1.00 0.50 1.00 0.60 0.50 1.00 0.20 0.50 0.80 0.40 0.50 0.60 0.60 0.30
B. Quotation
1. Quotation Price B1 0.55 0.60 0.20 0.10 0.60 0.35 0.65 0.55 0.40 0.60 0.50 0.20 0.70 0.55 0.60
2 Methods of Payment B2 0.00 0.00 0.00 0.20 0.55 0.60 0.40 0.55 0.30 0.25 0.80 0.70 0.90 1.00 0.00
C. Technical Capabilities
1. Expertise of personnel C1 0.90 0.80 0.50 0.60 0.40 0.50 0.65 0.45 0.85 0.70 0.60 0.55 0.35 0.60 0.50
2. Specializes in working methods C2 0.50 0.60 0.90 0.60 0.80 0.10 0.50 0.30 0.80 0.70 0.30 0.60 0.50 0.80 0.50
3. Material specification C3 0.80 0.90 0.60 0.80 0.70 0.50 0.60 0.90 0.60 0.70 0.80 0.50 0.60 0.90 0.80
D. Execution Time
1. Compression of schedule D1 0.10 0.20 0.25 0.30 0.30 0.30 0.10 0.20 0.25 0.45 0.15 0.15 0.20 0.35 0.25
E. The Type of Project References
1. Number of similar work E1 0.80 0.50 0.60 0.90 0.80 0.70 0.30 0.65 0.65 0.60 0.35 0.30 0.60 0.20 0.40
Data for Target
A. Subcontractor Credibility 0.77 0.79 0.61 0.76 0.71 0.61 0.71 0.53 0.49 0.62 0.52 0.40 0.55 0.62 0.56
B. Quotation 0.46 0.50 0.17 0.12 0.59 0.39 0.61 0.55 0.38 0.54 0.55 0.23 0.73 0.63 0.50
C. Technical Capabilities 0.73 0.77 0.67 0.67 0.63 0.37 0.65 0.55 0.75 0.70 0.57 0.55 0.48 0.77 0.60
D. Execution Time 0.10 0.20 0.25 0.30 0.30 0.30 0.10 0.20 0.25 0.45 0.15 0.15 0.20 0.35 0.25
E. Type of Project References 0.80 0.50 0.60 0.90 0.80 0.70 0.30 0.65 0.65 0.60 0.35 0.30 0.60 0.20 0.40
Final Score of The Expert Judgment 0.55 0.55 0.33 0.35 0.62 0.44 0.57 0.54 0.47 0.57 0.51 0.33 0.64 0.59 0.50
Output of ANN Learning 0.4794 0.5621 0.3512 0.3643 0.6309 0.5335 0.4115 0.5861 0.471 0.5725 0.594 0.2605 0.7941 0.7197 0.5357

Fachrurrazi Jurnal 9428

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fachrurrazi Jurnal 9428

Uploaded by

Copyright:

Available Formats

12/5/2019 The Subcontractor Selection Practice using ANN-Multilayer

Vol 8, No 4 (2017) > Civil Engineering

The Subcontractor Selection Practice using ANN-Multilayer

Fachrurrazi , Saiful Husin, Munirwansyah , Husaini

Corresponding email:  fachrurrazi@unsyiah.ac.id (mailto:fachrurrazi@unsyiah.ac.id)

Published at : 31 Jul 2017

Cite this article as:

DOWNLOAD PDF  (HTTP://IJTECH.ENG.UI.AC.ID/DOWNLOAD/ARTICLE/591)

WHO CITE THIS PAPER  (HTTPS://SCHOLAR.GOOGLE.COM/SCHOLAR?Q=THE+SUBCONTRACTOR+SELECTION+PRACTICE+USING+ANN-MULTILAYER)

Copyright © 2017 Faculty of Engineering

Vol 10, No 7 (2019) SE

Vol 10, No 6 (2019) SE

Vol 10, No 5 (2019)

Vol 10, No 4 (2019)

Vol 10, No 3 (2019) SE

Vol 10, No 2 (2019)

Vol 10, No 1 (2019)

 About the Journal

Editorial Board (http://ijtech.eng.ui.ac.id/people)

Not as user? LOGIN

Metrics by SCOPUS 2018

Acceptance rate: 28.32 %

 IJTech is indexed in:

Copyright © 2017 Faculty of Engineering

Fostering Partnerships and Strategic Alliances in Sustainable Infrastructure

Publication Date (Online):

Electrolytic Refining of Lead in Molten Chloride Electrolytes

Publication Date (Online):

Synthesis and Characterization of Hematite Nanoparticles using Ultrasonic

Munawar Khalil, Ning Liu, Robert L. Lee

Publication Date (Online):

Publication Date (Online):

Sin Activation Structural Tolerance of Online Sequential Circular Extreme Learning

Publication Date (Online):

A Preliminary Study on Shifting from Virtual Machine to Docker Container for

Publication Date (Online):

Publication Date (Online):

Calorific Value Analysis of Azadirachta Excelsa and Endospermum Malaccense as

Publication Date (Online):

Synthesis and Magnetic Characterization of Mn-Ti Substituted SrO.6Fe2-

Publication Date (Online):

Effect of Post-Hydrothermal Treatments on the Physical Properties of ZnO Layer

Publication Date (Online):

Experimental Study on Mechanical Properties of Aluminum Alloys under Uniaxial

Publication Date (Online):

Kinetic Model Development for Biogas Production from Lignocellulosic Biomass

Publication Date (Online):

Publication Date (Online):

Publication Date (Online):

TRIGA-2000 Research Reactor Thermal-hydraulic Analysis using

Publication Date (Online):

Controlling Unmanned Surface Vehicle Rocket using GPS Tracking Method

Publication Date (Online):

Publication Date (Online):

Strength Development of High-Performance Concrete using Nanosilica

Publication Date (Online):

Measuring the Effect of Strengthened Concrete on the Fracture Characteristics of

Publication Date (Online):

Shaking Table Test of Soil Liquefaction in Southern Yogyakarta

Publication Date (Online):

The Subcontractor Selection Practice using ANN-Multilayer

Publication Date (Online):

LIST OF REVIEWERS CONTRIBUTING IN THIS ISSUE (http://ijtech.eng.ui.ac.id/reviewer/33)