a r t i c l e i n f o a b s t r a c t
Article history: Optimal network design is a key factor in the enhancement of the economic, environmental, and social
Received 11 March 2016 performance and efﬁciency of the biomass supply chain (BSC), and this is why it has become quite popular
Received in revised form 5 September 2016 with the academia and practitioners. The great number of the related papers published in the scientiﬁc
Accepted 12 September 2016
journals in recent years is the proof of the claim; therefore, to make a framework of the past works and
Available online 7 October 2016
specify the future directions, a comprehensive review of the state-of-the-art papers deems necessary.
The objective of this paper is to review the papers regarding the biomass supply chain network design
Keywords:
(BSCND) models published in the scientiﬁc journals. A total number of 146 papers, published from Jan.
Biomass
Biofuel
1997 to Jul. 2016 are reviewed, analyzed and classiﬁed based on their modeling approaches, decisions,
Bioenergy uncertainties, solution methodologies, sustainability, model features, entities, data, and regions of the
Supply chain case studies. To determine the research opportunities and future directions, the gaps existing in the
Network design present literature have been clearly explained as well.
Optimization models © 2016 Elsevier B.V. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 973
2. Research motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 973
3. Research methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 973
3.1. Material collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974
3.2. Descriptive analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974
3.3. Category selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974
3.4. Material evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974
4. Detailed analyses of the literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974
4.1. Decisions in biomass supply chain network design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974
4.1.1. Facility-related strategic decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974
4.1.2. Biomass-related strategic decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 974
4.1.3. Final product-related strategic decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 982
4.2. Modeling approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 983
4.2.1. Mathematical programming approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 984
4.2.2. Multi-criteria decision making. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .989
4.2.3. Heuristic and meta-heuristic approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 989
Abbreviations: BSC, Biomass Supply Chain; BSCND, Biomass Supply Chain Network Design; GIS, Geographical Information Systems; EDSS, Environmental Decision Support
System; LP, Linear Programming; NLP, Non-linear Programming; MINLP, Mixed Integer Non-linear Programming; MCDM, Multi-Criteria Decision Making; MADM, Multi-
Attribute Decision Making; MODM, Multi-Objective Decision Making; BHBF, Binary Honey Bee Foraging; GA, Genetic Algorithm; PSO, Particle Swarm Optimization; SAA,
Sample Average Approximation; LA, Lagrangian Relaxation; CA, Continuum Approximation; CS, Commercial Solver; EA, Exact Algorithm; HMA, Heuristic, Meta-heuristic and
Approximation algorithms.
∗ Corresponding author.
E-mail addresses: moini@iust.ac.ir, moiniam@gmail.com (A. Moini).
http://dx.doi.org/10.1016/j.indcrop.2016.09.027
0926-6690/© 2016 Elsevier B.V. All rights reserved.
H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000 973
steps: (1) material collection, (2) descriptive analysis, (3) category 4. Detailed analyses of the literature
selection, and (4) material evaluation.
In this section, the selected papers are evaluated and analyzed
based on the classiﬁcation criteria so as to identify the gap in the
3.1. Material collection
literature and the potential opportunities for future researches.
The description of step 1, namely material collection, is
4.1. Decisions in biomass supply chain network design
described in this part. During the search process, the BSCND related
papers accepted in English scientiﬁc journals (available online)
In the supply chain management, planning decisions can be cat-
from the beginning of Jan. 1997 to Jul. 2016 were covered. First, a
egorized in three levels including (1) strategic, (2) tactical, and
maximum of one keyword was selected from each keyword group
(3) operational with respect to planning time horizon (Mula et al.,
in Table 1 and different combinations were made. Next, the word
2010). Basically, BSCND decisions are of the ﬁrst category (strategic-
combinations were searched in the Google Scholar Search Engine
level) because they cannot be changed in short term, and they have
(https://scholar.google.com/) for the time period from Jan. 1997
long-term effects on the entire supply chain. Also, they generally
to Jul. 2016. After ﬁnding the papers, their abstracts were glanced
require high investments, and they may need revision after about
online and archived (if related) to be studied more precisely later in
3–8 years (De Meyer et al., 2014). All network design decisions
the preparation of the present review; this resulted in the selection
affect one another and this is why they should be integrated. Table 2
of 146 papers.
shows the evaluation of BSCND papers based on such strategic
decisions as sourcing, allocation of biomass between facilities, the
3.2. Descriptive analysis biomass feedstock and ﬁnal product type in the case studies, pro-
duction technology as well as the capacity and location decisions
A signiﬁcant effort has been made in this study, to evaluate considered in BSCND models (Awudu and Zhang, 2012). Consid-
the 146 selected papers considering the features introduced in the ering their nature and to analyze them better, strategic decisions
BSCND discussion. Our early investigations show that these papers are classiﬁed, in this paper, into three general classes of (1) facility-
have been published in 65 different Journals (138 papers), 5 Con- related, (2) biomass-related, and (3) ﬁnal product-related.
ferences (5 papers), 2 Symposiums (2 papers), and 1 Congress (1
paper) (see Fig. 1) which are obtained from various publishers such 4.1.1. Facility-related strategic decisions
as ScienceDirect (http://www.sciencedirect.com/), Informs (http:// This class includes decisions regarding the technology, capac-
pubsonline.informs.org/), Elsevier (www.elsevier.com), Springer ity, and location of different facilities including supply, collection,
(www.springerlink.com), Taylor & Francis (http://taylorandfrancis. preprocessing, processing, intermediate processing, blending, and
com/), Wiley (http://www.wiley.com), IEEE (http://ieeexplore.ieee. distribution sites (Sharma et al., 2013a). These decisions can specif-
org/Xplore/home.jsp), ACS (www.pubs.acs.org), AIDIC (www.aidic. ically and separately determine the location and/or the capacity of
it), Inderscience (http://www.inderscience.com/), LBNL (https:// each facility. Fig. 4 shows the strategic decision variables related
commons.lbl.gov/display/rst/LBNL+Journals), Finish Society of For- to each facility. As shown in Fig. 4, the processing facility has the
est Science (http://www.metsatieteellinenseura.ﬁ/english). As highest number of decision variables, and the location of this facil-
shown in Fig. 1, Biomass and Bioenergy with 18 papers (12.33%) ity has attracted more attention than the other facilities. In most
ranks ﬁrst in publishing such papers. The ﬁrst paper on BSCND was of mathematical models existing in the literature, this decision is
published in 1997. Fig. 2 shows the number of the papers published stated in the form of a binary variable that shows the establish-
each year; as shown in Fig. 2, the number of BSCND-related pub- ment of a processing site in a potential location in a known period
lished papers has an increasing trend, and in 2016 it reaches its of time. Notably, locations of the processing sites directly affect
highest value (21 papers, 14.38%). the allocation of the biomass feedstock. Nearness of the supply and
processing sites is a factor that results in competition among the
3.3. Category selection biomass supply regions (Chen and Önal, 2014). The reason goes
back to its low bulk density; biomass feedstock occupies much
The classiﬁcation applied in this review is formed based on space, which causes an increase in the transportation costs espe-
different features and properties of BSCND models. Also, the clas- cially in long distances (Richard, 2010). Since such costs depend
siﬁcations used in (De Meyer et al., 2014; Sharma et al., 2013a) much on the processing site location, the corresponding facility
and (Govindan et al., 2015) are used as benchmark to check the location decision should be optimized. Similarly, optimal location
comprehensiveness of the proposed classiﬁcation framework. The and capacity of other facilities can decrease biomass supply chain
main topics and categories used to classify the papers are shown in costs. However, this review concludes that the little attention paid
Fig. 3. As shown in Fig. 3, the selected papers are classiﬁed based to the strategic decisions related to the other facilities is a gap in the
on such topics as decision variables, entities, modeling approaches, literature. With an increase in the model ﬂexibility, such decisions
solution methodologies, model features, uncertainty, sustainabil- will lead to the design of BSCs with lesser total costs.
ity, data and region of the case studies.
4.1.2. Biomass-related strategic decisions
The biomass-related strategic decisions include the determina-
3.4. Material evaluation tion of biomass type, sourcing and allocation decisions (see Table 2).
Evaluations have revealed that in 54.8% of the published papers, the
The use of deductive/inductive methods in the validation stage developed models have been multi-feedstock; however, only 45.2%
enhances the credibility of the research. Also, the application of of the case studies have studied multi-feedstock supply chains.
spreadsheet software such as Microsoft Excel and Access in the Fig. 5 shows the application trend of the single and multi-feedstock
evaluation and analysis phase reduces errors (Govindan et al., models in different years. As shown in Fig. 5, single-feedstock mod-
2015). To make sure of the completeness of the set of collected els have had a growing trend in recent years. Although multiple
papers and of the precision of the Google Scholar Search Engine, the sources of biomass decreases such lateral supply chain costs as
keywords are searched again in the WOS and SCOPUS data bases, transportation, insurance, and communication as well as price (due
and those papers not found in the ﬁrst search are added to the list. to the suppliers’ rivalry), it increases the need for various produc-
H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000 975
Table 1
Keyword groups for the searching process in Google scholar.
40
35 38
30
25
20
15 18
10
11 10 10
5 8 9
3 2 3 2 5 3 4 2 3 3 7 5
0
25
21
20 19
18
17
15 14
11 11
10 10
10
5
3 3
2 2 2
1 1 1
0 0 0
0
Fig. 2. Distribution of publications per year across the period of the study (146 papers: 1997–2016).
tion technologies in various supply chain facilities. These are the characterized by the increase in the number of papers published
main factors which caused less popularity of multiple sourcing in on the second-generation biomass in recent years.
BSCND models. Table 3 shows the evaluation and classiﬁcation of Sourcing can be deﬁned as specifying the sources, assuring the
the reviewed papers based on the biomass feedstock, ﬁnal prod- supply, insuring supply source diversiﬁcation, and collecting infor-
uct and its generation according to the classiﬁcation presented in mation from procurable sources (Driedonks et al., 2010). Based on
(Sharma et al., 2013a). As shown in Table 3, the second-generation Table 2, only 43.15% of the papers have sourcing decision mean-
biomass is the most widely studied generation in BSCND papers. ing that the supplier selection and management have not been of
The agricultural and animal residues and woody crops are also much importance in BSCND models. Allocating biomass to various
among the popular second-generation biomass in BSCND studies. production facilities depends on the distances between supply sites
It should be noted that the ﬁrst generation biomass is based on the and production facilities, the biomass type, the production technol-
materials that can also be used as human food (such as corn, sugar ogy, and the capacity of each facility. Table 2 shows that in 78% of
cane and soybean) and the consumption of such materials endan- the existing papers, the biomass allocation is considered in BSCND
gers the human foods supply. The researchers’ concern regarding models.
the substitution of the food biomass by the non-food biomass is
976
Table 2
Strategic decisions in the biomass supply chain network design models.
Location Capacity Technology Sourcing Allocation Multi Feedstock Biomass type in case Multi Product Product types in case
studies studies
977
978
Table 2 (Continued)
Location Capacity Technology Sourcing Allocation Multi Feedstock Biomass type in case Multi Product Product types in case
studies studies
979
980
Table 2 (Continued)
Location Capacity Technology Sourcing Allocation Multi Feedstock Biomass type in case Multi Product Product types in case
studies studies
Tong et al. (2014a) LPP, LP CPP, CP x x x x Crop residues (e.g. Corn x Gasoline, diesel, jet fuel
stover), energy crops
(e.g.
Switchgrass), wood
residues (e.g. Forest
residues), crude oil
Marvin et al. (2012) LP CP x x x Corn, corn stover, x Bioethanol, biodiesel
forest biomass, wood
waste, energy crops
Bai et al. (2012) LP x Corn Bioethanol
LS: location of supply site; CS: capacity of supply site; LC: location of collection site; CC: capacity of collection site; LPP: location of preprocessing site; CPP: capacity of preprocessing site; LP: location of processing site; CP:
capacity of processing site; LIP: location of intermediate processing site; CIP: capacity of intermediate processing site; LB: location of blending site; CB: capacity of blending site; LD: location of distribution site; CD: capacity of
distribution site; LF: location of facilities, CF: capacity of facilities.
981
982 H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000
Multi-feedstock Single-feedstock
14 13
12 12
12 11
10
8
8 7 7 7 7
7 6
6 5 7
6
5 5
4 3 3
2 2 2
2
2 1 1 1 1
0 0 0 1
0 0 0 1
0 0
0
4.1.3. Final product-related strategic decisions trend of the single and multi-product models in different years.
In the third part of Table 2, the BSC ﬁnal products have been The computational complexities associated with multi-feedstock
studied in terms of their types and number. An investigation of and multi-product models can be accounted for their low appli-
the selected papers shows that more than 40% of the developed cability. An investigation of the biofuels in Table 3 shows that
models are multi-product, but only about 36% of the case stud- researches on BSCND with second-generation biofuels are increas-
ies are related to multi-product BSCs. Fig. 6 shows the application ing. As mentioned earlier, concern regarding the production of the
H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000 983
Table 3
Publications based on biomass feedstock, ﬁnal product and its generation.
First generation bioethanol, Starch and sugar based Morrow et al. (2006); Dal-Mas et al. (2011); Mas et al. (2010); Giarola et al. (2011a,
biodiesel, heat and power 2011b, 2012, 2013); Mele et al. (2011); Kostin et al. (2012); Roni et al. (2014); Singh
et al. (2014); Srisuwan and Dumrongsiri (2012); Zamboni et al. (2009a, 2009b);
Corsano et al. (2011); Bai et al. (2011, 2012, 2015, 2016); Xie et al. (2014); McLean and
Li (2013); Lam et al. (2011); Ren et al. (2013, 2015); Tursun et al. (2008); Ekşioğlu et al.
(2010); Marvin et al. (2012); Akgul et al. (2010, 2011, 2012); Čuček et al. (2010, 2012);
Bairamzadeh et al. (2015); Gonela et al. (2015a, 2015b); d’Amore and Bezzo (2016);
Santibañez-Aguilar et al. (2016); Miret et al. (2016); Yue et al. (2016)
Agricultural residues and Nagel (2000); Vlachos et al. (2008); Čuček et al. (2010); Alex Marvin et al. (2012); You
livestock products et al. (2012); Zhu and Yao (2011); Avami (2012); Aksoy et al. (2011); Zhang and Hu
(2013); Singh et al. (2011); Ma et al. (2005); Tong et al. (2013, 2014a, 2014b);Giarola
et al. (2011a, 2011b, 2012, 2013); Ebadian et al. (2013); Mapemba et al. (2008);
Rentizelas et al. (2009a); Ekşioğlu et al. (2009); Huang et al. (2010); Bai et al. (2011);
You and Wang (2011, 2012); Perpina et al. (2009); Dunnett et al. (2008); Parker et al.
(2010); Li and Hu (2014); Yue et al. (2013, 2014, 2016); Walther et al. (2012);
Rentizelas and Tatsiopoulos (2010); Celli et al. (2008); Izquierdo et al. (2008); Vera
et al. (2010); Yue and You (2014); Marvin et al. (2012); Akgul et al. (2012);
Gebreslassie et al. (2012); Čuček et al. (2012); Murillo-Alvarado et al. (2015);
Hombach et al. (2016); Ng and Maravelias (2016); Woo et al. (2016); Balaman and
Selim (2016); d’Amore and Bezzo (2016); Poudel et al. (2016); Lim and Lam (2016)
Urban and industrial woody You et al. (2012); Aksoy et al. (2011); Natarajan et al. (2012, 2014); Balaman and Selim
wastes and landﬁlls (2014); Ekşioğlu et al. (2009); An et al. (2011); Aksoy et al. (2011); You and Wang
(2011, 2012); Chen and Fan (2012); Frombo et al. (2009a); Perpina et al. (2009);
Kanzian et al. (2009); Leduc et al. (2008); Parker et al. (2010); Tong et al. (2013);
Izquierdo et al. (2008); Marvin et al. (2012); Gebreslassie et al. (2012); Yue et al.
(2013); Akgul et al. (2014); Čuček et al. (2012); Marufuzzaman et al. (2014a);
Babazadeh et al. (2016); Duarte et al. (2016); Woo et al. (2016); Cambero and Sowlati
(2016); Lim and Lam (2016)
Forest biomass Mapemba et al. (2007); Nagel (2000); Freppaz et al. (2004); Gunnarsson et al. (2004);
Frombo et al. (2009a, 2009b); Panichelli and Gnansounou (2008); Aksoy et al. (2011);
Leduc et al. (2010); Čuček et al. (2010); Natarajan et al. (2012, 2014); Elia et al. (2013);
Palak et al. (2014); Mansoornejad et al. (2013); Ekşioğlu et al. (2009); Huang et al.
(2010); Kim et al. (2010, 2011a,b); Kanzian et al. (2009); Parker et al. (2010); Leduc
et al. (2008); Tong et al. (2013, 2014a, 2014b); Walther et al. (2012); Ayoub et al.
(2007); Izquierdo et al. (2008); López et al. (2008); Shi et al. (2008); Marvin et al.
(2012); Chinese and Meneghetti (2009); Gebreslassie et al. (2012); Yue et al. (2013);
Akgul et al. (2014); Čuček et al. (2012); Cambero et al. (2015); Sharifzadeh et al.
(2015); Paulo et al. (2015); Hombach et al. (2016); Woo et al. (2016);
Santibañez-Aguilar et al. (2016); Miret et al. (2016); Poudel et al. (2016); Cambero
et al. (2016); Zhang et al. (2016); Lim and Lam (2016)
Herbaceous energy crops Cundiff et al. (1997); Morrow et al. (2006); Dunnett et al. (2007); Grigoroudis et al.
(2014); Zhu et al. (2011); You et al. (2012); Zhu and Yao (2011); Wang et al. (2012);
Rozakis et al. (2001); Zhang et al. (2013); Osmani and Zhang (2013); Mapemba et al.
(2008); An et al. (2011); Bai et al. (2011); Giarola et al. (2012, 2013); You and Wang
(2011, 2012); Sharma et al. (2013b); Tong et al. (2014a); Akgul et al. (2011);
Gebreslassie et al. (2012); Gonela et al. (2015a); Paolucci et al. (2016); Roni et al.
(2016); Hombach et al. (2016); Ng and Maravelias (2016); Balaman and Selim (2016);
Santibañez-Aguilar et al. (2016); Yue et al. (2016); De Meyer et al. (2016)
Short rotation woody crops Walther et al. (2012); Tatsiopoulos and Tolis (2003); Rentizelas et al. (2009a); Huang
et al. (2010); Avami (2012); Rentizelas and Tatsiopoulos (2010)
First generation biodiesel, heat Oily crops Andersen et al. (2012); Bowling et al. (2011); Leão et al. (2011); Foo et al. (2013);
and power Rentizelas and Tatsiopoulos (2010); Santibañez-Aguilar et al. (2016); Azadeh and
Arani (2016)
Second generation biodiesel, Energy crops Andersen et al. (2012); Leduc et al. (2009); Srisuwan and Dumrongsiri (2012); Parker
heat and power et al. (2010); Ren et al. (2013); Wang et al. (2013); Tong et al. (2013); Marvin et al.
(2012); Yue et al. (2013); Tong et al. (2014b); Babazadeh et al. (2015); Babazadeh et al.
(2016); Santibañez-Aguilar et al. (2016)
Third generation biodiesel, algae Gong and You (2014); Ahn et al. (2015); Mohseni et al. (2016)
heat and power
Not determined biomass Mol et al. (1997); Tembo et al. (2003); Bai et al. (2011); Tittmann et al. (2010);
Marufuzzaman et al. (2014b); Bruglieri and Liberti (2006); Venema and Calamai
(2003); Azadeh et al. (2014); An and Wilhelm (2014); Giarola et al. (2014)
ﬁrst generation biofuels, due to their foodstuff consumption, is the classiﬁed in three main categories including mathematical pro-
main reason for the increasing interest of researchers on studying gramming, multi-criteria decision making, as well as heuristic and
the second-generation biofuels. meta-heuristic approaches. It is worth noting that these classes
may have overlaps; for instance, a paper with mathematical pro-
4.2. Modeling approaches gramming approach can be a multi-criteria decision making model
too. Fig. 7 shows the distribution of different modeling approaches
In this section, different modeling approaches used in BSCND used in BSCND in different years. According to these results, the
are evaluated and analyzed in an integrated framework. After eval- mathematical programming, with 92.5% usage, is the most applied
uating the selected papers, the used modeling approaches are approach in BSCND models. Again, in spite of the fact that the real
984 H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000
Multi-product Single-product
16
14
14
12
12
10 10
10
8 8 8
8 7 7 7 7
6 6 6
6
5
3 4 3 3
4
2 2
2 1 1 1 1 1
0 0 0 1 0
0 0 0 0
0
35
1
30
25 1 10
1 2 6
20 0
4 3
4
1 0 4
15 2 2 3
2
2
2
10 1 2 0
1
1 1 17
3
15 15 10 15
1
5 7 9
9
1 5
01 011 0 01 1
0 1 01 10 1 2 01 1 2 1 2 2 1 1
0 0 0 0
world problems are often complicated and cannot be modeled by 4.2.1.1. Linear programming. Linear programming (LP) is a speciﬁc
linear methods, with 88.1% usage, linear approach is the dominat- type of mathematical programming methods wherein the objective
ing approach in BSCND models. Notably, only 11.8% of the published function and the constraints are both linear. It is worth mentioning
papers propose nonlinear models for BSCND. Therefore, it is sug- that linear models constitute 7.4% of the mathematical and 6.8%
gested that the modeling approaches that conform better to reality of all types of modeling approaches in BSCND. Table 5 shows the
be used in the future studies. Next, each modeling approach will be evaluation of the papers wherein linear programming models are
evaluated and analyzed based on different features such as objec- developed. All the linear BSCND models are single-objective and
tive function type (see Table 4), the number of objective functions, ignore the issue of sustainability.
the number of time periods, sustainability, uncertainty, uncertain Facility location optimization highly reduces the transporta-
factor, and solution method. tion costs in the BSC. As a powerful tool, Geographical Information
Systems (GIS) can help to determine the appropriate location of
the facilities in a speciﬁc area through integrating the informa-
4.2.1. Mathematical programming approach tion of different factors (population and the main and secondary
Mathematical programming models include an objective func- routes of the supply chain). Panichelli and Gnansounou (2008)
tion, constraints, and decision variables. They are applied to obtain presented a GIS-based approach that is able to determine the opti-
the optimum value for the objective function and decision vari- mal location of the processing sites and the optimum allocation
ables while satisfying all the constraints (Winston and Goldberg, of biomass based on marginal delivery costs and resource compe-
2004). Considering the objective function, decision variables, and tition between facilities when there is a high price-variability of
constraints, the mathematical programming models used in the biomass. In this approach, the biomass allocation is done through
selected papers include linear, nonlinear, mixed integer linear, and BIOAL wherein use is made of Dijkstra algorithm to ﬁnd the shortest
mixed integer nonlinear programming models. As shown in Fig. 7, route between the harvesting and processing sites.
the mixed integer linear programming (MILP) models are the most Frombo et al. (2009a) presented a GIS-based Environmental
widely used one in the literature. Decision Support System (EDSS) for an optimal logistics strategic
planning to produce energy from woody biomass. The proposed
H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000 985
Table 4 tion of strategic variables such as the location and capacity of the
Objective function coding and classiﬁcation.
bioreﬁneries with the objective of reducing the total supply chain
Objective function class Code Objective function costs. Morrow et al. (2006) proposed a linear programming model
Economic 1 Minimize (total) costs to determine the optimal locations of the bioreﬁneries in a supply
2 Maximize (total) proﬁt. chain that produces bioethanol from switchgrass and corn in the
3 Maximize net present value United States.
4 Maximize ﬁnancial revenue Ren et al. (2013) proposed a model for the design of sustain-
5 Minimize risk on investment.
able bioethanol supply chain wherein multiple biomass feedstocks,
6 Minimize transportation cost
8 Minimize Logistics cost of production technologies, transportation modes and patterns, and
biomass collection disposal methods are all possible. There are certain constraints in
15 Minimize transport distance this model that affect the objective function. Notably, the objective
16 Minimize unit cost
function aims to minimize the total ecological footprint.
17 Maximize economic potential
18 Minimize conditional Uncertainty is an important challenge in the development of
value-at-risk BSCs (Awudu and Zhang, 2012). Considering uncertainty in the
19 Maximize efﬁciencies level of biomass production due to varying climatic conditions in
20 Minimize marginal delivery different periods of growth and harvesting, Cundiff et al. (1997) pro-
cost
posed a scenario-based multi-period linear programming model
Environmental 7 Minimize greenhouse gas to determine the capacity of the collection sites in a herbaceous
emissions
biomass delivery system. Sharma et al. (2013b) also proposed a
9 Maximize the total greenhouse
gas emission savings
scenario-based multi-period linear programming model to opti-
11 Maximize net energy output mize the design of a biofuel supply chain under uncertain climatic
12 Minimize environmental conditions.
impact
13 Minimize global warming
potential
4.2.1.2. Mixed integer linear programming. Similar to LP models, the
MILP models have linear objective functions and constraints with
Social 10 Maximize the number job
this difference that in the latter, all or some of the decision vari-
opportunities
14 Minimize social impact
ables are integers. MILP models are useful tools for the solution
21 Minimize number of workers of problems wherein integer variables are necessary because of
22 Maximize total service level the presence of discrete phenomena. Out of 146 papers reviewed,
109 papers (74.6%) had made use of MILP models. Therefore, MILP
is a modeling technique used most widely in BSCND problems.
EDSS has three main levels including GIS, database, and optimiza- Table 6 shows the evaluation results of MILP models used in
tion (at strategic, tactical, and operational levels). This system is BSCND problems from different aspects. Facility location decision
able to optimize the decision variables of the bioreﬁnery capacity is an important reason for using MILP models since the related
and the annual biomass harvesting based on minimization of the binary decision variables can appropriately model establishment
total costs. of facilities in different capacities and production technologies.
Since harvesting and processing sites are far apart and biomass Since BSCND affects biomass logistics, decision makers determine
is scattered in a vast area, its procurement is costly (Möller and optimal BSCND and optimum material ﬂow between facilities
Nielsen, 2007). Different algorithms (e.g. Dijkstra) are proposed simultaneously (Vlachos et al., 2008). Accordingly, the continuous
in the literature to determine the shortest distance between the variables of MILP are used to show the material ﬂow between facil-
biomass harvesting and processing sites. Using Dijkstra-based ities (Bowling et al., 2011; De Meyer et al., 2014; Vlachos et al.,
GIS and focusing on the transportation and logistics strategies, 2008). Optimization, in these models, is done either under eco-
Perpina et al. (2009) developed a methodology capable of locat- nomic, environmental, or social objectives (De Meyer et al., 2014),
ing a network of processing sites around a region. Tatsiopoulos and or in some cases, the economic, environmental, and social aspects
Tolis (2003) developed a multi-period linear programming model are optimized simultaneously (You et al., 2012; Yue et al., 2014).
wherein they investigated the possibility of generating heat and MILP models have been used in different problems. Mol et al.
electricity from chopped cotton-plant stalks through the optimiza- (1997) proposed an MILP model for BSCND wherein optimum loca-
Table 5
Publications applying linear programming with identiﬁcation of different characteristics.
Ref. Objective Multi period Multi Objective Sustainability Uncertainty Uncertain Factor Solution method
Table 6
Publications applying mixed integer linear programming with identiﬁcation of different characteristics.
Ref. Objective Multi period Multi objective Sustainability Uncertainty Non-deterministic parameters Solution method
Table 6 (Continued)
Ref. Objective Multi period Multi objective Sustainability Uncertainty Non-deterministic parameters Solution method
tions of the processing sites are determined through minimizing trical energy from biomass at the regional level (Freppaz et al.,
the logistic costs of collecting biomass. Nagel (2000) proposed an 2004); design and scheduling of supply chain processes to generate
MILP model that is able to ﬁnd the most economic and ecological heat from biomass (Dunnett et al., 2007); design and evaluation of
structure for supplying biomass through the dynamic evaluation waste biomass sustainable supply chains (Vlachos et al., 2008); the
of the economic performance and efﬁciency. MILP models with strategic planning of bioethanol supply chain systems (Huang et al.,
the objective function of cost minimization are used in different 2010); optimal design of bioethanol supply chains together with
problems including the evaluation of generating heat and elec- determining the optimum harvesting and bioethanol production
988 H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000
rates (Akgul et al., 2010); and BSC strategic planning and optimum in the BSCND models (Awudu and Zhang, 2012); however, there
allocation of feedstock under uncertain location decisions (Chen are other sources of uncertainty which can be considered in BSCND
and Fan, 2012). models. In this regard, Dal-Mas et al. (2011) proposed a stochastic
As mentioned in the previous section, GIS, as a powerful tool, dynamic MILP model for the design and planning of multi-echelon
can be used in facility location and BSCND. Geijzendorffer et al. biofuel supply chains under market uncertainty. Giarola et al.
(2008) combined GIS and MILP to calculate the expected value of (2012) also proposed a multi-echelon, multi-period MILP model
the biomass supply and the transportation distance. In addition to for the design of a bioethanol supply chain considering biomass
facility location, GIS can be used, considering different factors, in and carbon market uncertainties.
selecting the optimal route for the feedstock ﬂow between facili- Giarola et al. (2013) considered market uncertainties in a
ties. To achieve this, ﬁrstly the most inexpensive routes between risk constrained multi-objective stochastic MILP model in order
different facilities are found through the GIS (Fernando et al., 2015; to determine the strategic planning decisions and design of a
Frombo et al., 2009b; Parker et al., 2010; Tittmann et al., 2010), and bioethanol supply chain. proposed a two-stage, mixed integer
then the results are used in the MILP model for BSCND. stochastic programming model to ﬁnd the optimal design of BSC in
Various objectives are used in BSCND models. Table 4 shows the southeastern region of the United States wherein the biomass
different economic, environmental, and social objective functions supply, market demand, market price, and the production technol-
used in BSCND models. Although all of them affect BSCND deci- ogy vary in different scenarios. In this model, the production sites’
sions, most MILP models have considered only the economic aspect. strategic variables (location and capacity) are optimized in the ﬁrst
In the meantime, some authors have focused on the environmen- stage and, in the second stage, the optimum values of biomass feed-
tal optimization. For example, Lam et al. (2010) presented a new stock and biofuel ﬂow rates are determined. In the mixed integer
method for regional energy clustering and Foo et al. (2013) pro- stochastic programming model presented by Chen and Fan (2012),
posed some robust models for the synthesis of empty fruit bunch the strategic planning and feedstock allocation decisions of bioen-
allocation networks in the regional BSC to generate energy from ergy supply chain systems are associated with uncertainties in the
palm oil. The only objective in both models is the minimization biomass supply and bioethanol demand. Kim et al. (2011b) pro-
of greenhouse gas emissions. Economic and environmental objec- posed a scenario-based robust MILP model for the biofuel supply
tives usually are modeled by minimizing the total supply chain chain design problem under varying demand conditions that could
costs and greenhouse gas emissions in most of MILP models (You specify the optimum value of the feedstock ﬂow between differ-
and Wang, 2011, 2012; Zamboni et al., 2009a,b). The maximization ent supply chain facilities as well as the optimum values of the
of net present value and minimization of environmental impacts bioreﬁnery-related strategic variables (location and capacity). In
(Giarola et al., 2013; Mele et al., 2011) are also considered as the order to design a second generation synthetic biodiesel supply
economic and environmental objectives in MILP models. A small chain, Walther et al. (2012) proposed a multi-period, scenario-
number of papers consider all the three aspects of sustainability in based MILP model which determines the optimal values of the
BSCND problems. You et al. (2012) and Yue et al. (2014) proposed location, capacity, and production technology decisions.
multi-objective MILP models which consider cost minimization, Considering uncertainties in the biomass supply and biofuel
greenhouse gas emission minimization, and job opportunity max- demand as the functions of price, Azadeh et al. (2014) pro-
imization as objective functions. Results of the present study show posed a dynamic, stochastic MILP model for the multi-product
that among the MILP models, the 83 (76%) single-objective and BSCND wherein both the market and disruption risks are taken
26 (24%) multi-objective models are developed in the literature, into account. Considering uncertainties in supply and demand,
respectively. Gebreslassie et al. (2012) also proposed a multi-period, stochas-
In the real world problems, sometimes all or a number of param- tic MILP model for the optimal design of a hydrocarbon bioreﬁnery
eters undergo some changes over time according to a foreseeable supply chain that is able to simultaneously minimize annual costs
trend. Therefore, it is necessary to consider such dynamism in and ﬁnancial risks. Design and planning of the hydrocarbon biofuel
problem parameters since it can signiﬁcantly affect the planning supply chain has been studied by Tong et al. (2014b) too. Kostin
decisions. In such cases, the time horizon is divided into sev- et al. (2012) carried out the strategic design and planning of the
eral periods, and a multi-period model is used to conform to the sugar and ethanol supply chain by presenting a stochastic MILP
dynamic changes (Melo et al., 2009). This means that some deci- model under demand uncertainty. Tong et al. (2013) proposed a
sions related to the location and capacity of the facilities as well multi-period, stochastic MILP model to cope with uncertainty and
as the biomass and biofuel ﬂows are to be made for each period. dynamism of biomass availability, biofuel price, crude oil demand,
The investigation results on MILP models show that out of 109 and production technology. Advanced biofuel supply chains mini-
papers, 68 (62.38%) have developed multi-period BSCND models. mize the biomass transportation costs, and they have the advantage
The binary variables in a multi-period MILP model in BSCND show of economies to scale for the bio-oil gasiﬁcation facilities. In this
either a facility is open or closed and in most of the cases are inde- regard, Li and Hu (2014) proposed a two-stage stochastic MILP
pendent from the time periods (De Meyer et al., 2014). Giarola et al. based on bio-oil gasiﬁcation wherein they assumed such factors
(2011a) have presented a multi-objective multi-period model for as biomass availability, technology advancement, and biofuel price
the design and planning of a bioethanol supply chain considering as uncertain parameters.
both ﬁrst and second generation bioreﬁneries. The capacity and A comprehensive review of the literature shows that uncer-
technology of facilities are strategic decisions that can vary in dif- tainty has been considered in some of BSCND models by the
ferent time periods. Mele et al. (2011) presented a multi-period authors; however, most of them have employed stochastic pro-
MILP model for the design of a synthetic sugar-bioethanol supply gramming to deal with uncertainty. Notably, in many real-life
chain which considers both economic and environmental objec- problems, the uncertainty could not be modeled by probability
tives. In this model, the number and capacity of collection sites and distributions and the use of stochastic programming is con-
bioreﬁneries are the main decisions which should be determined fronted with signiﬁcant difﬁculties according to lack of knowledge
for each time period. and historical data about uncertain parameters. To overcome
Uncertainty is an important factor which can signiﬁcantly these obstacles, a number of studies have been done which
inﬂuence the performance of BSC. Uncertainties in the biomass have employed fuzzy mathematical programming approach. Tong
feedstock supply, biofuel demand, bioenergy production, price, et al. (2014a) proposed a multi-period MILP model for designing
logistics, and transportation are common instances of uncertainty an advanced hydrocarbon biofuel supply chain integrated with
H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000 989
existing petroleum reﬁneries. In order to deal with uncertain area modeling approach to evaluate the possibility of building new
parameters, they presented a possibilistic programming approach biomass power plants and determining their optimal locations in a
in which possibility, necessity and credibility measures are applied supply chain located in Guangdong, China. Grigoroudis et al. (2014)
based on decision makers’ preferences. Bairamzadeh et al. (2015) proposed a recursive DEA model wherein facility selection is done
developed a multi-objective MILP model for the design of a sus- based on the minimum cost and maximum efﬁciency.
tainable bioethanol supply chain under multiple uncertainties. In MODM models, several objectives are simultaneously taken
Moreover, they developed a novel multi-objective robust possi- into account within a mathematical programming model. The eval-
bilistic programming approach to account for the data uncertainty uation criterion for an objective might be quite different from those
of the problem. of the others and it is also possible that the objectives may be
conﬂicting. In BSCND models, these objectives include the com-
4.2.1.3. Non-linear programming models. Non-linear programming bination of two economic objectives (Gebreslassie et al., 2012;
(NLP) models are special types of mathematical programming mod- Mas et al., 2010; Tong et al., 2014b), or the combination of eco-
els that have non-linear constraints and/or objective functions. nomic and environmental objectives (You and Wang, 2011, 2012;
Depending on the presence or absence of integer variables, the Giarola et al., 2011a,b, 2013; Rozakis et al., 2001; Akgul et al.,
model is either of the mixed integer non-linear programming 2014; Yue et al., 2013, 2016; Mele et al., 2011; Zamboni et al.,
(MINLP) type, or simply of the non-linear programming type. Out 2009a; Wang et al., 2013; Gong and You, 2014; Marufuzzaman et al.,
of 146 papers, 14 papers (9.59%) proposed non-linear program- 2014a; Murillo-Alvarado et al., 2015; Paolucci et al., 2016; Cambero
ming models including 13MINLP and 1 NLP models. Since real et al., 2016; Santibañez-Aguilar et al., 2016; d’Amore and Bezzo,
world problems are complicated, linear modeling is not applicable 2016; Babazadeh et al., 2016), or the combination of economic and
in many cases. On the other hand, the complexities of non-linear social objectives (Balaman and Selim, 2016), or the combination of
models and the simplicity of the linear ones encourage researchers economic, environmental, and social objectives (You et al., 2012;
to develop the latter in different problems. Ayoub et al., 2007; Yue et al., 2014; Čuček et al., 2012; Bairamzadeh
As shown in Table 7, most non-linear models (Bai et al., 2011, et al., 2015; Roni et al., 2016; Miret et al., 2016; Cambero and
2012, 2015, 2016; Singh et al., 2011, 2014; Marufuzzaman et al., Sowlati, 2016). These results show that in MCDM models developed
2014a; Yue and You, 2014; Corsano et al., 2011; Bruglieri and for BSCND, the researchers have focused mainly on the economic
Liberti, 2006; Poudel et al., 2016) have only economic objective and environmental aspects and only a little part of the literature
functions. Some other research works such as (Wang et al., 2013; covers all the aspects of sustainability (only 8 papers out of 35,
Yue et al., 2013; Akgul et al., 2014; Gong and You, 2014; Babazadeh 22.86%). An investigation of MCDM models reveals that 20 papers
et al., 2016) consider the environmental and economic optimiza- (57.14%) considered multiple time periods and 8 papers (22.86%)
tion simultaneously. The work by (Čuček et al., 2012) is the only used stochastic programming models to deal with uncertainty.
non-linear model which considers all the three sustainability pillars
including economic, environmental, and social objective simulta- 4.2.3. Heuristic and meta-heuristic approaches
neously. Among the MINLP and NLP models (Singh et al., 2014; In general, optimization methods and algorithms are either
Babazadeh et al., 2016) consider multiple time periods in their mod- exact or approximate; the former that includes commercial solvers
els. In order to handle the data uncertainty, Babazadeh et al. (2016) and exact solution algorithms is capable of ﬁnding exact optimum
developed a multi-objective possibilistic programming model to solutions, but it is incapable of dealing with complex optimiza-
design a second generation biodiesel supply chain under risk. In tion problems (i.e., NP-hard and NP-complete problems) because
this paper, by deﬁning the possibilistic mean and absolute deviation the computational time increases exponentially. Approximation
of fuzzy numbers, a new formulation of possibilistic programming methods that include heuristic and meta-heuristic algorithms are
method is developed. capable of ﬁnding near optimum solutions for complex problems in
a short period of time. The main problem with heuristic algorithms
4.2.2. Multi-criteria decision making is that they may stick in local optimums and cannot be used in var-
Multi-Criteria Decision Making (MCDM) models usually deal ious type of problems. Meta-heuristic algorithms are proposed to
with conﬂicting objectives or attributes in a decision making prob- solve this problem; in fact, they are a type of approximate optimiza-
lem. These models are either Multi-Attribute Decision Making tion methods that have ways out of local optimums and can be used
(MADM) or Multi-Objective Decision Making (MODM). In the for- in a wide range of problems. Our review study on BSCND shows
mer, the selection of one from among the existing choices is that 9 papers with meta-heuristic approach including 6 Genetic
considered. In general, MADM includes such speciﬁc decisions (of Algorithms (GA), 2 Particle Swarm Optimization (PSO) algorithms,
preferential type) as evaluation, prioritization, or selection from and 1 Binary honey bee foraging (BHBF) algorithm are published
among the existing choices under conﬂicting criteria (Malczewski, up to now. Notably, all the three algorithms are population-based
1999; Wang et al., 2009). Table 8 shows the evaluation of MCDM which are able to provide a set of solutions at each iteration. Table 9
models used for BSCND. As an unparalleled and dynamic technol- shows the results of our review on heuristic and meta-heuristic
ogy, and with the objective of optimal management of spatial data, approaches used in BSCND literature. As shown in Table 9, GA is the
GIS is being widely used by the users of different sciences and most popular meta-heuristic algorithm in BSCND problems. GA is a
technologies. Combining GIS capabilities (in data collection, stor- speciﬁc type of evolution algorithm that makes use of such natural
age, retrieval, and analysis) with those of MCDM, can affect the evolution techniques as heredity and mutation to ﬁnd approximate
conversion of geographical data and decision makers’ preferences solutions for optimization and search problems (Mitchell, 1998).
to unidimensional values of different alternatives and enhance Venema and Calamai (2003) used GA in a P-median problem that
their ranking (Malczewski, 1999). In this regard, Ma et al. (2005) is actually a location-routing problem, for the determination of the
proposed a GIS model wherein the evaluation of the land appropri- optimum location of biomass supply sites and bioreﬁneries.
ateness and the determination of optimal locations for the potential Singh et al. (2014) presented an MINLP model with black-box
bioenergy production sites are done according to social and envi- functions for designing bioethanol supply chain under corn com-
ronmental constraints and economic factors. In this model, the petitive markets. To solve the proposed optimization problem, they
analytic hierarchy process is used to determine the factors’ weights developed a GA. Ayoub et al. (2007) integrated data mining tech-
in the ﬁrst stage and selecting the most appropriate locations in niques and GA in order to determine the conversion and storing
the second stage. Using GIS, Shi et al. (2008) proposed a supply capacities in such a way that the costs of transportation, carbon
990 H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000
Table 7
Publications applying non-linear programming approaches with identiﬁcation of different characteristics.
Ref. Type Objective Multi-Period Multi-Objective Sustainability Uncertainty Non-deterministic parameters Solution method
Table 8
Publications applying multi-criteria decision making approaches with identiﬁcation of different characteristics.
Ref. Type Objective Multi-Period Sustainability Uncertainty Non-deterministic parameters Solution method
dioxide emissions, and the number of workers were minimized. ber and capacity of production facilities in a third generation BSC
Celli et al. (2008) proposed a solution method which combines GA with multiple biomass sources. In this method, ﬁrst, many near
and GIS to ﬁnd the optimum number and locations of hybrid heat- global optimal solutions are found using GA, and then the obtained
power plants in Sardinia, Italy. Integrating GA and SQP, Rentizelas solution is used as the starting point for the SQP convergent model
et al. (2009a) and Rentizelas and Tatsiopoulos (2010) proposed a to ﬁnd the closest global optimal solution. This method decreases
decision support system that could determine the optimum num-
H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000 991
Table 9
Publications applying heuristic and meta-heurisitc approaches with identiﬁcation of different characteristics.
Multi-period Single-period behavior of honey bees are used to ﬁnd the optimal solutions. BHBF
18 16 that has binary variables for location decisions, is the adjusted form
16 of HBF. Vera et al. (2010) proposed a BHBF algorithm that deter-
14 12
mines the location, capacity, and supply area of biomass power
12 11
10 10
plants with the highest proﬁtability for the investors.
10 9 9
8 9
8 7 7
5 6 5 4.3. Solution methodologies
6
4
4
1 1 1 2 1 1
2
2 1 1 1
Solution of BSCND problems can be achieved through different
2 0 0
0
0 1 0 1 solution methods. In this section, the methods used by researchers
0 0 0
for this purpose are classiﬁed in “exact solution algorithms”,
“commercial solvers”, and “heuristic, meta-heuristic, and approxi-
mation algorithms”. The ﬁrst group includes “branch-and-bound”,
Fig. 9. Single-period and multi-period models in different years.
“branch-and-cut”, and “decomposition algorithms”. Branch-and-
bound is the most widely used algorithm among the exact solution
the possibility of ﬁnding a local optimal instead of a global solution algorithms which, in some cases, is integrated with heuristic or
which usually happens in the SQP method. Lagrangian relaxation methods to determine the bounds. Decom-
PSO which was ﬁrstly introduced by Eberchart and Kennedy position methods are also among useful methods in supply chain
(1995), imitates the social behavior of organisms such as the move- location problems (Mirchandani and Francis, 1990). They are highly
ments of the ﬂocks of birds or schools of ﬁsh. It is a computational applied in BSCND problems because the variables in network design
method wherein the problem optimization is done through the problems can be decomposed into binary and continuous variables
improvement of candidate solutions considering the pre-speciﬁed (Melo et al., 2009). Nevertheless, decomposition methods has been
criteria. Using this method, it is possible to deal with the prob- used in few studies such as (Marufuzzaman et al., 2014b), (Roni
lems, the solutions of which are either a single point or an area in et al., 2014) and (Chen and Fan, 2012). Notably the multiple layers in
an n-dimensional space (López et al., 2008). Izquierdo et al. (2008) BSCs increase the number of strategic decisions, and consequently
used the PSO algorithm to do the strategic planning of a BSC. This the decomposition into different sub-problems makes the solution
optimization problem is in the form of a non-linear model that has method more difﬁcult and complex. In these cases, use is usually
both binary and continuous variables; the latter are related to the made of heuristic, metaheuristic, and approximation algorithms
strategic-level decisions, and the former, to the production tech- (GA, PSO, BHBF, SAA, LA, CA) that obtain near optimal solutions
nology selection. López et al. (2008) proposed a binary PSO-based in shorter time durations.
approach that is used to determine the optimal location of biomass Some other researchers use commercial solvers such as Lingo,
power plants in a supply area. GAMS, or CPLEX to ﬁnd the exact solutions. The distribution of
The HBF algorithm which was ﬁrstly introduced by Pham et al. BSCND problem solution methodologies is shown in Fig. 8. As
(2005), is a population-based algorithm wherein the food foraging shown in Fig. 8, commercial solvers are the most widely used solu-
992 H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000
Multi-objective Single-objective
16
14 14 14
14
12 11 11 11
10
10 9 10
8
8
6 5
4
4 3 3 3 3
2 2 2 2 2
2 1 1 1
0 0 0 0 0 0
0 0 0 0 0 0
0 0
tion methodology used in BSCND problems; they have been used 4.4.2. Objectives
in the solution of 100% LP models, 92.6% of MILP models, 62.5% of Single-objective optimization models obtain only one opti-
MINLP/NLP models, and 91.4% of MCDM models. Fig. 8 also shows mal solution while multi-objective ones consider various and
that exact solution algorithms have been used in the solution of 4.6% conﬂicting objectives and yield a set of Pareto solutions (Deb,
MILP models, 12.5% of MINLP/NLP models, and 2.8% of MCDM mod- 2014). In multi-objective BSCND models, the objectives are usu-
els. Therefore, commercial solvers and exact solution algorithms ally either economic-environmental or economic-environmental-
are used as a solution method in 81.5% of the models developed social. Fig. 10 shows the application trend of single and
in the area of BSCND. In the remaining 18.5%, use has been made multi-objective models in different years, and indicates a share of
of heuristic, meta-heuristic, and approximation algorithms to ﬁnd 77.4% for single-objective models versus 22.6% for multi-objective
approximate optimal solutions. These methods have been used for ones. In addition to the economic aspects, the real world BSC oper-
the solution of 2.7% of MILP models, 25% of MINLP/NLP models, and ation affects the environmental aspects (due to the use of land,
5.7% of MCDM models. water, forest, and greenhouse gas emissions because of the use of
There is always a disagreement among researchers interested biofuel) and the social aspects (due to the creation/elimination of
in exact solution methods and those who prefer heuristic, meta- job opportunities). Therefore, the little share of the multi-objective
heuristic, and approximation algorithms because, in many cases, models in BSCND problems is a proof of the existence of a gap in
the use of both is possible. For instance, in many complicated, the literature, and this is why the researchers are expected to con-
sizeable, real world, and non-linear problems, the use of heuris- centrate more on the development of multi-objective models in the
tic, meta-heuristic, and approximation algorithms is unavoidable; future studies.
however, in these methods, the amount of closeness of the obtained
solution and the optimal one is not measurable, and the opti-
mality is not guaranteed. Therefore, there is a big gap in the
solution methodologies that should be attended to in the future 4.5. Uncertainty
studies. Making use of hybrid methods, using speciﬁc solutions
for speciﬁc problems and, also, improving the commercial solvers, Uncertainty has been always a serious challenge in BSCND prob-
exact solution algorithms as well as heuristic, meta-heuristic, and lems (Subrahmanyam et al., 1994). Uncertainty in a BSCND problem
approximation algorithms are among the research directions which includes changes in biomass supply, biofuel demand, and their
can be useful in the future researches. prices due to seasonal ﬂuctuations, geographical conditions, unsta-
ble economic conditions, population growth, and other unforeseen
events that may happen in the supply chain. Uncertainty may affect
4.4. Model characteristics the feasibility and efﬁciency of a BSC and may lead to infeasibility
or sub-optimality of BSCND solution. Therefore, taking into account
4.4.1. Period the uncertain nature of parameters and developing an appropri-
Time is an effective factor in the planning of the BSC and always ate non-deterministic model are the requirements of an optimal
affects the tactical and operational planning decisions. In addition, design of BSC (Gebreslassie et al., 2012). Different approaches
it affects the biomass supply and biofuel demand in long-term used to tackle uncertainty in BSCND models include stochastic
strategies and causes BSCND decisions to vary. Therefore, dividing programming (e.g. scenario-based, two-stage, probabilistic, and
the time horizon into several time periods leads to improvement chance-constrained) and possibilistic programming. The distribu-
in decision making process (De Meyer et al., 2014). Fig. 9 shows tion of different deterministic and non-deterministic approaches
the trend of the use of single and multi-period models in differ- in the selected papers is shown in Fig. 11. Fig. 12 shows the num-
ent years. As shown in Fig. 9, despite a relative increase in the use ber of papers used deterministic and non-deterministic models in
of multi-period models in recent years, the share of single-period different years. The 21.2% share of non-deterministic models is an
models in BSCND problems is approximately equal to that of multi- indication of lack of sufﬁcient research in this area. The analysis of
period models (49.3% for single-period and 50.7% for multi-period). the uncertain parameters is the complementary step in this sec-
Considering the dynamic nature of some parameters such as the tion. Fig. 13 shows the results of such analysis; as shown in Fig. 13,
biomass supply and biofuel demand in different periods, and their the biomass supply and biofuel demand parameters are the most
effects on the supply chain strategic decisions, it is expected that important uncertain parameters in BSCND models.
more efforts will be given to development of multi-period models.
H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000 993
22
18
12
4
3 3
2 2
Table 10
Entities, data type and region of the reviewed papers.
Table 10 (Continued)
SS: supply site; CS: collection site; PPS: preprocessing site; PS: processing site; IPS: intermediate processing site; BS: blending site; DS: distribution site; DC: demand center.
996 H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000
140
88.4%
120
100
80
129
60
40
11.6%
20
17
0
Case study Numerical example
Fig. 15. Distribution of the publications based on the use of case study and numerical
example.
60
52
50
40
30
20 16
10 5 5 5 6 6
3 3 4
1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2
0
Fig. 16. Distribution of the publications based on the regions of the case studies.
been mostly orientated towards single-feedstock, single-product, uncertainty may include biomass and biofuel price, feedstock cost
single-period, single-objective, and deterministic models without and production technologies.
considering all the dimensions of sustainability. Practically, a sustainable supply chain needs to integrate consid-
This study shows that the most of BSCND papers apply eration of economic, environmental, and social objectives (Amigun
mathematical programming approaches. Among the mathemat- et al., 2011; Duku et al., 2011; Treleven and Schweikhart, 1988).
ical programming approaches, MILP is the high usage approach The results of our studies indicate that only eight papers pro-
applied to optimize binary variables that determine whether a facil- posed multi-objective models in which the above-mentioned three
ity is open at a certain place and continuous variables, which are aspects are considered. Therefore, it can be concluded that the
related to the biomass or bioenergy ﬂows between facilities. How- absence of sustainability is strongly felt in BSCND models. Lack of
ever, real world problems are always complicated; therefore, they sustainability in BSCND models results in making decisions which
cannot be modeled using simple MILP approaches. Indeed, the com- are risky or will not beneﬁt from the opportunities that higher levels
plexities of non-linear models and the simplicity of the linear ones of sustainable BSCND models put forward. Accordingly, developing
encourage researchers to develop MILP models for BSCND prob- models that combine economic, environmental and social sus-
lems. Using non-linear approaches compatible with real-world tainability concepts is a future direction for researchers who are
problems, instead of simple linear methods, and more ﬂexible con- interested in sustainable BSCND.
vex optimization methods and solution methodologies could help In terms of analyzing strategic decisions (facility-related deci-
researchers to deal with real world problems. sion), it is shown that a majority of papers have focused mostly on
Solution methodology in BSCND problems is a challenging issue the processing sites; however, the optimization of BSCND models
according to computational complexity of such problems. In many is determined by interrelationship and interdependence between
complicated, sizeable, real world, and non-linear problems, the locations of all facilities and their capacities. This means that more
use of heuristic, meta-heuristic and approximation algorithms is efforts are required to propose integrated and holistic BSCND mod-
unavoidable; however, in these methods, the amount of closeness els that give enough emphasis to all facilities in the entire supply
of the obtained solution and the optimal solution is not measurable chain.
in most of the cases. Developing exact solution algorithms such as According to the results of this study, multi-objective and multi-
decomposition–based algorithms or hybrid solution methods are period approaches are sadly lacking in BSCND models. Time affects
among the most attractive future research directions. the biomass supply and biofuel demand in long-term strategies and
Regarding some qualitative and quantitative analyses, causes BSCND decisions to vary. Considering the dynamism and
researchers may consider the parameters of their models as changes in such parameters and variables and their impacts on the
deterministic or non-deterministic. The little share of non- supply chain strategic decisions is highly needed to be followed
deterministic models is an indication of unwillingness due to in future research efforts. Moreover, the little share of the multi-
the computational complications in their applications. Different objective models in BSCND models is a proof of the existence of
approaches utilized to tackle uncertainty in BSCND models include gap in the literature, and this is why the researchers are expected
stochastic (scenario-based, two-stage, probabilistic, and chance- to concentrate more on the development of multi-objective models
constrained) and possibilistic programming. This study indicates in the future studies.
that researchers almost considered stochastic approaches to Finally, biofuel production is limited to some speciﬁc countries
handle uncertainties. However, fuzzy logic, robust optimization, in which the production process is based on the type of biomass
interval approaches, and chaos theory are useful approaches feedstock available in the country and other economic and political
which can be utilized to cope with data uncertainty. Regarding factors. Conducting case studies of BSCND in various countries with
a stochastic way of dealing with uncertainties, the results of different climates can be also regarded as another future research
analyses indicates that researchers should consider multi-stage opportunity.
stochastic approaches and robust optimization techniques instead
of traditional stochastic programming models as future oppor-
tunities of research. In terms of uncertain parameters, supply of References
biomass and demand of biofuel are the most important parameters
that are addressed as uncertain parameters in most of related Ahn, Y.-C., Lee, I.-B., Lee, K.-H., Han, J.-H., 2015. Strategic planning design of
microalgae biomass-to-biodiesel supply chain network: multi-period
papers. However, these are not the only imprecise parameters
deterministic model. Appl. Energy 154, 528–542.
in BSCND problems; some other important parameters under Akgul, O., Mac Dowell, N., Papageorgiou, L., Shah, N., 2014. A mixed integer
nonlinear programming (MINLP) supply chain optimisation framework for
998 H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000
carbon negative electricity generation using biomass to energy with CCS Celli, G., Ghiani, E., Loddo, M., Pilo, F., Pani, S., 2008. Optimal location of biogas and
(BECCS) in the UK. Int. J. Greenh. Gas Control 28, 189–202. biomass generation plants. In: Universities Power Engineering Conference,
Akgul, O., Shah, N., Papageorgiou, L., 2011. An MILP Model for the Strategic Design 2008, UPEC 2008. 43rd International, pp. 1–6 (IEEE).
of the UK Bioethanol Supply Chain. Chen, C.-W., Fan, Y., 2012. Bioethanol supply chain system planning under supply
Akgul, O., Shah, N., Papageorgiou, L.A., 2012. A spatially-explicit, multi-Period MILP and demand uncertainties. Trans. Res. Part E 48, 150–164.
modelling framework for the optimal design of a hybrid biofuel supply chain. Chen, X., Önal, H., 2014. An economic analysis of the future US biofuel industry
Symposium on Computer Aided Process Engineering, 20. facility location, and supply chain network. Trans. Sci. 48, 575–591.
Akgul, O., Zamboni, A., Bezzo, F., Shah, N., Papageorgiou, L.G., 2010. Chinese, D., Meneghetti, A., 2009. Design of forest biofuel supply chains. Int. J.
Optimization-based approaches for bioethanol supply chains. Ind. Eng. Chem. Logist. Syst. Manage. 5, 525–550.
Res. 50, 4927–4938. Corsano, G., Vecchietti, A.R., Montagna, J.M., 2011. Optimal design for sustainable
Aksoy, B., Cullinan, H., Webster, D., Gue, K., Sukumaran, S., Eden, M., Sammons, N., bioethanol supply chain considering detailed plant performance model.
2011. Woody biomass and mill waste utilization opportunities in Alabama: Comput. Chem. Eng. 35, 1384–1398.
transportation cost minimization optimum facility location, economic Čuček, L., Lam, H.L., Klemeš, J.J., Varbanov, P.S., Kravanja, Z., 2010. Synthesis of
feasibility, and impact. Environ. Prog. Sustain. Energy 30, 720–732. regional networks for the supply of energy and bioproducts. Clean Technol.
Alex Marvin, W., Schmidt, L.D., Benjaafar, S., Tiffany, D.G., Daoutidis, P., 2012. Environ. Policy 12, 635–645.
Economic optimization of a lignocellulosic biomass-to-ethanol supply chain. Čuček, L., Varbanov, P.S., Klemeš, J.J., Kravanja, Z., 2012. Total footprints-based
Chem. Eng. Sci. 67, 68–79. multi-criteria optimisation of regional biomass energy supply chains. Energy
Allen, J., Browne, M., Hunter, A., Boyd, J., Palmer, H., 1998. Logistics management 44, 135–145.
and costs of biomass fuel supply. Int. J. Phys. Distrib. Logist. Manage. 28, Cundiff, J.S., Dias, N., Sherali, H.D., 1997. A linear programming approach for
463–477. designing a herbaceous biomass delivery system. Bioresour. Technol. 59,
Amigun, B., Musango, J.K., Stafford, W., 2011. Biofuels and sustainability in Africa. 47–55.
Renew. Sustain. Energy Rev. 15, 1360–1372. D’amore, F., Bezzo, F., 2016. Strategic optimisation of biomass-based energy supply
An, H., Wilhelm, W.E., 2014. An exact solution approach based on column chains for sustainable mobility. Comput. Chem. Eng. 87, 68–81.
generation and a partial-objective constraint to design a cellulosic biofuel Dal-Mas, M., Giarola, S., Zamboni, A., Bezzo, F., 2011. Strategic design and
supply chain. Comput. Chem. Eng. 71, 11–23. investment capacity planning of the ethanol supply chain under price
An, H., Wilhelm, W.E., Searcy, S.W., 2011. A mathematical model to design a uncertainty. Biomass Bioenergy 35, 2059–2071.
lignocellulosic biofuel supply chain system with a case study based on a region De Meyer, A., Cattrysse, D., Rasinmäki, J., Van Orshoven, J., 2014. Methods to
in Central Texas. Bioresour. Technol. 102, 7860–7870. optimise the design and management of biomass-for-bioenergy supply chains:
Andersen, F., Iturmendi, F., Espinosa, S., Diaz, M.S., 2012. Optimal design and a review. Renew. Sustain. Energy Rev. 31, 657–670.
planning of biodiesel supply chain with land competition. Comput. Chem. Eng. De Meyer, A., Cattrysse, D., Van Orshoven, J., 2016. Considering biomass growth
47, 170–182. and regeneration in the optimisation of biomass supply chains. Renew. Energ.
Avami, A., 2012. A model for biodiesel supply chain: a case study in Iran. Renew. 87, 990–1002.
Sustain. Energy Rev. 16, 4196–4203. Deb, K., 2014. Multi-objective optimization. In: Search Methodologies. Springer.
Awudu, I., Zhang, J., 2012. Uncertainties and sustainability concepts in biofuel Driedonks, B.A., Gevers, J.M., Van Weele, A.J., 2010. Managing sourcing team
supply chain management: a review. Renew. Sustain. Energy Rev. 16, effectiveness: the need for a team perspective in purchasing organizations. J.
1359–1368. Purch. Supply Manage. 16, 109–117.
Ayoub, N., Martins, R., Wang, K., Seki, H., Naka, Y., 2007. Two levels decision system Duarte, A., Sarache, W., Costa, Y., 2016. Biofuel supply chain design from Coffee Cut
for efﬁcient planning and implementation of bioenergy production. Energy Stem under environmental analysis. Energy 100, 321–331.
Convers. Manage. 48, 709–723. Duku, M.H., Gu, S., Hagan, E.B., 2011. A comprehensive review of biomass resources
Azadeh, A., Arani, H.V., 2016. Biodiesel supply chain optimization via a hybrid and biofuels potential in Ghana. Renew. Sustain. Energy Rev. 15, 404–415.
system dynamics-mathematical programming approach. Renew. Energ. 93, Dunnett, A., Adjiman, C., Shah, N., 2007. Biomass to heat supply chains:
383–403. applications of process optimization. Process Saf. Environ. Prot. 85, 419–429.
Azadeh, A., Arani, H.V., Dashti, H., 2014. A stochastic programming approach Dunnett, A.J., Adjiman, C.S., Shah, N., 2008. A spatially explicit whole-system model
towards optimization of biofuel supply chain. Energy 76, 513–525. of the lignocellulosic bioethanol supply chain: an assessment of decentralised
Babazadeh, R., Razmi, J., Pishvaee, M.S., Rabbani, M., 2016. A sustainable processing potential. Biotechnol. Biofuels 1, 1–17.
second-generation biodiesel supply chain network design problem under risk. Ebadian, M., Sowlati, T., Sokhansanj, S., Townley-Smith, L., Stumborg, M., 2013.
Omega, in press. Modeling and analysing storage systems in agricultural biomass supply chain
Babazadeh, R., Razmi, J., Rabbani, M., Pishvaee, M.S., 2015. An integrated data for cellulosic ethanol production. Appl. Energy 102, 840–849.
envelopment analysis–mathematical programming approach to strategic Eberchart, R., Kennedy, J., 1995. Particle swarm optimization. In: Proceedings of
biodiesel supply chain network design problem. J. Clean. Prod. the IEEE International Conference on Neural Networks, Perth, Australia, pp.
Bai, Y., Hwang, T., Kang, S., Ouyang, Y., 2011. Biofuel reﬁnery location and supply 1942–1948.
chain planning under trafﬁc congestion. Trans. Res. Part B 45, 162–175. Ekşioğlu, S.D., Acharya, A., Leightley, L.E., Arora, S., 2009. Analyzing the design and
Bai, Y., Li, X., Peng, F., Wang, X., Ouyang, Y., 2015. Effects of disruption risks on management of biomass-to-bioreﬁnery supply chain. Comput. Ind. Eng. 57,
bioreﬁnery location design. Energies 8, 1468–1486. 1342–1352.
Bai, Y., Ouyang, Y., Pang, J.-S., 2012. Biofuel supply chain design under competitive Ekşioğlu, S.D., Li, S., Zhang, S., Sokhansanj, S., Petrolia, D., 2010. Analyzing impact of
agricultural land use and feedstock market equilibrium. Energy Econ. 34, intermodal facilities on design and management of biofuel supply chain. Trans.
1623–1633. Res. Rec. 2191, 144–151.
Bai, Y., Ouyang, Y., Pang, J.-S., 2016. Enhanced models and improved solution for Elia, J.A., Baliban, R.C., Floudas, C.A., Gurau, B., Weingarten, M.B., Klotz, S.D., 2013.
competitive biofuel supply chain design under land use constraints. Eur. J. Hardwood biomass to gasoline diesel, and jet fuel: 2. Supply chain
Operat. Res. 249, 281–297. optimization framework for a network of thermochemical reﬁneries. Energy
Bairamzadeh, S., Pishvaee, M.S., Saidi-Mehrabad, M., 2015. Multiobjective robust Fuels 27, 4325–4352.
possibilistic programming approach to sustainable bioethanol supply chain Fernando, A.L., Duarte, M.P., Vatsanidou, A., Alexopoulou, E., 2015. Environmental
design under multiple uncertainties. Ind. Eng. Chem. Res. 55, 237–256. aspects of ﬁber crops cultivation and use. Ind. Crops Prod. 68, 105–115.
Balaman, Ş.Y., Selim, H., 2014. A network design model for biomass to energy Foo, D.C., Tan, R.R., Lam, H.L., Aziz, M.K.A., Klemeš, J.J., 2013. Robust models for the
supply chains with anaerobic digestion systems. Appl. Energy 130, 289–304. synthesis of ﬂexible palm oil-based regional bioenergy supply chain. Energy
Balaman, S.Y., Selim, H., 2016. Sustainable design of renewable energy supply 55, 68–73.
chains integrated with district heating systems: a fuzzy optimization Freppaz, D., Minciardi, R., Robba, M., Rovatti, M., Sacile, R., Taramasso, A., 2004.
approach. J. Clean. Prod. Optimizing forest biomass exploitation for energy supply at a regional level.
Banerjee, S., Mudliar, S., Sen, R., Giri, B., Satpute, D., Chakrabarti, T., Pandey, R., Biomass Bioenergy 26, 15–25.
2010. Commercializing lignocellulosic bioethanol: technology bottlenecks and Frombo, F., Minciardi, R., Robba, M., Rosso, F., Sacile, R., 2009a. Planning woody
possible remedies. Biofuels Bioprod. Bior. 4, 77–93. biomass logistics for energy production: a strategic decision model. Biomass
Bowling, I.M., Ponce-Ortega, J.M., El-Halwagi, M.M., 2011. Facility location and Bioenergy 33, 372–383.
supply chain optimization for a bioreﬁnery. Ind. Eng. Chem. Res. 50, Frombo, F., Minciardi, R., Robba, M., Sacile, R., 2009b. A decision support system for
6276–6286. planning biomass-based energy production. Energy 34, 362–369.
Bruglieri, M., Liberti, L., 2006. Optimal design of a biomass-based energy Gebreslassie, B.H., Yao, Y., You, F., 2012. Design under uncertainty of hydrocarbon
production process. In: Operation Research Models and Methods in the Energy bioreﬁnery supply chains: multiobjective stochastic programming models
Sector, Coimbra, Portugal. decomposition algorithm, and a comparison between CVaR and downside risk.
Cambero, C., Sowlati, T., 2016. Incorporating social beneﬁts in multi-objective AIChE J. 58, 2155–2179.
optimization of forest-based bioenergy and biofuel supply chains. Appl. Energy Geijzendorffer, I., Annevelink, E., Elbersen, B., Smidt, R., De Mol, R., 2008.
178, 721–735. Application of a GIS-BIOLOCO tool for the design and assessment of biomass
Cambero, C., Sowlati, T., Marinescu, M., Röser, D., 2015. Strategic optimization of delivery chains. Proceedings of 16th European Biomass Conference &
forest residues to bioenergy and biofuel supply chain. Int. J. Energy Res. 39, Exhibition.
439–452. Giarola, S., Bezzo, F., Shah, N., 2013. A risk management approach to the economic
Cambero, C., Sowlati, T., Pavel, M., 2016. Economic and life cycle environmental and environmental strategic design of ethanol supply chains. Biomass
optimization of forest-based bioreﬁnery supply chains for bioenergy and Bioenergy 58, 31–51.
biofuel production. Chem. Eng. Res. Des. 107, 218–235.
H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000 999
Giarola, S., Patel, M., Shah, N., 2014. RETRACTED: biomass supply chain Mapemba, L.D., Epplin, F.M., Taliaferro, C.M., Huhnke, R.L., 2007. Bioreﬁnery
optimisation for Organosolv-based bioreﬁneries. Bioresour. Technol. 159, feedstock production on conservation reserve program land. Appl. Econ.
387–396. Perspect. Policy 29, 227–246.
Giarola, S., Shah, N., Bezzo, F., 2012. A comprehensive approach to the design of Marufuzzaman, M., Ekşioğlu, S.D., Hernandez, R., 2014a. Environmentally friendly
ethanol supply chains including carbon trading effects. Bioresour. Technol. supply chain planning and design for biodiesel production via wastewater
107, 175–185. sludge. Trans. Sci. 48, 555–574.
Giarola, S., Zamboni, A., Bezzo, F., 2011a. Spatially explicit multi-objective Marufuzzaman, M., Eksioglu, S.D., Li, X., Wang, J., 2014b. Analyzing the impact of
optimisation for design and planning of hybrid ﬁrst and second generation intermodal-related risk to the design and management of biofuel supply chain.
bioreﬁneries. Comput. Chem. Eng. 35, 1782–1797. Trans. Res. Part E 69, 122–145.
Giarola, S., Zamboni, A., Bezzo, F., 2011b. Supply chain design and capacity Marvin, W.A., Schmidt, L.D., Daoutidis, P., 2012. Bioreﬁnery location and
planning: from ﬁrst to second generation biofuel systems. Chem. Eng. Trans. technology selection through supply chain optimization. Ind. Eng. Chem. Res.
24, 253–258. 52, 3192–3208.
Gonela, V., Zhang, J., Osmani, A., 2015a. Stochastic optimization of sustainable Mas, M.D., Giarola, S., Zamboni, A., Bezzo, F., 2010. Capacity planning and ﬁnancial
industrial symbiosis based hybrid generation bioethanol supply chains. optimization of the bioethanol supply chain under price uncertainty. Comput.
Comput. Ind. Eng. 87, 40–65. Aided Chem. Engine 28, 97–102.
Gonela, V., Zhang, J., Osmani, A., Onyeaghala, R., 2015b. Stochastic optimization of Mayring, P., 2011. Qualitative Inhaltsanalyse.
sustainable hybrid generation bioethanol supply chains. Trans. Res. Part E 77, Mclean, K., Li, X., 2013. Robust scenario formulations for strategic supply chain
1–28. optimization under uncertainty. Ind. Eng. Chem. Res. 52, 5721–5734.
Gong, J., You, F., 2014. Global optimization for sustainable design and synthesis of Mele, F.D., Kostin, A.M., Guillén-Gosálbez, G., Jiménez, L., 2011. Multiobjective
algae processing network for CO2 mitigation and biofuel production using life model for more sustainable fuel supply chains: a case study of the sugar cane
cycle optimization. AIChE J. 60, 3195–3210. industry in Argentina. Ind. Eng. Chem. Res. 50, 4939–4958.
Govindan, K., Soleimani, H., Kannan, D., 2015. Reverse logistics and closed-loop Melo, M.T., Nickel, S., Saldanha-Da-Gama, F., 2009. Facility location and supply
supply chain: a comprehensive review to explore the future. Eur. J. Oper. Res. chain management–a review. Eur. J. Oper. Res. 196, 401–412.
240, 603–626. Merriam, S.B., 1988. Case Study Research in Education: A Qualitative Approach.
Grigoroudis, E., Petridis, K., Arabatzis, G., 2014. RDEA: a recursive DEA based Jossey-Bass.
algorithm for the optimal design of biomass supply chain networks. Renew. Mirchandani, P.B., Francis, R.L., 1990. Discrete Location Theory.
Energy 71, 113–122. Miret, C., Chazara, P., Montastruc, L., Negny, S., Domenech, S., 2016. Design of
Gunnarsson, H., Rönnqvist, M., Lundgren, J.T., 2004. Supply chain modelling of bioethanol green supply chain: comparison between ﬁrst and second
forest fuel. Eur. J. Oper. Res. 158, 103–123. generation biomass concerning economic: environmental and social criteria.
Hombach, L.E., Cambero, C., Sowlati, T., Walther, G., 2016. Optimal design of supply Comput. Chem. Eng. 85, 16–35.
chains for second generation biofuels incorporating European biofuel Mitchell, M., 1998. An Introduction to Genetic Algorithms. MIT press.
regulations. J. Clean. Prod. Mohseni, S., Pishvaee, M.S., Sahebi, H., 2016. Robust design and planning of
Huang, Y., Chen, C.-W., Fan, Y., 2010. Multistage optimization of the supply chains microalgae biomass-to-biodiesel supply chain: a case study in Iran. Energy
of biofuels. Trans. Res. Part E 46, 820–830. 111, 736–755.
Izquierdo, J., Minciardi, R., Montalvo, I., Robba, M., Tavera, M., 2008. Particle Swarm Mol, R., Jogems, M., Van Beek, P., Gigler, J., 1997. Simulation and optimization of
Optimization for the biomass supply chain strategic planning. Proceedings of the logistics of biomass fuel collection. NJAS Wagen. J. Life Sci. 45, 217–228.
the International Congress on Environmental Modelling and Software, Möller, B., Nielsen, P.S., 2007. Analysing transport costs of Danish forest wood chip
1272–1280. resources by means of continuous cost surfaces. Biomass Bioenergy 31,
Kanzian, C., Holzleitner, F., Stampfer, K., Ashton, S., 2009. Regional energy wood 291–298.
logistics–optimizing local fuel supply. Silva Fennica 43, 113–128. Morrow, W.R., Grifﬁn, W.M., Matthews, H.S., 2006. Modeling switchgrass derived
Kim, J., Realff, M., Lee, J., 2010. Simultaneous design and operation decisions for cellulosic ethanol distribution in the United States. Environ. Sci. Technol. 40,
bioreﬁnery supply chain networks: centralised vs. distibuted system. In: 2877–2886.
Proceedings of the 9th International Symposium on Dynamics and Control of Mula, J., Peidro, D., Díaz-Madroñero, M., Vicens, E., 2010. Mathematical
Process Systems, IFAC Laxenburg, Austria, pp. 73–78. programming models for supply chain production and transport planning. Eur.
Kim, J., Realff, M.J., Lee, J.H., 2011a. Optimal design and global sensitivity analysis J. Oper. Res. 204, 377–390.
of biomass supply chain networks for biofuels under uncertainty. Comput. Murillo-Alvarado, P.E., Guillén-Gosálbez, G., Ponce-Ortega, J.M., Castro-Montoya,
Chem. Eng. 35, 1738–1751. A.J., Serna-González, M., Jiménez, L., 2015. Multi-objective optimization of the
Kim, J., Realff, M.J., Lee, J.H., Whittaker, C., Furtner, L., 2011b. Design of biomass supply chain of biofuels from residues of the tequila industry in Mexico. J.
processing network for biofuel production using an MILP model. Biomass Clean. Prod. 108, 422–441.
Bioenergy 35, 853–871. Nagel, J., 2000. Determination of an economic energy supply structure based on
Kostin, A., Guillén-Gosálbez, G., Mele, F., Bagajewicz, M., Jiménez, L., 2012. Design biomass using a mixed-integer linear optimization model. Ecol. Eng. 16,
and planning of infrastructures for bioethanol and sugar production under 91–102.
demand uncertainty. Chem. Eng. Res. Des. 90, 359–376. Natarajan, K., Leduc, S., Pelkonen, P., Tomppo, E., Dotzauer, E., 2012. Optimal
Lam, H.L., Klemeš, J.J., Kravanja, Z., 2011. Model-size reduction techniques for locations for methanol and CHP production in Eastern Finland. Bioenergy Res.
large-scale biomass production and supply networks. Energy 36, 4599–4608. 5, 412–423.
Lam, H.L., Varbanov, P., Klemeš, J., 2010. Minimising carbon footprint of regional Natarajan, K., Leduc, S., Pelkonen, P., Tomppo, E., Dotzauer, E., 2014. Optimal
biomass supply chains. Resour. Conserv. Recycl. 54, 303–309. locations for second generation Fischer Tropsch biodiesel production in
Leão, R.R.D.C.C., Hamacher, S., Oliveira, F., 2011. Optimization of biodiesel supply Finland. Renew. Energy 62, 319–330.
chains based on small farmers: a case study in Brazil. Bioresour. Technol. 102, Ng, R.T., Maravelias, C.T., 2016. Design of biofuel supply chains with variable
8958–8963. regional depot and bioreﬁnery locations. Renew. Energy.
Leduc, S., Natarajan, K., Dotzauer, E., Mccallum, I., Obersteiner, M., 2009. Osmani, A., Zhang, J., 2013. Stochastic optimization of a multi-feedstock
Optimizing biodiesel production in India. Appl. Energy 86, S125–S131. lignocellulosic-based bioethanol supply chain under multiple uncertainties.
Leduc, S., Schwab, D., Dotzauer, E., Schmid, E., Obersteiner, M., 2008. Optimal Energy 59, 157–172.
location of wood gasiﬁcation plants for methanol production with heat Palak, G., Ekşioğlu, S.D., Geunes, J., 2014. Analyzing the impacts of carbon
recovery. Int. J. Energy Res. 32, 1080–1091. regulatory mechanisms on supplier and mode selection decisions: an
Leduc, S., Starfelt, F., Dotzauer, E., Kindermann, G., Mccallum, I., Obersteiner, M., application to a biofuel supply chain. Int. J. Prod. Econ. 154, 198–216.
Lundgren, J., 2010. Optimal location of lignocellulosic ethanol reﬁneries with Panichelli, L., Gnansounou, E., 2008. GIS-based approach for deﬁning bioenergy
polygeneration in Sweden. Energy 35, 2709–2716. facilities location: a case study in Northern Spain based on marginal delivery
Li, Q., Hu, G., 2014. Supply chain design under uncertainty for advanced biofuel costs and resources competition between facilities. Biomass Bioenergy 32,
production based on bio-oil gasiﬁcation. Energy 74, 576–584. 289–300.
Lim, C.H., Lam, H.L., 2016. Biomass supply chain optimisation via novel biomass Paolucci, N., Bezzo, F., Tugnoli, A., 2016. A two-tier approach to the optimization of
element life cycle analysis (BELCA). Appl. Energy 161, 733–745. a biomass supply chain for pyrolysis processes. Biomass Bioenergy 84, 87–97.
López, P.R., Galán, S.G., Reyes, N.R., Jurado, F., 2008. A method for particle swarm Parker, N., Tittmann, P., Hart, Q., Nelson, R., Skog, K., Schmidt, A., Gray, E., Jenkins,
optimization and its application in location of biomass power plants. Int. J. B., 2010. Development of a bioreﬁnery optimized biofuel supply curve for the
Green Energy 5, 199–211. Western United States. Biomass Bioenergy 34, 1597–1607.
Ma, J., Scott, N.R., Degloria, S.D., Lembo, A.J., 2005. Siting analysis of farm-based Paulo, H., Azcue, X., Barbosa-Póvoa, A.P., Relvas, S., 2015. Supply chain
centralized anaerobic digester systems for distributed generation using GIS. optimization of residual forestry biomass for bioenergy production: the case
Biomass Bioenergy 28, 591–600. study of Portugal. Biomass Bioenergy 83, 245–256.
Malczewski, J., 1999. GIS and Multicriteria Decision Analysis. John Wiley & Sons. Perpina, C., Alfonso, D., Pérez-Navarro, A., Penalvo, E., Vargas, C., Cárdenas, R., 2009.
Mansoornejad, B., Pistikopoulos, E.N., Stuart, P.R., 2013. Scenario-based strategic Methodology based on Geographic Information Systems for biomass logistics
supply chain design and analysis for the forest bioreﬁnery using an operational and transport optimisation. Renew. Energy 34, 555–565.
supply chain model. Int. J. Prod. Econ. 144, 618–634. Pham, D., Ghanbarzadeh, A., Koc, E., Otri, S., Rahim, S., Zaidi, M., 2005. The Bees
Mapemba, L.D., Epplin, F.M., Huhnke, R.L., Taliaferro, C.M., 2008. Herbaceous plant Algorithm Technical Note. Manufacturing Engineering Centre, Cardiff
biomass harvest and delivery cost with harvest segmented by month and University, UK, pp. 1–57.
number of harvest machines endogenously determined. Biomass Bioenergy 32,
1016–1027.
1000 H. Ghaderi et al. / Industrial Crops and Products 94 (2016) 972–1000
Poudel, S.R., Marufuzzaman, M., Bian, L., 2016. Designing a reliable bio-fuel supply Venema, H.D., Calamai, P.H., 2003. Bioenergy systems planning using
chain network considering link failure probabilities. Comput. Ind. Eng. 91, location–allocation and landscape ecology design principles. Ann. Oper. Res.
85–99. 123, 241–264.
Ren, J., Manzardo, A., Toniolo, S., Scipioni, A., Tan, S., Dong, L., Gao, S., 2013. Design Vera, D., Carabias, J., Jurado, F., Ruiz-Reyes, N., 2010. A Honey Bee Foraging
and modeling of sustainable bioethanol supply chain by minimizing the total approach for optimal location of a biomass power plant. Appl. Energy 87,
ecological footprint in life cycle perspective. Bioresour. Technol. 146, 771–774. 2119–2127.
Ren, J., Dong, L., Sun, L., Goodsite, M.E., Tan, S., Dong, L., 2015. Life cycle cost Vlachos, D., Iakovou, E., Karagiannidis, A., Toka, A., 2008. A strategic supply chain
optimization of biofuel supply chains under uncertainties based on interval management model for waste biomass networks. 3rd International Conference
linear programming. Bioresour. Technol. 187, 6–13. on Manufacturing Engineering, 797–804.
Rentizelas, A., Tatsiopoulos, I., Tolis, A., 2009a. An optimization model for Walther, G., Schatka, A., Spengler, T.S., 2012. Design of regional production
multi-biomass tri-generation energy supply. Biomass Bioenergy 33, 223–233. networks for second generation synthetic bio-fuel? A case study in Northern
Rentizelas, A.A., Tatsiopoulos, I.P., 2010. Locating a bioenergy facility using a hybrid Germany. Eur. J. Oper. Res. 218, 280–292.
optimization method. Int. J. Product. Econ. 123, 196–209. Wang, B., Gebreslassie, B.H., You, F., 2013. Sustainable design and synthesis of
Rentizelas, A.A., Tolis, A.J., Tatsiopoulos, I.P., 2009b. Logistics issues of biomass: the hydrocarbon bioreﬁnery via gasiﬁcation pathway: integrated life cycle
storage problem and the multi-biomass supply chain. Renew. Sustain. Energy assessment and technoeconomic analysis with multiobjective superstructure
Rev. 13, 887–894. optimization. Comput. Chem. Eng. 52, 55–76.
Richard, T.L., 2010. Challenges in scaling up biofuels infrastructure. Science Wang, J.-J., Jing, Y.-Y., Zhang, C.-F., Zhao, J.-H., 2009. Review on multi-criteria
(Washington) 329, 793–796. decision analysis aid in sustainable energy decision-making. Renew. Sustain.
Roni, M.S., Eksioglu, S.D., Cafferty, K.G., Jacobson, J.J., 2016. A multi-objective, Energy Rev. 13, 2263–2278.
hub-and-spoke model to design and manage biofuel supply chains. Ann. Oper. Wang, S., Hastings, A., Smith, P., 2012. An optimization model for energy crop
Res., 1–30. supply. GCB Bioenergy 4, 88–95.
Roni, M.S., Eksioglu, S.D., Searcy, E., Jha, K., 2014. A supply chain network design Winston, W.L., Goldberg, J.B., 2004. Operations Research: Applications and
model for biomass co-ﬁring in coal-ﬁred power plants. Trans. Res. Part E 61, Algorithms. Duxbury press, Boston.
115–134. Woo, Y.-B., Cho, S., Kim, J., Kim, B.S., 2016. Optimization-based approach for
Rozakis, S., Soldatos, P.G., Kallivroussis, L., Nicolaou, I., 2001. Multiple criteria strategic design and operation of a biomass-to-hydrogen supply chain. Int. J.
analysis of bio-energy projects: evaluation of bio-electricity production in Hydrogen Energy 41, 5405–5418.
Farsala Plain: greece. J. Geogr. Inf. Decis. Anal. 5, 49–64. Xie, F., Huang, Y., Eksioglu, S., 2014. Integrating multimodal transport into
Sabri, E.H., Beamon, B.M., 2000. A multi-objective approach to simultaneous cellulosic biofuel supply chain design under feedstock seasonality with a case
strategic and operational planning in supply chain design. Omega 28, 581–598. study based on California. Bioresour. Technol. 152, 15–23.
Santibañez-Aguilar, J.E., Morales-Rodriguez, R., González-Campos, J.B., You, F., Tao, L., Graziano, D.J., Snyder, S.W., 2012. Optimal design of sustainable
Ponce-Ortega, J.M., 2016. Stochastic design of bioreﬁnery supply chains cellulosic biofuel supply chains: multiobjective optimization coupled with life
considering economic and environmental objectives. J. Clean. Prod. cycle assessment and input–output analysis. AIChE J. 58, 1157–1180.
Sathre, R., 2014. Comparing the heat of combustion of fossil fuels to the heat You, F., Wang, B., 2011. Life cycle optimization of biomass-to-liquid supply chains
accumulated by their lifecycle greenhouse gases. Fuel 115, 674–677. with distributed–centralized processing networks. Ind. Eng. Chem. Res. 50,
Sharifzadeh, M., Garcia, M.C., Shah, N., 2015. Supply chain network design and 10102–10127.
operation: systematic decision-making for centralized distributed, and mobile You, F., Wang, B., 2012. Multiobjective Optimization of Biomass-to-Liquids
biofuel production using mixed integer linear programming (MILP) under Processing Networks. Foundations of Computer-Aided Process Operations,
uncertainty. Biomass Bioenergy 81, 401–414. Savannah, Georgia.
Sharma, B., Ingalls, R., Jones, C., Khanchi, A., 2013a. Biomass supply chain design Yue, D., Kim, M.A., You, F., 2013. Design of sustainable product systems and supply
and analysis: basis overview, modeling, challenges, and future. Renew. Sustain. chains with life cycle optimization based on functional unit: general modeling
Energy Rev. 24, 608–627. framework, mixed-integer nonlinear programming algorithms and case study
Sharma, B., Ingalls, R.G., Jones, C.L., Huhnke, R.L., Khanchi, A., 2013b. Scenario on hydrocarbon biofuels. ACS Sustain. Chem. Eng. 1, 1003–1014.
optimization modeling approach for design and management of Yue, D., Pandya, S., You, F., 2016. Integrating hybrid life cycle assessment with
biomass-to-bioreﬁnery supply chain system. Bioresour. Technol. 150, 163–171. multiobjective optimization: a modeling framework. Environ. Sci. Technol. 50,
Shi, X., Elmore, A., Li, X., Gorence, N.J., Jin, H., Zhang, X., Wang, F., 2008. Using 1501–1509.
spatial information technologies to select sites for biomass power plants: a Yue, D., Slivinsky, M., Sumpter, J., You, F., 2014. Sustainable design and operation of
case study in Guangdong Province, China. Biomass Bioenergy 32, 35–43. cellulosic bioelectricity supply chain networks with life cycle economic
Singh, A., Chu, Y., You, F., 2014. Bioreﬁnery supply chain network design under environmental, and social optimization. Ind. Eng. Chem. Res. 53, 4008–4029.
competitive feedstock markets: an agent-Based simulation and optimization Yue, D., You, F., 2014. Game-theoretic modeling and optimization of multi-echelon
approach. Ind. Eng. Chem. Res. 53, 15111–15126. supply chain design and operation under Stackelberg game and market
Singh, J., Panesar, B., Sharma, S., 2011. Geographical distribution of agricultural equilibrium. Comput. Chem. Eng.
residues and optimum sites of biomass based power plant in Bathinda, Punjab. Zamboni, A., Bezzo, F., Shah, N., 2009a. Spatially explicit static model for the
Biomass Bioenergy 35, 4455–4460. strategic design of future bioethanol production systems. 2. Multi-objective
Srisuwan, P., Dumrongsiri, A., 2012. Mathematical model of production and environmental optimization. Energy Fuels 23, 5134–5143.
logistics planning for crops producing E-20 biofuel. In: Second International Zamboni, A., Shah, N., Bezzo, F., 2009b. Spatially explicit static model for the
Conference on Economics, Trade and Development, Bangkok, Thailand, pp. strategic design of future bioethanol production systems. 1. Cost minimization.
17–22. Energy Fuels 23, 5121–5133.
Subrahmanyam, S., Pekny, J.F., Reklaitis, G.V., 1994. Design of batch chemical Zhang, F., Johnson, D.M., Wang, J., 2016. Integrating multimodal transport into
plants under market uncertainty. Ind. Eng. Chem. Res. 33, 2688–2701. forest-delivered biofuel supply chain design. Renew. Energy 93, 58–67.
Tatsiopoulos, I., Tolis, A., 2003. Economic aspects of the cotton-stalk biomass Zhang, J., Osmani, A., Awudu, I., Gonela, V., 2013. An integrated optimization model
logistics and comparison of supply chain methods. Biomass Bioenergy 24, for switchgrass-based bioethanol supply chain. Appl. Energy 102, 1205–1217.
199–214. Zhang, L., Hu, G., 2013. Supply chain design and operational planning models for
Tembo, G., Epplin, F.M., Huhnke, R.L., 2003. Integrative investment appraisal of a biomass to drop-in fuel production. Biomass Bioenergy 58, 238–250.
lignocellulosic biomass-to-ethanol industry. J. Agric. Resour. Econ., 611–633. Zhu, X., Li, X., Yao, Q., Chen, Y., 2011. Challenges and models in supporting logistics
Tittmann, P., Parker, N., Hart, Q., Jenkins, B., 2010. A spatially explicit system design for dedicated-biomass-based bioenergy industry. Bioresour.
techno-economic model of bioenergy and biofuels production in California. J. Technol. 102, 1344–1351.
Trans. Geogr. 18, 715–728. Zhu, X., Yao, Q., 2011. Logistics system design for biomass-to-bioenergy industry
Tong, K., Gleeson, M.J., Rong, G., You, F., 2014a. Optimal design of advanced drop-in with multiple types of feedstocks. Bioresour. Technol. 102, 10936–10945.
hydrocarbon biofuel supply chain integrating with existing petroleum
reﬁneries under uncertainty. Biomass Bioenergy 60, 108–120.
Tong, K., Gong, J., Yue, D., You, F., 2013. Stochastic programming approach to Web references
optimal design and operations of integrated hydrocarbon biofuel and
petroleum supply chains. ACS Sustain. Chem. Engine 2, 49–61. RFA. World Fuel Ethanol Production. Available from: <http://
Tong, K., You, F., Rong, G., 2014b. Robust design and operations of hydrocarbon
biofuel supply chain integrating with existing petroleum reﬁneries considering www.ethanolrfa.org/resources/industry/statistics/world/>, 2014
unit cost objective. Comput. Chem. Eng. [cited: 01.03. 15].
Treleven, M., Schweikhart, S.B., 1988. A risk/beneﬁt analysis of sourcing strategies: Statista. The world’s biggest biodiesel producers in 2014, by
single vs. multiple sourcing. J. Oper. Manage. 7, 93–114.
Tursun, U.D., Kang, S., Onal, H., Ouyang, Y., Scheffran, J., 2008. Optimal bioreﬁnery
country. Available from: <http://www.statista.com/statistics/
locations and transportation network for the future biofuels industry in 271472/biodiesel-production-in-selected-countries/>, 2014
Illinois. In: Environmental and Rural Development Impacts Conference, [cited: 01.03. 15].
October 15–16, 2008, St. Louis, Missouri. Farm Foundation, Transition to a Bio
Economy Conferences.
Bien plus que des documents.
Découvrez tout ce que Scribd a à offrir, dont les livres et les livres audio des principaux éditeurs.
Annulez à tout moment.