+1 (208) 254-6996 [email protected]
  

Write a two-page paper, plus the title page and a reference page on the following statement:

Describe one direct and one indirect impact of reverse logistics operations on some local or regional area and compare it to some aspect of the world environment.

Don't use plagiarized sources. Get Your Custom Essay on
Number 3
Just from $13/Page
Order Essay

Incorporate at least one peer reference from articles listed within the online APUS library.  You must cite the peer reviewed reference within the paper.

•Written Communication: Written communication is free of errors that detract from the overall message. •APA formatting: Resources and citations are formatted according to APA style and formatting. •Length of paper: typed, double-spaced pages with no less than a two-page paper. •Font and font size: Times New Roman, 12 point.

lable at ScienceDirect

Journal of Cleaner Production 201 (2018) 1081e1091

Contents lists avai

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

A goal programming model for sustainable reverse logistics operations planning and an application

Alperen Bal a, Sule Itir Satoglu b, *

a Yalova University, Engineering Faculty, Industrial Engineering Department, Yalova, Turkey b Istanbul Technical University, Faculty of Management, Industrial Engineering Department, Istanbul, Turkey

a r t i c l e i n f o

Article history: Received 2 January 2018 Received in revised form 15 July 2018 Accepted 10 August 2018 Available online 13 August 2018

* Corresponding author. Istanbul Teknik Universite Macka, Sisli, Istanbul, Turkey.

E-mail address: [email protected] (S.I. Satoglu).

https://doi.org/10.1016/j.jclepro.2018.08.104 0959-6526/© 2018 Elsevier Ltd. All rights reserved.

a b s t r a c t

Global concerns about climate change and its environmental consequences, social factors and economic constraints require pursuit of a new approach to the supply chain planning at the strategic, tactical and operational levels. Recovery of waste electric and electronic equipment (WEEE) has become an important issue in the developing economies, as legislations that mandate manufacturers and importers to take back the wastes of their electrical and electronic products (WEEE) has been promulgated. Therefore, this study addresses the process of collecting WEEE products from service points, transporting them to recycling facilities, and recovery of the waste materials. Our framework considers triple-bottom-line approach and employs goal programming to reach economic, social and environmental targets. A multi-facility, multi-product and multi-period mathematical model is proposed, considering the real conditions, for the first time in the literature. In addition, this goal programming approach is illustrated on a WEEE reverse supply chain of the household appliances.

© 2018 Elsevier Ltd. All rights reserved.

1. Introduction

Global concerns about climate change and its environmental consequences, social factors and economic constraints require pursuit of a new approach to the supply chain planning at the strategic, tactical and operational levels. In this day and age, it is not enough to think only from economic perspective. Especially private companies aim cost minimization, but it is also necessary for them to consider the environmental protection and the social impact. At this point, governments consider to amplify social benefit and make legislations to reduce unfavorable environmental impact.

Sustainability is either used to sustain an implementation or to emphasize environmental awareness both in academic and non- academic resources. Although both definitions are correct, they are incomplete. The approach used in analyzing sustainability is described as triple bottom line (TBL) accounting. The TBL concept states that for a system to be sustainable, economic, environmental and social requirements must be reached at a minimum (Jeurissen, 2000). As Linton et al. (2007) expressed, to achieve a sustainable supply chain, each fragment of it should have environmentally

si, Isletme Fakultesi, 34369,

friendly procedures including product design, manufacturing, us- age, recycling, and transporting among suppliers, manufacturers, and customers.

Recovery of WEEE products has become an important issue in the developing economies. However, companies are reluctant or not capable of entering this market. Manufacturers, on the other hand, are under pressured according to the market trend and obliged to implement environmental regulations (Kumar and Putnam, 2008). A regulatory control of waste electric and elec- tronic equipment that mandates manufacturers and importers to take back their products has been promulgated (Ministry of Environment and Urbanization, 2012).

Based on the aforementioned considerations, this paper ad- dresses the issue of the collecting waste products from service points and transporting them to recycling facilities. Our framework considers triple-bottom-line approach and employs goal pro- gramming to reach economic, social and environmental targets. A goal-programming model has been developed in tactical opera- tions planning of the reverse supply chains with multi-facility, multi-product and multi-period. The proposed mathematical model can be used for transportation and recovery operations de- cision making with a TBL perspective and it can be extended to different types of reverse supply chains. In addition, we illustrate our approach on a WEEE reverse supply chain of the householdmailto:[email protected]http://crossmark.crossref.org/dialog/?doi=10.1016/j.jclepro.2018.08.104&domain=pdfwww.sciencedirect.com/science/journal/09596526http://www.elsevier.com/locate/jcleprohttps://doi.org/10.1016/j.jclepro.2018.08.104https://doi.org/10.1016/j.jclepro.2018.08.104https://doi.org/10.1016/j.jclepro.2018.08.104

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e10911082

appliances. The paper is structured as follows. Section 2 offers a sustain-

ability based literature review to assess the optimization papers in forward/reverse logistics network design. To further explain the TBL accounting, the proposed framework is illustrated and eco- nomic, environmental and social perspectives are discussed in section 3. To design reverse logistics network, a goal-programming model is developed in section 4. Section 5 explains Augmented ε-constraint methodology and section 6 presents the Case Study. Later, results and discussion on the case study are presented. Lastly, conclusion of the paper and further research are explained in sec- tion 8.

2. Literature review

A significant amount of sustainable supply chain research has been conducted considering various sustainability indicators related to TBL for managerial decision making in supply chain management (SCM) (Carter and Rogers, 2008) and operations management (Drake and Spinler, 2013; Kleindorfer et al., 2005), in particular. Compared to the extensive research on environmental aspects and especially economic issues, the social aspects are neglected in the sustainable SCM literature. Yura (1994) elaborated social issues, Brent et al. (2007) and Abreu and Camarinha-Matos (2008) studied on socio-economic issues and Clift (2003) detailed socio-environmental interfaces. Paksoy et al. (2010) proposed a closed loop supply chain design using multi-objective mixed- integer linear programming. The model minimizes cost and greenhouse gas emissions at the same time. Tseng and Hung (2014) considered both social costs caused by the carbon dioxide emis- sions and operational cost in an apparel manufacturing supply chain network. Transportation planning in reverse and closed-loop supply chain design is also discussed at the tactical level, and

Table 1 Reviewed Articles about sustainable supply chain design. (Notation e BD: benders decom GP: goal programming, MILP: mixed integer linear programming, MINLP: mixed integer n optimization, S: social, SMILP: stochastic mixed integer linear programming SO: stochas

Network structure

Anvari and Turkay, 2017 Forward Arampantzi and Minis, 2017 Forward Feit�o-Cesp�on et al., 2017 Reverse Safaei et al., 2017 Closed-loop Sarkar et al., 2017 Closed-loop Yu and Solvang, 2017 Reverse Demirel et al., 2016 Reverse Govindan et al., 2016a Closed-loop Govindan et al., 2016b Reverse Shaw et al., 2016 Closed-loop Ene and Oztürk, 2015 Reverse Zhou and Zhou, 2015 Reverse Hashemi et al., 2014 Closed-loop Ozceylan et al., 2014 Closed-loop Roghanian and Pazhoheshfar, 2014 Reverse Soleimani and Govindan, 2014 Reverse Amin and Zhang, 2013 Closed-loop Diabat et al., 2013 Closed-loop Ozceylan and Paksoy, 2013 Closed-loop Ramezani et al., 2013 Closed-loop Alumur et al., 2012 Reverse Das and Chowdhury, 2012 Reverse Kannan et al., 2012 Reverse Ozkır and Başligıl, 2012 Closed-loop Fonseca et al., 2010 Reverse Ramudhin et al., 2010 Forward Lee and Dong, 2009 Reverse Aras and Aksen, 2008 Reverse Demirel and Gokçen, 2008 Closed-loop Pati et al., 2008 Reverse

mathematical models are proposed (Dekker et al., 2013). A large number of multi-facility multi-product deterministic

facility location problems were studied in the literature. However, operational planning has attracted little attention. Also, sustain- ability approach requires multiple objectives to be achieved. Gonela et al. (2015) proposed a stochastic mixed integer linear program- ming model for bioethanol supply chain and evaluated the results under different sustainability concerns. Krumwiede and Sheu (2002) developed a reverse logistics decision-making model for third-party logistics providers to help engage in the reverse logis- tics business. Hung Lau and Wang (2009) investigated the feasi- bility of current reverse logistics theories and models for electronics industry taking into account developing countries like China. A mixed integer programming model is proposed by Shih (2001) for reverse logistics network design considering cost pa- rameters including sale revenue of reclaimed materials. Kara et al. (2007) calculated collection cost of waste appliances in reverse logistics network using discrete event simulation. Mutha and Pokharel (2009) designed a multi-echelon network including a consolidation warehouse into the system before they are sent to the reprocessing center for inspection or dismantling. Tuzkaya et al. (2011) proposed a two staged multi objective model for reverse the logistics network design problem and presented its application in the Turkish white appliances industry.

Bal and Satoglu (2017) used sustainability perspective as well as legal requirements to set up a goal-programming model to opti- mize a global white appliance manufacturers’ reverse logistics system. Coskun et al. (2016) proposed a goal-programming model to re-design green supply chain network considering three different customer segments. The results demonstrated that the increase in the number of green consumers expanded the tendency of the retailers to cooperate with the suppliers to redesign the supply chain, to fit the consumers’ expected greenness level. A

position, E: economic, En: environmental, FMoO: fuzzy multi objective optimization on-linear programming, MoSO: multi objective stochastic optimization, RO: Robust tic optimization).

Objective Modelling approach Case study

EEnS MILP Yes EEnS GP Yes EEnS MoSO Yes E MILP Yes EEn MINLP No EEn MoSO No E MILP Yes EEnS MILP Yes EEnS FMoO Yes En BD No E MILP No E MINLP Yes E MILP No E MINLP No E MILP No E SO No EEn MoSO No EEn MILP No E MILP No E MoSO No E MILP Yes E MILP No En MILP Yes E MILP No ES MoSO Yes EEn GP No E SMILP No E MINLP No E MILP No EEn GP Yes

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e1091 1083

similar research was carried out by Ghosh and Shah (2015). They verified that supply chain stakeholders are provided better op- portunities to launch green initiatives by green consumer markets.

In Table 1, the papers are summarized concerning economic (E), environmental (En) and social (S) objectives. Economic objectives are considered at all of the papers. Sixteen papers used only eco- nomic objectives. Both economic and environmental objectives are used by Sarkar et al. (2017), Yu and Solvang (2017), Pati et al. (2008), Amin and Zhang (2013), Diabat et al. (2013), Ramudhin et al. (2010). However, these studies but do not have any social objective. Especially recently published papers are using social objectives in addition to economic and environmental objectives (Anvari and Turkay (2017), Arampantzi and Minis (2017), Feit�o-Cesp�on et al. (2017), Govindan et al. (2016a), Govindan et al. (2016b)). In spite of many papers with economic objectives, sustainability approach is not widely studied in the literature. Multi-objective optimization has been used increasingly in recent years.

In addition, there are only a few papers using goal programming (Arampantzi and Minis (2017), Ramudhin et al. (2010), Pati et al. (2008)), in the literature. Arampantzi and Minis (2017) consid- ered only forward logistics network design, especially facility

Fig. 1. Schematically representatio

location problem and capacity extension decisions. Anvari and Turkay (2017) also studied the facility location problem that in- corporates the TBL approach for sustainability.

Colapinto et al. (2015) presented a comprehensive review of the GP studies. This technique has been frequently used for solving multi-criteria decision problems concerned with engineering design, management and social sciences. Design of the hybrid manufacturing systems (Satoglu and Suresh, 2009), paper recycling system (Pati et al., 2008), closed-loop battery supply chains (Subulan et al., 2015a), tire closed-loop supply chains (Subulan et al., 2015b) were performed by means of GP or fuzzy GP.

To the best of authors’ knowledge, this is the first study in the literature that proposes a goal programing model for reverse supply chains based on a TBL approach and performs a case study in household goods (WEEE) recovery industry. Our detailed literature review supports this finding.

3. Proposed framework for operations planning

In this section, we present the proposed decision making framework (see Fig. 1). This decision-making framework can be

n of the decision framework.

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e10911084

applied for operations planning in any sustainable supply chain. In our case, operations planning problem requires definition of the system boundary and determination of the model assumptions utilizing the knowledge in the related literature and experts working on reverse supply chain networks. In our network we consider customers, WEEE collection sites, recycling facilities, raw material markets, and government. The framework of the model is completed by considering economic, environmental and social factors. Since we optimize conflicting objectives in the Pareto optimal set we need to implement a decision maker strategy where experts are involved.

One of the challenging sides of analyzing sustainability is the conflict between essential factors. It is absolutely necessary for companies to maintain their profitability so that they can sustain their existence, but also the responsibility for nature and society must not be ignored. Some guidance (e.g. ISO 14001) and regula- tions exist regarding environmental responsibility which force the companies. As for the performance of the supply chain, not only economic and environmental aspects but also social factors are important (Ramudhin et al., 2010). Nevertheless, social perspective remains an area that received less attention (Seuring and Müller, 2008).

Economic parameters: The most important thing for investors is the profit of the investment. We did the primary considerations for parameters selection in the economic dimension in this context. We focused on the cost of running reverse logistics operations, but not the initial investment cost. The cost items include the fixed cost of recycling the products, labor cost, transportation cost and pen- alty cost of uncollected products. Besides, revenue item is consid- ered as the monetary value of the recycled materials such as aluminum, copper etc. (Shih, 2001).

Environmental parameters: We derived emission and waste rates used in the model from recognized data sources including web sites (www.myclimate.org, footprint.wwf.org.uk, www.nature. org), research articles (Eskandarpour et al., 2015; Neumüller et al., 2015) and reports (Trends in Global CO2 emissions: 2016, In- ventory of US greenhouse gas emissions and sinks: 1990e2015). In addition, we considered performance characteristics of the trucks from manufacturers such as Volvo (Martersson, 2010). Based on (Martersson, 2010), we have calculated the emission of carbon di- oxide for a 40-ton truck for which the payload is 27 tons and the fuel consumption is 0.35 L per kilometer, as follows:

0.35 l/km � 2.7 kg/l per 27 tons z 0.035 kg/ton-km. The proposed model considers not only emission from trans-

portation but also facility operations. Because a facility creates emission from power consumption, employee transportation, pa- per consumption and use of computers as well.

In Table 2, we show compared data of Euro6 and Euro 5 emission standards for the heavy duty engines (Williams and Minjares, 2016).

Social parameters: International Guidance Standard on Social Responsibility-ISO 26000 (ISO, 2010) is a good reference for social criteria identification. ISO 26000 sets the frameworks of social

Table 2 Euro 5 and Euro 6 standards for heavy-duty diesel engines: steady state testing (Notation e CO: carbon monoxide, ELR: European load response, ESC: European stationary cycle, HC: hydrocarbon, NOx: nitrogen oxide, PM: particulate matter, PN: particle number, WHSC: world harmonized stationary testing).

Test CO (g/kWh)

HC (g/kWh)

NOx (g/kWh)

PM (g/kWh)

PN (1/kWh)

Euro VI WHSC 1.5 0.13 0.40 0.01 8.0 � 1011 Euro V ESC&ELR 1.5 0.46 2.0 0.02

PM ¼ 0.13 g/kWh for engines < 0.75 dm3 swept volume per cylinder and a rated power speed > 3000 min�1.

responsibility in seven major topics: organizational governance, human rights, labor practices, the environment, fair operating practices, consumer issues, community involvement and develop- ment. Also, Anvari and Turkay (2017) divides social factors into five categories: demand satisfaction, resource equity, job opportunity, regional development, security level at the location, medical facility access level. These factors are used for selection of a facility loca- tion. However, it can also be considered as the local development goal, and we determined keeping the number of workers at a certain level for an operating facility as a social goal. The main reason behind this is the variability in the demand of WEEE to be recycled (Bal et al., 2018) which can cause layoffs at some period. For both employers and governments, it is important to keep workforce level at a certain level. Governments do not wish the workforce level to decrease. On the other hand, employers often do not wish to pay compensation by layoffs and confront with unions. Thus, we aimed to provide more regular work opportunity to employees.

4. Proposed goal programming model

In this paper, we address the operations planning problem for a reverse supply chain considering four goals. The proposed model determines the timing and amount of WEEE collection from the pre-determined points considering cost & revenue items, emission, available workforce, collection target, capacity of the recycling fa- cility and distance. There are four goals determined according to the cost minimization, environmental effect reduction, workforce balance and catching legal targets. The model handles operations planning problem, which has a TBL accounting perspective within a multi-product, multi-facility and multi-period case.

Model Assumptions.

� Products are collected from the central point of the city. Inner- city routing is out of the scope of this study.

� Cost of recycling does not change with years. � The numbers of collection sites are known and locations of the

recycling facilities are predetermined. � The specific facility that can recover a product type is

predetermined. � Cost parameters are foreknown as material, operation, recycling,

transportation, hiring, laying off and fixed cost. � The holding cost, stock out cost and storage cost are disregarded.

Model notation.

Sets:i: set of all types of products, i21:::I j: set of all types of raw recycled materials, j21;:::J k: set of all cities, k21:::K l: set of all facilities, l21:::L t: set of all periods, t21:::T p: set of periods, p21:::T

Scalars: BG : big number EOQ : economic order quantity for a city ðA full � truck loadÞ FTLðvehicleÞ : full truck load per transport vehicle Gðgram=unit vehicle=KmÞ : amount of emission per unit transport per km MRð%Þ : minimum collection rate RBð$=personÞ : employment cost of a worker RCð$=personÞ : hiring cost of a worker in $ RDð$=personÞ : layoff cost of a worker in $ RTð$=vehicleÞ : fixed cost of transportation in $ WMð%Þ : maximum workforce level; ðWM � 100%Þhttp://www.myclimate.orghttp://footprint.wwf.org.ukhttp://www.nature.orghttp://www.nature.org

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e1091 1085

MH (Hour): Number of work hours per month per one worker.

Parameters: ßtðgramÞ : target value of CO2 emission due to all transportation process CAPltðunit productÞ : capacity of facility l in period t dklðkmÞ : distance between demand location k and recycling facility location l DMiktðunitÞ : amount of product i sold in region k at period t EiðgramÞ : amount of emission stem from recycling of a product i εtðgramÞ : target value of CO2 emission due to all recycling processes at period t FAiðhourÞ : required person � hour workforce to recycle product i FtðunitÞ : collection target at period t LtðpersonÞ : target number of worker in period t MSijð$=kg mterialÞ : monetary value of material j recycled from product i Oitð%Þ : the percentage at which product i should be collected accordin to the legislation PRið$=vehicle=KmÞ : cost of transportation of product i RAið$=unit productÞ : recycling operation cost of product i RPið$=unit productÞ : penalty cost of uncollected product i RSijðkg=unit productÞ : amount of material j recycled from product i SMitðunit productÞ : amount of product i sold in period t TRitð$Þ : target total cost of recycling of product i in period t Employment targetlt : Number of people targeted to be employed at facility � l;in period � t:

Decision variables: Hlt : number of workers hired at facility l during period t Mlt : number of redundant workers at facility l during period t Wlt : number of workers employed at facility l in period t Xikt : collected � recycled number of product i in period t from region k

Ykt

� 1; if product i is collected in period t from region k 0; otherwise

f þit , f � it ; tr

þ it , tr

� it ; e

� t ;e

þ t , p

þ lt ;p

� lt : Deviational variables.

Objectives:

Min Z1 ¼ X t

X i

f �it (1)

Min Z2 ¼ X t

X i

trþit (2)

Min Z3 ¼ X l

X t p�lt (3)

Min Z4 ¼ X t eþt (4)

Subject to : X i

X k

X l

XiktRAi þWltRBþHltRC þMltRDþ X i

X k

X l

dkl PRi xikt FTL

þ X i

X k

xikt FTL

RT þ X i

X k

ðDMikt �XiktÞRPi

� X i

X j

X k

XiktRSijMSij ¼ X i

� TRit þtrþit �tr

� it

� ; ðct2TÞ

(5)

X k

X t Xikt ¼

X t OitSMit þ f þit � f

� it ; ðci2IÞ (6)

X i

X k

XiktEi þ X k

X l

Xikt FTL

Gdkl ¼ εt þ bt þ eþt � e�t ; ðct2TÞ

(7)

Wlt ¼ Employment targetlt þ pþlt � p � lt ; ðct2T; cl2LÞ (8)

Wlt � X i

XiktFAi=MH; ðct2T; cl2LÞ (9)

Wlt � X i

XiktFAi � WM=MH; ðct2T; cl2LÞ (10)

Wlt�1 þ Hlt � Mlt ¼ Wlt; ðct2T; cl2LÞ (11) X i

X k

Xikt � CAPlt; ðct2T; cl2LÞ (12)

EOQ � X i

Xikt � BGð1 � YktÞ ; ðct2T; ck2KÞ (13)

X i

Xikt � BG � Ykt; ðct2T; ck2KÞ (14)

X i

Xikt � MR � X i

DMikt; ðct2T; ck2KÞ (15)

Xp t¼1

Xikt � Xp t¼1

DMikt; ðcp ¼ 1; …; 12Þðci2I ; ck 2KÞ (16)

All variables � 0 (17)

Xikt; Hlt; Mlt; Wlt 2Z þ ðci; k; l; tÞ; Ykt2f0; 1g (18)

Objective functions (1), (2), (3), (4) minimizes the negative de- viation from WEEE collection target, minimizes positive deviation from cost target, negative deviation from employment target and positive deviation from total emission target that stems from both transportation and recycling operations. Constraint (5) defines cost that the manufacturer must pay for. The fact that reverse supply chain may not (always) make revenue, the objective is to minimize the reverse logistics cost. Cost items are composed of the fixed cost of recycling operation in the recycling facilities, employment cost, fixed and variable cost of transportation, penalty cost of uncollected items. On the other hand, income is earned out of sales of the material obtained from recycled WEEE. TR defines the target cost, and since the goal is to catch a break-even point, this is set to zero. Constraint (6) denotes legal collection goal taking into account actual sales (Smit) and the amount of product (Xikt) decided to collect in that period. There is a legal requirement that at least Oit percent of the sold goods are recycled. Constraint (7) describes environmental effect of each products’ recycling operation in the facility and each truck sent to collect the products. The total emission goal ðεtÞ is set with regard to total emission expected from all operations.

Minimization of the negative deviation ðp�lt Þ from the employ- ment target is aimed at the third objective. Related with this objective, Constraint (8) stipulates that the number of workers in

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e10911086

each facility (Wlt) should be close to the employment target. The structural Constraints (9), (10) ensures workforce is greater than required person-hour work and does not exceed the allowed maximum workforce level. Constraint (11) implies that sum of the workers employed in the previous period and those hired in the current period minus the redundant workers is equal to the current number of workers employed. Here, Wlt denotes the number of workers employed and Hlt denotes the number hired, in period-t. Constraint (12) defines the capacity for each facility. Thus, recy- cled products in each period cannot exceed the capacity of the fa- cilities. Economic order quantity is provided by Constraints (13), (14). These two constraints are modeled as conditional con- straints and ensure that a truck is sent to a collection point if at least the amount of products is equal to the economic order quantity. Here, Ykt is a binary variable and makes constraint (14) equal to Xikt which is greater than EOQ. Then, the constraint (13) becomes 0 due to ð1 � YktÞ. Otherwise, 0 is assigned to Xikt. EOQ is determined as a full-truck load that must be satisfied to collect WEEE from a city. Constraint (15) provides that at least some certain percent of the demand is collected in each city and each period. This constraint prevents the model to collect no products so as to produce zero emission. On the other hand, constraint (16) ensures that the total collected amount of product in a period cannot exceed the total demand from first period to a relevant period. Constraint (17) im- poses non-negativity restrictions while set of integrality re- strictions for decision variables Xikt; Hlt; Mlt; Wlt; Ykt are imposed by constraint (18).

5. Solution methodology

In the literature, many different and improved versions of the Augmented ε-constraint method exist (Ehrgott and Ryan, 2002; Laumanns et al., 2006; Hamacher et al., 2007; Mavrotas, 2009; Mavrotas and Florios, 2013). Since the Augmented ε-constraint method 2 (AUGMECON2) (Mavrotas and Florios, 2013) was proved to have better performance than the others, we preferred to use this method as our solution algorithm. We applied AUGMECON2 as shown below (Mavrotas and Florios, 2013):

min f1ðxÞ þ eps � �

s2 fmax2 � fmin2

þ 10�1 � s3 fmax3 � fmin3

þ 10�2 � s4 fmax4 � fmin4

(19)

Subject to : (20)

f2ðxÞ þ s2 ¼ fmin2 þ t � ðfmax2 � fmin2Þ=q2 (21)

f3ðxÞ þ s3 ¼ fmin3 þ t � ðfmax3 � fmin3Þ=q3 (22)

f4ðxÞ þ s4 ¼ fmin4 þ t � ðfmax4 � fmin4Þ=q4 (23)

x2S and si2R þ: (24)

In this formulation, f1 corresponds to ‘Legal Function’, f2 corre- sponds to ‘Cost Function’, f3 corresponds to ‘Social Function’ and f4 corresponds to ‘Environmental Function’. Surplus variables of the respective constraints are represented by s2, s3, and s4, respectively. The maximum and minimum value of objective functions from the payoff table are fmaxi and fmini respectively. The range of fi is fmaxi � fmini , t is the counter of the interval (if fi is divided to 4 then t changes from 1 to 4) and qi is the length of the equal intervals

of the objective function fi, and ε is relatively a small number be- tween 10�6 and 10�3 (Mavrotas and Florios, 2013). The identifica- tion of Pareto-optimal solutions is essential in multi-objective optimization. Thus we used CPLEX solver of the GAMS® software to generate a set of Pareto optimal solutions.

6. Case study

We illustrate our proposed model on a case study with real data and analyze the results. Some operational decisions are made for a global white appliances manufacturer. From manufacturers view- point, recycling of WEEE is very important in terms of fulfilling the responsibilities. In addition, WEEE collection rates are rising at a rapid pace due to the legal regulations (Ministry of Environment and Urbanization, 2012).

In the current setting, the customers have the right to return their used electrical-electronic products when they purchase a new one. So, as soon as the technical service provider of the manufac- turer delivers the new product to the customer, he/she should take back the old one (WEEE), if the customer demands. Besides, electrical-electronic product manufacturers are obliged to take back some certain percent of their sales amount (legal target). Otherwise, high monetary penalties incur to the manufacturer. So, these conditions are valid in this case study. In Table 3, we present the required information about the case.

The network comprises customers, local collection sites of WEEE, and recycling centers, as well as transportation links among them. The reverse logistics system works in such a way that when customers buy a new product, they have the right to deliver waste appliances. Therefore, technical service of the manufacturer re- ceives and stores the WEEE in its own warehouse. A third-party logistics provider is responsible for collecting the WEEE from all service points and delivering to recycling facility. In the recycling facility, products are initially stored in the collection center, which is in the same location with the facility. When they are ready to be processed, they are sorted and recycling operations start. Different types of output materials are obtained such as iron, aluminum, copper, compressor, plastic, wire, recyclable CFC (chlorofluoro- carbon), glass. These materials are sent to different facilities ac- cording to their specifications. Fig. 2 describes the focal point of the optimization model in reverse supply chain network. It starts with the collection of WEEE and ends with the recycling processes at the facilities. Therefore, a collection decision is made for each period by considering the pre-determined goals in the proposed model.

In this case, there are two recovery facilities. The location of the recovery facilities is pre-determined. So, no facility location deci- sion is made. WEEE products accumulate in the local collection points of the 81 cities. 12 months-period is considered. The dis- tances between the city centers and the facilities are used in the model. There are three types of products’ WEEE to be recovered, namely refrigerator, washing machine and dish-washer. Refriger- ator is recovered at the first facility, where the other two types of WEEE are recovered in the second one.

Monthly target recovery ratios for all products are assumed as 5%, which means at least 5% of the sold products should be recovered (Oit ¼ 5%). The total sales quantities of these three products are known. The target employment levels are 8 and 12 for the first and second facilities, respectively (Employ- ment_target1t ¼ 8, Employment_target2t ¼ 12). The target cost is assumed as zero (TRit ¼ 0), so that the operational and trans- portations costs are offset by the sales revenue earned out of the materials recovered. The total CO2 emission target for the recovery & transportation operations is 55,000 kg/month ðεt þ bt ¼ 55; 000 kg=monthÞ: This emission target is determined in accordance with

Table 3 Settings of the case study.

# of city 81 all cities over the country

# of facility 2 1: refrigerator, 2: washing machine and dish-washer

# of product type 3 1: refrigerator, 2: washing machine, 3: dish-washer

collection target (%) 5 legal target for 2016 cost target ($) 0 set as breakeven point environmental target

(kg/month) 55000 emission stem from in-facility

and transportation operations employment target 8 (1. facility), 12 (2. facility) transportation vehicle engine Euro5 or higher recycled materials iron, aluminum, copper, compressor,

plastic, wire, recyclable CFC (chlorofluorocarbon), glass

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e1091 1087

the WEEE recovery demand during a year, the expected number of vehicles, and the energy consumption of the recovery operations in the facilities. The model forces not to exceed this CO2 emission target.

In this case, the stock out cost is not needed to be considered since there is no demand to be satisfied by the customers in the supply chain to recover the waste material in a certain period. Be- sides, inventory holding cost does not actually incur to the recovery facility/company, since there is no investment in the WEEE that is collected. There is no purchasing cost of WEEE to the recovery company. Besides, there is enough available space to keep the waste inventory.

7. Results and discussion

The model was solved lexicographically using the GAMS® soft- ware, CPLEX solver, by a notebook computer with a 1.7 GHz Intel core i5 CPU and 4 GB RAM running Windows 10. Table 4 presents

Fig. 2. Reverse logi

Table 4 Model solution statistics for case study.

Number of

Blocks of equations Blocks of variables Single equations

Goal programming 36 18 42044 ε-constraint model 40 22 42048

the statistics and the solution report of the model. The goal pro- gramming model includes 42044 equations, and 7030 variables where 6876 of them are discrete variables. When we solved the model for the legal objective, it took only 0.188 s to reach the op- timum. Then the goal programming model is converted into an ε-constraint model. It has four additional equations, and additional (slack) variables. When this model is also solved, it took around 0.328 s. The computational times are very short.

Based on the solution of the ε-constraint model, the payoff table is obtained as shown in Table 5. We can see from the pay-off table (Table 5) that if we optimize the legal target, we can obtain mini- mum 8948 uncollected items. In the meantime, total cost becomes 687,163.844$, the total deviation from employment target becomes 7 persons in total and, the total deviation from environmental target becomes 203,234.95 kg. When we optimize the economic target, we can obtain minimum 667,692.730$ total cost with 21,956 uncollected items, the total deviation of the employment target of 22 persons and the total deviation of 72,152 kg emission. In this case the target emission level is exceeded approximately by %15. Therefore, Euro5 vehicles cannot reach the emission target, when the problem is solved according to legal, economic and social ob- jectives. Also, social target is obtained 5 that is minimum which indicates the deviation from employment target during all periods. Lastly, we obtained zero deviation from the desired emission level when we solved according to the environmental target. By using the parameter values presented in the payoff table, the range values of each objective function were determined, based on the AUG- MECON 2 (Mavrotas and Florios, 2013). Based on the intervals of fmaxi � fmini values for f2, f3, and f4, the solution of the problem generated 60 unique Pareto Optimal solutions, and those are shown in Appendix A, Table A.

Because our model has four different goals, the results are influenced by each goal. Therefore, we solved a bi-objective version of the model by fixing the rest of the two other goals to observe the

stics network.

Single variables Nonzero elements Discrete variables Execution time (s)

7030 268829 6876 0.188 7034 268911 6876 0.328

Table 5 Pay-off table obtained.

Z1 (legal)

Z2 (economic)

Z3 (social)

Z4 (environmental)

Min Z1 (legal) 8948 687163.844 7 203234.95 Min Z2 (economic) 21956 667692.730 22 72152.00 Min Z3 (social) 56075 978431.485 5 206964.3 Min Z4 (env.) 56066 959096.742 39 0

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e10911088

relationship between economic goal with the other goals. Fig. 3(a) shows the economic goal (net cost) versus legal goal (number of uncollected items). The number of uncollected items gradually in- creases as the cost decreases, in a non-linear way. Fig. 3(b) shows the economic goal versus environmental goal. The cost remains

Fig. 3. The effect of economic goal (net cost) versus (a) number of uncollected items (legal target: 17500, social target: 18) (c) social goal (legal target: 17500, environmental target: 1

Fig. 4. Comparison of workers in the both facilities with the number of products recovered (b) number of workers at white goods recycling facility.

approximately constant over a range of emission values. If the emission value gets closer to the target emission value, the cost increases. If the emission value is far from the emission target, the cost is also increasing. The reason behind this increase is the longer routes that the trucks use to collect the products. Fig. 3(c) shows the economic goal versus social goal. Starting from least possible social goal, the total cost sharply decreases and then gradually continues to decrease, as the social target reached becomes worse. A small improvement in the social goal requires a higher increase in the cost, when the (social) target is reached.

The results from the Pareto-optimal solutions table in the appendix A show that when we want to reach WEEE collection target with the social and environmental goals, the total cost is increasing considerably. For example, at the 36th Pareto optimal solution total cost is 745,377 $ while total number of uncollected

target) (environmental target: 102500, social target: 18), (b) environmental goal (legal 02500).

(WEEE demand) in each period. (a) Number of workers at refrigerator recycling facility

Fig. 5. Amounts of WEEE Collected at the 15th Pareto optimal solution for the 3rd period.

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e1091 1089

items is 11,072 items, total emission value is 152,355 kg CO2 and negative deviation from employment target is 39 persons. On the other hand, total cost is 823,062 $ at the 48th Pareto optimal so- lution while total number of uncollected items is 10,551, total emission value is 155,223 kg CO2 and negative deviation from employment target is 22 persons. It is also worth noting that the total cost does not vary linearly, as one target decreases and the other increases. For this reason, the obtained Pareto optimal results must be evaluated by experts.

Fig. 4 represents the number of workers at two of the facilities and the number of products recycled in each period (WEEE de- mand) according to the 15th Pareto optimal solution. The first fa- cility only recycles refrigerators and the demand peaks during months 8, 9 and, 10 (see Fig. 4(a)). The exact number of refrigerator recycled in Fig. 4(a) is around 4800 items during the first six months although it reaches 12,500 items at the eighth month. On the other hand, the second facility recycles rest of appliances. It recycles around 240,000 items in a year. For this facility the demand decreases at the mid of a year. Although the demand (number of products recycled) varies considerably, it appears that the vari- ability in the number of workers in both facilities are less. The reason for this situation is to reach the social goal.

Fig. 5 describes an example of the obtained the results for collection decision at the 3rd period. Results indicate that 59 re- frigerators from Kutahya city, 77 refrigerators from Corum city and 102 refrigerators should be collected from Malatya city to refrig- erator recovery facility and from Kastamonu city 83 products should be collected to the white appliances recovery facility on March (3rd period).

8. Conclusions

This paper optimizes operations planning of reverse supply chain network with the Triple Bottom Line accounting perspective. In this study, for a case study of durable goods recovery company, the goals were defined by the legal regulations and sustainability dimensions. According to the laws, the manufacturers are required to recover a minimum percent of their goods sold, and loyalty to the regulations is important. Therefore, the legal target is set to catch the collection target. Besides, an economically sustainable business is vital for the companies. For this reason, the goal of not to incur

loss (zero cost target) is determined as an economic objective. Because of the fact that it is not highly possible to make profits in recycling systems, loss minimization was defined instead of profit maximization. When the companies are considered to be respon- sive to the employees, it is very important that they can provide permanent employment to their employees. In spite of variability in the demand of WEEE, the social goal is set to catch employment target which is defined according to the capacity of facilities and demand. In order to provide also environmental responsibility, an emission target is determined while optimizing the network. So the goal is not to exceed the determined CO2 emission level. Therefore, manufacturers are driven to change transport vehicles with more efficient and greener engines (Euro6 engines). The solution of the resulting mixed-integer linear goal programming model was pro- vided by an implementation of the AUGMECON-2 in GAMS®

software. For all that, this model is used for a case of a global white ap-

pliances manufacturer; it can be generalized for producers like all types of electric and electronic equipment (EEE) producers, auto- motive supply industry or health sector. Nevertheless, some sectors do not have legislations on waste products. In this case, the opti- mization model can also be run without a legal goal. In addition, some additional environmental constraints can be added according to the product structure. In future studies, due to the uncertainties in number of products to be recycled (demand), and prices of the recovered materials, stochastic optimization techniques can be employed. Besides, reverse logistics of other types of products other than WEEE can be planned, and different social objectives can be set, in future studies.

Disclosure statement

A potential conflict of interest was not reported by the authors.

Acknowledgement

This study has been financially supported by the Istanbul Technical University, with the project numbered 39770.

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e10911090

Appendix A

Table A Pareto Optimal Solutions for the case study.

Experiment # Z1 Z2 Z3 Z4 Experiment # Z1 Z2 Z3 Z4

1 15.377 745.377 39 51.741 31 12.454 978.431 39 103.482 2 15.033 978.431 31 51.741 32 12.443 900.747 14 103.482 3 14.858 900.747 22 51.741 33 11.414 823.062 31 154.524 4 14.858 978.431 22 51.741 34 11.307 900.747 14 155.223 5 14.805 823.062 22 51.741 35 11.219 823.062 14 155.223 6 14.763 823.062 14 51.741 36 11.072 745.377 39 152.355 7 14.722 745.377 14 51.741 37 10.974 823.062 39 155.223 8 14.559 900.747 31 51.741 38 10.866 978.431 22 155.223 9 14.532 823.062 31 51.734 39 10.812 745.377 14 155.223 10 14.521 900.747 14 51.740 40 10.728 900.747 39 155.223 11 14.516 978.431 39 51.718 41 10.649 745.377 22 154.273 12 14.514 823.062 39 51.741 42 10.645 900.747 31 155.223 13 14.496 745.377 22 51.733 43 10.645 978.431 31 155.223 14 14.490 978.431 14 51.741 44 10.625 745.377 31 155.209 15 14.488 900.747 39 51.741 45 10.608 900.747 22 154.064 16 14.485 745.377 31 51.741 46 10.561 978.431 39 155.223 17 13.518 823.062 22 103.482 47 10.560 978.431 14 155.137 18 13.349 900.747 31 98.190 48 10.551 823.062 22 155.223 19 13.349 978.431 31 98.190 49 9.209 900.747 22 206.471 20 12.959 900.747 22 103.482 50 9.097 745.377 14 206.964 21 12.942 823.062 39 103.482 51 9.096 978.431 22 205.640 22 12.839 978.431 22 103.482 52 9.089 900.747 14 206.964 23 12.766 745.377 39 103.482 53 9.089 978.431 14 206.963 24 12.731 745.377 14 103.482 54 9.045 823.062 14 206.964 25 12.566 978.431 14 103.482 55 9.043 823.062 39 206.964 26 12.507 823.062 14 103.482 56 9.043 900.747 31 205.906 27 12.493 745.377 31 103.482 57 9.004 900.747 39 206.964 28 12.484 823.062 31 103.482 58 8.977 823.062 22 206.964 29 12.474 745.377 22 103.482 59 8.967 745.377 39 206.964 30 12.454 900.747 39 103.482 60 8.951 978.431 31 206.964

References

Abreu, A., Camarinha-Matos, L.M., 2008. On the role of value systems to promote the sustainability of collaborative environments. Int. J. Prod. Res. 46 (5), 1207e1229.

Alumur, S.A., Nickel, S., Saldanha-da-Gama, F., Verter, V., 2012. Multi-period reverse logistics network design. Eur. J. Oper. Res. 220 (1), 67e78.

Amin, S.H., Zhang, G., 2013. A multi-objective facility location model for closed-loop supply chain network under uncertain demand and return. Appl. Math. Model. 37 (6), 4165e4176.

Anvari, S., Turkay, M., 2017. The facility location problem from the perspective of triple bottom line accounting of sustainability. Int. J. Prod. Res. 55 (21), 1e22.

Arampantzi, C., Minis, I., 2017. A new model for designing sustainable supply chain networks and its application to a global manufacturer. J. Clean. Prod. 156, 276e292.

Aras, N., Aksen, D., 2008. Locating collection centers for distance-and incentive- dependent returns. Int. J. Prod. Econ. 111 (2), 316e333.

Bal, A., Satoglu, S.I., 2017. A goal programming model for the WEEE products re- covery operations planning. In: Proceedings of International Conference on Computers & Industrial Engineering: CIE47.

Bal, A., Sarvari, P.A., Satoglu, S.I., 2018. Analyzing recycling operations data of white appliances industry in Turkish market. In: Çalışır, Fethi, Camgoz Akda�g, Hatice (Eds.), Industrial Engineering in the Industry 4.0 Era. Springer Science & Busi- ness Media.

Brent, A.C., Rogers, D.E., Ramabitsa-Siimane, T.S., Rohwer, M.B., 2007. Application of the analytical hierarchy process to establish health care waste management systems that minimise infection risks in developing countries. Eur. J. Oper. Res. 181 (1), 403e424.

Carter, C.R., Rogers, D.S., 2008. A framework of sustainable supply chain manage- ment: moving toward new theory. Int. J. Phys. Distrib. Logist. Manag. 38 (5), 360e387.

Clift, R., 2003. Metrics for supply chain sustainability. Clean Technol. Environ. Policy 5 (3e4), 240e247.

Colapinto, C., Jayaraman, R., Marsiglio, S., 2015. Multi-criteria decision analysis with goal programming in engineering, management and social sciences: a state-of- the art review. Ann. Oper. Res. 251 (1), 7e40.

Coskun, S., Ozgur, L., Polat, O., Gungor, A., 2016. A model proposal for green supply chain network design based on consumer segmentation. J. Clean. Prod. 110, 149e157.

Das, K., Chowdhury, A.H., 2012. Designing a reverse logistics network for optimal collection, recovery and quality-based product-mix planning. Int. J. Prod. Econ.

135 (1), 209e221. Dekker, R., Fleischmann, M., Inderfurth, K., van Wassenhove, L.N., 2013. Reverse

Logistics: Quantitative Models for Closed-loop Supply Chains. Springer Science & Business Media.

Demirel, E., Demirel, N., Gokçen, H., 2016. A mixed integer linear programming model to optimize reverse logistics activities of end-of-life vehicles in Turkey. J. Clean. Prod. 112, 2101e2113.

Demirel, N.O., Gokçen, H., 2008. A mixed integer programming model for rema- nufacturing in reverse logistics environment. Int. J. Adv. Manuf. Technol. 39 (11), 1197e1206.

Diabat, A., Abdallah, T., Al-Refaie, A., Svetinovic, D., Govindan, K., 2013. Strategic closed-loop facility location problem with carbon market trading. IEEE Trans. Eng. Manag. 60 (2), 398e408.

Drake, D.F., Spinler, S., 2013. OM forumdsustainable operations management: an enduring stream or a passing fancy? Manuf. Serv. Oper. Manag. 15 (4), 689e700.

Ene, S., Oztürk, N., 2015. Network modeling for reverse flows of end-of-life vehicles. Waste Manag. 38, 284e296.

Ehrgott, M., Ryan, D.M., 2002. Constructing robust crew schedules with bicriteria optimization. J. Multi-Criteria Decis. Anal. 11 (3), 139e150.

Eskandarpour, M., Dejax, P., Miemczyk, J., P�eton, O., 2015. Sustainable supply chain network design: an optimization-oriented review. Omega 54, 11e32.

Feit�o-Cesp�on, M., Sarache, W., Piedra-Jimenez, F., Cesp�on-Castro, R., 2017. Redesign of a sustainable reverse supply chain under uncertainty: a case study. J. Clean. Prod. 151, 206e217.

Fonseca, M.C., García-S�anchez, �A., Ortega-Mier, M., Saldanha-da-Gama, F., 2010. A stochastic bi-objective location model for strategic reverse logistics. Top 18 (1), 158e184.

Ghosh, D., Shah, J., 2015. Supply chain analysis under green sensitive consumer demand and cost sharing contract. Int. J. Prod. Econ. 164, 319e329.

Gonela, V., Zhang, J., Osmani, A., 2015. Stochastic optimization of sustainable in- dustrial symbiosis based hybrid generation bioethanol supply chains. Comput. Ind. Eng. 87, 40e65.

Govindan, K., Jha, P.C., Garg, K., 2016a. Product recovery optimization in closed-loop supply chain to improve sustainability in manufacturing. Int. J. Prod. Res. 54 (5), 1463e1486.

Govindan, K., Paam, P., Abtahi, A.R., 2016b. A fuzzy multi-objective optimization model for sustainable reverse logistics network design. Ecol. Indicat. 67, 753e768.

Hamacher, H.W., Pedersen, C.R., Ruzika, S., 2007. Finding representative systems for discrete bicriterion optimization problems. Oper. Res. Lett. 35 (3), 336e344.

Hashemi, V., Chen, M., Fang, L., 2014. Process planning for closed-loop aerospacehttp://refhub.elsevier.com/S0959-6526(18)32446-6/sref1http://refhub.elsevier.com/S0959-6526(18)32446-6/sref1http://refhub.elsevier.com/S0959-6526(18)32446-6/sref1http://refhub.elsevier.com/S0959-6526(18)32446-6/sref1http://refhub.elsevier.com/S0959-6526(18)32446-6/sref2http://refhub.elsevier.com/S0959-6526(18)32446-6/sref2http://refhub.elsevier.com/S0959-6526(18)32446-6/sref2http://refhub.elsevier.com/S0959-6526(18)32446-6/sref3http://refhub.elsevier.com/S0959-6526(18)32446-6/sref3http://refhub.elsevier.com/S0959-6526(18)32446-6/sref3http://refhub.elsevier.com/S0959-6526(18)32446-6/sref3http://refhub.elsevier.com/S0959-6526(18)32446-6/sref4http://refhub.elsevier.com/S0959-6526(18)32446-6/sref4http://refhub.elsevier.com/S0959-6526(18)32446-6/sref4http://refhub.elsevier.com/S0959-6526(18)32446-6/sref5http://refhub.elsevier.com/S0959-6526(18)32446-6/sref5http://refhub.elsevier.com/S0959-6526(18)32446-6/sref5http://refhub.elsevier.com/S0959-6526(18)32446-6/sref5http://refhub.elsevier.com/S0959-6526(18)32446-6/sref6http://refhub.elsevier.com/S0959-6526(18)32446-6/sref6http://refhub.elsevier.com/S0959-6526(18)32446-6/sref6http://refhub.elsevier.com/S0959-6526(18)32446-6/sref7http://refhub.elsevier.com/S0959-6526(18)32446-6/sref7http://refhub.elsevier.com/S0959-6526(18)32446-6/sref7http://refhub.elsevier.com/S0959-6526(18)32446-6/sref7http://refhub.elsevier.com/S0959-6526(18)32446-6/sref8http://refhub.elsevier.com/S0959-6526(18)32446-6/sref8http://refhub.elsevier.com/S0959-6526(18)32446-6/sref8http://refhub.elsevier.com/S0959-6526(18)32446-6/sref8http://refhub.elsevier.com/S0959-6526(18)32446-6/sref8http://refhub.elsevier.com/S0959-6526(18)32446-6/sref8http://refhub.elsevier.com/S0959-6526(18)32446-6/sref8http://refhub.elsevier.com/S0959-6526(18)32446-6/sref9http://refhub.elsevier.com/S0959-6526(18)32446-6/sref9http://refhub.elsevier.com/S0959-6526(18)32446-6/sref9http://refhub.elsevier.com/S0959-6526(18)32446-6/sref9http://refhub.elsevier.com/S0959-6526(18)32446-6/sref9http://refhub.elsevier.com/S0959-6526(18)32446-6/sref10http://refhub.elsevier.com/S0959-6526(18)32446-6/sref10http://refhub.elsevier.com/S0959-6526(18)32446-6/sref10http://refhub.elsevier.com/S0959-6526(18)32446-6/sref10http://refhub.elsevier.com/S0959-6526(18)32446-6/sref11http://refhub.elsevier.com/S0959-6526(18)32446-6/sref11http://refhub.elsevier.com/S0959-6526(18)32446-6/sref11http://refhub.elsevier.com/S0959-6526(18)32446-6/sref11http://refhub.elsevier.com/S0959-6526(18)32446-6/sref12http://refhub.elsevier.com/S0959-6526(18)32446-6/sref12http://refhub.elsevier.com/S0959-6526(18)32446-6/sref12http://refhub.elsevier.com/S0959-6526(18)32446-6/sref12http://refhub.elsevier.com/S0959-6526(18)32446-6/sref13http://refhub.elsevier.com/S0959-6526(18)32446-6/sref13http://refhub.elsevier.com/S0959-6526(18)32446-6/sref13http://refhub.elsevier.com/S0959-6526(18)32446-6/sref13http://refhub.elsevier.com/S0959-6526(18)32446-6/sref14http://refhub.elsevier.com/S0959-6526(18)32446-6/sref14http://refhub.elsevier.com/S0959-6526(18)32446-6/sref14http://refhub.elsevier.com/S0959-6526(18)32446-6/sref14http://refhub.elsevier.com/S0959-6526(18)32446-6/sref15http://refhub.elsevier.com/S0959-6526(18)32446-6/sref15http://refhub.elsevier.com/S0959-6526(18)32446-6/sref15http://refhub.elsevier.com/S0959-6526(18)32446-6/sref16http://refhub.elsevier.com/S0959-6526(18)32446-6/sref16http://refhub.elsevier.com/S0959-6526(18)32446-6/sref16http://refhub.elsevier.com/S0959-6526(18)32446-6/sref16http://refhub.elsevier.com/S0959-6526(18)32446-6/sref17http://refhub.elsevier.com/S0959-6526(18)32446-6/sref17http://refhub.elsevier.com/S0959-6526(18)32446-6/sref17http://refhub.elsevier.com/S0959-6526(18)32446-6/sref17http://refhub.elsevier.com/S0959-6526(18)32446-6/sref18http://refhub.elsevier.com/S0959-6526(18)32446-6/sref18http://refhub.elsevier.com/S0959-6526(18)32446-6/sref18http://refhub.elsevier.com/S0959-6526(18)32446-6/sref18http://refhub.elsevier.com/S0959-6526(18)32446-6/sref19http://refhub.elsevier.com/S0959-6526(18)32446-6/sref19http://refhub.elsevier.com/S0959-6526(18)32446-6/sref19http://refhub.elsevier.com/S0959-6526(18)32446-6/sref19http://refhub.elsevier.com/S0959-6526(18)32446-6/sref20http://refhub.elsevier.com/S0959-6526(18)32446-6/sref20http://refhub.elsevier.com/S0959-6526(18)32446-6/sref20http://refhub.elsevier.com/S0959-6526(18)32446-6/sref21http://refhub.elsevier.com/S0959-6526(18)32446-6/sref21http://refhub.elsevier.com/S0959-6526(18)32446-6/sref21http://refhub.elsevier.com/S0959-6526(18)32446-6/sref22http://refhub.elsevier.com/S0959-6526(18)32446-6/sref22http://refhub.elsevier.com/S0959-6526(18)32446-6/sref22http://refhub.elsevier.com/S0959-6526(18)32446-6/sref22http://refhub.elsevier.com/S0959-6526(18)32446-6/sref24http://refhub.elsevier.com/S0959-6526(18)32446-6/sref24http://refhub.elsevier.com/S0959-6526(18)32446-6/sref24http://refhub.elsevier.com/S0959-6526(18)32446-6/sref24http://refhub.elsevier.com/S0959-6526(18)32446-6/sref24http://refhub.elsevier.com/S0959-6526(18)32446-6/sref24http://refhub.elsevier.com/S0959-6526(18)32446-6/sref24http://refhub.elsevier.com/S0959-6526(18)32446-6/sref25http://refhub.elsevier.com/S0959-6526(18)32446-6/sref25http://refhub.elsevier.com/S0959-6526(18)32446-6/sref25http://refhub.elsevier.com/S0959-6526(18)32446-6/sref25http://refhub.elsevier.com/S0959-6526(18)32446-6/sref25http://refhub.elsevier.com/S0959-6526(18)32446-6/sref25http://refhub.elsevier.com/S0959-6526(18)32446-6/sref26http://refhub.elsevier.com/S0959-6526(18)32446-6/sref26http://refhub.elsevier.com/S0959-6526(18)32446-6/sref26http://refhub.elsevier.com/S0959-6526(18)32446-6/sref27http://refhub.elsevier.com/S0959-6526(18)32446-6/sref27http://refhub.elsevier.com/S0959-6526(18)32446-6/sref27http://refhub.elsevier.com/S0959-6526(18)32446-6/sref27http://refhub.elsevier.com/S0959-6526(18)32446-6/sref28http://refhub.elsevier.com/S0959-6526(18)32446-6/sref28http://refhub.elsevier.com/S0959-6526(18)32446-6/sref28http://refhub.elsevier.com/S0959-6526(18)32446-6/sref28http://refhub.elsevier.com/S0959-6526(18)32446-6/sref29http://refhub.elsevier.com/S0959-6526(18)32446-6/sref29http://refhub.elsevier.com/S0959-6526(18)32446-6/sref29http://refhub.elsevier.com/S0959-6526(18)32446-6/sref29http://refhub.elsevier.com/S0959-6526(18)32446-6/sref30http://refhub.elsevier.com/S0959-6526(18)32446-6/sref30http://refhub.elsevier.com/S0959-6526(18)32446-6/sref30http://refhub.elsevier.com/S0959-6526(18)32446-6/sref31

A. Bal, S.I. Satoglu / Journal of Cleaner Production 201 (2018) 1081e1091 1091

manufacturing supply chain and environmental impact reduction. Comput. Ind. Eng. 75, 87e95.

Hung Lau, K., Wang, Y., 2009. Reverse logistics in the electronic industry of China: a case study. Supply Chain Manag.: Int. J. 14 (6), 447e465.

ISO, 2015. ISO14001, Environmental Management Systems. Final Draft). ISO, 2010. ISO26000, Guidance on Social Responsibility. Final Draft. http://www.

uobaghdad.edu.iq/uploads/pics13/qaa/iso26000.pdf. Jeurissen, R., 2000. John Elkington, Cannibals with forks: the triple bottom line of

21st century business. J. Bus. Ethics 23 (2), 229e231. Kannan, D., Diabat, A., Alrefaei, M., Govindan, K., Yong, G., 2012. A carbon footprint

based reverse logistics network design model. Resour. Conserv. Recycl. 67, 75e79.

Kara, S., Rugrungruang, F., Kaebernick, H., 2007. Simulation modelling of reverse logistics networks. Int. J. Prod. Econ. 106 (1), 61e69.

Kleindorfer, P.R., Singhal, K., Wassenhove, L.N., 2005. Sustainable operations man- agement. Prod. Oper. Manag. 14 (4), 482e492.

Krumwiede, D.W., Sheu, C., 2002. A model for reverse logistics entry by third-party providers. Omega 30 (5), 325e333.

Kumar, S., Putnam, V., 2008. Cradle to cradle: reverse logistics strategies and op- portunities across three industry sectors. Int. J. Prod. Econ. 115 (2), 305e315.

Laumanns, M., Thiele, L., Zitzler, E., 2006. An efficient, adaptive parameter variation scheme for metaheuristics based on the epsilon-constraint method. Eur. J. Oper. Res. 169 (3), 932e942.

Lee, D.H., Dong, M., 2009. Dynamic network design for reverse logistics operations under uncertainty. Transport. Res. E Logist. Transport. Rev. 45 (1), 61e71.

Linton, J.D., Klassen, R., Jayaraman, V., 2007. Sustainable supply chains: an intro- duction. J. Oper. Manag. 25 (6), 1075e1082.

Martersson, L., 2010. Emissions from Volvo’s trucks. Regulation no: 2064010005. Volvo Truck Corp. 1, 1e5.

Mavrotas, G., 2009. Effective implementation of the ε-constraint method in multi- objective mathematical programming problems. Appl. Math. Comput. 213 (2), 455e465.

Mavrotas, G., Florios, K., 2013. An improved version of the augmented ε-constraint method (AUGMECON2) for finding the exact pareto set in multi-objective integer programming problems. Appl. Math. Comput. 219 (18), 9652e9669.

Ministry of Environment and Urbanisation, 2012. Regulation for control of waste electric and electronic equipment. Official J. Turkish Republic 28300.

Mutha, A., Pokharel, S., 2009. Strategic network design for reverse logistics and remanufacturing using new and old product modules. Comput. Ind. Eng. 56 (1), 334e346.

Neumüller, C., Kellner, F., Gupta, J.N., Lasch, R., 2015. Integrating three-dimensional sustainability in distribution centre selection: the process analysis method- based analytic network process. Int. J. Prod. Res. 53 (2), 409e434.

Ozceylan, E., Paksoy, T., 2013. A mixed integer programming model for a closed-loop supply-chain network. Int. J. Prod. Res. 51 (3), 718e734.

Ozceylan, E., Paksoy, T., Bektaş, T., 2014. Modeling and optimizing the integrated problem of closed-loop supply chain network design and disassembly line balancing. Transport. Res. E Logist. Transport. Rev. 61, 142e164.

Ozkır, V., Başlıgıl, H., 2012. Modelling product-recovery processes in closed-loop supply-chain network design. Int. J. Prod. Res. 50 (8), 2218e2233.

Paksoy, T., Ozceylan, E., Weber, G.W., 2010, June. A multi objective model for opti- mization of a green supply chain network. In: AIP Conference Proceedings. AIP,

pp. 311e320 vol. 1239, No. 1. Pati, R.K., Vrat, P., Kumar, P., 2008. A goal programming model for paper recycling

system. Omega 36 (3), 405e417. Ramezani, M., Bashiri, M., Tavakkoli-Moghaddam, R., 2013. A new multi-objective

stochastic model for a forward/reverse logistic network design with respon- siveness and quality level. Appl. Math. Model. 37 (1), 328e344.

Ramudhin, A., Chaabane, A., Paquet, M., 2010. Carbon market sensitive sustainable supply chain network design. Int. J. Manag. Sci. Eng. Manag. 5 (1), 30e38.

Roghanian, E., Pazhoheshfar, P., 2014. An optimization model for reverse logistics network under stochastic environment by using genetic algorithm. J. Manuf. Syst. 33 (3), 348e356.

Safaei, A.S., Roozbeh, A., Paydar, M.M., 2017. A robust optimization model for the design of a cardboard closed-loop supply chain. J. Clean. Prod. 166, 1154e1168.

Sarkar, B., Ullah, M., Kim, N., 2017a. Environmental and economic assessment of closed-loop supply chain with remanufacturing and returnable transport items. Comput. Ind. Eng. 111, 148e163.

Satoglu, S.I., Suresh, N.C., 2009. A goal-programming approach for design of hybrid cellular manufacturing systems in dual resource constrained environments. Comput. Ind. Eng. 56 (2), 560e575.

Seuring, S., Müller, M., 2008. From a literature review to a conceptual framework for sustainable supply chain management. J. Clean. Prod. 16 (15), 1699e1710.

Shaw, K., Irfan, M., Shankar, R., Yadav, S.S., 2016. Low carbon chance constrained supply chain network design problem: a Benders decomposition based approach. Comput. Ind. Eng. 98, 483e497.

Shih, L.H., 2001. Reverse logistics system planning for recycling electrical appliances and computers in Taiwan. Resour. Conserv. Recycl. 32 (1), 55e72.

Soleimani, H., Govindan, K., 2014. Reverse logistics network design and planning utilizing conditional value at risk. Eur. J. Oper. Res. 237 (2), 487e497.

Subulan, K., Tasan, A.S., Baykasoglu, A., 2015a. A fuzzy goal programming model to strategic planning problem of a lead/acid battery closed-loop supply chain. J. Manuf. Syst. 37, 243e264.

Subulan, K., Taşan, A.S., Baykasoglu, A., 2015b. Designing an environmentally conscious tire closed-loop supply chain network with multiple recovery options using interactive fuzzy goal programming. Appl. Math. Model. 39 (9), 2661e2702.

Tseng, S.C., Hung, S.W., 2014. A strategic decision-making model considering the social costs of carbon dioxide emissions for sustainable supply chain manage- ment. J. Environ. Manag. 133, 315e322.

Tuzkaya, G., Gulsun, B., Onsel, Ş., 2011. A methodology for the strategic design of reverse logistics networks and its application in the Turkish white goods in- dustry. Int. J. Prod. Res. 49 (15), 4543e4571.

Williams, M., Minjares, R., 2016. A Technical Summary of Euro 6/VI Vehicle Emission Standards. ICCT International council on clean transportation.

Yu, H., Solvang, W.D., 2017. A carbon-constrained stochastic optimization model with augmented multi-criteria scenario-based risk-averse solution for reverse logistics network design under uncertainty. J. Clean. Prod. 164, 1248e1267.

Yura, K., 1994. Production scheduling to satisfy worker’s preferences for days off and overtime under due-date constraints. Int. J. Prod. Econ. 33 (1), 265e270.

Zhou, X., Zhou, Y., 2015. Designing a multi-echelon reverse logistics operation and network: a case study of office paper in Beijing. Resour. Conserv. Recycl. 100, 58e69.http://refhub.elsevier.com/S0959-6526(18)32446-6/sref31http://refhub.elsevier.com/S0959-6526(18)32446-6/sref31http://refhub.elsevier.com/S0959-6526(18)32446-6/sref31http://refhub.elsevier.com/S0959-6526(18)32446-6/sref32http://refhub.elsevier.com/S0959-6526(18)32446-6/sref32http://refhub.elsevier.com/S0959-6526(18)32446-6/sref32http://refhub.elsevier.com/S0959-6526(18)32446-6/sref33http://www.uobaghdad.edu.iq/uploads/pics13/qaa/iso26000.pdfhttp://www.uobaghdad.edu.iq/uploads/pics13/qaa/iso26000.pdfhttp://refhub.elsevier.com/S0959-6526(18)32446-6/sref35http://refhub.elsevier.com/S0959-6526(18)32446-6/sref35http://refhub.elsevier.com/S0959-6526(18)32446-6/sref35http://refhub.elsevier.com/S0959-6526(18)32446-6/sref36http://refhub.elsevier.com/S0959-6526(18)32446-6/sref36http://refhub.elsevier.com/S0959-6526(18)32446-6/sref36http://refhub.elsevier.com/S0959-6526(18)32446-6/sref36http://refhub.elsevier.com/S0959-6526(18)32446-6/sref37http://refhub.elsevier.com/S0959-6526(18)32446-6/sref37http://refhub.elsevier.com/S0959-6526(18)32446-6/sref37http://refhub.elsevier.com/S0959-6526(18)32446-6/sref38http://refhub.elsevier.com/S0959-6526(18)32446-6/sref38http://refhub.elsevier.com/S0959-6526(18)32446-6/sref38http://refhub.elsevier.com/S0959-6526(18)32446-6/sref39http://refhub.elsevier.com/S0959-6526(18)32446-6/sref39http://refhub.elsevier.com/S0959-6526(18)32446-6/sref39http://refhub.elsevier.com/S0959-6526(18)32446-6/sref40http://refhub.elsevier.com/S0959-6526(18)32446-6/sref40http://refhub.elsevier.com/S0959-6526(18)32446-6/sref40http://refhub.elsevier.com/S0959-6526(18)32446-6/sref41http://refhub.elsevier.com/S0959-6526(18)32446-6/sref41http://refhub.elsevier.com/S0959-6526(18)32446-6/sref41http://refhub.elsevier.com/S0959-6526(18)32446-6/sref41http://refhub.elsevier.com/S0959-6526(18)32446-6/sref42http://refhub.elsevier.com/S0959-6526(18)32446-6/sref42http://refhub.elsevier.com/S0959-6526(18)32446-6/sref42http://refhub.elsevier.com/S0959-6526(18)32446-6/sref43http://refhub.elsevier.com/S0959-6526(18)32446-6/sref43http://refhub.elsevier.com/S0959-6526(18)32446-6/sref43http://refhub.elsevier.com/S0959-6526(18)32446-6/sref44http://refhub.elsevier.com/S0959-6526(18)32446-6/sref44http://refhub.elsevier.com/S0959-6526(18)32446-6/sref44http://refhub.elsevier.com/S0959-6526(18)32446-6/sref45http://refhub.elsevier.com/S0959-6526(18)32446-6/sref45http://refhub.elsevier.com/S0959-6526(18)32446-6/sref45http://refhub.elsevier.com/S0959-6526(18)32446-6/sref45http://refhub.elsevier.com/S0959-6526(18)32446-6/sref45http://refhub.elsevier.com/S0959-6526(18)32446-6/sref46http://refhub.elsevier.com/S0959-6526(18)32446-6/sref46http://refhub.elsevier.com/S0959-6526(18)32446-6/sref46http://refhub.elsevier.com/S0959-6526(18)32446-6/sref46http://refhub.elsevier.com/S0959-6526(18)32446-6/sref46http://refhub.elsevier.com/S0959-6526(18)32446-6/sref47http://refhub.elsevier.com/S0959-6526(18)32446-6/sref47http://refhub.elsevier.com/S0959-6526(18)32446-6/sref73http://refhub.elsevier.com/S0959-6526(18)32446-6/sref73http://refhub.elsevier.com/S0959-6526(18)32446-6/sref73http://refhub.elsevier.com/S0959-6526(18)32446-6/sref73http://refhub.elsevier.com/S0959-6526(18)32446-6/sref48http://refhub.elsevier.com/S0959-6526(18)32446-6/sref48http://refhub.elsevier.com/S0959-6526(18)32446-6/sref48http://refhub.elsevier.com/S0959-6526(18)32446-6/sref48http://refhub.elsevier.com/S0959-6526(18)32446-6/sref49http://refhub.elsevier.com/S0959-6526(18)32446-6/sref49http://refhub.elsevier.com/S0959-6526(18)32446-6/sref49http://refhub.elsevier.com/S0959-6526(18)32446-6/sref50http://refhub.elsevier.com/S0959-6526(18)32446-6/sref50http://refhub.elsevier.com/S0959-6526(18)32446-6/sref50http://refhub.elsevier.com/S0959-6526(18)32446-6/sref50http://refhub.elsevier.com/S0959-6526(18)32446-6/sref50http://refhub.elsevier.com/S0959-6526(18)32446-6/sref51http://refhub.elsevier.com/S0959-6526(18)32446-6/sref51http://refhub.elsevier.com/S0959-6526(18)32446-6/sref51http://refhub.elsevier.com/S0959-6526(18)32446-6/sref51http://refhub.elsevier.com/S0959-6526(18)32446-6/sref51http://refhub.elsevier.com/S0959-6526(18)32446-6/sref51http://refhub.elsevier.com/S0959-6526(18)32446-6/sref51http://refhub.elsevier.com/S0959-6526(18)32446-6/sref52http://refhub.elsevier.com/S0959-6526(18)32446-6/sref52http://refhub.elsevier.com/S0959-6526(18)32446-6/sref52http://refhub.elsevier.com/S0959-6526(18)32446-6/sref52http://refhub.elsevier.com/S0959-6526(18)32446-6/sref53http://refhub.elsevier.com/S0959-6526(18)32446-6/sref53http://refhub.elsevier.com/S0959-6526(18)32446-6/sref53http://refhub.elsevier.com/S0959-6526(18)32446-6/sref55http://refhub.elsevier.com/S0959-6526(18)32446-6/sref55http://refhub.elsevier.com/S0959-6526(18)32446-6/sref55http://refhub.elsevier.com/S0959-6526(18)32446-6/sref55http://refhub.elsevier.com/S0959-6526(18)32446-6/sref56http://refhub.elsevier.com/S0959-6526(18)32446-6/sref56http://refhub.elsevier.com/S0959-6526(18)32446-6/sref56http://refhub.elsevier.com/S0959-6526(18)32446-6/sref57http://refhub.elsevier.com/S0959-6526(18)32446-6/sref57http://refhub.elsevier.com/S0959-6526(18)32446-6/sref57http://refhub.elsevier.com/S0959-6526(18)32446-6/sref57http://refhub.elsevier.com/S0959-6526(18)32446-6/sref58http://refhub.elsevier.com/S0959-6526(18)32446-6/sref58http://refhub.elsevier.com/S0959-6526(18)32446-6/sref58http://refhub.elsevier.com/S0959-6526(18)32446-6/sref59http://refhub.elsevier.com/S0959-6526(18)32446-6/sref59http://refhub.elsevier.com/S0959-6526(18)32446-6/sref59http://refhub.elsevier.com/S0959-6526(18)32446-6/sref59http://refhub.elsevier.com/S0959-6526(18)32446-6/sref60http://refhub.elsevier.com/S0959-6526(18)32446-6/sref60http://refhub.elsevier.com/S0959-6526(18)32446-6/sref60http://refhub.elsevier.com/S0959-6526(18)32446-6/sref60http://refhub.elsevier.com/S0959-6526(18)32446-6/sref61http://refhub.elsevier.com/S0959-6526(18)32446-6/sref61http://refhub.elsevier.com/S0959-6526(18)32446-6/sref61http://refhub.elsevier.com/S0959-6526(18)32446-6/sref62http://refhub.elsevier.com/S0959-6526(18)32446-6/sref62http://refhub.elsevier.com/S0959-6526(18)32446-6/sref62http://refhub.elsevier.com/S0959-6526(18)32446-6/sref62http://refhub.elsevier.com/S0959-6526(18)32446-6/sref63http://refhub.elsevier.com/S0959-6526(18)32446-6/sref63http://refhub.elsevier.com/S0959-6526(18)32446-6/sref63http://refhub.elsevier.com/S0959-6526(18)32446-6/sref64http://refhub.elsevier.com/S0959-6526(18)32446-6/sref64http://refhub.elsevier.com/S0959-6526(18)32446-6/sref64http://refhub.elsevier.com/S0959-6526(18)32446-6/sref65http://refhub.elsevier.com/S0959-6526(18)32446-6/sref65http://refhub.elsevier.com/S0959-6526(18)32446-6/sref65http://refhub.elsevier.com/S0959-6526(18)32446-6/sref65http://refhub.elsevier.com/S0959-6526(18)32446-6/sref66http://refhub.elsevier.com/S0959-6526(18)32446-6/sref66http://refhub.elsevier.com/S0959-6526(18)32446-6/sref66http://refhub.elsevier.com/S0959-6526(18)32446-6/sref66http://refhub.elsevier.com/S0959-6526(18)32446-6/sref66http://refhub.elsevier.com/S0959-6526(18)32446-6/sref66http://refhub.elsevier.com/S0959-6526(18)32446-6/sref67http://refhub.elsevier.com/S0959-6526(18)32446-6/sref67http://refhub.elsevier.com/S0959-6526(18)32446-6/sref67http://refhub.elsevier.com/S0959-6526(18)32446-6/sref67http://refhub.elsevier.com/S0959-6526(18)32446-6/sref68http://refhub.elsevier.com/S0959-6526(18)32446-6/sref68http://refhub.elsevier.com/S0959-6526(18)32446-6/sref68http://refhub.elsevier.com/S0959-6526(18)32446-6/sref68http://refhub.elsevier.com/S0959-6526(18)32446-6/sref68http://refhub.elsevier.com/S0959-6526(18)32446-6/sref69http://refhub.elsevier.com/S0959-6526(18)32446-6/sref69http://refhub.elsevier.com/S0959-6526(18)32446-6/sref70http://refhub.elsevier.com/S0959-6526(18)32446-6/sref70http://refhub.elsevier.com/S0959-6526(18)32446-6/sref70http://refhub.elsevier.com/S0959-6526(18)32446-6/sref70http://refhub.elsevier.com/S0959-6526(18)32446-6/sref71http://refhub.elsevier.com/S0959-6526(18)32446-6/sref71http://refhub.elsevier.com/S0959-6526(18)32446-6/sref71http://refhub.elsevier.com/S0959-6526(18)32446-6/sref72http://refhub.elsevier.com/S0959-6526(18)32446-6/sref72http://refhub.elsevier.com/S0959-6526(18)32446-6/sref72http://refhub.elsevier.com/S0959-6526(18)32446-6/sref72

  • A goal programming model for sustainable reverse logistics operations planning and an application
    • 1. Introduction
    • 2. Literature review
    • 3. Proposed framework for operations planning
    • 4. Proposed goal programming model
    • 5. Solution methodology
    • 6. Case study
    • 7. Results and discussion
    • 8. Conclusions
    • Disclosure statement
    • Acknowledgement
    • Appendix A
    • References

Order your essay today and save 10% with the discount code ESSAYHELP