Designing a closed-loop supply chain network for citrus fruits crates considering environmental and economic issues

Highlights

• A six-level MILP model is considered to determine the optimal location and number of facilities needed.

• A new mixed linear mathematical model for a CLSC is developed and the case of citrus fruits crates is firstly considered.

• Taxes on emissions are considered in the case to control and reduce environmental pollution.

• Three leading metaheuristics and two hybrid methods are utilized and tuned to solve the proposed model.

Abstract

Most cities, notably major and agricultural ones, are faced with environmental and waste problems. Distribution and collection of agricultural crops can be challenging duties as world demand and production are substantially increased. Accordingly, resource depletion, environmental concern, and the importance of the circular economy have convinced this research group to focus on a Closed-Loop Supply Chain (CLSC) network design. In this study, a new mixed linear mathematical model for a CLSC was developed which minimizes the CLSC’s total costs and which tackles and controls air pollution. Contrary to previous works about supply chain network design, we firstly consider citrus fruits’ crates in our model. To solve the model, two leading algorithms, Genetic Algorithm and Simulated Annealing, are employed and a third recently successful method, Keshtel Algorithm, is utilized. Further, two hybridization algorithms stemmed from mentioned ones are applied. Finally, the results are assessed by different criteria and compared, and then the two best algorithms are chosen in this case. Consequently, in order to achieve the most effective result, a real case study of crates was conducted. The results obviously presented applicability and efficiency of the proposed model. Thus, the most suitable network for CLSC of citrus fruits’ crates was designed in which the costs and emissions were reduced.

Introduction

Over the past decades, supply chains have attracted much more attention in both academia and industries [1]. Initially, waste management and environmental issues were not taken into account in supply chains, but more recently, reverse logistics has become a most interesting topic. A closed-loop supply chain has evolved over time, one that considers both forward and backward supply chains, simultaneously [2]. While many researchers have carried out studies on a wide range of subjects, we are currently faced with serious environmental problems such as air pollution, waste, and water pollution. Furthermore, the rise in population parallels the increase in the amount of refuse so that a CLSC plays a key role in such an area [[3], [4], [5]].

The importance of reverse logistics (RL) is widely recognized, especially when industrialization and overpopulation create excesses of household and industrial waste. Despite enforcing stricter laws, the environment is in danger [6]. Consequently, treatment options have been extended to include approaches such as plastic and paper recycling, vegetable, fruit, and garden waste composting and so on. Due to the growth of cling-film, researchers have often focused on collecting and recycling facilities as well as locations of those [3,5,[7], [8], [9], [10]]. A suitable location can also cut costs and emissions because of reducing distance, trips, and so forth.

Some problems still correspond to the optimal number and location of facilities which result in increased transportation costs and worsened air pollution due to extra trips. To design supply chain networks, facility locations and demand allocations have raised a topical subject among researchers [[11], [12], [13], [14], [15]].

Concerning CLSCs, the whole supply chain from suppliers to recyclers is under consideration [5]; thus, its aim could be minimizing the cost of SC and boosting profits from it. While forward SCs have been used for satisfying the demand of customers, RL seeks to address the problem of waste [16]. Subsequently, using perfect CLSC networks can recycle waste materials and adopt eco-friendly transportation systems [17]. Hence, CLSCs have had widespread uses in various areas from household items to agricultural products ([1], [2], [5], and [15]).

Regarding the latest survey by authors on some databases, citrus fruits were the most important and dominant fruits in the world by approximately 133 million metric tons in 2016 according to the FAO, 2018¹ report (Fig. 1). Regarding official statistics by the FAO published in 2018, Asia was the largest producer of citrus fruits all over the globe (Fig. 2). Farmers harvest fruits and then package them for distribution. Some countries, especially developing ones, use plastic crates for packaging because of their accessibility and cheap raw materials. Afterwards, plastic packages have been utilized greatly as validated by statistical evidences provided from Office for National Statistics (UK) (Fig. 3). Both the huge waste and the significance of distribution systems make a closed-loop supply chain necessary to lessen the detrimental effect of citrus fruits distribution. Hence, after consuming fruits, producers should collect crates from end customers and then prepare them for recycling. Nevertheless, the bulk of plastic packages are not recycled in many countries, even some developed countries (Euro Statistics²) (Fig. 4).

According to the literature review detailed in the next section and to the best of our knowledge, no work has considered recycling plastic citrus crates in CLSCs. Our research method is to develop a mathematical model and solve the model by tuned algorithms. Hence, we firstly develop a new mathematical model for a CLSC in which minimizing the costs of recycling, transportation, and controlling emissions is sought. The taxes on emissions are also considered to control the air pollution. Further, to have and design the most efficient supply chain network, a location-allocation problem is considered. For reasons given earlier, a closed-loop supply chain is suggested by considering real assumptions and a real case, featuring a case study of crates carried out in Juybar, Mazandaran, Iran. The data were gathered not only from Iranian organizations but also from internet databases like Google maps. Finally, three superior metaheuristics (Genetic, Simulated Annealing, Keshtel algorithms) and two hybrid methods will be employed to solve the model. Afterwards, the proposed algorithms’ results are analyzed and compared, and the most suitable approaches are identified for this specific matter.

The rest of the paper is organized as follows. The following section provides the state-of-the-art literature review of CLSCs and RL. Section 3 illustrates the proposed model and defines the problem. In section 4, algorithms are explained, and subsequently analytical results are shown in section 5, and finally, section 6 presents our conclusions and suggests future research directions.