عنوان مقاله [English]
In recent years, the supply chain network (SCN) design problem has been gaining importance due to increasing competitiveness introduced by market globalization. One of the most important and strategic issues in supply chain management is configuration of the logistics network, which has a significant effect on the total performance of the supply chain. Nowadays, the emphasis on productivity and customer satisfaction leads firms to focus on the supply chain and logistics. Due to national and international rules, waste management, waste minimization, reuse, and material recycling have received increased attention over the last decade. Waste management is an important and rapidly growing industry for developing countries. Attention to reverse logistics networks has increased during the last decade, since their economic impact has become increasingly important and as environmental legislation has become stricter. It is necessary to design an effective product recovery network to minimize the total cost. Therefore, most companies only put their efforts into designing a logistics network that efficiently moves the products from seller to buyer. But, due to increasing environmental concerns and reduction of resources, issues like reverse logistics, product recovery, remanufacturing and reusing have received growing attention. The design of a product recovery network is an important and challenging problem in the field of reverse logistics, and some models have been formatted by researchers under a deterministic environment. In this paper, we present a stochastic integer programming model for a reverse logistics network. In this network, it is assumed that demand and production rates of returned products are stochastic. In this network, different products are distributed through different stages of the supply chain. Each supply center has a specific capacity for each product. The objective function is cost minimization. To analyze this model, some test problems are designed and then solved using GAMS software. The numerical results show the performance of the model.