Different issues in designing closed loop supply chair network

Abstract

Converting raw materials into a final product, distributing it to consumers and then disposing of it by landfilling or incineration for any reason is called a classic or a traditional supply chain. The generated waste has increased and resulted in huge issues harming the environment. Therefore, to mitigate these issues, governments have started to enforce strict rules and researchers have proposed various ideas; one of which is integrating a reverse flow, to apply recovery processes on returned products through reuse, remanufacturing, or disposal, with the traditional supply chain to form a closed loop supply chain (CLSC). In addition to the environmental benefits, implementing the closed loop supply chain has its benefits on the economy as well. In this dissertation, we propose different mathematical models to design a closed loop supply chain network considering different issues and under several circumstances.In the first model, we propose a robust optimization (RO) scenario-based model with a single objective to minimize the total cost of the CLSC network under the carbon cap policy, the carbon tax policy, the carbon cap-and-trade policy and the carbon offset policy. We use downward product substitution policy while the uncertainty considered is in the product demand and the number of returned products. In the second model, we introduce linear physical programing (LPP) to model the problem of the CLSC design network considering deterministic product demand and the number of returned products. We consider economical objective to minimize the total cost of the CLSC network, environmental objective to minimize the emitted carbon using the carbon cap and tax policy, and service level objective to maximize the customer satisfaction using maximal covering locations problem (MCLP) to measure the service level score. In this model, we use the downward substitution and we consider the uncertainties from the product demand and the number of returned products. The third model integrates LPP and scenario-based RO to address the CLSC design problem to consider the economic, environmental and service level objectives, while using MCLP to find the score level for the service objective. The proposed model is built under the uncertainty of product demand and the number of returned products. We also use the downward product substitution policy. In the fourth model, we consider multi-period planning horizon and multiple products with different generations. We use an integrated model consisting of LPP and RO uncertainty set. This model considers economic, environmental and service level objectives. The service level scores used in the service level objective are calculated using the technique for order of preference by similarity to ideal solution (TOPSIS). Besides the uncertainty considered in the previous models concerning the product demand and the number of returned products, we consider an additional uncertainty that stems from the product substitution fraction. The product substitution policy in this model allows products from different generations to be substituted with new and remanufactured products. In addition, it allows the new products to substitute the remanufactured products from the same generation in one-direction. The last model uses goal programming and scenario-based RO for a single period. The model considers economic, environmental and service level objectives. We use TOPSIS to find the service level score for the service objective. The uncertainties in product demand and the number of returned products are considered in this model. Finally, we suggest some ideas for future studies.--Author's abstract