Belirsiz talep ve arz altında jenerik kapalı döngülü tedarik zincirindeki demontaj, yenileme ve yeniden üretim operasyonlarının planlanması

There is growing interest over closed loop supply chain management according to thecost, the legislation issues and environmental considerations. In this thesis, we areinterested in disassembly, remanufacturing and refurbishing operations of closedloop supply chains for products with modularity properties.We develop a generic mathematical model for commercial retail products suchas computers, cell phones etc. We use stochastic optimization and robustoptimization approaches to assist in making more accurate decisions underuncertain environment conditions.First we present the mathematical model in detail and then we use twodifferent approaches which are Robust Optimization and Stochastic Optimizationfor the generic model. The generic disassembly and refurbishing model hastwo stages to give decisions. In the first stage we determine the numberof product disassembly and part refurbishing sites which are given strategically.In the second stage, we determine the operational decisions such as the quantityof disassembled products, refurbished parts etc. We give first stage decisions withStochastic Optimization and Robust Optimization Approaches. Stochastic Optimizationminimizes the expected cost while giving the first stage decisions for the combinedproblem. Robust Optimization Approach which minimizes the deviation betweenoptimum cost and robust cost for each scenario while giving the first stage decisionsfor the combined problem.The generic model has complexity because it is a large scale mixed integer problem.In addition, it has two different uncertain factors which are the distribution of demandand returns. Thus to give strategic decisions like opening disassembly and refurbishingsites , we should be aware of the expected costs and what we lose under differentsituations if we give the first stage decisions at the beginning of the planning horizon.In the numerical results section, we solve different scenarios in a combined problemand give strategic decisions at the beginning of the time horizon using StochasticOptimization and Robust Optimization and compare the results of these two differentapproaches. The numerical results part show us which approach is appropriate under different conditions.The variability of demand and returns affect the first stagedecisions directly. We conclude that variability of returns has more effects overstrategic decisions rather than variability of demand.The second important result that we obtain is if the costs and capacities ofopening disassembly and refurbishing sites are high (inflexible environment),Stochastic Optimization Approach gives better solutions and if the costs andcapacities of opening disassembly and refurbishing sites are low(flexible environment),Robust Optimization Approach gives better solutions. Finally, we conclude thatdetermining the length of planning horizon has a critical importance in giving strategic decisions.