Remanufacturing Reverse Logistics Network Design In Random Environment

Currently, studies of reverse logistics network mainly focus on a single-period single-objective level. However, a few studies have addressed the multi-period or multi-objective problem, while studies of multi-period multi-objective problem are even fewer. The background of this paper is the subproject of the National Natural Science Foundation which my tutor undertakes. To deal with optimal design problem of remanufacturing reverse logistics network, single-period multi-objective, multi-period multi-objective and multi-enterprise combined recovery reverse logistics networks are studied, based on the uncertainty of the returned quantities of waste products. Main work and results are as follows.(1) A single-period multi-objective remanufacturing reverse logistics network model was established considering the uncertainty of the returned quantities of waste products. From the perspectives of the economy and society, a bi-objective mixed-integer programming model was proposed. One objective of the model was to minimize the total cost of construction, and the other was to minimize the disutility to people. The algorithm for solving the model was proposed. And the validity of the model was illustrated by a numerical example. Example analysis showed that without considering the disutility to people had an impact on model result of location.(2) On the basis of single-period multi-objective remanufacturing reverse logistics network model, a multi-period static location model for remanufacturing reverse logistics network was built. The algorithm for solving the model using the method of chance constrained programming and goal programming was proposed. And the validity of the model was illustrated by a numerical example. Example analysis showed that the change on the confidence level and permission to stock both had an impact on model results. And benefit trade offs among two objectives were analyzed by example analysis based on objective achievement scale.(3) On the basis of multi-period multi-objective remanufacturing reverse logistics network model, a multi-enterprise combined recovery reverse logistics network model was built. A mixed-integer programming model for the facility location and flows distribution was established. And the validity of the model was illustrated by a numerical example. The solution of model showed that establishing a joint reverse logistics network could not only reduce the investment pressure on building reverse logistics network by single-enterprise, but also implement the scale operation with saving the recovery cost. That is a multi-win situation.