Research On Location Routing Problem For Fresh Agri-Products Cold Chain Distribution Center Based On Bi-Level Programming

In the fresh market of domestic goods,agricultural products is one of the major consumer goods.The total volume of cold chain logistics transactions of agricultural products in China reached 4 trillion yuan in 2017,while the total volume of social logistics reached 252.8 trillion yuan in the same year,accounting for about 1.6%of the total volume,so the cold food chain market has a huge space to improve.However,by stark contrast to the huge potential market of fresh agricultural products,China's current cold chain logistics development is in a stagnant state.The unreasonable choice of cold chain distribution center node and the unreasonable planning of cold chain distribution route have become the main factors restricting the development of cold chain logistics.So,the location selection of fresh agricultural cold chain logistics distribution center is of great significance for improving the efficiency of the whole cold chain logistics.At the same time,due to the current limitations of new energy vehicles in driving distance and refrigeration power,they cannot undertake the task of cold chain logistics and transportation.Therefore,it is of great research value to optimize the distribution route of refrigerated vehicles so as to reduce the use cost of vehicles in the distribution process.The research on the location and path of fresh agricultural products cold chain logistics distribution center has become the focus of research on cold chain logistics at the present stage.Firstly,this thesis studies the domestic and foreign research results status quo on cold chain logistics and location-routing problem,and proposes a bi-level planning method.Aiming at the characteristics of the location-routing problem of cold chain logistics,a bi-level programming model is established.In the upper model,the thesis mainly considers the location problem of cold food distribution center,and the game-based relationship between the penalty cost of transportation without full load and the storage cost of distribution center is introduced to optimize this location model.In the lower model,the thesis introduces carbon emission constraint to deeply study the distribution route problem.The solution of the model is to give the site selection scheme of the upper model in the first place.Based on the preliminary scheme given by the upper model,the lower model selects its own decision variables and selects the best distribution path plan.In the decision-making process of the lower level,intermediate variables feeds back to the upper level model as part of the upper level model.Finally,the optimization is achieved through the upper and lower level collaborative optimization iteration.In addition,a bi-level genetic algorithm is designed to solve it.The main body of the algorithm is the upper model,and the algorithm of the lower model is nested in the upper model.At the same time,in the upper genetic algorithm,memory hank operation is set to speed up the speed of the algorithm.In the lower genetic algorithm,generation gap operation is added to preserve excellent population individuals.Finally,a specific case study is designed for W enterprises in HEB city to solve the optimal location-route result and verify the effectiveness of the model algorithm.