Research On Optimization Of Terminal Distribution Network Of Fresh Agricultural Products Under Community Group Purchase Mode

With the rapid development of Internet technology,new retail has gradually prevailed,and more and more retail models have begun to emerge.As an online purchase mode of fresh food e-commerce emerging in recent years,community group purchase is an important platform to meet the living needs of residents during the epidemic period,and has developed rapidly throughout the country.In the post-epidemic era,under the restriction of the new policy of community group purchase,the optimization of logistics distribution plays a key role in the development of community group purchase.Studying the terminal network optimization of fresh agricultural products under the community group purchase mode has certain practical significance for enterprises carrying out community group purchase in reducing the damage of fresh agricultural products,shortening the distribution distance and reducing the operating costs of enterprises.This paper first introduces the related theories of community group purchase,cold chain logistics terminal distribution of fresh agricultural products,cold chain logistics distribution center location and path optimization,and lays a theoretical foundation for studying the optimization of terminal distribution network of fresh agricultural products.Secondly,based on the logistics distribution mode of community group purchase and the characteristics of terminal distribution of fresh agricultural products,considering the factors such as distribution time limit,road congestion and carbon emission,a bi-level programming model of location and path optimization is constructed with the goal of minimizing the total cost of operation cost,transportation cost,penalty cost and carbon emission cost.The upper model is the location problem of the grid warehouse,and the lower model is the vehicle routing problem with time window.Then,on the basis of the above research,the simulated annealing algorithm and the improved genetic algorithm are used to solve the upper location model and the lower path planning model respectively.Then,considering the vehicle routing scheme of the lower model and the location result of the upper model,the terminal location and path optimization scheme of fresh agricultural products under the community group purchase mode are obtained.Finally,an empirical study was conducted based on the terminal distribution system of M company in N city.The relevant parameters are substituted into the model for solution,and the best grid warehouse location scheme and vehicle path optimization scheme are obtained,which verifies the correctness of the model and solution algorithm in this paper.At the same time,the solution process and specific results of the standard genetic algorithm and the improved genetic algorithm are analyzed and compared,which shows the advantages of the improved genetic algorithm with fast convergence speed and strong optimization ability,and verifies the superiority of the algorithm.The research results show that the bi-level programming model of the location-routing problem of fresh agricultural products constructed in this paper has certain practicability,and has certain guiding significance for the current fresh e-commerce enterprises participating in community group purchase in the optimization of terminal distribution network.At the same time,the model has strong versatility.Community group-buying enterprises in different regions can also adjust the model according to their own conditions,and obtain the terminal distribution network optimization scheme suitable for themselves by changing the parameters.This paper proposes the application of the simulated annealing algorithm and improved genetic algorithm to solve the bi-level programming model,which can effectively obtain the optimal solution of the location-path optimization problem.In addition,the improved genetic algorithm can better solve the location-routing bi-level programming model established in the paper than the standard genetic algorithm,which is conducive to the practical application of the model.