Goods Distribution Based on Simulation Annealing and Dynamic Programming

The optimization problem of goods distribution is an important issue in the modern logistics industry. In this paper, we propose a novel and effective mathematical model to solve the optimization problem. Furthermore, we present a novel method that can effectively combine the simulation annealing algorithm with dynamic programming algorithm to achieve the optimized results of the mathematical model. First, the minimum number of vehicles is roughly estimated. According to the minimum number of vehicles, we group the customers with clustering algorithm. Secondly, the shortest distance between any two vertexes for every group is calculated by the Floyd algorithm. The equivalent network graph is constructed based on the shortest distance. In addition, a Hamilton circle is solved by the simulating annealing algorithm from the equivalent network graph. An approximate optimal solution of the vehicles delivery route is obtained by integrating the points in both ends. Finally, the optimal route of every vehicle is integrated and performs local adjustment by the dynamic programming. Therefore, the globally optimal solution of the whole problem is obtained. The experimental results show that the proposed method is more effectively solve the goods distribution problems than other traditional algorithms.