Research On Logistics Routing Optimization Problem Based On Improved Ant Colony Algorithm

Logistics routing optimization is an important part of logistics scheduling which usually affects the cost and efficiency. According to official data, logistics transport cost accounts for more than half of the total cost of logistics in China, which is much higher than that of the developed countries. Improving transport efficiency and optimizing logistics and distribution routes have been becoming a critical problem. In order to save transportation cost, improve vehicle efficiency, and rationally allocate resources, various solutions have been proposed. However, to solve the problems of lower accuracy and the algorithm premature convergence, the paper improves ant colony algorithm which is applied to solve the routing optimization problem of logistics.This paper firstly analyzes the logistics routing optimization research status, studies the algorithms of solving routing optimization, and summarizes the problems of these algorithms, especially focusing on VRPDOP and VRPSDP. Then, under the case of multi-distribution centers, a decomposition method for pretreatment is proposed. The multi-distribution centers will be transformed into multiple single-distribution centers with this method. Furthermore, as to the VRPDOP, we develop a mathematical model, and adopt an improved ant colony algorithm to solve it. The improved algorithm combines the genetic algorithm, and improves transfer strategies of the ants and the pheromone update method. Finally, for the VRPSDP, we add a heuristic function and design a transfer strategy to improve ant colony algorithm, which is able to take composite factors into account to comprehensively and accurately search the next node, and consequently find a better route.Through theoretical analysis and experimental results, this paper effectively solves the logistics routing problem. From the simulation and analysis results, the improved algorithms are considered to be correct. The results of comparisons with other algorithms demonstrate that, the improved methods have certain advantages over the others.