The Optimization Of Multiple-Depot Vehicle Scheduling Based On Adaptive Genetic Algorithm

With the rapid development of market economy, the logistics impact on economic activity is increasingly obvious, at the same time also more and more attention. In logistics distribution vehicle scheduling problem is a core problem, a direct impact on logistics service quality and economic benefits. Logistics distribution vehicle scheduling is under the condition of meet the demand of customers, distribution is reasonable for the customer the quantity of the goods, send the least vehicle, and assign distribution vehicle transportation time and transportation cost the province route. Therefore, adopting the reasonable and scientific methods for logistics distribution vehicle scheduling optimization, is one of the very important link in logistics distribution. In the actual situation, there are often multiple logistics distribution center, therefore distribution center logistics distribution problem more optimization is of great significance. Specific studies are as follows:First, the current logistics industry development present situation and the related theoretical analysis, understand the current distribution center vehicle scheduling optimization model, the more the more details the distribution center of vehicle scheduling problem, on the basis of related concepts, to establish distribution center, the mathematical model of vehicle scheduling, according to the need to optimize the target calculation formula of objective function is determined, as well as the need to consider in the process of delivery to the constraint condition, etc.Second, according to the distribution center for vehicle scheduling more complicated mathematical model, to solve the long wait for a characteristic, selected as the optimization method, genetic algorithm and traditional genetic algorithm, on the basis of introduction of more suitable and efficient adaptive genetic algorithm. In order to accelerate the solving speed of genetic algorithm, improve its convergence performance, the encoding of genetic algorithm and the way of selection and heredity, mutation operator to improve. Coding mode adopted include configuration center distribution vehicle number and customer number, order value of chromosome gene sequence; Selection method using the optimal chromosome directly copied to the next generation, the rest of the chromosome using the roulette bets method; Heredity and variation process, using adaptive genetic operator and mutation operator of genetic algorithm.Finally, the adaptive genetic algorithm is adopted to more distribution center the vehicle scheduling model of case analysis and verify, and comparing with the traditional methods, through the experimental analysis, adaptive genetic algorithm in solving speed and the algorithm convergence is better than traditional methods.