"遗传算法优化物流配送车辆调度问题研究"

ty, the traditional VSP algorithm can no longer meet the requirements of customers for logistics distribution, so the concept of time windows has emerged. The vehicle optimization scheduling problem with time windows is a more complex NP hard problem than VSP. In this paper, on the basis of studying the vehicle optimization scheduling problem of logistics distribution, the vehicle optimization scheduling problem with time windows is analyzed. The basic theory of the genetic algorithm used is also discussed. For the non-full load VSP problem with time windows, the freight volume constraints and soft time window constraints are transformed into target constraints, and a non-full load VSP model is established. A genetic algorithm based on natural number coding, using techniques such as maximum preservation crossover and improved reverse mutation is designed. Experimental analysis has achieved good results. Since this problem was researched by the team members collaboratively, this paper focuses on the part completed by the author regarding the fitness function and mutation operation. Keywords: logistics distribution, vehicle optimization scheduling, genetic algorithm, time window