Research On Multi Depot Vehicle Routing Problem With Fuzzy Time Windows

As one of the hottest and most advanced topics in the field of modern logistics,vehicle routing problem plays a key role in logistics distribution optimization.Due to the expansion of transportation network and the increase in demand for logistics distribution,it is not easy for an enterprise with single distribution center to get the complex dispatching job done.Hence,the multi depot vehicle routing problem gets promising application in real life.What's more,the change in consumer attitudes towards logistic service leads customers to care more about the delivery time and the service,and service level improvement is becoming increasingly important in logistics.This paper focuses on the multi depot vehicle problem with fuzzy time windows,taking into account transportation cost and customer satisfaction.Studies on this problem are shown as follows:(1)Based on the discussion in different types of customer service time windows,a fuzzy membership function of delivery time is introduced to express customer satisfaction level as well as customer needs.(2)Considering minimizing transportation cost and maximizing customer satisfaction,three multi depot vehicle scheduling models are established.The first model is established with minimizing transport cost as its goal,named Model I;then the second model is constructed with maximizing customer satisfaction as its goal,named Model II;the last one takes into account both the transportation cost and customer satisfaction,named Model III.A minimum service level is set to see the influence provoked in the transportation cost and customer service level.(3)To work out the answer,Particle Swarm Optimization algorithm is proposed to solve the three optimization models.In order to solve multi depot vehicle routing problem,the particle encoding is newly designed based on job shop scheduling problem.Discretizing the particle vector in a sorting way to make the algorithm more effective in solving the models.As to the multi objective model,Multi Objective Particle Swarm Optimization is applied,and a evaluation method based on analytical hierarchy process is presented to select the most rational plan in the pareto optimal solutions.(4)Experiment has been conducted,simulations results show that setting a minimum service level has an impact on the result of transportation cost and customer satisfaction.The average customer satisfaction can be improved by setting a minimum service level,thus reducing the risk of customer loss.The two objectives of Model III,transportation cost and customer satisfaction,have contradictory relationship with each other and can not achieve optimal value at the same time.However,the AHP evaluation method proposed in the paper can help to make a decision.