Research On Models And Algorithms For Vehicle Routing Problem Of Third Party Logistics

With the development of the market economy, logistics whose impact on the economic activities is increasingly obvious has received more and more attention. Distribution is an important part of logistics directly connecting to consumers. The quality and the service level of distribution directly affect customer satisfaction with the entire logistics system. The core of distribution is the cargo delivery vehicles, cargo sorting and delivery process. At the same time, the optimization of delivery path is essential for transportation speed, cost and efficiency of the whole logistics system. As an important part of logistics optimization system, the vehicle routing problem has gotten extensive attention.In this thesis, with consideration of the rapid development of third-party logistics, we study vehicle routing problems of third-party distribution. And we divide this problem into two kinds:vehicle routing problem and open vehicle routing problem. Analyzing models, solutions and application under different situations, we expect to offer a theoretical basis and decision support for the development of third-party logistics.Firstly, this thesis briefly introduces the content, features, classification and the current study condition of vehicle routing problem and open vehicle routing problem respectively. And then, after making a summary of the research at home and abroad, the deficiencies of the current study are pointed out. The research situation of heuristic algorithms for solving open vehicle routing problem is also concluded.After that, a case of large-scale vehicle routing problem about tobacco distribution of third party logistics enterprise is studied. The research situation for tobacco distribution problem is introduced. The heuristic algorithm to solve the problem is proposed and the algorithm is tested by real date in the thesis..Secondly, a large-scale vehicle routing problem with depot closed, which means third party logistics enterprise has enough vehicles for distribution service, is analyzed. And a hybrid genetic algorithm is proposed to solve the problem. The procedure to solve the problem is divided two stages. In the first stage, retail stores were divided in different regions by using location based heuristic, such as Genetic Algorithm, Neighborhood Search. In the second stage, order for distribution road was determined by using optimal algorithms.Thirdly, an open vehicle routing problem with time window is analyzed and the mathematical model is proposed with vehicle number uncertainty for distribution service. The model is solved by an improved particle swarm optimization algorithm. The proposed algorithm, in which the initial population was built by using ordinal number coding method based on consumer and evolution equation of particle swarm optimization was improved, make search procedure of particle swarm optimization having self-adaptability. The validity of the model and the algorithm was tested by numerical examples based on real situation of the third logistics and different objective of the problem.Fourthly, an open vehicle routing problem with fuzzy demand, in which third party logistics enterprise must perform distribution service by leasing vehicle and consumer demand is fuzzy number, is analyzed. And the fuzzy chance constrained programming mathematical model is proposed by applying fuzzy credibility theory. The stochastic simulation algorithm and improved differential evolution algorithm were used to solve the complex uncertainty problem. The validity of the model and the algorithm was tested by numerical examples.Finally, the efforts made for this study and the findings in this thesis are summarized briefly. And the issues and direction for further study are proposed.