The Study Of Half-open Vehicle Routing Problem With Multi-depot And Multi-vehicle

The vehicle routing problem is a typical combinatorial optimization problem, which is to reach the goal of objective function optimization in the service of transportation operations by the way of optimizing the vehicle routing based on the system optimization. With the accelerated pace of the economic requirements of the proposed transport operations services diversification and in-depth study of vehicle routing problem, vehicle routing problem spawned a myriad of species and made huge achievements, also more step closer to the scene what the actual transport services are facing.However, the vast majority of scholars is focus on the objective function optimization of the current transport operations and ignored the non-time characteristics of transportation services in a large number of research results. Moreover, they ignoring that the objective function optimization of the current transport operations perhaps make inconvenience to the next cycle transport operations,which possible take adverse effects to the cost control of long-term.This paper proposed a improved concept of half-open vehicle routing, which is bring the minimize the impact on the next cycle of transport operations to consider based on minimize the cost of transport operations services in current cycle. In other word, we have another objective function than stander vehicle routing problem, which is minimize the variance between the expected shipments of distribution center in next cycle and transport capacity of distribution center. This paper established a mathematical model of multi-objective optimization based on multi-distribution centers and multi-vehicle, which is named half-open vehicle routing problem mathematical model whit multi-distribution centers and multi-vehicle. Applied the concept of dominated solutions and Pareto solutions on standard particle swarm optimization, designed a multi-objective particle swarm optimization algorithm to solve the problem that proposed in the paper. Solve a classic example of vehicle routing problem rely on the advantages of particle swarm optimization algorithm that have small dependence on the initial solution and easy to control the simple parameter to improve the feasibility of routing strategy, the validity of the mathematical model and solve the final solution.