随机需求的车辆路线问题的新模型

本主要研究随机需求的VRP问题，其中服务需求量满足二项式分布，根据期望值的大小我们提出了在一条路线上理想最大服务点数的新概念，并在此基础上建立了三种VRP问题的新模型，由于允许服务失败两次和部分服务使得模糊能适应多种实际问题，以模拟退火思想为基础的两阶段方法经修正后用于解新模型并取得较好的数值结果，理论分析和数值结果表明，新模型较好地描述随机需求的VRP问题，并且容易求解。

New Models for the Vehicle Routing Problems in the Case of Stochastic Demand

The vehicle routing problems(VRP)in the stochastic demand is studied in this paper,where the demand at service nodes is subjected to the binomial distribution.According to the expected value,we develop three new models for VRP in the case of stochastic demand.Because the two route failures and partial services are allowed in the node,the new models are suitable for many kinds of the practical problems.The simulated annealing algorithm for solving VRP is modified and is used to solve the new models.Theoretical analysis and numerical results show that the new models describe VRP well in the case of stochastic demand and the solution is easy to find.