Generalized Expected Value Model for Stochastic Programming and its Application in Vehicle Routing Problems

In this paper, propose a quasi-linear pattern based on expectation and variance to process the random constraints, after analyzing the essence of stochastic programming and the deficiencies of existing methods. Give a stochastic programming pattern (generalized expected value model) with good operability, and establish its corresponding generalized expected value model in stochastic vehicle routing problem. Its performance is then discussed through an example. All these indicate that this generalized model generalizes the existing methods to some extent, and can effectively solve the stochastic vehicle routing problem with unknown distribution of the random variable under random environment. It is worth to point out that the solution reflects the consciousness of the decision maker, so it enriches the theory and methods of stochastic programming.