求解动态随机综合运输问题的仿真优化方法

The Simulation-optimisation Approach for Solving the Dynamic Stochastic Coordinated Transportation Problem

解决动态随机综合运输问题的一种有效途径是系统仿真方法与优化技术的结合。根据问题的特点，提出解决该问题的仿真优化结构，给出寻找最优货物调拨方案的最佳步长探测仿真优化算法，并对算法的收敛性进行了证明。实例分析表明，所提出的优化结构及算法十分有效。研究问题的思路能为其他动态随机多目标仿真优化问题的研究提供有价值的启迪。

An effective way for solvig the dynamic stochastic coordinated transportation (DSCT)problem is the combination of system simulation and optimisation techniques. Based on DSCT characteristics, the simulation-optimisation structure for DSCT and the optimisation algorithm, the exploratory method of optimum step (EMOS) of searching the optimal plans of commodity allocations are presented,with the proof of EMOS convergence. It is shown by a real example studies that the structure and algorithm are quite effective. The train of thought researching the problem provides some valuable enlightenment for other kind of dymanic stochastic multi-objective simulation-optimisation problems.