在竞争环境中的拥塞设施截流选址问题

Flow interception location problem of congestion facilities in competitive system

研究了在竞争环境下考虑到顾客等待时间的截流选址问题.产生于网络路径上的顾客流按一定概率分布来选择各个设施,概率分布由设施容量,偏移原路径距离和排队等待时间共同决定.在已有设施的情况下,通过选址决策使新建立的设施截得的顾客流最大.结合贪婪算法和Dogleg路径信赖域方法来求解该问题,经过大量计算试验,表明该方法求解质量和求解速度都较理想,可应用于大规模的网络选址.

This paper studied flow an interception location problem with consideration of customers＇ waiting time under competitive surrounding. Customer flows on the paths obtain service from facilities according to a probabilistic distribution, which is determined by facility capacities, deviation distance from the given path and customer waiting time. With some facilities existed, the objective is to maximize the customer flows intercepted by new facilities through their locations. We employ greedy heuristic and trustregion dogleg method to solve this model. Through a great of computation tests, we find this method is efficient and can be applied to large network location.