Capacity Allocation in Queueing Networks

Xuefeng GAO (ISYE) (Prof Ton Dieker (ISYE). The objective of the proposed research is to develop methods to do capacity optimization in a large feedforward queueing network. That is, given a cost constraint on service personnel, how do we allocate service capacity in the network in order to minimize our objective function, which is the total number of customers in the system? A feedforward network is a network where we can number the stations in an increasing order and customers always flow from lower numbers to high numbers of stations. This problem is challenging. We have no explicit formulas for the objective as a function of the capacity allocation vector in general. Besides, numerical methods and simulations suffer from high computational cost especially in the case of large network. Therefore, it is our goal to devise efficient and provable algorithms to tackle the capacity allocation problem in large queueing network setting.