Performance analysis based Markov theory for hybrid control serial production lines

Abstract

This paper studied a multiple-stage tandem production system. The control strategy is Hybrid which combines Kanban and CONWIP strategy together. The objective here is to evaluate the performance of the system by Markov chain models. We focus on Work-In-Process (WIP) and lost rate of the system. In general it is hard to obtain an exact analytical result of the system. Thus some approximation methods are proposed here. A simple situation is considered and formulated as a Markov chain process. Then we derive the steady-state probability of the system for two and three machines case, and obtain the WIP and the lost rate. The formulations proposed show the connection of control cards and performance measures. The method can extended to the case of multiple machines. We also present the results of a simulation study that tests the performance of our approach. ©2010 IEEE.