Transient Analysis of Downtime Event in Manufacturing Systems

Abstract

In manufacturing industry, downtimes have been considered as major impact factors of production performance. To improve the system performance in real-time and properly allocate limited resources/efforts to different stations, it is necessary to quantify the impact of each station downtime event on the production throughput of the whole transfer line. Complete characterization of the impact requires a careful investigation of the transients of the line dynamics disturbed by the downtime event. We study in this paper, the impact of single isolated downtime event on the performance of inhomogeneous serial transfer lines. Our mathematical analysis suggests that the impact of any isolated downtime event is only apparent in the relatively long run when the duration exceeds a certain threshold called opportunity window. The size of the opportunity window not only depends on the initial buffer levels but also depends on the location and the processing speed of the slowest station. Analytical solutions of the opportunity window and the loose upper bounds for the recovery time are also provided in the paper as two separate theorems. Despite the fact that the paper focuses on the transient analysis of single isolated downtime event, the results are applicable to opportunistic maintenance scheduling, personnel staffing and downtime cost estimation.