基于几何可靠性机器模型的装配系统实时性能分析

Assembly system is one of the most fundamental production system structures. It can be widely seen in practical manufacturing environments (e.g., automomobile, appliance, consumer electronics). Compared to the extensive investigations of serial lines in the existing literature, investigations on assembly systems, especially transient-based realtime performance analysis, are still largely unexplored. In the framework of assembly systems with three geometric machines and finite buffers, a mathematical model for real time performance analysis of such systems is developed in this paper. The formulas for exact performance analysis are derived by using the Markovian approach first. Then, a decomposition-based computationally efficient algorithm for real-time performance evaluation is proposed. Specifically, closed-form formulas are derived for calculating real-time production rate, consumption rates, as well as work-in-process, completion time on completing a production run, of assembly systems with three geometric machines. Finally, the accuracy of the algorithm is justified by numerical experiments.