Models for reliability assessment of reconfigurable manufacturing system regarding configuration orders

Reliability performance is a crucial indicator in manufacturing systems. Different configurations of a manufacturing system can have profound impacts on its reliability performance. However, there has not been many in-depth studies showing how the two factors interact. This paper proposes a novel framework that can apply to scenarios with different failure distributions to explore how configuration orders impact reliability performance. Reliability assessment models are built for configuration orders of Reconfigurable Manufacturing Tools (RMTs) in a Reconfigurable Manufacturing System (RMS), where the reconfiguration process is defined by Markov states and piecewise-defined failure rate. First a model of component failure mode following Poisson process is presented which concluded that reconfiguration order does not impact the reliability performance in terms of failure rate, reliability function and unreliability function. Further, two case studies are presented: a Monte Carlo simulation for further verification, and a demonstration of model application. Then a model of component failure following Weibull distribution is proposed with failure rate continuity assumption, relationship between Weibull parameters and processing parameters is mathematically calculated. The relationship contributes to modeling of causalities among configurations. A case study demonstrates that configuration order has a significant impact on the reliability performance when failure mode follows Weibull distribution.