An integrated quality and maintenance model for two-unit series systems

Abstract Statistical process monitoring and maintenance are two major concepts to increase system reliability and conforming products. To simplify mathematical modeling, the most of existing researches in the literature suffer from three significant disadvantages: (1) statistical process monitoring and maintenance often have been implemented on the single unit systems, in spite of the fact that most of real production processes consists of multiple units; (2) they apply control charts with fixed parameters that are slow in detecting small and moderate shifts; and (3) they design control charts by minimizing the quality cost ignoring statistical properties that decreases extremely the power of control chart. To overcome these drawbacks, this research integrates the economic-statistical design of a VP-Shewhart chart with condition-based maintenance for two-unit series systems. The particle swarm optimization algorithm is applied to minimize the expected total cost per time unit subject to specific statistical constraints. Eventually, the cost savings from monitoring the system with the flexible VP-Shewhart chart are compared to the FP-Shewhart chart through several instances based on a case study. Also, a sensitivity analysis is implemented on the decision variables to extend insights into the matter. The attained results demonstrate that applying the VP-Shewhart chart instead of FP-Shewhart chart leads to a significant increase in the cost savings.