Evaluating the Relationship between System Output and Station Reliability in Serial Production Lines: A Methodological Approach

Accurately predicting system output performance is essential for manufacturing system design. However, traditional approaches that rely solely on the reliability of individual workstations may lead to inaccuracies, especially for a serial production line that includes several workstations with low reliability. This study investigates how prediction inaccuracies arise from intricate interactions among workstations, as revealed through discrete event simulation (DES). To resolve this issue, a novel methodology is proposed, leveraging DES and dual regression analyses to develop an adjustment factor. This paper details the application of this methodology across a range of reliability levels (85–99%) and workstation counts (5–10), conducting sensitivity analyses and validating the results. This approach offers a practical tool for improving the efficiency and precision of a system output prediction in design contexts. Additionally, the paper discusses the methodology’s limitations and suggests avenues for future research.