Lead Time Analysis of Manufacturing Systems with Time-Driven Rework Operations

Products whose quality features deteriorate with the time they spend in the manufacturing system are common in industry. Usually, they are perishable products that loose their performance or value whenever their sojourn time in the system exceeds a certain threshold. A typical example is the food industry where such products need to be identified and scrapped. In other cases, perishable products can be reworked for restoring their functions when their sojourn exceeds a threshold. For instance, automotive painting processes suffer of dust settling on parts before the drying process and semiconductor wafers risk contamination between the cleaning and heating processes. This work develops an analytical method to evaluate the performance of systems with rework operations triggered by an extended residence time. The approach is based on the exact computation of the lead time distribution in production lines featuring general markovian machines. By exploiting the developed method, specific production policies aiming at minimizing the rework rate and maximizing the effective throughput in such systems are investigated. The method is validated within a real case showing high benefits through industrial applications.