Lead time prediction using machine learning algorithms: A case study by a semiconductor manufacturer

The accurate prediction of manufacturing lead times (LT) significantly influences the quality and efficiency of production planning and scheduling (PPS). Traditional planning and control methods mostly calculate average lead times, derived from historical data. This often results in the deficiency of PPS, as production planners cannot consider the variability of LT, affected by multiple criteria in today's complex manufacturing environment. In case of semiconductor manufacturing, sophisticated LT prediction methods are needed, due to complex operations, mass production, multiple routings and demands to high process resource efficiency. To overcome these challenges, supervised machine learning (ML) approaches can be employed for LT prediction, relying on historical production data obtained from manufacturing execution systems (MES). The paper examines the use of state-of-the-art regression algorithms and their effect on increasing accuracy of LT prediction. Through a rea l industrial case study, a multi-criteria comparison of the methods is provided, and conclusions are drawn about the selection of features and applicability of the methods in the semiconductor industry.