Discovering Dispatching Rules in a Semiconductor Fab Using Interpretable Machine Learning

Recent studies have been conducted in the application of machine learning (ML)-based dispatching methods. Unfortunately, the internal dispatching behavior of such ML-based models is difficult to interpret. Therefore, this study transforms the ML-based model to a rule-based dispatching model that is fast and interpretable. An ML-based dispatching model is first trained using job-pair data. The model is then transformed to a rule-based dispatching model by identifying the rules through a post-hoc interpretable algorithm called RuleCOSI+. The proposed method is evaluated using a dataset that was obtained from a commercial scheduling engine used in semiconductor fabs. The experimental results showed that both ML-based and rule-based models could obtain exactly the same dispatching results as the original dispatching rules, but the rule-based model was faster and more interpretable than the ML-based model.