Change detection and identification in Simple Serial Lines

According to the Industry 4.0 paradigm, Digital Twins of production systems must be always aligned with the real systems to guarantee an effective decision making process in a continuously changing environment. To keep the alignment, digital process models can be updated with Process Mining techniques through data collected by sensors. This thesis proposes a model-update procedure and addresses the issue of detecting and identifying changes occurring in Simple Serial Lines, a particular type of production lines. Using event logs, which are data structures generated by line embedded sensors, KPIs are computed, plotted, and analyzed to get insights of the system behavior. Simulation is used to test the effectiveness of the procedure.