Mining Standard Operation Times for Real-time Advanced Production Planning and Scheduling from RFID-enabled Shopfloor Data

Abstract

Production planning and scheduling require standard operation times (SOTs) which have been obtained from time studies or based on past experiences. Wide variations exist and frequently cause serious discrepancies in executing plans and schedules. Radio frequency identification (RFID) technology has recently been applied to create a real-time ubiquitous manufacturing environment, where real-time shopfloor operational data about men, machines, materials, and orders could be captured and collected. Such data carry invaluable information and knowledge which might be used for supporting advanced production planning and scheduling (APS). APS usually needs precise SOTs for perfect decision-making within the RFID-enabled real-time ubiquitous manufacturing environment. This paper proposes a data mining model to estimate realistic SOTs and their standard deviations from RFID-enabled shopfloor data. Key impact factors on SOTs are examined, including working shifts, different machines, gender, and technology complexity. It is observed that working shifts and the learning curves of three types of operators (junior, intermediate, and senior) greatly influence the SOTs. The other factors have minor affection in this case. Considering the two significant impact factors, precise and reasonable SOTs could be worked out, aiming at improving the quality and stability of production plans and schedules.