The construction industry is adopting off-site construction for its superior quality, short cycle time, and minimal environmental impact. Among the different off-site construction methods, panelized construction follows a manufacturing approach where houses are built in smaller panels in an assembly line. Due to the complex and unique nature of the home building process, existing manufacturing concepts such as automated production planning and control system are not applicable to the panelized home production facility. This paper utilizes radio frequency identification (RFID) technology, data mining, and simulation-based optimization to develop a production planning and control system for a panelized home production facility. The RFID system is used to automatically collect the production data; however, the RFID data in this study contains considerable noise. Thus, it is crucial to extract the relevant data from the raw data. A RANSAC model is developed for this purpose to automatically clean the RFID data and then use it to build the simulation model, which is then integrated with an optimization algorithm to generate the production schedule and real-time performance monitoring. The proposed framework cleans the noisy RFID data to extract records that represent a steady flow of assembly line. These clean records are then used to build PDFs that are required to develop discrete event simulation model that can help managers to capture an almost real-time snapshot of the production of the assembly line as well as automatically optimize the production schedule. The proposed system is implemented and validated in a panelized wall production facility operated by ACQBUILT Inc., a prefabricated home builder in Edmonton, Canada.