Multistage self-adaptive decision-making mechanism for prefabricated building modules with IoT-enabled graduation manufacturing system

Prefabricated production Planning, Scheduling, and Execution (PSE) is crucial for the on-time delivery of prefabricated modules in Modular Prefabricated Construction (MPC). In prefabricated production, various uncertainties decrease the effectiveness of production PSE, increasing construction periods and costs. However, a stable and resilient fixed-position production mode with high applicability that meets the characteristics of prefabricated production has yet to be realized. In the context of Industry 4.0, a Graduation Intelligent Manufacturing System (GiMS) is described to facilitate the flexible handling of uncertainties in real time. Internet of Things (IoT)-enabled prefabricated production is designed to achieve real-time visibility and traceability. A multistage self-adaptive decision-making mechanism comprising spectral clustering for planning and real-time ticket sequencing for scheduling and execution is developed to improve the performance levels of PSE decisions under large fluctuations in operation time. A numerical study is conducted to evaluate the performance of the GiMS. According to the results, >70% improvement in tardy job reduction and a 16.1% average decrease in mold table idle time are achieved relative to two traditional methods. Sensitivity analysis is carried out to investigate the impacts of fluctuations in the operation time and the number of mold tables. GiMS possesses excellent potential to cope with uncertainties, needing up to four mold tables for modular prefabricated production. This proposed methodology fills the gap in fixed-position prefabricated production PSE and increases the applicability of the decision-making mechanism in real construction projects.