Scheduling optimization for production of prefabricated components with parallel work of serial machines

Scheduling optimization of manufacturing prefabricated components is critical for construction in enhancing efficiency and productivity. However, problems of low formulation efficiency and low resource utilization still exist, impeding productivity and increasing costs. Although numerous efforts have been devoted to scheduling optimization, limited research has considered the parallel work of serial machines in production lines, leading to a deviation between actual situations and expectations. Therefore, a new optimization method is proposed with two major contributions. First, a Prefabricated Components Production Scheduling (PCPS) model considering the parallel work of serial machines is established, containing 3 new constraint conditions. a Position Algorithm (PA) is also designed to facilitate a constraint related to the locations of component groups. Second, a genetic algorithm-based method is designed to seek potential optimal scheduling schemes. The model is demonstrated and evaluated in an actual case. The results show that: (1) Optimal schedules calculated by the PCPS model can reduce penalty cost by 28.6% and total completion time by 14.9% compared to the case without considering the parallel work of serial machines. (2) Optimal schedules can effectively reduce penalty cost by 63.9% and total completion time by 34.4% compared to the empirical method of ascending order by the due date. The optimal schedules can provide a solution for more detailed and practical scheduling. For future improvement, other factors affecting productivity, such as operator proficiency, will be added to the model constraints.

Automated summary

Scheduling optimization of manufacturing prefabricated components is critical for enhancing efficiency and productivity in construction. A new optimization method is proposed, which establishes a Prefabricated Components Production Scheduling (PCPS) model considering the parallel work of serial machines and designs a genetic algorithm-based method to seek potential optimal scheduling schemes.