การจัดตารางการผลิตที่เหมาะสมที่สุดสำหรับโรงงานผลิตอาหารพร้อมบริโภค

Abstract

Optimal production scheduling is a key method for planning and improving the efficiency of manufacturing processes. Scheduling in a flexible job-shop system (FJS) is more complex than traditional job-shop systems because each type of product has different manufacturing operations which can be processed on multiple machines, making the problem more intricate. This research presents a mathematical model of the flexible job-shop scheduling problem with sequence-dependent setup times (FJSP-SDST) with the objective of minimizing the makespan. Testing the problem with an exact method approach proved infeasible for large-scale problems due to its NP-hard nature, making it impractical within practical time limits. Therefore, this research applies the Differential Evolution (DE) algorithm, a popular and efficient metaheuristic, to find solutions. After testing and fine-tuning suitable parameters for the FJSP-SDST problem, it showed that the 'current-to-best/1/bin' strategy performed better than other strategies. As a result, this research applies the DE algorithm to the production scheduling problem in a case study of a food manufacturer and distributor with a flexible job-shop production system. The results showed that, on average, the DE method reduces the total makespan time by 36.07% compared to the company's traditional scheduling approach for all case study problems.