Research on Master Production Schedule Optimization of Fast-Moving Consumer Goods Companies Based on the APO-NSAGA Algorithm

Abstract

This study addresses several issues related to the Master Production Schedule (MPS) in companies, such as over-reliance on manual scheduling by planners, failure to consider the overall perspective of the company, lack of consideration for the impact of seasonal demand fluctuations on the MPS, overlooking potential losses due to stockouts and the resulting loss of market share, inability to adjust the MPS in a timely manner to respond to complex and rapidly changing market demands, and the suboptimal outcomes of the manually compiled production plans. To address these problems, an MPS model is established with the goals of optimizing equipment utilization balance, production costs, inventory costs, and delay costs while comprehensively considering the annual product demand. Subsequently, the artificial protozoan optimization nondominated sorting adaptive genetic algorithm (APO-NSAGA) is developed based on the artificial protozoan optimization (APO) and the non-dominated sorting genetic algorithm (NSGA-II). In this algorithm, autotrophic, heterotrophic, and dormancy behaviors are used to enhance the local search capability and guide the algorithm’s evolutionary process. A self-adaptive crossover and mutation operator, which adjusts according to the number of iterations, is designed to allow the algorithm to converge faster in the early stages of operation and better preserve population diversity and high-fitness individuals in the later stages, thus improving algorithm performance. Finally, through simulation examples, an analysis is conducted on the monthly production volume of the company’s three major product categories, as well as the monthly production, available sales, and demand quantities, end-of-month inventory, stock-to-sales ratio, and inventory turnover rate of five typical products. The results demonstrate that the MPS obtained using the proposed model and algorithm achieves comprehensive optimization.OPEN ACCESS Received: 22/07/2024 Accepted: 01/11/2024 Published: 07/04/2025

Document