Optimal Decision-Making Process for Maintenance and Service Management System in the Manufacturing Industry

This real-time case study research employs a non-traditional optimization technique to enhance maintenance operations in small and medium enterprises (SMEs) to minimize investment costs. The study improves machine maintenance parameters through the application of a Genetic Algorithm (GA) to determine optimal solutions, thereby facilitating an effective decision-making framework for maintenance and service operations. Furthermore, the research analyzes variations in machine availability under both faulty and ideal operating conditions in SMEs using the Markov Decision Model. Mathematical formulations are developed based on the transition state diagrams of individual subsystems within the manufacturing plant, utilizing first-order differential equations. The primary objective of this research is to identify critical subsystems within the manufacturing plant and implement an optimal maintenance management system through a structured decision-making approach. The results obtained from the proposed algorithm indicate that machine availability in SMEs can be increased to 73.25% under faulty conditions, while system availability can reach 76.17% under ideal operating conditions. Based on these findings, a novel optimal decision support system is proposed for preventive maintenance management in SMEs, aiming to achieve maximum productivity with minimal maintenance and service investment through optimized planning and scheduling processes.