Enhancing circular economy in reconditioned spare parts through artificial intelligence and genetic algorithms

Abstract

The increasing global focus on sustainability and resource conservation has spurred industries to adopt circular economy (CE) principles, particularly in areas with significant environmental impact, such as spare parts management for industrial maintenance. This study explores a multi-objective optimization approach to maintenance scheduling, aiming to balance reliability, cost, environmental impact, and circular economy contributions when selecting between new and reconditioned spare parts. We propose a robust, AI-driven model based on genetic algorithms to optimize these criteria simultaneously, generating a set of Pareto-optimal solutions that highlight the trade-offs among cost, reliability, carbon footprint, and resource reuse. By integrating reconditioned components, the model enables notable reductions in environmental impact and cost, though with a slight compromise in reliability. This highlights the importance of well-defined thresholds and strategic decision-making. The approach empowers stakeholders to adopt tailored maintenance solutions that align with both economic objectives and sustainability goals. Empirical results reveal significant improvements in emissions reduction and waste minimization, validating the feasibility of incorporating CE principles into spare parts logistics without compromising operational performance. Overall, this work delivers a comprehensive, data-driven strategy that supports sustainable maintenance by aligning circularity goals with industrial constraints and decision-making priorities.