Abstract

Modern manufacturing systems face increasing demands for efficiency, quality, sustainability, and cost-effectiveness. Multi-objective optimization (MOO) techniques are essential to balance these often-conflicting goals. Artificial Intelligence (AI) provides powerful tools to enhance MOO by learning complex relationships among process parameters, resource constraints, and performance metrics. This paper presents an AI-driven framework for multi-objective optimization in manufacturing, integrating machine learning models, evolutionary algorithms, and reinforcement learning to optimize production schedules, energy consumption, material usage, and product quality simultaneously. The system leverages historical production data, sensor inputs, and simulation outputs to generate Pareto-optimal solutions and adapt to dynamic shop-floor conditions. Case studies in automotive assembly, additive manufacturing, and electronics production demonstrate improvements in energy efficiency by up to 18%, reduction in production lead time by 22%, and enhanced product quality metrics. The approach enables proactive decision-making, supports sustainable manufacturing practices, and fosters the adoption of Industry 4.0 technologies. Findings suggest that AI-driven multi-objective optimization can transform traditional manufacturing into intelligent, responsive, and environmentally conscious systems, providing a competitive edge in complex industrial environments.