Advancements in Optimization for Automotive Manufacturing: Hybrid Approaches and Machine Learning

Abstract

This paper addresses the need for innovative optimization solutions in automotive manufacturing. Through advanced algorithms, we review existing methods and introduce novel approaches tailored to this sector. Our literature review identifies gaps and limitations in current methodologies. We define a specific optimization problem within automotive manufacturing, emphasizing its unique challenges. Our key contributions include: (a) Exploring hybrid optimization algorithms, combining genetic algorithms with simulated annealing for a 15% improvement in convergence speed, (b) Integrating machine learning techniques, resulting in a 20% reduction in optimization error compared to static settings, (c) Incorporating multiobjective optimization, achieving a 25% improvement in simultaneous cost and efficiency optimization, and (d) Proposing dynamic optimization algorithms, reducing decision-making latency by 30% during rapid environmental changes. Case studies demonstrate practical application, with quantitative results highlighting the superiority of our approaches over traditional methods. Additionally, the data analysis was conducted using Python, contributing to the robustness and accuracy of our findings.

Article history: Received (March 21, 2024); Revised (April 24, 2024); Accepted (July 3, 2024); Published online (August 23, 2024)