Machine Learning Integration into Automotive Quality Processes: An Innovative Approach for Scrap Reduction and Quality Enhancement at Minebea AccessSolution

Quality is a critical factor in the automotive industry, as meeting high standards is essential for producing safe and reliable products that enhance a company's competitiveness in the global market. To achieve this, many companies are increasingly adopting new technologies in their production and quality control processes. This thesis investigates the application of artificial intelligence in the automotive industry, specifically focusing on the benefits of integrating machine learning techniques with traditional quality control tools. The research was conducted during the internship period at Pianezza site of the Japanese multinational company Minebea AccessSolution, which is specialized in the production of automotive components, particularly door handles. The aim of the activity was to reduce the amount of scrap produced within the assembly department, thereby enhancing company performance and minimizing waste. To this end, the traditional problem-solving process was employed, but it was augmented with the use of machine learning models. Various classification algorithms were implemented, and the most effective one was selected for further analysis. In particular, this optimal model was then used to identify the factor that had the greatest influence on scrap generation. Once this key factor was determined, subsequent analyses could focus more precisely on it and this could lead to great simplifications in the process for identifying the root cause and implementing solutions to address the issue. The study highlights the effectiveness of using machine learning to improve product quality, reduce scrap, and lower production costs, yielding significant benefits for the company.