Otimização de processos em uma montadora automotiva: eficácia geral do equipamento (OEE) através da análise de dados com Python

Abstract

The industry has evolved with new work approaches and increased the relevance of integrating advanced data analytics and machine learning techniques in engineering, in search of competitive prices by eliminating waste to increase profits. Therefore, this work addresses the integration of Overall Equipment Effectiveness (OEE) in a stamping process in an automotive assembly plant through data analysis with Python. OEE is a crucial metric in industrial engineering responsible for measuring overall effectiveness using three components: availability, performance and quality. Python is a tool with comprehensive data processing and machine learning capabilities. The methodology adopted involves pre-processing data, calculating OEE components and applying machine learning techniques to predict and improve operational efficiencies. This process aims to calculate OEE and establish the basis for predictive maintenance and process optimization strategies. The use of machine learning allows you to analyze and interpret data from machines and processes, enabling failure predictions, optimizations of maintenance schedules and improvements in operational efficiency. The results show which parts demonstrate greater efficiency in their production, the OEE of general production and an example of prediction of 10 batches of parts. In this way, the study contributes to the transformative potential of data sciences in mechanical engineering setting a precedent for future research and applications in industry.