Deep Learning and Neuro Evolution Method for Enhancing Productivity Analysis in Indonesia's Automotive Industry

This study explores the application of  Deep Learning and Neuro Evolution to enhance productivity analysis in Indonesia's automotive industry. Deep Learning and Neuro Evolution an advanced AI method, is employed due to its superior capability in automating the selection, optimization, and tuning of machine learning models, making it the most suitable approach compared to other AI methods developed worldwide. By leveraging Deep Learning and Neuro Evolution, we aim to identify and optimize the key factors affecting manufacturing productivity, including foreign direct investment, energy use, gross fixed capital formation, research and development expenditure, and total labor force. Using data spanning from 2003 - 2022, the automated approach facilitates the handling of complex and large datasets, ensuring a comprehensive analysis of how these variables impact value added per worker in the industry. The results indicate that Deep Learning and Neuro Evolution models outperform traditional methods, yielding a Mean Squared Error (MSE) of 0.012 and a Mean Absolute Percentage Error (MAPE) of 1.6 %. These findings provide actionable insights for policymakers and industry stakeholders to foster a more productive and competitive automotive sector in Indonesia.