enhancing part identification and maintenance efficiency in the steel industry using artificial neural networks

In the steel industry, accurate identification and tracking of parts are critical for efficient inventory management and operational planning. an innovative approach using artificial neural networks (ann) has been developed to identify parts with multiple identification codes. this method not only enhances identification accuracy but also significantly reduces time and cost, making it highly effective for maintenance operations, including preventive maintenance (pm), corrective maintenance (cm), and emergency maintenance (em). a synthetic dataset representing various parts with unique and multiple identification codes was generated and used to train an ann model to predict the relationship between these codes and the parts they represent. the results demonstrate high accuracy in identifying parts, highlighting the potential application of this method in improving inventory management and operational efficiency within the steel industry. to evaluate the performance of the network, 15 datasets, each containing 100,000 data points, were utilized.