Develop A Novel Cluster-Based Framework For Improving Steel Production Quality

Abstract

The steel sector has had difficulties in finding solutions for quality control of goods using data mining methods, notwithstanding recent progress. “This study presents a steel quality prediction system that integrates real-world data with in-depth data analysis conclusions. The main process is carefully designed as a regression problem, which is therefore best handled by integrating various learning algorithms with their huge repository of historical production data. A comprehensive examination and comparison of the characteristics of the most often utilized learning models in regression problem analysis has been conducted. The efficacy of our steel quality control prediction system, which utilizes an ensemble machine learning model, showcases promising outcomes. This system offers great usability for local businesses in addressing production problems via the use of machine learning methods. Moreover, the practical implementation of this system is shown and analyzed. The proposed method attained high accuracy, precision, recall and f1 score, mean absolute error, root mean square error as compared to other different technique. Lastly, this study highlights the future prospects and sets out the anticipated level of performance.”