Integrated AI Models for Simultaneous Quality Improvement and Risk Reduction in Production Processes

Abstract

The application of artificial intelligence in the manufacturing sector, over the last years, transformed it, with better risk mitigation about the production system as well as its quality. This research is proposed to find an integrated AI-driven method that applies the use of various machine learning (ML) algorithms, namely SVM, Random Forests (RF), LSTM, and DRL, to address real-time operation risks in order to enhance the quality control for the manufacturing processes. High defect rates and unscheduled downtime are most often the result of traditional approaches to risk management and quality assurance failing to adjust to dynamic production systems. However, AI models' predictive powers and data-driven decision-making procedures make proactive control of production line inefficiencies possible. The research paper describes the application of the models SVM, RF, LSTM, and DRL for predicting defects through production data as well as in optimizing machine maintenance to ultimately make production processes efficient. Experiential results show that, compared with more conventional techniques, the AI system is significantly accurate, able to manage risks efficiently, reduce defect rates, hence increase outputs and lower running costs. All of these algorithms become embedded in industrial systems and thus shape the next form of intelligent manufacturing. The research, therefore, underlines the importance of AI for better sustainability and quality in manufacturing processes by focusing on both quality control and risk mitigation.