Exploring smart quality predictive modelling approach: a case study of the injection-molding industry

Technology-driven quality monitoring and control can effectively predict, prevent and reduce defects in the manufacturing industry and improve productivity. The study aims to explore real-time injection molding process monitoring and demonstrate intelligent quality control through a case study. The process environment is monitored to capture variabilities through which the relationship of variables with the quality characteristics of molded parts is derived. The defects are represented as a function of process variables using statistical analysis of the past process and product data employing appropriate machine learning methods. From the fitted models, decision rules are retrieved, and desirable process conditions required for making defect-free molded parts are recommended for quality control practice. Further, these models are deployed to predict the defects in the parts during production by observing the real-time process conditions in the manufacturing process environment. This study has drawn significant research and practical implications for the manufacturing industry as it can effectively control quality and automatically fine-tune the process for better quality and productivity.