Conceptual Framework for Development of Intelligent Control Systems for Thermoplastics Injection Molding

Abstract

Injection molding is one of the most widely used processes for production of plastic parts. It allows manufacturing of products of various shapes and sizes with high efficiency and production rates. However, setting up process parameters for manufacturing high-quality items and no in-process variation is often a tricky task, which is usually solved using the trial-and-error method and the experience of injection molding machine operators. Such an approach has several disadvantages, as it requires a significant amount of time, also it might lead to production of scrap. An intelligent monitoring and control system for thermoplastics injection molding can be implemented to avoid this. The article presents a conceptual framework for development of such system using a systems engineering (SE) approach. The system would allow the collection, storage and analysis of injection molding process data in order to predict the quality of produced parts without the need for the application of the trial-and-error method. Machine Learning methods are proposed to be a part of the computational core. The article lists stakeholders’ needs relevant for the system, defines the problem at hand, introduces what needs to be done and what has been done to find a relevant solution. Lack of use of the SE during design and development of similar systems might lead to creation of a system that is not workable, lacks required functionality or parts of it fail to work together.