Automating quality control through an expert system

In this article, we present findings from an interventional study conducted within a small enterprise in northern Italy, focused on automating quality control in press-in operation for the production of reduction gearboxes. Guided by Organizational Information Processing Theory, we developed an expert system to automate quality control and facilitate early fault detection. This novel approach enhances quality control within this production stage and could potentially impact other levels of the supply chain. We contribute to the theory by providing a revised version of the Organizational Information Processing Theory framework which integrates technological advancements and variability of the task over time as critical factors affecting information processing, and shows the iterative nature of the digitalization process in SMEs. Operationally, the solution increases defect identification from 6% at end-of-line to 15% through step-by-step checks. It provides a cost-effective, practical example of AI-driven quality control, advocating for data-driven decision-making demonstrating a scalable pathway for SMEs to adopt AI with limited resources.