Abstract
Artificial Intelligence plays a major role in the development, implementation, and diffusion of advanced manufacturing applications. To support scholars and practitioners from industry in gathering knowledge in this field, Learning Factories play a crucial role, representing the joining link between theory and practice, between academia and industry. The Smart Mini Factory Laboratory of the Free University of Bozen-Bolzano welcomes students and industry professionals to discuss these topics, among other. This work presents the development of a visual inspection system based on computer vision to achieve image classification and Machine Learning techniques to achieve path planning optimization. The inspection station has been deployed within a lab-scaled Digital Twin smart manufacturing environment. Among other machinery and robots, an intelligent transfer system manages the motion shuttles that convey semifinished products manually loaded by a human operator in chaotic way. With this solution, the human operator who loads the shuttles does not have to follow a precise loading sequence, decreasing the risk of errors. In a Digital Twin-managed context, this also enables dynamic reconfiguration of normal operations at run time. The presence of this workstation within the Learning Factory allows teachers and professor to deconstruct its functioning principle and exploit it to spread knowledge about a variety of cutting-edge technologies and methods characterizing advancement of the contemporary manufacturing industry.