Mutualistic and Adaptive Human-Machine Collaboration Based on Machine Learning in an Injection Moulding Manufacturing Line

This paper proposes an adaptive human-machine collaboration paradigm based on machine learning. Human-machine collaboration requires more than letting humans and machines interact according to fixed rules. A decision-maker is needed to assess production status and to activate adaptations that improve productivity and workers’ well-being.

The proposed solution has been tested in an injection moulding manufacturing line. By introducing a physiological monitoring system and a smart decision-maker, relief from fatigue and mental stress is pursued by adjusting the level of support offered through a cobot. Results reported a reduction of operators’ physical and mental workload as well as productivity increase.