A Strain Based Model for Adaptive Regulation of Cognitive Assistance Systems—Theoretical Framework and Practical Limitations

In order to manage increasing complexity so called cognitive assistance systems are integrated into assembly systems. On the basis of real-time measurement and analysis of physiological signals, these assistance systems help to coordinate efficient behavior and to prevent states of long lasting detrimental workload and strain. With measurement technology getting smaller, more powerful and wearable it’s possible to collect and analyze personal physiological data in real-time and detect significant changes at the workplace. It is intended to use these data to control a cognitive assistance systems which as a consequence of a monitored detrimental workload leads to adaptive changes in assembly processes and to a reduction of workload. The underlying principle can be a self-actualizing machine learning algorithm. We want to present a theoretical framework to sketch possibilities of such data-controlled, adaptive systems and to describe some obstacles which have to be overcome before they’re ready for use.