Systematic Approach to Develop a Flexible Adaptive Human-Machine Interface in Socio-Technological Systems

Modern automatic machines in production have been becoming more and more complex within the recent years. Thus, human-machine interfaces (HMI) reflect multiple different functions. An approach to improve human-machine interaction can be realised by adjusting the HMI to the operators' requirements and complementing their individual skills and capabilities, supporting them in self-reliant machine operation. Based on ergonomic concepts of information processing, we present a systematic approach for developing an adaptive HMI after the MATE concept (Measure, Adapt & Teach). In a first step, we develop a taxonomy of human capabilities that have an impact on individual performance during informational work tasks with machine HMI. We further evaluate three representative use cases by pairwise comparison regarding the classified attributes. Results show that cognitive information processes, such as different forms of attention and factual knowledge (crystalline intelligence) are most relevant on average. Moreover, perceptive capabilities that are restricted by task environment, e.g. several auditory attributes; as well as problem solving demand further support, according to the experts' estimation.