(in German)

The scope of this PhD-Thesis was to elaborate the originally qualitative cognitive task load model (Neerincx, 1995 und 2003) into a quantitative model to predict mental workload in process control tasks. The quantitative cognitive task load model was verified, validated and optimized.

In industry 4.0 settings, communication networks between materials, machinery, and a global optimizing production regime are growing. Furthermore, process control as supervisory control tasks including diagnosis of disturbances is gaining more and more attention and takes effect on workplaces in control rooms as well as in context of industy 4.0. This situation poses two questions in early stages of the design process of tasks, work equipment, and workplaces:

1. With how many parallel tasks are operators able to cope? and
2. How does the design of work equipment, e. g. human-machine-interface, affect the quantitative workload?

To answer these questions, the original cognitive task load model was quantitatively elaborated with emphasis on conformity to international standards and improvement of the prediction algorithm by reducing systematic error sources as well as employing efficient fuzzy logic algorithms. A fuzzy inference system was parametrized based on the theoretical foundation of the original cognitive task load model. Further on, the fuzzy inference system was verified and validated. For its verification the data of a field study was used. The data of a lab study with 29 operators was designed for validation. The operators were asked to perform 41 supervisory control tasks. Along with performing the different tasks, mental workload was recorded multi-dimensionally using subjective, objective, and various performance data. At the end, examples of how to use the elaborated fuzzy cognitive task load model to inform the design process of tasks and work equipment were deducted.

Please download the complete report "Development of an analytical model for predicting mental workload in process control" (in German).