Model-based training for manual procedures in automated production systems
The research paper has been published in the IFAC/Elsevier Mechatronics journal
Maintenance engineers deal with increasingly complex automated production systems, characterized by increasing computerization or the addition of robots that collaborate with human workers. The effects of changing or replacing components are difficult to assess since there are complex interdependencies between process parameters and components. The introduction of models that describe such dependencies into training systems could support the understanding of these interdependencies and the formation of a correct mental model of a maintenance procedure and the machine and thereby improve the training success. This paper proposes a model-based training system that introduces domain-independent SysML models that formalize such dependencies. The training system consists of a virtual training system for initial training and an online support system for assistance during the procedures. The on- and offline training systems visualize the state of the machine at a certain step of the procedure using structural SysML models. An evaluation of the system against a paper-based manual validated the motivations and showed promising results regarding effectiveness, usability and attractiveness.