Research Topic

In the light of frequently changing system configurations in modular cyber-physical production systems (CPPS), it will no longer be sufficient for human operators to learn pre-defined processes and examples as it was common practice in traditional chemical process industry. Instead, it will be essential to equip human operators with a deep understanding of underlying causal structures and principles, in order to enable successful problem solving in variety of system configurations. Due to short operating times of specific system configurations, operators will not have sufficient opportunity to learn the functioning of every new system configuration from experience. Therefore, the ability to draw on knowledge, skills and strategies learnt in one situation and apply it to a new system configuration is vital in order to deal with the increased flexibility and changeability of the novel systems.

This doctoral project addresses this issue and aims at investigating possibilities to support the acquisition of relevant portable knowledge required for effective problem solving in the domain of modular CPPS in the chemical process industry. For this, I am combining research approaches from experimental cognitive psychology, instructional psychology as well as chemical process engineering.

Deeper learning is a process that facilitates the recognition of when and how new problems are related to previously encountered ones, and therefore constitutes the basis for transfer of knowledge. Experts have been demonstrated to organize their knowledge around underlying principles and causal structures, whereas novices tend focus on salient superficial features (Chi et al., 1981; Rottman et al., 2012). Hence, experts form relational categories, whereas novices organize their knowledge using feature-based categories. Knowledge organization along relational categories facilitates drawing connections across domains and situations, and consequently makes transfer more likely.

• In the first step, it was my aim to identify underlying principles or causal structures relevant in the domain of modular CPPS in the chemical process industry. These are important as they can serve as the basis for expert-like knowledge organization in terms of relational categories. I identified the key causal models common effect, common cause, causal chain, positive feedback system.
• In a next step, I drew on findings of a study by Goldwater and Gentner (2015) that identified learning experiences that support the acquisition of abstract mental representations of key causal models (i.e., common effect, common cause, causal chain, positive feedback system). This encompasses the building of relational categories of the causal models and thus enhanced likelihood of recognizing the causal models across different situations. It was my aim to go beyond that finding and investigate whether the competence to recognize underlying key causal models can facilitate the solving of complex problems that are based on the according key causal models. For this, I combined the intervention with a complex problem solving task.
• In a follow-up study, I explored means to enhance the positive effects beyond the recognition of the key causal models, by linking the causal model categories with knowledge on their dynamic behavior. 

Literatur:
Chi, M. T. H., Feltovich, P. J., & Glaser, R. (1981). Categorization and Representation of Physics Problems by Experts and Novices*. Cognitive Science, 5(2), 121–152. https://doi.org/10.1207/s15516709cog0502_2
Goldwater, M. B., & Gentner, D. (2015). On the acquisition of abstract knowledge: Structural alignment and explication in learning causal system categories. Cognition, 137, 137–153. https://doi.org/10.1016/j.cognition.2014.12.001
Rottman, B. M., Gentner, D., & Goldwater, M. B. (2012). Causal Systems Categories: Differences in Novice and Expert Categorization of Causal Phenomena. Cognitive Science, 36(5), 919–932. https://doi.org/10.1111/j.1551-6709.2012.01253.x