| Title |
Analyzing the Experience of Biological Process Experts Using XAI |
| DOI |
https://doi.org/10.5370/KIEE.2025.74.5.942 |
| Keywords |
ANN; XAI; SHARP; Expert Operation Experience; Wastewater Treatment System |
| Abstract |
Biological processes are characterized by complex interactions between environmental conditions and process variables, and therefore the operational experience of skilled experts plays an essential role in process optimization. However, such expert experience is often conveyed as implicit rules and is often not explicitly systematized, requiring a data-based verification process. In this paper, in order to verify the expert operation rules of biological processes, we modeled a biological process based on an artificial neural network and verified the expert operation rules from this model using the SHARP technique of XAI. Through this, we aim to determine whether the direction of the expert operation rules is appropriate and what additional factors should be considered. The analysis results using the XAI technique through simulation using three years of data from a sewage treatment plant using a biological process in the metropolitan area showed that the expert operation rules of biological processes are basically in the direction and that additional supplementary elements of expert operation rules are needed.
These results of this paper are believed to provide a reliable basis for experts when applying ANN models to biological process operations, and can contribute to improving the transparency of AI-based decision-making systems. In addition, it was confirmed that it is possible to further supplement expert operating rules. |