Modern chemical plants are increasingly complex, placing significant demands on operators to diagnose faults effectively. This study aims to incorporate human fault diagnosis abilities into a computational approach, providing a software tool to aid operators in fault diagnosis scenarios. The Abstraction Decomposition Hierarchy offers a suitable two-dimensional representation of system functionality and part-whole relationships, aligning with human-machine interaction principles. The Tennessee Eastman Process was selected as a benchmark system. To formally represent the Abstraction Decomposition Hierarchy of the process, OntoCAPE ontology was selected to model both the structural and behavioral aspects of the process. The developed Knowledge Graph was imported into Python, and queries were used to generate PFD, P&ID, and process behavior graphs. These graphs formed the basis for a software tool named PlantGraphExpert. This tool enables users to explore process flow, filter component types, navigate plant layouts, analyze control loop components, and investigate relationships between physico-chemical phenomena and process parameters. The results indicate how ontological modeling, combined with graph-based representations, can be used in representing complex industrial environments and form a foundation for advanced decision-support tools.