Ontological model of preventive maintenance of industrial enterprise equipment

This paper addresses the problem of formalizing expert knowledge in the field of preventive maintenance of industrial equipment, which arises when interpreting the results of visual inspections taking into account the operating conditions. The focus of the study is the development of an ontological model that provides a logically grounded transition from unstructured descriptions of technical condition indicators to specific equipment maintenance actions. A multi-level ontological model is proposed, including the following levels: inspection, indicators, operating context, technical condition, and action. The results are intended for use in decision support systems for equipment maintenance. The novelty of the study lies in combining contextual parameters (such as equipment age, criticality level, aggressiveness of the operating environment, and others) with a logical inference mechanism, as well as in applying a method for structuring recommendations as a composition of a basic action and context-dependent refinements. The model was built in a Python environment and tested on a specific dataset, where equipment states were automatically determined, adaptive recommendations were generated, and reports were produced. The model showed robustness to incomplete information. The applied ontological approach to technical diagnostics contributed to the achievement of practically significant results for the digitalization of preventive maintenance processes of industrial enterprise equipment.