Physics-Informed Knowledge-Driven Decision-Making Framework for Holistic Bridge Maintenance

Abstract

Bridge maintenance is a highly intricate task that involves considering a wide range of factors in order to achieve optimal decisions that align with multiple objectives, criteria, and the entire lifecycle of the bridge. While physics-informed analysis, such as the finite element method (FEM), can simulate complex and closely coupled scenarios, such as bridge structural analysis, it cannot account for some loosely coupled discrete factors, which could be addressed by ontological reasoning. Therefore, this paper presents a knowledge-driven decision-making framework that combines static knowledge reasoning with dynamic FEM analysis results to support holistic bridge maintenance decisions. One significant contribution of this research is the development of a comprehensive bridge maintenance ontology that incorporates knowledge derived from bridge maintenance standards. Another key contribution is the ability to employ complex runtime rules-based reasoning to tackle intricate bridge maintenance scenarios. To enable automatic knowledge-driven reasoning, an integrated workflow is developed to orchestrate semantic modeling with numerical modeling through a Python-based Web Ontology Language application programming interface (OWL API). This integration facilitates the efficient orchestration of the framework. A case study is presented to demonstrate the potential for the developed framework in assisting with the complex holistic decisions required for bridge maintenance.