Ontology-Based Safety Management for Next-Generation Gaseous Fuel Infrastructure: A Review

Next-generation gaseous fuel infrastructure for low- and zero-carbon gaseous fuels (hydrogen, ammonia, and hydrogen-blended natural gas) is planned to expand, but most announced hydrogen pipeline and storage projects have not reached a final investment decision. Hydrogen infrastructure safety research has centered on material integrity, risk assessment, and standards, while systematic syntheses of ontologies and knowledge graphs for this infrastructure remain relatively limited. This review classifies approaches to safety knowledge by their basis for risk judgment into traditional risk assessment, data-driven, and knowledge-structure-based approaches. Across these approaches, the basis for judgment tends to be fixed in advance, largely during design, training, or knowledge encoding. Computer science, by contrast, has a methodologically mature line of research reasoning only over the parts a query needs. Comparing the two lines reveals a direction that appears promising, in which the ontology remains the knowledge structure and only the necessary analyses run at query time. Prerequisite challenges include confirmation of response times, a shared safety vocabulary, scarce accident data, and verification methods. Close applications were found only outside gaseous fuel infrastructure. These observations rest on abstract-level screening and hold only within the search scope of this review, and the direction has not been demonstrated.