Ontology-based automated knowledge identification of safety risks in Metro shield construction

Le résumé fourni par la source

Safety risk identification of metro tunnel construction is a knowledge-intensive process involving various stakeholders and communities. Currently, safety risk information related to decision making in metro tunnel construction is ill-structurally stored in various disordered formats, which hinders knowledge sharing and reuse. This study develops a domain ontology to formalize safety risk knowledge in metro tunnel construction to support safety risk identification. The method integrates ontology and rules inference technology, utilizing semantic inference hazards and corresponding measures to promote security management. An ontology development method with five steps is adopted. The domain ontology organizes safety risk knowledge into five unified classes (i.e. projects, construction activities, precursor information, risk events risk prevention measures). Defined classes, together with corresponding properties and relations, are coded using the Protégé platform. The domain ontology is evaluated theoretically and practically, using criteria-based and application-based evaluations respectively. Finally, a shield axial deviation example was selected to verify the theoretical and technical feasibility of developing the ontology model. The results show that the operation results of the ontology model are consistent with the thinking of safety managers. This approach creates a holistic knowledge base that integrates a wide range of domain knowledge. The model can facilitate real-time hazard identification and prevention at construction sites, providing important guidance and decision support to construction managers.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.