Ontology-based Conceptual Model Development for the Railway Domain: A Maintenance Case Study

Abstract

The railway sector is currently acting in a fragmented way and in silos corresponding most often to physical or functional subsystems or use cases, and the different owners/managers of the overall infrastructure at regional/national level, without global extensive view or full control of the global system involved by rail operations. A number of digital models however do exist, for different subsystems, and by different stakeholders (railway company’s managers, railway manufacturing industry, national policy makers). They are generally not communicating among each other: then, there is currently no global view, no possibility of assessing the effect of design changes of one subsystem on the other interacting subsystems, and more generally, on the overall system operation performance. Furthermore, the development of new systems involving various stakeholders (railways/industry/engineering) is currently based on non-formal requirements coming from Railway Infrastructure Managers (IM) or Railway Undertakings (RU), rather than on a verified performance of a component/subsystem into the whole system, before tests are carried out in situ at the very end of the development phase. This lack of global performance and formal verification at different stages of the design can possibly entail costly redesigns if the required system performance is not met. With the aim to avoid design errors and to have a shared knowledge view between stakeholders, there is a need to establish conceptual and semantic clarifications of the railway concepts.