DMrail - digital maintenance for a sustainable railway infrastructure
| Reference number | |
| Coordinator | RISE Research Institutes of Sweden AB - RISE RESEARCH INSTITUTES OF SWEDEN AB, VÄSTERÅS |
| Funding from Vinnova | SEK 4 411 951 |
| Project duration | June 2019 - August 2022 |
| Status | Completed |
| Venture | The strategic innovation program InfraSweden |
| Call | Solutions for sustainable transport infrastructure |
Purpose and goal
DMrail aimed at improving existing state-of-practice maintenance systems in order to increase the operational uptime of rail infrastructure. Innovative methods for continuous health monitoring, data driven decision making, and efficient allocation of maintenance resources wanted to be utilized to this aim. DMrail aimed at addressing the digitalization dimension of the InfraSweden2030 and brings us one step closer to a connected transportation vision.
Expected results and effects
The innovation potential has been built around creating simple innovative measuring methods, and trouble-free installation and rational (dynamic) data transfer that is sufficient for an automated remote condition diagnostics and analysis that trigger maintenance activities as needed. The validation of project results was key driver. Real Bombardier data sets have been used for assessing the validity of the solutions including Cloud and Edge ML anomaly detections. The Trafikverket’s Ofelia database allowed us to analyze actual faults and compare them with our algorithms.
Planned approach and implementation
The project was performed though the collaboration between the actors in the project consortium - toward a highly innovative technological solutions for train communication infrastructure and railway signaling systems (with enhanced survivability support). The DMrail dynamic data-driven predictive maintenance methods has been defined and validated by the infrastructure owners’ and based on their needs - primarily aimed at reducing the fault outcome on the key functions of switches and track circuits.