In this sub project, a feasibility study of the use of AI-based methods for analysis of cause-and-effect relations connected to auxiliary heat on the Iron Ore Line has been performed. At the overall level, the sub project aims to support a more sustainable and robust railway infrastructure by investigating an analysis method for a more dynamic and condition based maintenance program.
By collecting, integrating, washing and processing information from various data sources, a set of measures was created that capture factors with potential impact on faults in the switches along the Iron Ore Line. A machine-learning model was trained to determine the explanatory value of each factor for reported errors.
The result is a rich source of data for further analysis and a compilation of the identified cause-and-effect relationships. In addition, the model developed has shown potential for identifying vulnerable switches, early warnings and risk planning based on different scenarios. All in all, the results of the projects represent a step towards a more condition based and proactive maintenance program.
Stakeholders: Luleå tekniska universitet, Kairos Future, Trafikverket, LKAB och InfraNord
Project leader: Helene Olsson, Kairos.