Temporal association rule mining for the preventive diagnosis of onboard subsystems within floating train data framework

The increasing interest in preventive maintenance strategies for railway transportation systems and the emergence of telecommunication technologies have both led to the development of floating train data (FTD) systems. Commercial trains are being equipped with both positioning and communications systems as well as onboard intelligent sensors monitoring various subsystems all over the train. The sizable collected amounts of real-time spatio-temporal data can be used to leverage the development of innovative diagnosis methodologies based on temporal and sequential data mining. This paper presents a temporal association rule mining approach named T-patterns, applied on highly challenging floating train data. The aim is to discover temporal associations between pairs of timestamped alarms, called events, that can predict the occurrence of severe failures within a complex bursty environment. Experiments carried out on Alstom´s TrainTracer™ data show promising results.