The maintenance of massive transport vehicles is very important to guarantee the quality of this social service. This paper presents a software tool named ISPMAT able to perform an automatic detection and diagnosis of possible faults in two main components of a train: the compressor and the traction bogie. ISPMAT is based on artificial intelligent techniques: neural networks and expert systems. This paper includes the description of the ISPMAT architecture and some experimental results.
Keywords: Predictive maintenance, Diagnosis expert system, Neural networks, Health of components, Electrical trains
16th International Euromaintenance 2002. Helsinki (Finlandia). 3-5 June 2002
Published: June 2002.