Mining data in the transportation sector

Televic Rail, a global technology company with over 30 years of experience, investigates how to collect and present insightful data about train fleets to enable predictive maintenance. By doing so it takes research about mining data in the transportation sector a huge step further.

By Televic

This new research, led by Televic Rail, falls under the imec.icon research program and as always brings together industry players and research groups with different specializations. As each of those projects, it has a duration of two years. This imec.icon research project focuses on designing novel techniques for mining the explosive amounts of collected sensor data on trains. “On-board sensor systems are getting more and more complex”, says Dirk Van Den Wouwer, R&D manager at Televic Rail. “Since it is getting easier and cheaper to produce sensors, more of them are effectively being used: we have gone from 3 or 4 per coach to a good 50.”

Predictive maintenance

All these sensors collect data about a variety of parameters, such as the temperature of components, vibrations, speed, acceleration and load on the train’s metal structure. Investigating how to collect and present this vital information will help Televic Rail to provide an optimal tool for predictive maintenance. And that is of key importance for train operators. Seeing from a distance what is wrong with a train set, gives them the opportunity to send it to the right repair location immediately. There, the right fitters, parts and tools are available, and as a result, downtime is limited to a minimum.