Introducing Predictive Maintenance
CHALLENGE
Maintenance contractor for Rail Infrastructure executes maintenance based on human inspection of rail track videos.
Introduce predictive maintenance technology to streamline and improve inspection.
APPROACH
- Gather data on inspections executed to create a large test set of observations.
- Develop a deep learning algorithm (based on neural networks) to automate the inspection.
- Minimise false positives (judged okay but in fact not okay) and false negatives (judged not okay but in fact okay).
- Test and validate the software in the live environment.
- Transfer the software to the maintenance organisation.
Inspection with minimal number of false positives (very important for quality of maintenance). Inspection with small number of negative positives to be checked visually (important for reduction of workload).
Fewer false positives leading to better-maintained rail tracks.
Automated inspection leading to sharp reduction in inspection workload.
Cases
