In this study the opportunity to introduce PHM (prognostic
and health monitoring) concepts into a cracked railway axle
management is investigated.
The performances of two different prognostic algorithm are
assessed on the basis of their RUL (remaining useful life)
predictions accuracy: a prognostic model based on the
Bayesian theory and a physical prognostic model. Both
models rely on the measured crack size. The measured crack
growth measure has been built from simulated probabilistic
crack growth path by adding measurements errors. The
effect of monitoring frequency and the measurement HW
and SW infrastructure size error to RUL predictions’
accuracy is assessed as well, trying to evaluate the
hypothetical measuring infrastructure capabilities’ (sensors
+ layout) effect on the overall PHM predictions.
Furthermore the PHM approach is compared to the classical
preventive maintenance approach to railway axle
maintenance management based on expensive and regular
NDT.