Validation of a Physics-based Prognostic Model with Incomplete Data: A Rail Wear Case Study

While the development of prognostic models is nowadays rather feasible, its implementation and validation can still cre-ate many challenges. One of the main challenges is the lack of high-quality input data like operational data, environmental data, maintenance data and the limited amount of degradation or failure data. The uncertainty in the output of the prognos-tic model needs to be quantified before it can be utilised for either model validation or actual maintenance decision sup-port. This study, therefore, proposes a generic framework for prognostic model validation with limited data based on uncertainty propagation. This is realised by using sensitiv-ity indices, correlation coefficients, Monte Carlo simulations and analytical approaches. For demonstration purposes, a rail wear prognostic model is used. The demonstration con-cludes that by following the generic framework, the prognos-tic model can be validated, and as a result, realistic main-tenance advice can be given to rail infrastructure managers, even when limited data is available.

Automated summary

While developing prognostic models is relatively feasible nowadays, implementing and validating these models still face many challenges, including the lack of high-quality input data and limited degradation or failure data. Hence, this study proposes a generic framework for validating prognostic models with limited data based on uncertainty propagation. By using sensitivity indices, correlation coefficients, Monte Carlo simulations, and analytical approaches, the uncertainty in the output of the model can be quantified. A rail wear prognostic model is used as a demonstration, showing that by following this generic framework, the model can be validated and realistic maintenance advice can be provided to rail infrastructure managers even with limited data.