Assessing the Economic Life of Rail Using a Stochastic Analysis of Failures

Abstract

In this article, a model is developed to evaluate the economic life of rail using a stochastic analysis of rail failures. The occurrence of rail defects and failures is analysed and issues related to rail maintenance are addressed. The hazard rate of alumino-thermic weld failures is predicted for continuously welded rail, where the expected number of welds is increasing in tonnage as a result of replacement. Imperfect inspections are modelled for predicting the expected number of failures by employing a filtered non-homogeneous Poisson process. The impact of maintenance on failure occurrence and whole life-cycle cost (LCC) is analysed and integrated within the model. The application and capability of the proposed model and sensitivity of parameters are illustrated through the use of a practical example. It is shown that an optimal economic life of rail can always be found, as the repair cost and the risk cost of accident are increasing with tonnage that passed through it. In addition, the inspection frequency has a significant impact on rail LCC. The results also suggest that the proposed model is also valid for optimizing the interval between rail inspections.