The relatively long life cycle of railway infrastructure means that maintenance and renewal decisions significantly influence the total life cycle cost (LCC) associated with the infrastructure. A decision support tool such as life cycle costing assists infrastructure managers in making maintenance and renewal decisions. A shift from qualitative to quantitative decision making is possible using decision support tools and modelling approaches based on appropriate data. Most LCC maintenance models in the literature are deterministic in nature. However, there is inherent uncertainty present within the reliability and maintainability (R&M) parameters. The uncertainty within the R&M parameters can be characterised through appropriate statistical distributions or using bootstrapping in conjunction with available data. A maintenance modelling approach based on stochastic methods and Monte Carlo simulation is presented in this paper with specific attention to a model developed for the rail component. The proposed model allows quantification of inherent uncertainty within the calculated LCC which is coupled to the uncertainty within the input R&M parameters. This modelling approach is flexible in nature and supports the use of large input data sets, capturing variability within the real-world situation of maintenance management. The flexibility of the modelling approach is demonstrated using an example which incorporates risk to assist an infrastructure manager in deciding whether to use flash butt or alumino-thermic welding during rail maintenance.

Monte Carlo simulation, life cycle cost, uncertainty, maintenance modelling, rail