A variety of decision-making tools can help railway infrastructure managers plan timely, effective, and efficient maintenance. Most of these tools address project-level maintenance on short sections of railway track, where the level of benefits tends to be discrete and definable. However, it is at the network level where the benefits are less tangible and the consideration of the maintenance investment returns plays an important role in helping frame policy. A major issue concerns how network-level benefits may be presented in a concise and straightforward manner that is meaningful to politicians and senior decision makers. To this end, this paper describes research that was carried out to develop a tool based on probabilistic models, which are capable of determining the effects of maintenance on network condition under any budget scenario. The robustness and viability of the system is demonstrated by means of an example.
- genetic algorithm
- railway track