Title

New Decision Support System for Optimization of Rail Track Maintenance Planning Based on Adaptive Neurofuzzy Inference System

Document Type

Journal Article

Publication Date

2008

Subject Area

operations - scheduling, infrastructure - track, land use - planning, ridership - commuting, organisation - management, mode - rail

Keywords

Tracks, Track rehabilitation, Track maintenance, Scheduling, Railroad tracks, Rail maintenance, Optimization, Optimisation, Neural networks, Maintenance of way, Maintenance management, Fuzzy logic, Deterioration, Decision support systems, Artificial neural networks, ANNs (Artificial neural networks)

Abstract

It is well known that maintenance planning affects, in general, the life of the structures, material wear, and quality of service. In particular, the maintenance of rail tracks affects the traffic volume as well, and therefore it is an important issue for the management of a railway system. Accurate maintenance planning is necessary to optimize resources. The condition of railways is checked by special diagnostic trains. Because of the vast amount of data that these trains record, it is necessary to analyze these data through an appropriate decision support system (DSS). However, the most up-to-date DSSs, such as EcoTrack, are based on a binary logic with rigid thresholds and complicated algorithms with a large number of rules that restrict their flexibility in use. In addition, they adopt considerable simplifications in the rail track deterioration model. In this paper, a neurofuzzy inference engine has been implemented for a DSS to overcome these drawbacks. Based on fuzzy logic, it was able to handle thresholds expressed as a range, an approximate number, or even a verbal value. Moreover, through artificial neural networks, it was possible to obtain more precise rail track deterioration models. The results obtained with the proposed model have been clustered through a fuzzy procedure to optimize the maintenance schedule, thus grouping the interventions in space and in time.