Abstract

Subway networks are the most crucial transit systems of large cities. According to reports, ridership is growing and will continue to do so in the following decade. The 2013 America’s Infrastructure Report Card evaluated transit infrastructure with a grade of D which is translated in poor condition. Large amount of capital investment is required, for instance, the “Société de Transport de Montréal” has assigned around $500 Million for renovation works of its infrastructure. Despite that, transit authorities so far have been relying on empirical management approaches based on engineering judgment and decision makers’ preference. Few models are currently found in which, they either focus on stations solely or examine structural performance only. Taking into account the deterioration severity and the amount of passengers, the duty of proper subway asset management becomes a critical public safety issue. New models are required since they will ensure passenger safety, assist in repair planning and optimize budget allocation.
This research is aiming at developing a model for subway network performance assessment including structural, electrical and mechanical infrastructure. To achieve this objective, a typical breakdown is introduced including network, lines, stations, tunnels and components. The methodology passes through two main phases. First, a condition assessment model for components is developed based on identified defects. Subsequently, the condition index of stations and tunnels is calculated, followed by the rating of subway lines and the entire network. The Analytic Hierarchy and Network Processes, along with the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) and systems reliability are utilized in the computation procedure. Second, a condition prediction model is developed using Weibull theory, which constructs deterioration curves for all the network levels. In addition, the above two phases are incorporated in a user-friendly computerized application.
Data for the relative importance weights are collected through on-line surveys sent to subway experts. It is concluded that components related to passenger safety and client services, such as emergency lightning and escalators, are the most important of the subway network. The developed methodology is also applied to a sub-network of Athens Metro system. The results show that stations are recording condition indexes of more than 7/10 and the network has satisfactory performance. Repair actions need to be planned for 2020. This research provides a new subway network asset management tool that considers all aspects of infrastructure, measures the condition based on actual defects and offers future condition prediction. The outcomes of this research are relevant to transit authorities, asset managers, engineers and researchers.