Subway station diagnosis index (SSDI) : a condition assessment model

Condition assessment of subway stations is a major issue facing public transit authorities worldwide. In 2002, the Societe de Transport de Montreal (STM) valued its subway station replacement cost at 2.6 CAD Billions. While its stations are becoming aged, the STM requires a rehabilitation budget of 643.6 CAD Million between the year 2006 and 2010. Nevertheless, the STM lacks a planning strategy reflecting this increase. The principal obstacle to the development of effective planning strategies is the lack of condition assessment models of subway stations. This research develops a condition assessment model, the 'Subway Station Diagnosis Index (SSDI)', and a scale. The SSDI model is used to diagnose a specific subway station and assess its condition using an index (0 to 10). Based on the SSDI, the condition scale describes the station's condition state, its deterioration level (%), and proposed consequent actions. The new model identifies and evaluates the different functional/operational criteria for subway stations; mainly structural, architectural, mechanical, electrical, security and communications criteria. It uses specific decision analysis tools in order to evaluate a 'Functional Diagnosis Index' (FDI), and a global 'Station Diagnosis Index' (SDI). In other words, the SSDI model uses the Analytical Hierarchy Process (AHP) in order to determine the criteria weights. It also utilizes the Preference Ranking Organization METHod of Enrichment Evaluation (PROMETHEE) in order to aggregate the multi-criteria. Finally, the SSDI applies the Multi-Attribute Utility Theory (MAUT) to determine the FDI and the SDI values. Data were collected from experts through interviews, phone calls and questionnaires as well as STM inspection reports. The targeted interviewees were transit authority experts in both Canada and U.S.A. Statistical and sensitivity analyses were performed on the collected data. Analyses show that structural and security/communication criteria are the most important (36.1% and 27.3%, respectively). The newly developed model is applied to seven stations from the STM network. Results show that these stations are deficient, with an average SDI of 4.4 out of 10. Ranking of the seven stations is compared to that of PROMETHEE, which shows similar results. In addition, the SDI values are confirmed by STM engineers with 80% agreement. This research is relevant to industry practitioners (management, engineers, and field inspectors) and researchers, since it develops, a multi-criteria condition assessment model and scale for subway stations