Performance Measurements on Mass Transit: Case Study of New York City Transit Authority

Document Type

Journal Article

Publication Date


Subject Area

operations - performance, planning - service quality, organisation - performance, mode - bus, mode - rail, mode - mass transit, mode - subway/metro


Waiting time, Underground railways, Transit, Subways, Service quality, Schedule maintenance, Quality of service, Public transit, Performance measurement, Performance indicators, Passenger service quality, On time performance, New York City Transit Authority, New York City Transit, Mass transit, Local transit, Intracity bus transportation, Headways, Data quality, Data integrity, Data collection, Data acquisition, Continuous quality improvement, Case studies, Bus transit operations, Bus transit


For all organizations, public or private, it is essential to establish measurements to ensure that the services provided are being done well and, when they are not, that the organization can diagnose problems. The mission of New York City Transit (NYCT) is to provide timely and reliable mass transit to more than 7 million daily riders. NYCT has established three main performance indicators (PIs) to ascertain how closely this mission is being met: en route schedule adherence, headway regularity, and wait assessment. Whereas en route schedule adherence (−1 to +5 min) and wait assessment are easily explained, regularity (±50%) is a useful diagnostic for operations management. NYCT selected 23 subway routes and 42 borough-representative principal bus routes for performance analysis. A stratified sample, designed to prevent undesirable sample bias, is generated by using a fully automated system and achieves an accuracy of 95 ±5% at the route level. Computerized data processing and analysis were implemented in 1995. Recently, paperless data collection was initiated; this further decreases reporting lag and improving data quality. Indicators are reported semiannually to the public, and detailed internal diagnostic reports are issued frequently to help operations management improve service performance. PI statistics are now used by senior management for setting goals and by rider advocacy groups to assess agency performance. A partnership and spirit of cooperation has developed between operating areas and analytical staff in troubleshooting delay issues and continuous quality improvement. The PI infrastructure is tapped by a pilot program to assess performance of operations improvement initiatives.