Objective

To determine whether Wait Assessment performance measurements reported to the Metropolitan Transportation Authority (MTA) Board and customers are accurate and consistently calculated. The audit covered the period between July 1, 2017 and November 30, 2018.

About the Program

Under the Public Authorities Law, the MTA is required to issue an annual report on its mission statement, measurements, and performance indicators. The MTA’s 2017 and 2018 calendar year Mission Statement and Performance Indicator reports list Subway Wait Assessment (WA) as an indicator of the MTA’s progress toward its goal of providing reliable service for its customers and federal, State, and local government partners.  

Performance measure results are often reviewed by the Board, which provides feedback, guidance, and recommendations to Transit on the success of its operations. On July 25, 2017, in response to the Board’s demand for performance improvements, the MTA released the Subway Action Plan to improve service. One of the changes in the Plan was the implementation of new customer-focused performance measures intended to be more relevant and easier to understand. In September 2017, Transit introduced the new metrics: Additional Platform Time, the average time that customers wait at a station beyond their scheduled wait time; Additional Train Time, the average time customers spend on board a train beyond their scheduled travel time; and the sum of these, Additional Journey Time.

With the introduction of these new metrics, WA became a “legacy indicator,” but continues to be reported monthly to the MTA Board’s Transit and Bus Committee (Committee) and posted on the MTA website’s Subway Performance Dashboard (Dashboard).

Key Findings

• According to Transit Operations Planning (OP), subway WA is calculated at time points, which should consist of between 25 and 50 percent of all stops along each route. We found that time points only cover between 11 and 30 percent of all stops on each route per direction. Therefore, the majority of the service provided is not captured in the WA calculation. For example, the Number 1 train, a local service train that runs from 242nd Street in the Bronx to South Ferry in downtown Manhattan, calculates WA based on only 5 of the 38 stops (13.2 percent).
• Even though the time points were selected based on customer traffic, OP did not include key subway stations such as Penn Station, 59th Street-Columbus Circle, and 14th Street. These stations had the second, third, and fourth highest rates of ridership, respectively, in 2017 and on average from 2012–2017.
• For November 2018, more than 3.1 million actual train records were excluded from the WA calculation because they were recorded at stations that are not time points. As a result, OP may not be providing its customers and government partners with complete performance results. For example, WA for the E train shows that it was met about 72 percent of the time based on OP’s data set, but including all of the data in WA calculation results in a rate of about 24 percent.
• The MTA does not disclose to the Committee or on the Dashboard that WA is only based on certain stations. Based on our review, only one train service line, Number 4, calculated WA on more than 25 percent of scheduled stops per direction. Additionally, WA calculations may be affected by trains opting to skip non-time point stops in order to avoid arriving late at time points. WA is reported for each of the 24 subway service lines and is presented separately in the Committee reports for both weekday and weekend service. However, on the Dashboard, an individual can filter WA data by line, division, and peak/off-peak service for only weekday service – weekend statistics are not provided on the Dashboard.

Key Recommendations

• Calculate system-wide WA performance based on all subway stations rather than time point stations. In the interim, disclose that WA is calculated based on only time point locations.
• Provide the public with the supplement schedules that are used to calculate WA. Only include intervals that are in accordance with Guidelines in the WA calculation and, in the interim, disclose that headways may not be in accordance with the Guidelines.
• Provide weekend WA statistics on the Dashboard.