MTA replies to on-time complaint
Getting There thanks Terry Owens, spokesman for the Maryland Transit Administration, for his prompt reply to the complaint about the No. 1 bus route registered by rider Youssef Mahmoud. Here goes:
We appreciate input from our passengers. We are also proud of the fact that the MTA moves over 350,000 people every day with the majority arriving at their destinations on time. However, when someone like Mr. Mahmoud experiences a service delay we understand his frustration, and seek to resolve performance issues to the best of our ability.
We have heard from Mr. Mahmoud often in recent months via social media and your blog. His concerns are being taken seriously. I have asked our Service Quality Division to investigate his latest issue with MTA Local Bus Service.
The following is an explanation of On Time Performance and how it is calculated at the MTA from our Performance Management Team.
The MTA, like most transit agencies, uses a wide variety of performance measures to monitor and track the delivery of service. No single measure is perfect—each has its flaws and limitations. A typical measure of service quality is on-time performance or OTP. As was previously stated, MTA’s OTP measure is calculated as the total number of timepoints arrived at on time (between 1 minute early and 5 minutes late) divided by the total number of timepoints that were measured. This measuring process is done by our Automatic Vehicle Location (AVL) system, which basically gets bus location data from GPS and compares it to schedule data. In any given week, a little over 300,000 timepoints are measured on our bus service, with varying numbers of timepoints being measured on each line depending on the time separation between timepoints, how long each bus line is, and how many trips are made on each line.
The AVL system that we use, which is not unlike most other transit agencies’ systems, has its limitations. As we mentioned earlier, OTP does not take into account cut service or when a bus breaks down on the line. This is not because the MTA desires to overlook these events, but because the system that calculates OTP is not capable of measuring these events as an OTP factor. Simply put, if the bus that is supposed to be on the line is NOT on the line, that bus’s performance cannot be measured under the current system design. The bus is not crossing timepoints and therefore cannot contribute to the overall OTP percentage.
One way we compensate for this measurement problem is that we measure and track runs (bus pullouts) that are cut and delayed as well as the number of major road calls (any instance where service is affected by a mechanical failure, sick patron, or other event that necessitated replacement or repair of an in-service bus). We also look to data from customer complaints, from the Rate Your Ride project (www.rateyourride.org), and reports of service issues from operators.
In summary, the OTP statistic is not a perfect measure of bus service reliability, and we recognize that. It also does not adequately represent the experience of the largest proportion of riders, since overall OTP is mostly better during off-peak periods when fewer people are riding. OTP data is currently best used to compare lines to each other
We share Mr. Mahmoud’s concerns, and are currently investigating better ways to measure and present OTP data that more accurately reflects the experiences of the public.