Photo by brownpau on Flickr.

A proposal before the WMATA Board tomorrow would define a new, formal standard for the minimum acceptable service level. But it sets the bar far too low: 15 minutes during peak times, and 30 minutes off-peak.

Another proposal would a establish passengers per car loading thresholds of 120 people maximum and 80 people minimum.

Having a standard is a good idea. It formalizes decisions such as where to put the 8-car trains or how to adjust schedules to meet customer demand. Formally defined service criteria also let customers and regional governments know what to expect from Metro. But Metro should never find service this infrequent to be acceptable.

Frequency matters

Customers should expect better than a train every 15 minutes during the peak, and better than 30 minutes during off-peak. One of the most important factors for customers is frequent service. In the words of transportation writer Jarrett Walker, “Frequency is Freedom.” Low frequency means people wait longer for trains, wait longer for connections, and have to build in extra time in their schedules if they connect to buses.

This is just a minimum, and it doesn’t mean Metro will cut service tomorrow, but Metro should never allow service to get anywhere near this infrequent outside of planned or unplanned track work (and even then, such frequencies are a burden for riders).

The 2012 budget calls for service every 6 minutes for every line during peak periods (and provides even more frequent service in the core where lines double up), and every 12-15 minutes during non-peak periods. This new standard would reduce the amount of service that Metro considers the minimum allowable.

Most troubling is that the presentation says that the Board would delegate authority to the General Manager/CEO to make minor service changes based on the service criteria. That could let the GM/CEO reduce service to this unacceptably low level without asking.

Standards could lead to less off-peak service

This authority also means that there are consistently fewer than 80 passengers per railcar on a route, he could start cutting service. A railcar typically has about 60 seats.

The thresholds for passengers per car also need adjusting. During peak times, customers expect railcars to be crowded, and in any case there often isn’t anything Metro can do about crowded trains because there aren’t any extra railcars available. But during off-peak times, Metro has options to reduce railcar loading (longer trains, more frequent trains), and Metro is competing against less congested roads and more available parking.

It doesn’t make sense to expect customers to pack onto trains that are as crowded during non-peak times. Metro’s presentation says that staff is looking into establishing other thresholds for off-peak. The Board should ensure they follow through on this effort.

The presentation also says that Metro will establish data monitoring and reporting procedures. Metro should make a part of this that they report this information to the public. In previous public information requests, I have found that this information is difficult to obtain.

Metro is doing the right thing to set defined criteria for frequency and passenger loading. But the frequency should be much higher than in the proposal. A train should leave from each terminal at least every 6 minutes during peak, and every 15 minutes during off-peak, since this is the budgeted service we have now. Metro should measure passengers per car during all time periods, not just during peak, and should establish a lower threshold loading during off-peak than during peak.

We've just launched our brand new website and are working out some kinks. Find something that looks like a bug? Please help out by sending us an email with the details!

Michael Perkins blogs about Metro operations and fares, performance parking, and any other government and economics information he finds on the Web. He lives with his wife and two children in Arlington, Virginia.