A worker marks a new railway tie so that it can be placed onto the track near Ballston during one of the summer SafeTrack projects. (Katherine Frey/The Washington Post)

Metro managers are making the case that the successor program to SafeTrack should not be SafeTrack 2.0.

Instead, they say, the rail system’s maintenance program needs to move away from today’s largely reactionary approach into a phase that emphasizes preventive maintenance — in other words, more hours of unfettered access to the tracks.

Andrew Off, assistant general manager for transit infrastructure and engineering services, is going to lay out this argument for Metro board members Thursday, in support of Metro General Manager Paul J. Wiedefeld’s request to get more maintenance time by giving up eight hours a week of rail service.

In effect, SafeTrack is Whack-a-Mole. It targets sections of rail that should not have been allowed to deteriorate as much as they have. Of course, that work is important to correct defects, but it’s not a model for a sustainable maintenance program.

“An overwhelming majority of the work we do is reactionary,” Off said in an interview. “We’re trying to get to a mode where we do preventive maintenance.”

What sort of maintenance? Among the things Metro wants to do with the extra overnight hours: test the traction power cables; test the tracks for stray current; test and maintain the ballast beneath the tracks; grind and weld switch points; inspect and maintain the joints where rails come together; inspect and repair tunnel lighting, drains and emergency exits; perform ultrasonic tests on the rails to detect problems before they become visible.

Let’s look at one of those jobs, the power-cable testing. The potential problems the crews are trying to detect are not likely to be spotted by track walkers as they pass by. This job requires the crews to disconnect and test each section of cable. That’s 13,529 cables to be tested every four years, or about 19 hours of track work every week. This type of testing can’t be done when the tracks are in service.

Why do we want them to do that? It could prevent fires and smoke incidents such as those that disrupted service this year at L’Enfant Plaza, McPherson Square and Metro Center, inconveniencing many thousands of riders.

The Metro staffers have made a calculation for each of the maintenance programs on that list. Rail problems stemming from stray current, for example, have caused 83 incidents along the tracks this year.

Metro wants a maintenance program that tests the entire rail system for stray current every four to five years. Doing a test requires shutting down the power to a section of track, then using a test voltage on the tracks to find any weaknesses in the electrical insulation.

Metro staffers calculate that each of those inspections on a section of track requires four periods of at least three hours, for the setup, the test and checking the results. Again, the trains can’t be running during such tests.

One last example from the list: The use of rail vehicles to test and maintain the track ballast beneath the rails preserves the tracks and reduces bumps. This type of inspection and maintenance requires 20 work hours a week so that all of the main tracks are checked biannually and all of the switches are examined annually. This work, too, can’t be done when trains are in service.

Metro says 117 disruptions on the tracks so far this year stemmed from track problems that might have been detected and corrected with such testing.

Metro has temporarily expanded its opportunity to do maintenance by six hours by suspending the late-night service on Fridays and Saturdays. Wiedefeld wants to keep those six hours, in some form, and add two more.

The transit staff and various consultants have said that the old window of unfettered access to the tracks was not sufficient to do the preventive maintenance that Metrorail needs. You can see what they have been talking about by looking at the chart below, which shows that the Friday-Saturday work windows before and after the start of SafeTrack in June.


Chart shows the maintenance window as it was on Friday and Saturday nights before SafeTrack suspended the late-night service on weekends. (WMATA)

Why can’t the work crews just get more efficient in their use of the regular maintenance windows? “We do agree there are some opportunities for efficiency,” Off said, “but certainly not enough to move the needle.” It’s possible, for example, that the crews could better manage the power-off and power-up timing, and Metro might be able to adjust train schedules to see whether it could clear the lines earlier for the work crews. But after all of the planning and testing to speed up those processes, Off said, the work time would not have been increased enough to make a significant difference in the track program.

“Waiting two to three years to find a half-hour more efficiency isn’t good,” he said.

Riders and civic leaders concerned about the inconveniences that would come with reduced hours have been asking about efficiency, and they have also been asking about alternative strategies. Some ask what is wrong with continuing to pick off sections of the system for concentrated maintenance work. In other words, why not SafeTrack 2.0?

We would wind up with dozens of mini-projects, stressing both the transit staff and the riders. Metro would have to publicize each disruption so that riders are not surprised, deploy extra staffers to help people confused by the new service pattern and send out the shuttle buses to bridge service gaps. Also, much of the SafeTrack work done during the 15 special maintenance projects are outside the core of the transit system. Doing special projects during the day time inside the core, the most heavily traveled part of the rail system, would be far more disruptive. The transit staff estimates the trains would need to be 20 minutes apart, amounting to an 85 percent reduction in service.

Okay then, why not add more work crews overnight during the maintenance window?

On an average night, 57 crews are working in the rail system. As one example of the extent of the work, Off cited the night of Sept. 7, when crews worked on 164 miles out of 234 miles of tracks.

The work plan is a more of a time and space problem than a problem with the number of people working, Off said. “We already have inefficiencies with 57 work crews. One way to become more inefficient is to add more work crews.”

So far this year, according to the transit staff calculations, Metrorail riders have experienced 639 service disruptions related to the state of the infrastructure. That’s the cracked rails, the deteriorating rail ties — all of the things you’ve learned to hate. The late-night service reduction would affect half a percent of all riders. The potential gain from a safer, more reliable service would affect hundreds of thousands of riders and the region’s economy.

There are no guarantees this program will work — and we’ve been disappointed before — but there’s no good-looking alternative.