The Making of a Restructure: The "Commuter" Routes

For the third year, I’m proposing a full network, in-service time neutral restructure of Sound Transit and King County Metro’s public transit service in Seattle. This year, I’m adjusting my approach slightly. Previously, after I established a budget of service hours available, I would test large, wide-ranging batches of changes against it, repeating this until I was close to the line without going over. This time, I’m parceling changes by theme, in hopes of better appreciating the savings and costs of different types of changes.

Also new this year is that I am balancing the service time budget not in terms of expenditure within Seattle, but throughout King County¹. The impetus for this is a change in how I’m computing access², but it’s fundamentally a more accurate assessment of expenditure regardless. The in-service time from segments of transit trips that crossed the Seattle border were unaccounted for in previous attempts. By expanding service-time measurement to the county level, this is only a concern for portions of service that leave the county, of which there are far fewer. This change is particularly germane to the routes featured in this post.

I’m calling these routes “commuter” routes, but, for some of them, this term is an awkward fit. Some King County Metro routes are clearly intended for commuters living in suburbs and working 9-to-5 jobs in downtown Seattle; routes like 102, 257, and 311 fit this mold. They run service only around the AM and PM rush hours, only on weekdays, only in the direction of commuter flow, and stop at points in downtown Seattle that are already well-served by local bus service. Some routes stretch this definition: route 212 is bidirectional, route 101 has all-day service every day, and route 556 serves the U District instead of downtown. Others stretch it to its breaking point: route 522 runs bidirectionally all-day, every day, serves points in Lake City, and terminates at Roosevelt Station. The unifying characteristic is that when these routes do offer service in Seattle, it’s not unique—it overlaps with other, typically more frequent, routes.

Like its predecessors, this restructure cuts the overlapping service in favor of redirecting the routes to 1 Line stations that provide capacious, fast, and frequent service connecting to their erstwhile terminals. In some cases, this process is straightforward. Under the restructure, all routes that serve the Evergreen Point Freeway Station terminate at University of Washington Station. Because some routes already do this, the new schedules can draw on existing data. Route 177 is on the opposite end of this spectrum. Presently, the route makes a few stops in Federal Way before a long trip up Interstate 5 connects it to the SODO Busway. The restructured route shortens the trip up the interstate, having the route connect at Angle Lake Station for service onward to Seattle. Since there’s no transit route directly connecting route 177’s stops in Federal Way to Angle Lake Station today, I had to estimate the time it would take to run a bus between them. I used Google’s vehicle directions for this—which provide a wide range of possible durations even for a fixed time of day—erring on the conservative side.

The specific changes to the “commuter” routes (as well as some other routes which will be discussed in a subsequent update) can be observed in the route viewer. Shortening them generates a savings of in-service time. The table below shows this for a weekday.

What does a total savings of 1,104,254 seconds really mean, though? It’s about 3% of the total in-service time expended by Metro and Sound Transit in King County on a weekday. More concretely, weekday 1 Line service expends 1,046,730 in-service seconds within King County. It wouldn’t be possible to make this change and instantly double the frequency of the vast majority of the 1 Line—there are barely enough trains and storage space for Sound Transit to operate its current schedule—but the comparison illustrates how transformational reallocating just some of the investment in commuter service could be. Arriving at a 1 Line station after just missing the train would feel totally different if it ran 4 minute headways at peak and 5 minute headways midday.

Rather than reinforcing the 1 Line, though, these in-service hours will ultimately be distributed among bus routes in Seattle. I suspect some would see this as unfair. This investment is currently employed to get people living outside of Seattle into and out of it, but they are being repurposed to move different people within the city. Wouldn’t a more fair approach be to invest the savings in increasing the frequency or span of the shortened commuter routes?

The in-service time of a route isn’t owned by its current riders, so I’m not compelled by the idea of fairness when allocating it. The outcomes are what should be fair—or at least no less fair than the status quo. For the people losing “their” service hours, the outcomes are unlikely to be negative; their commutes to and from Seattle will just involve a transfer. In some cases, trips might be marginally longer, but they will also be less sensitive to the volatility of traffic delays on I-5, and provide the option of reaching destinations other than downtown Seattle. Transferring these hours into Seattle allows something that more frequent commuter service won’t. Ultimately, the goal of this restructure is to reduce rates of car use and ownership. A few more trips serving Park and Rides in Federal Way and Woodinville won’t change that; these suburbs aren’t built in ways that allow a person to reach everything that they desire with transit. Would this service time alone, reallocated to Seattle, cause a substantial reduction in the city’s household car ownership rate? Probably not, but it makes more incremental progress toward that goal than any alternative that I can envision, and there are other pools of in-service hours that can be repurposed to build upon its impact.