A table summarizing the performance of several different bus and rail transit services around the world.
Over the past few days, I’ve been working on a new post on the Beaverton/Wilsonville Corridor, currently served by WES, and wondering what options might work to provide service for it in the future. One of the efforts of that column was a table comparing the performance of several different types of transit technologies, to help inform debates on the subject. The table appears to be sufficiently useful to merit a standalone post.

The table lists quite a few transit lines (or segments thereof), along with key attributes. Many of the examples are from Portland or elsewhere in the Northwest, but a few are taken from other parts of the world. The attributes focused on are those which most directly affect transit speed–the characteristics of the right of way (how exclusive is it), and the stop spacing. Some rather interesting systems are included on the list. The list is sorted by average speed, without regard to technology.

A few notes:

• Generally, peak-hour times are used for services where that matters.

• Service classes are defined as follows, inspired by this Human Transit post. In general:

  • Class A services are those that only need stop at stations. This can refer to grade-separated lines (els or subways) or lines where the transit vehicle has absolute priority at grade crossings, and other vehicles (and pedestrians) are kept away from the route other than at well-marked, well-guarded points. An example of the latter is MAX in Beaverton.
  • Class B services are those where transit may need to stop at crossings, but has an exclusive lane otherwise. A + is added for signal priority (which is different from the absolute priority above), a – if the line is in close proximity to parallel auto traffic or pedestrian environments.
  • Class C services run in mixed traffic; a + is earned for those with signal priority or other enhancements

  .
  Rather than say anything more, I instead simply present the above table as is, and ask readers to draw their own conclusions below.