Bus bunching is something that’s often mentioned as a problem spot for bus reliability and particularly frustrating when riders have to wait 20 minutes longer than expected only to find two buses rumbling along one after the other. As it turns out, however, bunching isn’t some systematic anomaly that no one has the answer to. While there are a lot of factors that end up fluctating actual headways (as opposed to scheduled headways), late buses only exacerbate tardiness, therefore resulting in bunching.
More after the jump.
Earlier last month, Daniel Holz had the break-down behind the rather common-sense reason:
Putting this all together: if a random fluctuation creates a slow bus, then it will get slower and slower, and the bus behind it will get faster and faster, until the two buses meet up. At this point, the buses stick together, and are essentially incapable of separating. Thus, in general, buses will bunch up. This will usually happen in pairs, though on occasion triples and even quads may occur.
The reason behind bunching isn’t a new one, but particularly intriguing for those of us who have never bothered to find out why long waits would always end up being accompanied by long strings of buses. Across the country, this problem permeates most bus systems operating routes with frequent enough headways, but is a particularly nasty stigma for Seattle where our city core is saturated with buses, especially during peak hours.
The main issue about bunching from a ‘data’ standpoint is that multiple buses on a run will end up having pretty shoddy on-time performance and service frequency. Not only do the buses fail to adhere to their time schedule, but the headway intervals are also thrown out of whack. Martin’s post about bus reliability metrics last year brings some good points to light. TFL in London measures reliability by targeting what it calls “excess waiting time,” as opposed to Metro and ST’s “on-time” arrival metric here.
For high-ridership inner-city routes like the 7, 36, 48, 49, etc., riders are generally less concerned about the schedule as they are frequent service, so addressing ”excess waiting time” is a better metric for those routes. Because traffic patterns are extraneously dynamic and dedicating transit-only lanes are more-or-less politically infeasible, GPS tracking and TSP (transit signal priority) are hugely important to address service frequency. For the city which had the highest transit mode share of all rail-less major cities (until last July), we have focused too much on service frequency at the expense of service quality. In retrospect, Metro and SDOT have some catching up to do.