A mathematical model has been used to investigate the effect of cancelled buses on passenger waiting times, and the improvements which could be brought about by rescheduling, using two different strategies. The first consisted of rescheduling to provide equal headways for the available buses in a schedule which would change from day to day. The second strategy provided a reserve pool of buses to replace cancelled buses in a new service scheduled at a lower frequency. Both strategies provided similar reductions in passenger waiting time for service headways below about 12 minutes, and the percentage reductions obtained were roughly equal to the mean percentage of buses missing from the original scheduled service. At longer headways the second strategy provided still larger waiting time reductions while the first strategy caused waiting times to increase. Rescheduling dramatically reduced that part of the waiting time which was due to service unreliability, and which is likely to be particularly annoying to passengers. It is estimated that rescheduling in situations where services are running consistently below complement can produce savings of some 25000 hours of passenger waiting time per year per bus missing from service, and that the improved level of service produced by rescheduling a service in which 10 per cent of scheduled buses are missing on average could attract 5 per cent, and possibly as much as 10 per cent, extra patronage. A full description of the model is given in an appendix. (Author/publisher)
Samenvatting