The most common service reliability measure--on-time performance--has drawbacks transit managers and consumers should be aware of. By converting on-time performance to odds ratios, a fresh look at what is meant by transit reliability is obtained. Although differences in on-time performance may seem slight, measurement with odds ratioscan indicate quite different service levels. Transit managers, if they rely exclusively on on-time performance and do not heed the lesson to be learned from odds ratios, may be deceived when they estimate how much improvement can be achieved with a particular policy or action. The magnitude of effects will vary with level of effort, depending on the initial level of reliability. This paper appears in transportation research record no. 1297, Public transit research: management and planning 1991 .
Abstract