Exploring Travel Time Reliability Under Different Circumstances: A Case Study

Travel time reliability has been recognized as an important transportation performance measure as well as a major factor affecting a traveler’s route choice behavior in recent years. Existing commonly used travel time reliability measures, such as planning time index and buffer index, tend to be calculated with travel times of all trips, without differentiating various sources of travel time unreliability. This paper uses an example to explore travel time reliability and variability under different circumstances, namely normal traffic without incidents or roadwork, normal traffic with incidents but no roadwork, normal traffic with roadwork but no incidents, and finally traffic for all conditions. Results show that different travel time reliability measures have different variations from one circumstance to another, and buffer index tend to be more sensitive to different circumstances compared to planning time index. Roadwork more likely deteriorates buffer index than incidents do, meaning disseminating accurate roadwork information is more important than incident information in order to reduce the additional travel time that is necessary for on-time arrival. Distinctions are also observed between travel time reliability and variability. Cautiousness should be taken when generalizing the case study results.