Travel time reliability benefits: Allowing for correlation. (June 2015)

• Record Type: Journal Article
• Authors: Nicholson, Alan
• Is Part Of: Research in transportation economics. Volume 49(2015)
• Page Start: 14
• Page End: 21
• Publication Date: 2015-06
• Subjects: Travel time -- Reliability -- Estimation -- Benefits -- Correlation; R42; Transportation -- Periodicals; 388.05
• DOI: 10.1016/j.retrec.2015.04.002
• Languages: English
• ISSNs: 0739-8859

Abstract

Abstract: The economies in developed countries depend heavily on their transport systems. That dependence has increased with the adoption of 'just-in-time' production methods, which involve reducing the costs associated with holding goods in stock and relying on required goods being delivered promptly once ordered. This has meant an increased emphasis on the reliability of the service and increased efforts to mitigate the impacts of disruption of road transport networks, and has led to changes in reliability being included within economic appraisals of road network improvement projects that are expected to result in an improvement in reliability. Current methods for estimating the benefits of such projects typically involve estimating the change in the standard deviation of trip time and assigning a monetary value to that reduction. While it can be shown that the estimate of the standard deviation of trip time is sensitive to correlation between the travel speeds on the segments of a trip, the methods do not allow for such correlation. The few studies to assess the level and extent of such correlation have led to different conclusions regarding the need to allow for such correlation. This paper will discuss why those studies have produced different conclusions. Another issue associated with estimating the travel time reliability benefits of projects is that current traffic assignment methods do not take account of travel time variability when predicting the flows on alternative paths through networks. This means inconsistency between the flow estimation and economic appraisal stages, and casts further doubt upon the robustness of estimates of the travel time reliability benefits of road network improvement projects. The paper will review research relating to these two issues and describe research which shows that ignoring correlation can result in substantial errors in estimates of the benefits of projects that are expected to result in an improvement in reliability. The paper will conclude with a discussion of how the process for estimating travel time reliability benefits can be improved.