Extensive, high-quality data resources are critical to the understanding of nonrecurring highway congestion and travel time reliability. The reliability of transportation facilities can only be assessed in the context of a statistical distribution of travel times. A number of things affect travel times on a day-to-day basis, including fluctuations in travel demand, inadequate base capacity, weather, traffic incidents, special events, work zones, and poorly functioning traffic control devices. Months and months of travel time data and related data such as weather conditions are needed to understand reliability problems and how they can be addressed on a regional or corridor basis. The report establishes several alternative information technology architectures that could be used to develop an online reliability data archive, and it analyses their advantages, disadvantages, and costs. The report recommends a solution based on cloud computing data storage and a mixture of open-source and commercial, off-the-shelf software. This alternative was further assessed through the construction of a prototype data archive, which is also described in the report. The prototype archive was populated with a variety of data from SHRP 2 Reliability Project L03, which created a large quantity of data holdings of different types. Finally, the report finds that a SHRP 2 Reliability Archive is feasible and recommends that SHRP 2 move ahead with projects to design, build, and populate an online archive for the Reliability focus area. Work on the data archive is planned to be under way in 2011. It will ultimately be populated with data from all of the SHRP 2 Reliability research projects and closely related projects from other SHRP 2 research focus areas. (Author/publisher)
Abstract