Measuring Congestion: Learning from Operational Data

Abstract

As transportation agencies develop operations-based approaches to congestion management and carefully weigh system investments in operations as well as capital improvements, new types of performance information and communication are clearly needed. Citizens expect action on traffic congestion and demand performance information that relates to their concerns and is easy to understand. Traditional congestion measurements are based on modeled speed estimates generated from volume and capacity information. They are difficult to communicate, fail to capture subtle changes in real-world system performance, and are inadequate for many aspects of evaluating specific impacts of projects on congestion. The Washington State Department of Transportation (WSDOT), like other transportation agencies, has been grappling with new approaches to using operational data to monitor and measure system performance. This study describes progress the agency has made to date and how this work ties into national efforts. Speed and travel time data derived from the loop detector network on Puget Sound urban freeways are used by WSDOT to measure and communicate realtime travel times on 12 major commute routes. Continued work toward analyzing different types of congestion distinguishes between recurrent congestion caused by inadequate capacity and nonrecurrent congestion caused by incidents, inclement weather, and other factors such as travel to and from major sporting events. The agency also has made progress in measuring what is of most concern to commuters and haulers: travel time reliability.