Advance Warning of Stopped Traffic on Freeways: Current Practices and Field Studies of Queue Propagation Speeds

A major safety concern on freeways is traffic flowing at normal speed encountering unexpected slow or stopped traffic. Traffic can be queued due to recurrent congestion, work zones, or collisions and/or other incidents. Drivers encountering queues are often faced with rapidly changing conditions in terms of queue length, sight distance to the end-of-queue, terrain, and available warning devices for traffic control. The rear-end collision is the most common type of multi-vehicle freeway collision, often due to slow/stopped traffic on the main lanes. This report summarizes the first year of this effort, which was to conduct the literature review, to determine current practices for advance warning for stopped traffic, to observe field locations with traffic stopped due to congestion conditions, and to determine advance warning techniques applicable to Texas. In the observational field studies, researchers found instances of sustained, repetitive, and excessive queue propagation speeds. Additionally, in many instances, multiple lanes were impacted. Urban commuters, although generally aware of conditions encountered in their daily travels, might still be surprised by sudden and extensive queues. Unfamiliar drivers might experience conditions that tax their ability to respond without incident. All drivers are particularly vulnerable when geometric conditions unfavorably coincide with queue buildup. Queue warning systems, in order to be effective, should be installed in consideration of rapidly fluctuating queues. This axiom means that warning signs placed too close to queue tails might be overrun, with the possibility of drivers encountering the queue before seeing the sign. Warning signs placed too far from the queue, if the downstream location of the queue is mentioned, can become inaccurate between the time drivers view the sign and encounter the queue. Conditions change too quickly for human operators to handle appropriate warning sign adjustments, necessitating an automated system for real-time adjustment of queue position. Geolocated queues, for which drivers are advised of the distance to the queue tail, require multiple detection stations and warning sign locations. Many factors remain to be addressed in future research; however, observations conducted in this project can provide guidance to those testin and implementing and operating systems for advance warnin of slow/stopped traffic on freeways.