Safety hazards at construction/work zones are often associated with traffic crashes, aggressive driver behavior and traffic operational problems. A majority of traffic crashes in work zones occur at lane closure areas. For example, in Michigan approximately 6950 work zone crashes occurred in 1999, of which 47% occurred at lane closure areas. One situation that contributes to hazards at lane closure areas often pertain to the "late merge phenomenon". The Michigan Department of Transportation (MDOT) provided research funding to the authors of this paper to develop a traffic control system which mitigates the late lane merge phenomenon and aggressive driver behaviour at lane closures in highway work zones. The initial system installation included both traditional lane closure signs and a series of dynamic no passing zone signs mounted on trailers. The dynamic "Do Not Pass When Flashing" sign trailers are equipped with detectors to capture speed, volume and lane occupancies at the detection zone. These signs are located approximately 1,500 feet apart and are set to begin flashing indicating no passing zone of variable length (dynamic) in effect, based on the detected traffic volume and occupancies. The objective of this research was to develop a work zone traffic control system at lane closures which minimizes aggressive driver behavior and the late lane merge phenomenon. An initial test was performed using a traffic control system developed in the state of Indiana. Field testing of the system indicated some confusion among the motorists in terms of driver response. It was then redesigned to include a variable message sign with an arrow pointing to the direction in which to merge and at what point on the road to merge, followed by a series of dynamic lane merge sign trailers and a series of static "Do Not Pass" signs. A three-month long study at a suburban freeway work zone indicated a significant reduction in aggressive driver behavior, lane violations and a slight improvement in traffic flow. The number of aggressive driver maneuvers during the peak hours reduced dramatically at the test site. It reduced aggressive actions from an average of 68 per hour before the implementation of the system, to 32 per hour in the after period, at the northbound site. There was also an average of 38 per hour in the before period, to 9 per hour in the after period at the southbound site. The average peak period travel time delay decreased with the implementation of the dynamic lane merge traffic control system (LMTCS). At similar flow rates, the 'after' period average travel speed increased slightly. The average peak hour flow remained similar for the construction zones with and without the dynamic LMTCS. This paper describes the system of traffic control developed and the results of the effectiveness study. It also shows the results of a benefit cost analysis of implementing such a system at highway work zones. For the covering abstract see ITRD E124693.
Samenvatting