An Investigation into the Application of Dynamic Merge Control in Work Zones: A Simulation Study
View/ Open
Date
2015-08-19Type of Degree
Master's ThesisDepartment
Civil Engineering
Metadata
Show full item recordAbstract
With increasing highway traffic volume, construction and maintenance works are quite frequent. Upstream lane merging manoeuver and capacity bottleneck can result from a work zone lane closure, which pose increased safety risk and reduced traffic flow efficiency. Dynamic Merge Control is an application of Intelligent Transportation Systems (ITS) technology in work zones, which is expected to improve the safety and mobility of the through traffic movement by governing the lane change manoeuver of vehicles from closed lane to the open lanes. The objective of this study is to examine the effectiveness of the two forms of Dynamic Merge Control in work zones, the dynamic early merge and dynamic late merge, in comparison with the conventional lane closure scheme according to MUTCD. The measures of effectiveness to compare the three strategies are vehicle throughput and delay. Dynamic early merge encourages drivers to merge from the closed lane to the open lanes well in advance of the work zone lane closure to lower the chance of friction between the vehicles in the open lane and merging vehicles at the merge point of a lane closure. On the contrary, dynamic late merge encourages drivers to make the full use of roadway storage capacity by encouraging them to go all the way and merge immediately before the work zone taper. To evaluate the efficiency of a traffic system, micro-simulation models have been proved to be very effective in lieu of field study. Therefore, a micro-simulation model of a two-to-one freeway lane closure has been developed in the traffic micro-simulation software VISSIM. The layout of the dynamic merge control is varied by manipulating the number of dynamic message signs and the spacing between the signs. Then the layouts were tested under different traffic demands, and sensors’ threshold occupancy rates. After running a thorough statistical analysis, it was obtained that if the traffic demand volume is less than 2000 vehicles/hour, then irrespective of the truck percentages and measures of effectiveness, conventional merge performs significantly better than dynamic merge control. When the measure of effectiveness is vehicle throughput, the dynamic late merge is found to perform significantly better than conventional merge only in the case of high heavy vehicle percentage. On another note, if delay is the determining criterion for the performance and selection of temporary traffic control, then irrespective of the truck percentage, dynamic early merge performs significantly better than dynamic late merge.