The Contribution of Ramp Demand in the Capacity of Merge Bottleneck Locations

Transportation engineers rely on the Highway Capacity Manual (HCM) for estimating capacity at freeway segments. According to the HCM 2010, the capacity of the basic freeway segments is a function of the free-flow speed and it ranges from 2,400 passenger cars per hour per lane (pc/h/ln) for FFS 70 or 75 mi/h, to 2,250 pc/h/ln for FFS 55 mi/h. The freeway merge segments methodology in the HCM 2010 Update uses these same capacity values in the analysis procedure, although research has shown that capacities at these bottleneck locations are considerably lower. In addition to that, researchers have also observed that capacity varies significantly from day to day and from one site to the other. Researchers acknowledge that driver behavior and frequent interactions between mainline and ramp vehicles at these junctions are the causal factor of these variations in capacity and the low capacity values; however, this has not been reflected in the updated version of the HCM 2010. Furthermore, the HCM 2010 Update does not account for the conflicting movements and the contribution of the ramp vehicles on the overall merge junction capacity. This paper investigates the relationship between freeway and ramp demand and capacity at merge junctions. For the purposes of this research, historic data at merge bottleneck locations across North America with different geometric and operational characteristics were analyzed. The results of the analysis show that, there is a clear correlation between ramp demand, freeway demand and freeway capacity. More specifically, higher demand on the on-ramps produces lower overall capacity values. In addition, this paper proposes new capacity values for merge junctions as a function of the freeway and ramp demand and number of lanes.