Sub-problem 3c: The Southwestern Quadrant

At the point of the merge, we have a 4-lane basic freeway section with a free flow speed of 65 mph and a volume of 5,290 veh/hr. This produces a density of 20.8pc/mi/ln, which is LOS C. A similar analysis can be done further downstream, after the lane drop has occurred, where the density is 28.1 pc/mi/ln and the LOS is D. Here the freeway is more congested, as the traffic from the ramps is added and the number of lanes is reduced. Overall, the operation is still within an acceptable range for an urban freeway.

Where Route 7 ramps join I-787 south is a double-lane on-ramp. The volume on the ramp is the total of the traffic from the two sources, and the outside lane on the ramp ends 790 feet downstream of the merge. The remaining ramp continues on as a new, third freeway lane, similar to the issue that arose with the right-hand ramp from I-787 south to Route 7 west.

The HCM ramp procedure asks us to specify lengths for both the first and second acceleration lane. The first ramp ends 790 feet downstream of the initial merge, but the second lane doesn’t end, so we need to assume a long arbitrary distance. In Sub-problem 3b, the LOS for a merge is based on the density of the influence area.

This LOS assignment is subject to two conditions. 1) The HCM defines 4,600 veh/hr as the maximum number of vehicles per hour that can enter the influence area. If the volume entering the influence area is greater than 4,600 veh/hr, the merge is considered to be at LOS F. 2) The second condition restricts the volume exiting the influence area to be the appropriate merge capacity values from Exhibit 25-7 of the HCM. This creates inappropriate results when assuming a length for a lane which does not end. If we assume the number of lanes exiting the influence area is reduced by one, the capacity is reduced as well. This can cause a merge analysis to produce an inaccurate LOS F. If a large value is assumed for the acceleration lane length, the density produced by a merge analysis is a product of a regression equation that depends on the acceleration lane length. The resulting equation could misrepresent the actual densities that occur in the influence area by misconstruing the influence area itself.