A simulation approach for examining capacity and operational performance at freeway diverge areas

Freeway diverge areas, and particularly those in the proximity of exit ramps, are often viewed as potential bottlenecks in freeway operations. The existing diverge procedures within the 1994 and 1997 Highway Capacity Manual updates are limited in that they do not provide a direct estimate of freeway capacity nor do they model performance at oversaturated traffic conditions. Moreover, a parallel investigation on these procedures revealed some inconsistencies in predicting measures of performance at those critical areas. This paper describes the use of computer traffic simulation to explore the patterns of capacity and operational performance behaviour at these areas under the impact of some key geometric and traffic variables. For this purpose, the microscopic traffic simulation model INTEGRATION was selected to conduct an extensive experimental work on a typical ramp-freeway diverge section. Five control variables were investigated, namely, total upstream demand, off-ramp demand, length of deceleration lane, off-ramp free-flow speed, and number of lanes at mainline. The impact of upstream or downstream ramps is considered beyond the scope of this research. Except for off-ramp free-flow speed, the impact of other control variables on capacity and operational performance was shown to be significant. Also, the simulated trends of traffic behaviour showed considerable agreement with logic and expectations in light of the current state of knowledge on freeway operations.