Abstrato

Capacity of Interchange in Outer Connection Ramps

Jihad Obeidat

Interchanges are constructed in urban and suburban areas to ease traffic operation. Outer connection ramps constitute a major part of interchanges planning and design. The objective of this study was to estimate interchange ramp capacity and investigate the impacts of ramp geometry on the obtained capacity. To achieve this objective, 20 ramps were selected, including 10 circular ramps and 10 curve- straight-curve ramps. Capacity of ramp proper and ramp exit were investigated in the study. At ramp proper, data on traffic speed and traffic flow were collected at 1-minute intervals using video camera. Also, traffic data for ramp exit and mainline traffic were obtained under queued traffic condition on the ramp exit. An empirical approach using regression analysis was adopted to estimate capacity of the selected ramps and explore the impact of geometric design and traffic variables that might affect the estimated capacity. For ramp proper, the analyses revealed that the relationship between traffic speed and density is linear and between speed and traffic flow is parabolic irrespective of ramp configuration. For circular ramps, the capacity was found to vary from 1470 to about 2100 pc/hr./lane, and it was found that ramp radius is the most influencing factor in capacity estimation. For curve-straight-curve ramps, the results indicated that the obtained capacity varied from 1490 to 2200 pc/hr./lane, and both the straight segment length and the radius of the first curve are the most influencing factors. Also, it was found that ramp exit capacity is affected by flow and speed of traffic in the mainline street and degree of curvature of ramp exit curve. Finally, it was concluded that the use of curves with large radii at ramp proper and exit would increase ramp capacity, and customization of capacity values to reflect local conditions would provide more realistic estimates.

Isenção de responsabilidade: Este resumo foi traduzido usando ferramentas de inteligência artificial e ainda não foi revisado ou verificado