الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 155 |  |  |
| | | from | 155 | | |
Abstract
Uncontrolled on-ramp merging section is considered as bottleneck sections that causes severe traffic congestions at several locations on urban expressways in Greater Cairo Region (GCR). Ramp metering control strategy, which aims to regulate the entry of on-ramp vehicles onto the mainline, has proved several benefits in improving the traffic performance and safety on many freeways and expressways around the world. However, no studies have been conducted to assess the impacts of implementing such strategy on GCR roads network. Accordingly, the aim of this thesis is to examine the effects of applying ramp metering control system on urban expressways in GCR.
Four on-ramp sites with different geometric configurations were selected along the 6th of October corridor and the 15th of May corridor as a case study. The evaluation process was conducted based on the microsimulation methodology by using Vissim software package. The developed simulation models were calibrated and validated by using the traffic data collected during peak hours at the selected sites. Two different control strategies were tested at each site: fixed-time ramp metering strategy and fixed-time ramp signals integrated with mainline signals. The both two strategies were compared to “no control” condition and evaluated based on three different performance measures; average speed, average vehicle delay, and on-ramp queue length.
The results of the microscopic models indicated that implementing fixed-time ramp metering control strategy showed significant improvement in the traffic performance on the whole system (i.e., ramp and mainline traffic). However, fixed-time ramp metering control showed negative impacts on the traffic performance on the on-ramp. Therefore, the optimum signal timing was selected based on the reduction in the whole system delay. At three sites, fixed signal timing scenario with cycle length of 10 seconds and allow one car to enter per green was found to be the best scenarios in reducing the average vehicle delay on the whole system and mainline segments regardless the on-ramp queue length, although their different geometric configurations. It improved the overall traffic performance at site-1 by increasing the speed by 37% and reducing the delay by 38%. At site-2, the speed has been increased by 97% and the delay has been reduced by 48%, although it increased the on-ramp queue length significantly over the ramp storage length. Moreover, it increased the speed at site-3 by 62% and reduced the delay by 49%. At site-4 with multiple on-ramps with different traffic volumes, metering the on-ramp with the highest traffic volumes only with signal timing of 14 seconds and two cars are allowed per green increased the overall speed by 118% and reduced the delay by 57%.
Considering the on-ramp queue length, the optimum scenario was different based on the geometric configuration and on-ramp storage length for each site. Scenario with 16 seconds cycle length and 3-CPG was the optimum scenario at site-1 and scenario with 10 seconds cycle length and one car per green was the optimal scenario at site-3, while all scenarios resulted in queue spillback problem at site-2 and site-4. Therefore, at site-2, integrating the best fixed-timing ramp metering scenario with mainline signals was the optimal control technique in increasing the average speed by 28%, reducing the delay by 11% and reducing the on-ramp queue length to 130 m.
Based on the study findings, it could be concluded that applying ramp metering control systems at the examined sites will improve the network performance under a certain limit of ramp traffic volume regardless the geometric configurations of on-ramps. Otherwise, both ramp and mainline control strategy could be applied, if there is sufficient storage length on the mainline upstream segment with high on-ramp traffic and insufficient storage length. Further studies are required to test the multi-enters coordinated control systems and to develop on-line ramp control algorithm that accommodate with the local conditions of traffic and driver’s behavior.