| Summary: | 碩士 === 逢甲大學 === 交通工程與管理所 === 97 === Freeway congestion problem in Taiwan is most serious during Chinese Lunar New Year holidays. Even when the second freeway of north-south direction is constructed in 2004, the congestion phenomenon still exists. It is clear that relieving congestion problem can be achieved not only by road construction but also by traffic management and re-distribution strategy. Therefore, the subject of this research is focused on one strategy freeway ramp metering. A study area is defined on Zhongshan freeway from Wangtien north up to Fongyuan during an interested period Chinese Lunar New Year holidays. Whenever the data from vehicle detector are sent to the control centers, it is expected that whether a ramp metering strategy needs to be engaged and how to arrange can be decided. Hopefully, by using this ramp metering strategy, the density of mainline can be controlled in such a way that the traffic of mainline is moving smoother, and the queue length of ramps can be reduced in order to save travel time of road users.
The data from vehicle detector are not complete, this research uses flow conservation principle to estimate the entering flow demands from ramp to mainline at Wangtien, Nantun, Daya junctions. A regression model is formulated at Miaoli junction, where roadway geometry is similar to that of Taichung junction, to estimate the entering flow demand from ramp to mainline at Taichung junction. Furthermore, it is supposed that the merging flow demand just after the entering point is close to the average of flow on the very right lane and the entering flow demand from ramp to mainline. Moreover, based on the hypotheses that when the flow demand decreases, the density will decrease at the same rate and that the traffic density and speed relation at the merging segment is similar to that on the very right lane, the estimated speed can be obtained through a model regressing speed on density. Base on the above mentioned first hypothesis, the speed data of each lane of mainline can also be revised whenever flow demands on mainline are changed.
After getting the traffic data of each section (entering section, merging section and mainline section), three performance indicators of ramp metering strategy are chosen. One is efficiency. Efficiency measures the travel time for each ramp metering plan. A plan is decided as the best if the travel time by this plan is the shortest. The other is transporting capability, which is defined as accumulating figures of average travel speed multiplied by traffic volume. Whenever a plan can guarantee the highest value of transporting capability, it is selected as the best. Another is equity, which is evaluated by the standard deviation of travel speeds among all relevant mainline segments. Once a plan gives a result of smallest standard deviation, it is the most equitable. Since the efficiency of each plan is considered most important from the viewpoint of freeway administrative officiers, this thesis puts most efforts on establishing plans with best efficiency, i.e. shortest travel time for the Chinese Lunar New Year holidays.
|
|---|
