Abstract
The main goal of this dissertation is to offer Guidelines for Eco-Traffic Signal System Operations in Small and Medium Size City Environments.
The first part of this dissertation synthesizes and documents different aspects of emissions inventories, models, and optimized these models. This synthesis work focused on three areas: fuel consumption and emission modeling tools; sources of emissions inventory and data used in the models; and how to optimize these emission models.
The second part of this dissertation deals with vehicle performance modeling using advanced engine package software (GT-Suite). In this part of the dissertation, I will conduct vehicle performance modeling using an advanced engine package software (GT-Suite). The main goal is to determine the real fuel consumption, and environmental impact caused by the idling, cruising, stopping,/ and accelerating from different speeds at signalized intersection. The vehicle performance modeling was conducted on four different speed profiles as follows:
* Idling: vehicle is stopped at the intersection (0 mph) when the red display is shown
* Cruising: vehicle is cruising at different speeds (25, 35, 45, and 55 mph) when the green display is shown.,
* Accelerating from a Stop to Different Target Speeds: vehicle is accelerating from zero mph to different target speeds (25, 35, 45, and 55 mph)immediately after the green display is shown and using three different acceleration values as follows:
a) Mild acceleration, 40% of the maximum vehicle acceleration envelope, (4.7 ft. /s2),
b) Normal acceleration, 60% of the maximum vehicle acceleration envelope, (7.1 ft. /s2),
c) Aggressive acceleration, 100% of the maximum vehicle acceleration envelope, (11.8 ft. /s2).
* Accelerating from a Non-Zero Speed to Different Target Speeds: vehicle is accelerating from a non-zero speed to different target speeds (25, 35, 45, and 55 mph) when the signal display was red, and the driver was decelerating, then the signal display switched to green, and the driver is accelerating, but from a speed higher than zero mph. The same three acceleration values (4.7, 7.1, and 11.8 ft. /s2) were used in this case.
The third part of this dissertation deals with minimizing the environmental impact of corridor traffic operations in small and medium size cities. While corridor traffic in small and medium size cities does not experience the high levels of congestion typically present in large urban areas, it generates a considerable amount of emissions and vehicle pollutants, negatively impacting the environment. The primary objective of this research was to investigate different corridor traffic management plans and examine their potential benefit in reducing vehicle emissions and fuel consumption. Corridor optimizations and microscopic traffic simulation models were used to develop and test different corridor traffic management plans and to assess their operational and environmental impact. The methodology used in this research could be used to obtain environmentally friendly corridor management plans. The obtained fuel consumption and vehicle emission results showed that the timing plans optimized for excess fuel consumption appear to produce minimal effects on the environment.
In the fourth part of this dissertation, I discuss how to develop some signal control parameter guidelines for isolated intersections focusing on advanced settings like rest on red, rest on green, and delayed detection.
Throughout my Ph.D. work, I became interested in other non-emission topics. In the fifth and sixth parts of this dissertation, I discuss some of them.
The fifth part of this dissertation, appendix one, focuses on evaluating the long-term operation and safety impact of Differential Speed Limits (DSL) on rural freeways in Idaho. The analysis of speed data covering the periods 1996 to 1999 and 2009 to 2011 showed thatsince the implementation of the DSL policy, Idaho's speed trends have stabilized with no sizable change in speed data. The mean speed for trucks and passenger vehicles are very close to their respective posted speed limits (65.6 mph and 74.7 mph, respectively.) The 85th percentile speeds have also stabilized at about five mph above the respective speed limits (69.8 mph for trucks and 80.3 mph for passenger vehicles.) Implementing DSL also visibly improved the compliance rate of truck speed limits. The considerable reduction in the 85th percentile; the pace speeds for trucks; and the improved speed limit compliance rate all indicate that the DSL policy favorably impacted truck driver behavior by reducing the most extreme truck speeds. In influencing driver speed, the implementation of DSL policy has contributed to the improved safety conditions on rural freeways in Idaho. Crash rate analysis showed that DSL favorably affects safety. Before DSL was implemented, crash rates for all crash types were highest during the period 1996-1998 with a USL of 75 mph. After DSL policy was implemented in 1998, the crash rates decreased considerably and have continued to decline since that time. The results strongly suggest that DSL policy is improving traffic safety in Idaho's rural freeways.
The sixth and final part of this dissertation, appendix two, focuses on conducting a statistical comparison between the coordinates of Peachtree NGSIM left-turn vehicles and other simulated left-turn vehicles using VISSIM as a simulation tool. In this part, I began the work by plotting all left-turn vehicle co-ordinates on a NGSIM CAD diagram for the Peachtree street segment being studied, identifying four main comparison variables between the two datasets. These variables were effect of the time of day, intersection left-turn control type, intersection left turn angle, and vehicle position in the queue. The statistical comparison was performed using SAS statistical software, starting with transforming the data to follownormal Gaussian distribution, performing the ANOVA test, and slicing the interactions to determine the variables or interactions which have significant effects. The conclusion of this work showed that both of the variables of left-turn control type and intersection left-turn angle significantly affected the behavior of drivers turning left, while variables of time of the day, and vehicle position in the queue did not significantly affect behavior of drivers turning left.