Emergency Response in Advance Transportation Systems: Studying Data Capture and Routing Methodologies Using Simulation

The dissertation presents the development of a transportation system simulation model using Rockwell ARENA simulation software. The model used in evaluating alternative data capture scenarios of an advanced transportation system and different routing methodologies that could be used by emergency responders. The first part concentrates on the simulation model developed to realistically represent an approximately 3.5 by 3.5 mile area surrounding the University at Buffalo North Campus. The necessary data obtained for the area, such as average daily traffic counts, and information required to model the transportation network are presented. Using historical crash data, the modeling of the emergency vehicle is verified. The next part of the dissertation implements alternative data capture scenarios, and using a modified Dijkstra's algorithm routes emergency vehicles throughout the simulation model under various conditions. The results of these experiments demonstrate the performance of the routing methodology and the different levels of accuracy and placement of sensors. These results also show the importance of representing the travel of emergency vehicles accurately, as routing methodologies may perform very poorly if the wrong modeling is implemented. The third part of the dissertation focuses on a risk-averse routing technique knowing as the Variance-Constrained Shortest Path problem and how well it performs under the assumptions made in the simulation model. The results of this part show that certain limitations of the simulation model may hinder the results of such experiments, but do allow for demonstration of the data collection and routing methodologies. Finally, closing remarks and future research are presented.