Date of Award

Summer 2006

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Transportation - (Ph.D.)

Department

Executive Committee for the Interdisciplinary Program in Transportation

First Advisor

I-Jy Steven Chien

Second Advisor

Athanassios K. Bladikas

Third Advisor

Lazar Spasovic

Fourth Advisor

Janice Rhoda Daniel

Fifth Advisor

Kyriacos Mouskos

Abstract

Disaster response in areas of high population density is centered on the efficient evacuation of people and possibly goods. Developing evacuation plans suitable for different levels of urgency based on the intensity of threat is a challenging task. In case of densely populated cities (e.g., New York, Los Angeles), the level of threat is enhanced by the congestion of their transportation systems, and the decision to evacuate a region simultaneously or by dividing it into multiple stages (or zones) affects the required evacuation time and associated delays.

The evolution of the traffic conditions on the evacuation route can vary significantly based on the type of evacuation strategy employed (i.e., simultaneous or staged). In this dissertation, mathematical models are developed for estimating evacuation time and delay. Evacuation time is the time for evacuating all vehicles from a designated region, while delay includes queuing and moving delays incurred by evacuees. The base model handles a uniform demand distribution over the evacuation route and deterministic evacuees’ behavior. The relationship between delay and evacuation time is investigated, and the impact of a staged versus a simultaneous evacuation is analyzed. A numerical method is adopted to determine the optimal number of staging zones. A sensitivity analysis is conducted of parameters (e.g., demand density, access flow rate, and evacuation route length) affecting evacuation time and delay.

To account for the heterogeneous demand distribution over the evacuation region and evacuees’ behavioral responses to an evacuation order (e.g., fast, medium, and slow), a more realistic model is developed by enhancing the base model. Based on a numerical searching process, the enhanced model determines the optimal time windows and lengths of individual staged zones dependent on the demand distribution, behavioral response, and evolution of traffic conditions on the evacuation route. The applicability of the model is demonstrated with a numerical example. Results indicate that evacuation time and delay can be significantly reduced if a staged evacuation can be appropriately implemented.

Finally, the impact of compliance is investigated. Compliance is defined as the conformity of a staged zone to its demand loading pattern. It is found that the level of compliance and deviation from scheduled access time influence the effectiveness of staging. Further, a method to revise the optimal staging scheme to accommodate the noncompliant demand is illustrated.

The models developed in this research can serve as useful tools to provide suitable guidelines for emergency management authorities in making critical decisions during the evacuation process.

Share

COinS