Some developments in transportation network analysis and design with application to emergency management problems

This dissertation is concerned with three components in the area of transportation network analysis and design, and in the application of these components to emergency management systems. The first is the stochastic user equilibrium assignment problem. The second is the multiperiod network design problem. The third is the combined trip distribution and assignment problem. It is organized in two parts.; The first part deals with aftermath network restoration problems. This part includes a stochastic user equilibrium assignment model and a multiperiod network design model. In the traffic assignment model, a new methodology for generating an efficient set of alternative routes in the network is developed. In addition, a modification to Fisk's model is proposed. Then, an algorithm for solving the modified Fisk model is developed and tested. The multiperiod network design model includes an innovative methodology for solving the multiperiod network design problem. The presented methodology is unique, and can be applied to estimate highway project benefits as well as project interdependencies. A mathematical programming model was subsequently developed to generate an optimal schedule for the selected projects.; The second part deals with the before impact network evacuation problem. It includes an implementation of cellular automata for solving the combined trip distribution and assignment problem. The objective is to demonstrate the capabilities of models based on the cellular automata concept in solving complex problems. The combined trip distribution and assignment problem was selected as a test-bed due to its high level of complexity. A mathematical model for a simplified version of the problem was developed and solved. Then, a model that combines cellular automata and a greedy search method was developed and compared to the exact model.