Natural and man-made disasters are increasing in intensity, costing lives of thousands of people and millions of dollars in damage throughout the world. As the population density in major cities increases each year, the effects of these disasters can only increase. The intensity of these disasters may occasionally require an execution of an emergency evacuation order. In order to reduce the uncertainty before, during, and immediately following such an event, emergency response agencies are required to have evacuation plans. One element to include in these plans is the positioning of Emergency Response Vehicles (ERVs) to direct traffic away from the affected areas towards evacuation zones. The aim of this study is to generate a mathematical model to optimally assign and route ERVs to critical points in the road network. An effective and efficient algorithm is needed to route ERVs. This thesis extends the Dijkstra's Algorithm for dynamic travel times and integrates it with an assignment model. This combined approach minimizes the total travel time or latest arrival time of ERVs much faster than the mixed integer programming (MIP). The methodology is validated using an evacuation road network in a metropolitan area.