Traffic simulation models are frequently used to support decisions when an evacuation is planned. These models typically focus on traffic dynamics and the effect of traffic control measures to locate possible bottlenecks and predict evacuation times. However, a clear view of the crucial factors that determine evacuation time and emergent traffic states is lacking. In this paper, a structured and comprehensive sensitivity analysis identifies and quantifies the impact of variations in travel demand and network supply in the case of evacuation. The sensitivity analysis involves applying the macroscopic evacuation traffic simulation model EVAQ, in which aspects such as trip generation, departure rates, route flow rates, road capacities, and maximum speeds are systematically varied. That is accomplished using a case study that describes evacuation of the Rotterdam, Netherlands, metropolitan area. Departure rates and route flow rates are found to have a substantial nonlinear impact on network conditions and arrival pattern, particularly when the network load is relatively high, whereas trip generation and road capacities have a smaller quasilinear impact. Maximum speeds, independent of the effect on road capacities, have no significant impact on evacuation. The results, discussion, and conclusions presented can be used to identify the most important factors in (a) verifying, calibrating, and validating an evacuation model; (b) designing a network for evacuation studies; and (c) evaluating and testing the robustness of evacuation plans.