Man-made or natural disasters, either noticed or un-noticed, could result in severe life losses and property damages. Hurricane Sandy, for example, made landfall in a heavily populated urban area (New Jersey and New York City) in late 2012, causing117 deaths in the US and 69 more in Canada and the Caribbean, and leaving millions of people homeless and billions of dollars of damages. Emergency evacuation, a mass movement of people and their properties from disaster-impacted areas to safer ones, has been studied and practiced for decades as one means of countermeasures to mitigate these calamitous consequences. Numerous evacuation planning models have been developed in the past. Many of them are either too time consuming and hard to calibrate (e.g., for micro- or mesoscopic models) or too simple in terms of traffic flow dynamics (e.g., for macroscopic models) to produce accurate results.