Capturing Pre-evacuation Trips and Associative Delays: A Case Study of the Evacuation of Key West, Florida for Hurricane Wilma

The time it takes for the residents to evacuate an area is calculated as an evacuation time estimate (ETE). In theory, these time estimates are calculated based on a number of inputs, including clearance time, the impact of traffic management techniques, and the time for the public to prepare to evacuate (Dow, 2000). Evacuation models can calculate clearance times, as well as incorporate the temporal impact of traffic management techniques, like contra-flow traffic. However, these models do not include delays associated with pre-evacuation trips. Because these trips are not well represented in hurricane evacuation models, the evacuation time estimate may be miscalculated (Wilmot and Mei, 2004).

In order to capture pre-evacuation trip behavior, an online survey of residents’ responses to the evacuation order associated with Hurricane Wilma in 2005 was conducted. Survey data gathered from the residents of Key West, Florida indicate seven important aspects of preevacuation trip making behavior: (1) Socio-demographic variables, which have some association with evacuation behavior, were found to have a very weak to no association with pre-evacuation trip-making; (2) Socio-demographic variables were found to have an association with predicting evacuation behavior for respondents making pre-evacuation trips, and these associations are consistent to what has been found in other studies of evacuation behavior; (3) Delays at stops are longer than delays on links; (4) Trip delays are associated with trip purpose; (5) Residents did not travel from a single origin point directly to an evacuation point, but made various preevacuation trips, often exhibiting trip chaining that included traveling toward as well as away from the city; (6) Though the Key West mandatory evacuation for Hurricane Wilma consisted of a phased evacuation based on housing types, the residents did not evacuate based on housing type, and (7) the personal stories offered by respondents indicate that evacuating or not is often related to job requirements, economic opportunity, previous evacuation experience, and evacuation burnout.