A statistical approach to analyze user behaviour in road evacuation

The international literature related to evacuation conditions proposes many studies which focus on the hurricane emergency case and consider revealed preference (RP) surveys for demand model estimation. As RP data are not available for all dangerous events, such models, derived from observation of past evacuation behaviour, cannot be directly applied to other dangerous events. Prediction of user behaviour becomes essential. For this purpose, evacuation trials and stated preference (SP) surveys may be conducted. As a first step, statistical analysis of stated behaviour in emergency conditions could be carried out. In this work we propose several hypothetical scenarios, which could be classified as SP scenarios, considering three different dangerous situations, with delayed effects in time. A sample of users is considered and their responses are statistically analyzed to study user behaviour in evacuation conditions.