Hurricane is a natural disaster which could cause many deaths and considerable damage if improper emergency management was applied. Figuring out an efficient method to dynamically forecast the hurricane evacuation demand with high accuracy plays a crucial role in preparedness work of hurricane management. Recently, substantial studies and research exit on understanding hurricane evacuation behavior. However, in this thesis, some forecast covariates which were not mentioned before, are introduced into the prediction of hurricane evacuation rate. Moreover, two travel demand models are applied in this study: A Sequential Logit Model and a Cox proportional hazards model. These two models are used for estimating the probability of each household to evacuate in the specific time step. After applying the data from Hurricane Gustav (2008) in Louisiana, over 76% households’ dynamic evacuation behavior are predicted correctly.