عنوان مقاله [English]
Modeling the behavior of individuals in destination choice plays an important role in travel demand analysis and hence assessing the impact of transportation policies. Most researchers have used the multinomial logit model to examine this choice, which does not allow for the heterogeneity between individuals and alternatives. In the mixed logit model, for some coefficients of the utility function, random distribution is considered, hence the heterogeneity of individuals. In contrast, in the Heteroscedastic Extreme Value model an independent and non-uniform distribution of Gumbel is assumed. This assumption leads to a heterogeneity between alternatives. The purpose of this paper is to examine the heterogeneity between individuals and alternatives in choosing the destination of shopping trips. Therefore, the multinomial logit (as the base model, and without considering the heterogeneity between individuals and alternatives), mixed logit and Heteroscedastic Extreme Value are used and their results are compared with each other. In this study, unlike previous studies, the person chooses his destination based on the distance of the destination set to his residence. In order to achieve the goals of this article, information on shopping trips by residents of Qazvin city is used. The results of the comparison of the calibrated models for 1570 person-trips based on various criteria (such as Log likelihood and goodness of fit coefficient) indicate the superiority of the mixed logit model compared to the other two models. It is observed that the people behave in a heterogeneous way in order to choose their destination in the face of the distance variable. Also, people with a departure time before 12:00 have different views on the purpose of shopping away from their homes. On the other hand, the comparison of the multinomial logit model with the Heteroscedastic Extreme Value shows that during the decision-making process, non-observed factors in the utility function are considered independently and distributed with the same distribution. The results of this paper reduce the constraints of the conventional choice of destination models and enable the researcher to consider the heterogeneity and, consequently, bring the results closer to reality.