AUTHOR=Rezapour Mahdi , Ksaibati Khaled 

TITLE=Random regret minimization for analyzing driver actions, accounting for preference heterogeneity

JOURNAL=Frontiers in Built Environment

VOLUME=Volume 8 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/built-environment/articles/10.3389/fbuil.2022.1000289

DOI=10.3389/fbuil.2022.1000289

ISSN=2297-3362

ABSTRACT=Increasingly more studies have implemented the Random Regret Minimization (RRM) as an alternative to the Random Utility Maximization (RUM) for modeling travelers’ choice-making behaviors. While for the RUM, the focus is on the utility maximization, for the RRM, the emphasis is on the regret of not selecting the best alternative. This study presents the RRM and the RUM, for modeling actions made by drivers that resulted in crashes. The RRM method was considered as it is assumed that the actions before crashes might be more the resultants of avoidance of regrets across the alternatives rather than the maximization of the utility related to the considered attributes. In addition, the considered models were extended to account for the unobserved heterogeneity in the datasets. Finally, we changed the means of random parameters based on some observed attributes and observed heterogeneity. The results show that while the standard RUM outperforms the RRM, the standard mixed models and the extended mixed models, accounting for observed heterogeneity, outperform the other techniques. As expected from the methodological structure of the RRM, we found that the RRM performance is very sensitive to the included attributes. For instance, we found that by excluding attributes of the drivers’ condition and drivers under influence (DUI) attributes, the RRM significantly outperforms the RUM model. The impact might be linked to the fact that when drivers are under abnormal conditions or influenced by drugs or alcohol, based on the sum of pairwise regret comparison, the inclusion of those attributes deteriorate the goodness of the fit of RRM. It is possible that those parameters do not make a difference on regret pairwise comparison related to alternatives. The discussions at the end of this article examine possible reasons behind this performance.