Abstract This paper presents an alternative to Random Utility-Maximization models of travel choice. Our Random Regret-Minimization model is rooted in Regret Theory and provides several useful features for travel demand analysis. Firstly, it allows for the possibility that choices between travel alternatives may be driven by the avoidance of negative emotions, rather than the maximization of some form of payoff. Secondly, it acknowledges that traveler decision-making in the context of multiattribute alternatives may not be fully compensatory. Besides this, we show how the Random Regret-Minimization approach is straightforwardly extended towards the case of risky travel choice, using the notion of Expected Regret. Finally, our Random Regret-Minimization model provides a straightforward and intuitive way to incorporate the notion that travelers, when faced with knowledge limitations, may wish to postpone their choice and search for more information first. The developed model is estimated on data from a multimodal travel simulator, where participants could choose between travel alternatives with risky travel times and costs, and the option of travel information acquisition. Estimation results support the validity of the proposed model of Random Regret-Minimization.