A discrete shipment size choice model with latent classes of shipments' attributes

Abstract

Behavioral heterogeneity with respect to actors and shipments is a ubiquitous topic in the context of freight transport models. The necessity of consideration of these aspects and the challenges going along with the implementation into a freight transport model framework form a conflicting setting that needs to be analyzed in-depth in conjunction with solution approaches satisfying both aspects. In this study, we model the shipment size choice behavior by using a discretized total logistic costs approach as high-level logistics decision and account for the exceeding heterogeneity by forming segments of similarly behaving actors. This methodology enables the application in freight transport models as well as the analyses and prediction of policy impacts. We also give an outline how the latent segments can possibly be implemented into a German freight transport model.