A joint model of mode and shipment size choice using the first generation of Commodity Flow Survey Public Use Microdata

A behavior-based supply chain and freight transportation model was developed and implemented for the Maricopa Association of Governments (MAG) and Pima Association of Governments (PAG). This innovative, data-driven modeling system simulates commodity flows to, from and within Phoenix and Tucson Megaregion and is used for regional planning purposes. This paper details the logistics choice component of the system and describes the position and functioning of this component in the overall framework. The logistics choice model uses a nested logit formulation to evaluate mode choice and shipment size jointly. Modeling decisions related to integrating this component within the overall framework are discussed. This paper also describes practical insights gained from using the 2012 Commodity Flow Survey Public Use Microdata (released in 2015), which was the principal data source used to estimate the joint shipment size-mode choice nested logit model. Finally, the validation effort and related lessons learned are described.