Abstract
The framework of freight demand modeling has been the conventional four-step approach, in the past. Activity-based models that better capture the behavior and decision-making process are emerging in the passenger travel modeling to address the drawbacks of the four-step approach. This paper proposes an activity-based freight modeling framework and discuses its data needs. The framework consists of five modules: firm generation, supply chain replication, shipment forecasting, logistic decisions, and network analysis. Similar to the activity-based approach in the passenger travel modeling, in which individuals or households are the agents of the models, individual firms or a group of firms with similar characteristics, are the main players in the proposed framework. Public data sources that could be used in the U.S. to make this framework running are also introduced and a supplementary establishment survey is proposed as a cost-effective way to satisfy the data needs of the framework.