A behavioral mode choice microsimulation model for freight transportation in the U.S

This study embarked upon the development of a nationwide freight activity microsimulation as an acceptable analysis tool for policy assessments. Complexity of reproducing the freight shipment decisions is rooted into the lack of an acceptable freight modeling framework, and freight data scarcity both of which are targeted in this dissertation. A large-scale behavioral microsimulation framework, named Freight Activity Microsimulation Estimator (FAME) was introduced in this study. A new concept for firm-types is implemented in FAME to keep the computational burden at a reasonable level and to diminish the need for highly disaggregated data A total of 46,243 firm-types were generated in 130 domestic zones in the U.S., among which more than 10 billion tons of goods were simulated. FAME is heavily based on public freight data in the U.S. and therefore data collection costs are substantially mitigated. It is also one of the early efforts in freight demand modeling that has a separate component for supply chain configuration. Furthermore, this study proposed a cost-effective way of collecting disaggregate freight data, a valuable piece of information that considerably increases cost of such studies in many cases. Two major modeling efforts were also conducted as part of this research. Two freight mode choice models were calibrated based on the surveyed data, along with a fuzzy rule-based expert system for supplier selection. Since the primary objective of this simulation was to shed light on the mode choice behaviors in freight transportation, total tonnage, value, and ton-mile of commodities for each mode was obtained as the final output. FAME outputs, however, showed a satisfactory match with public freight data in the U.S.