California Statewide Freight Forecasting Model (CSFFM)

University of California, Irvine Institute
Goals and Objectives: 

The model is being developed to operate as a comprehensive freight analysis and modeling tool that identifies freight movements by truck, rail, and air.


A direct demand input-output model, an inventory-based firm microsimulation of vehicle flows, and use of truck GPS activity pattern data in sub-TAZ truck tour disaggregations and time of day O-D modeling.

Overview and Description: 

The California Statewide Freight Forecasting Model (CSFFM) is a policy-sensitive tool that is intended to forecast commercial vehicle and commodity flows within California. The model will address socioeconomic conditions, freight-related land use policies; environmental policy; and multimodal infrastructure investments. The model will allow for a better understanding of freight movement related policies, regulatory decisions, project decisions, and the impacts of these decisions. The core model is based on a variation of the traditional four steps: regression, logit share model, mode-route choice, and peak vehicle flow distribution. The steps for the enhanced model are listed below:

  • Direct Demand Input Output Model
  • Transshipment Network Model
  • Inventory-Based Firm Microsimulation
  • Truck Tour Disaggregation
  • Stochastic Assignment
Spatial Detail: 

The flow modeling is based on  96  TAZs in California; 7 TAZs in neighboring states; :46 TAZs in rest of United States, and 8 international TAZs and 51 air, water and rail “transshipment logistic nodes”: for a total of 208 origin/destination trip ends.

Mode Detail: 

Truck, Rail, Air.

Commodity Detail: 

14 commodity groupings based on aggregation of 2-digit SCTG commodity classes.


The model is meant for use by:

  • Planners and engineers in the Caltrans districts
  • Used by other state agency staff (i.e., California Air Resource and the California Energy Commission)
  • Used by planners in the MPOs and the Regional Transportation Planning Agencies

And to be a policy-sensitive model to forecast commodity flows and commercial vehicle flows within California addressing:

  • Socioeconomic conditions
  • Land use policies related to freight
  • Environmental policies
  • Multimodal infrastructure investments
System Platform: 

Citilabs CUBE

Data Sources: 

Public data sources through freight data repository, CALFRED:

  • Caltrans (AADTT; Motor Vehicle Stock Travel and Fuel Forecast)
  • Bureau of Transportation Statistics (Border Crossings Data; County Business Patterns; Transborder Surface Freight Data; Vehicle Inventory and Use Survey)
  • Department of Finance
  • California Air Resources Board (Estimated Annual Average Emissions)
  • FHWA (FAF2)
  • Pacific Maritime Association (Hours, Wages, and Shifts Report and Tonnage Report)
  • EPA (National Emission Inventory)
  • STB (Rail Carload Waybill Sample)
  • RAND California (Major Airport Operating Statistics)
  • Association of American Railroads (Rail Performance Measures Weekly Performance Report)
  • USCG (Marine Casualty and Pollution Investigation)
  • US Customs Vessels Entrances and Clearances
  • Waterborne Commerce Statistics Center
  • Waterborne Transportation Lines of the United States Vessel Characteristics
Example Inputs & Outputs: 
  • Inputs: socioeconomic data, exports/imports (commodity flows), commodity to vehicle distribution, transportation networks, transshipment data,
  • Outputs: spatial commodity flows, spatial commodity productions and consumptions, vehicle flows by mode (truck, rail, air).
Software Information: 

Anticipated completion is February 2014