Abstract:This paper discusses the disaggregation of the Federal Highway Administration's Freight Analysis Framework (FAF) database (version 3.0) on freight origin-destination data and the development of linear regression equations to describe the relationships between commodity-based freight trip productions/attractions to specific economic variables. Instead of generating a production/attraction equation for each commodity, commodities are grouped in certain ways to simplify model development and application. We consider three grouping methods and two model selection criteria (with and without intercepts), which are compared in terms of goodness of fit with two data sets (FAF versions 2.0 and 3.0). Furthermore, the freight generation models are validated using county-level economic data in California and applied to predict year 2015 commodity outputs. The results of this study can help city, county, metropolitan and state level planning agencies develop their own customized freight demand generation models without performing costly large-scale surveys.