This paper proposes a commodity freight generation demand model in such a way that the movement of commodities is explained by aggregate commodities production and attraction by zone. The proposed model is a modification of the first step of traditional four-step approach, which utilizes regression analysis. Regression analyses correlate freight production and freight attraction based on land use and natural resources productivity of the selected region. The analysis compares eight regression models within nine parameters. The research shows the using of multiple-linear regression and non-linear regression using inverse function to be more accurate and elastic than the other regression analysis.