Problem statement: Commodity Flow Survey (CFS) was launched to collect
comprehensive freight flow data throughout the kingdom of Thailand. The surveys database is the
most complete collection of commodity flow data in Thailand. The need to reveal interregional freight
characteristics using available data from the CFS led to the objectives of this research. Approach: An
origin destination matrix based on province was calibrated using a flexible Box-Cox function form. It
used maximum likelihood and the backward method for calibration and Root Mean Square Error
(RMSE) and Mean Relative Error (MRE) to verify the models performance. Independent variables
were classified into three groups: origin variable, destination variable and geographic variable. The
origin variable represented the behavior of the trip as generated at the place of origin. Some
consumption occurred at the origin. The employment and the average plant size variables were
selected for potential productivity while personal income per capita and total populations were
included to explain consumption behavior at the origin. Personal income per capita and total
populations were selected for destination variables which act as proxy for final demand at the
destination. The third category, distance, was the most conventional friction variable for geographical
variables. Results: The calibrated model revealed that origin income, origin average plant size and
origin population performed poorly. Therefore these variables were eliminated. The best developed
model included four strongly significant variables at a 5% level: origin employment, destination
population, destination income per capita and distance. Conclusion: The results showed that the
selected variables and the Box-Cox functional form were successful in explaining behavior of
interregional freight transportation in Thailand. The developed model was the first interregional freight
transportation model to be calibrated against Thailand commodity flow survey data