Freight can be efficiently transported between most locations in the U.S. using large trucks. Involvement of large trucks in crashes can cause much damage and serious injuries, due to their large sizes and heavy weights. The purpose of this study was to identify the relationships between large truck crashes and traffic and geometric characteristics on limited access highways. Crash and traffic and geometric-related data for Kansas were utilized to develop a Poisson regression model and a negative binomial regression model for understanding the relationships. Based on modelfitting statistics, the negative binomial model was found to be the better model, which was used to identify the important characteristics. By addressing identified factors, safety could be promoted through introduction of appropriate engineering improvements.