Abstract: To investigate the truck load probabilistic characteristics of Chinese first-class highway, the 289-day WIM data of G104 highway was introduced. The multi-normal distribution and multi-exponential distribution were used to describe the load parameter probabilistic distributions, including the gross truck weight, axle weight, axle spacing and truck speed etc. The types and parameters of the distributions were then obtained. The results show that the multi-normal distribution can accurately describe the distribution of all the load parameters except truck gap. The triple exponential distribution is one of the best choices for truck gap distributions. The trucks of G104 highway can be divided into 9 different types, in which the 2-axle truck is predominant. The percentage of semi-trailer is pretty low and type V8 is predominant.