Abstract: As a peculiar product in China today, farm transport vehicles play a very important role in economic construction and development of the countryside, but its work reliability remains low. In this paper truncated tracking is used to solve the low reliability of farm transport vehicles. Relevant reliability data were obtained by tracking a certain model vehicle and conducting reliability experiments. Data analysis revealed the weakest part of the vehicle system was the engine assembly. The theory of artificial neural network was employed to estimate a parameter of the reliability model based on self-adaptive linear neural network, and the reliability function educed by the estimation could provide important theory references for reliability reassignment, manufacture and management of farm transport vehicles.