Vehicle service loads need to be measured under actual driving conditions in order to ascertain the strength and durability requirements for customer usage. However, because road-load data are influenced by the vehicle category and specifications of a vehicle, it takes an enormous amount of time and labor to compare them with existing data and analyze the relationship between service loads and vehicle specifications. This paper presents a study concerning the use of the frequency response function, which differs from one vehicle to another, as a means of predicting the response of a particular target vehicle based on the road-load data, which are input by unevenness of roads, measured for a baseline vehicle. The results predicted in bench tests showed a margin of error of around plus or minus 20%, making them sufficiently accurate for practical use. Rice's formula was applied as an index for evaluating the prediction accuracy of the power spectral density. A method is also described for using this formula to predict the frequency even in the case of data which are not normal distributions as a whole. (A) For the covering abstract see ITRD E121867.