Generally, pass-by noise levels measured outdoors vary according to the influence of weather conditions, background noise and the driver’s skill. Manufactures, therefore, are trying to reproduce proving ground driving conditions on a chassis dynamometer. The tire noise that occurs on actual road surfaces, however, is difficult to reproduce in indoor tests. In 2016, new pass-by noise regulations (UN R51-03) will take effect in Europe, Japan and other countries. Furthermore, stricter regulations (2dB) will take effect in 2020. In addition to the acceleration runs required under current regulations, UN R51-03 will require constant speed runs. Therefore, an efficient measurement methods are necessary for vehicle development. To solve the above mentioned issues, an indoor evaluation system capable of reproducing the tire noise that occurs on road surfaces has been developed. Using a newly developed semi-anechoic chamber with a chassis dynamometer, pass-by noise at a 7.5 m microphone position can be measured. The facility has a background noise level of NC15 and an exhaust gas circulation system to accurately measure exhaust noise. By combining tire noise levels on actual road surfaces with indoor test levels, reproduction of pass-by noise in a chassis dynamometer can be achieved. In order to identify the tire noise on the actual road surface, the noise near the tires and the road is first measured outdoors, then all sources except for the tire noise are removed from a contribution analysis of then indoor measurement. For accurate measurement of tire noise during acceleration, a new technique was developed that can maintain the relative location of microphone and noise source. By applying a SUB matrix method which performs an inverse matrix operation on each noise source, the accuracy of the contribution analysis can be improved.