A brake durability experimental method is proposed to simulate a brake durability vehicle road test. Brake judder and noise often occur in brake durability road testing. Brake judder is difficult to address because of its many potential causes, such as assembly run out, component stiffness, lining characteristics, thermal coning/hot spot/thermal instability and corrosion. There are currently several test procedures to predict brake thermal roughness and pad cleaning corrosion performance for preventing brake judder. Brake durability vehicle road testing is performed to check brake NVH and wear; examples include the Mojacar test in Spain and the Huangshan test in China. Brake energy intensity and road vibration are the significant factors that cause brake rotor thickness variation, which generates brake judder in public road testing. This study is focused on brake pad wear depending on brake energy intensity and brake rotor temperature to simulate brake durability road testing and brake rotor thickness variation (RTV) generation induced by wear. A brake durability vehicle road test was performed to obtain the initial braking velocity, temperature, deceleration and brake pad wear during the test. A brake rotor temperature curve model and a brake pad wear model were developed on the basis of friction thermodynamics theory and the brake energy intensity wear principle. The brake pad wear and temperature were optimized to be equal to the results of the vehicle road test. A brake durability dynamometer test procedure was developed according to the mathematical model. In this study, the dyno test result is compared with the road result in brake pad wear, brake roughness and propensity for brake noise. The results are shown to be in fairly good agreement.