A majority of the approaches currently available for predicting rutting suggest limiting the vertical compressive strains on top of the subgrade in a pavement structure. Such approaches do not explicitly consider the rutting in individual layers. To factor the influence of axle loads, and environmental conditions on bituminous layer rutting, a rut depth prediction model is proposed in this investigation using the rut depth data generated from AASHTOWare Pavement ME Design. The global calibration constants were estimated for the mixes considered in this study using an improved creep and recovery test. Traffic and axle load data from 12 National Highways and weather data from 9 locations in India were used in this investigation. To identify the variables for the rut depth model, a sensitivity analysis was conducted on the rut depth generated for various case scenarios. It was seen that the annual average daily truck traffic (AADTT), speed of vehicles, and bituminous layer thickness were found to exert considerable influence for a given geographical location. Using the rut depth data from 216 simulations, a rut depth model was developed using response surface methodology. Calibration and validation of the rut depth model were carried out from the data collected for different traffic and climatic conditions existing in India. In addition, thickness optimization using the rut depth model was illustrated for extreme combinations of traffic and weather conditions. © 2020 American Society of Civil Engineers.