To ensure the high energy performance of a new building, its operational rating should be accurately estimated in the early design phase. Toward this end, this study developed an estimation methodology for the dynamic operational rating (DOR) of a new residential building using the advanced case-based reasoning (A-CBR) and stochastic approaches. This study was conducted in three steps: (i) establishment of a case database; (ii) retrieval of similar cases using the A-CBR approach; and (iii) estimation of the dynamic operational rating using the stochastic approach. The residential buildings located in Pusan, South Korea, were selected to validate the applicability of the developed methodology. Also, this study used the mean absolute percentage error (MAPE) to evaluate the prediction accuracy of the developed methodology (which means the difference between the predicted and measured energy performance). As a result, it was determined that the MAPE of the A-CBR model (i.e., 96.8% for electricity and 86.6% for gas energy) is superior to those of the other models (i.e., the basic CBR, multiple regression analysis, and artificial neural network models). In addition, based on the stochastic approach, it was estimated that cluster No.6, as a case study, would have the letter rating of 'B' grade (i.e., 25 < DOR <= 50) with 83.46% of probability; and the letter rating of 'C' grade (i.e., 50< DOR <= 75) with 16.54%. The developed methodology can be used to easily and accurately estimate the dynamic operational rating of a new residential building in the early design phase. (C) 2015 Elsevier Ltd. All rights reserved.