Air infiltration rates directly impact building energy consumption to a larger or smaller degree depending on the tightness of building enclosure and heating ventilation and air conditioning system operation. The relative importance of infiltration airflows has been increasing in total building energy consumption due to the improvements in building insulation and window products. The objective of this study is to compare the accuracy of building energy simulations associated with different air infiltration rates calculation approaches. This study used different sources of infiltration rate: time-dependent simulated data, AIVC database, and default settings in building energy simulations. A coupled framework associated with time-dependent infiltration rates is used by integrating computational fluid dynamics and multi-zone airflow modeling results into energy simulations. This framework is demonstrated with a case study for an office building in Michigan. The case study also uses the infiltration rates obtained from the database and default settings in energy simulation program. The comparison between simulation results and utility data shows that time-dependent infiltration rates could increase the accuracy of energy simulations with 3-11% reduction in the coefficient of variation of the root mean square error (CVRMSE), and 2-11% reduction in the normalized mean bias error (NMBE). (C) 2014 Elsevier B.V. All rights reserved.