Infiltration has traditionally been assumed to affect the energy load of a building byan amount equal to the product of the infiltration flow rate and the sensible enthalpydifference between inside and outside. However, laboratory and simulation research hasindicated that heat transfer between the infiltrating air and walls may be substantial, reducingthe impact of infiltration. In this paper, two- and three-dimensional CFD simulations areused to study the fundamental physics of the infiltration heat recovery process and a simplemacro-scale mathematical model for the prediction of a heat recovery factor is developed.CFD results were found to compare well (within about 10 percent) with limited publishedlaboratory data corresponding to one of the scenarios examined. The model, based on thesteady-state one-dimensional convection-diffusion equation, provides a simple analyticalsolution for the heat recovery factor and requires only three inputs: the infiltration rate, the Uvaluefor the building, and estimates of the effective areas for infiltration and exfiltration.The most difficult aspect of using the model is estimation of the effective areas, which isdone here through comparison with the CFD results. With proper input, the model givespredictions that agree well with CFD results over a large range of infiltration rates. Resultsshow that infiltration heat recovery can be a substantial effect and that the traditional methodmay greatly over-predict the infiltration energy load, by 80-95 percent at low leakage ratesand by about 20 percent at high leakage rates. This model for infiltration heat recovery couldeasily be incorporated into whole-building energy analysis programs to help provideimproved predictions of the energy impact of infiltration.