When applying computer simulation tools in practice uncertainties abound, for example in material properties and boundary conditions. To facilitate the quantification of the effects of uncertainties, the differential, factorial and Monte Carlo methods have been implemented within a simulation tool, ESP-r. These methods require multiple simulations to extract statistical measures of model uncertainty. An alternative approach is to embed uncertainty considerations within the simulation tool's algorithms. The principle advantages of this approach are that the uncertainty is quantified at all times and therefore requires only a single simulation. Coupled with this, it is possible to take control action based on the prevailing effects of uncertainties. This paper details the mathematical techniques required to integrate uncertainty considerations within the energy conservation equations when applied to the simulation of buildings. A comparison is made between the use of this novel approach and traditional mechanisms of assessing uncertainty. Crown Copyright (c) 2006 Published by Elsevier B.V All rights reserved.