Uncertainty analysis, based on Differential Sensitivity Analysis and Monte Carlo Analysis, has been used in several research projects in order to estimate reliability of results, especially in empirical validation-based projects where both measured and predicted uncertainty bands need to be evaluated However, such analyses have been time-consuming, involving either repetitive manual changing of the input parameters followed by a simulation run, or involving the writing of specific programs to automate the process. Their use has, in consequence, been restricted. This paper reviews the sources of uncertainty in the predictions from thermal simulation programs. It then describes how uncertainty analysis has been incorporated into the thermal simulation program ESP-r. An example of its application is given. (C) 2001 Elsevier Science B.V. All rights reserved.