Future application of probabilistic methods for building performance simulations

Many countries strive for continued reductions in energy demand from new buildings as a way of mitigating the impact of climate change. Consequently, the importance of accurate and efficient building performance simulations (BPS) increases to improve the prediction in energy demand during the design-process for buildings. As observed in earlier studies, there is currently a performance gap between the predicted annual energy demand from BPS based on deterministic methods compared to the measured annual energy use of a building. The performance gap is dependent on several parameters and types of uncertainties, as shown in previous studies. Using a probabilistic method, with stochastic input data using the Monte Carlo-method and dynamic BPS software, makes it possible to calculate a probability distribution for the energy use of a
specific building and indicate the probability of attaining a specific energy performance. This paper presents a summary of the overall results and conclusion from earlier papers regarding probabilistic methods for energy performance simulations and risk analysis, and how the results could improve future BPSs. This paper shows recent developments in available tools and how
these methods could be used in practice to include different types of uncertainties when performing risk analyses. Future research and developments needed to improve and broaden the application of the methods and how to include improved quality or quantity of data to improve additional or future simulations are discussed.