Overcoming data scarcity by a probabilistic approach to consider future retrofit rates in urban building energy modeling

As Urban Building Energy Modeling (UBEM) is central to assessing the transition to a carbon-neutral and sustainable building stock, it is essential to consider changes in the building stock due to future retrofits. Because this information is difficult to predict, a common approach is to use percentages without discrete mapping to buildings, which challenges bottom-up UBEM. This study wants to overcome this challenge, by using a probabilistic, archetype-based approach to consider future retrofit rates and estimate the future heating energy demand of districts. Therefore, a discrete weighted sampling approach generates probability-driven samples, where each sample represents one potential outcome based on the statistical information. Simulations are conducted using the City Energy Analyst software. The results are summarized by calculating the mean energy demand, along with standard deviation, and an expected range of values that represent probability distribution of future energy demands. A case study based on a German town center with known information for the current state is presented to demonstrate the probabilistic approach and shows how probability-based samples help define realistic ranges for future energy demand predictions. To account for different scenarios, four cases are modeled based on different retrofit rates. The results of these different cases demonstrate the range of potential future energy savings. The results show a range of potential energy savings, which depend heavily on the estimated retrofit rate. If the retrofit rate remains as it is now, the estimated savings in 2045 are approximately 8.4 %. As the retrofit rate increases, so do the savings. A full deep retrofit of the building stock achieves an average energy saving of approximately 66.3 %. The developed method presents a significant advancement in including statistical data to bottom-up UBEM, enhancing the reliability of UBEM outcomes for future urban energy planning.