Applicability of overheating analysis approaches for big indoor temperature monitoring data

Other conference contribution, 2025

During the summer of 2018, northern Europe endured a heat wave rarely seen before. The heat wave was exceptional, with high temperatures and long duration, leading to 700 reported heat-related deaths in Sweden. Such extreme events are projected to increase even more in the future, yet the extent of overheating occurrence in different buildings and dwellings in Sweden is unknown. Currently, the Public Health Agency of Sweden recommends an upper threshold of 26°C for sensitive occupants and that nighttime temperatures above the same threshold should be avoided for all occupants. However, it is not established how these recommendations should be used for the evaluation of overheating occurence in the building stock.

Over the last few years, several overheating assessment methods have been developed worldwide to address the growing overheating risk. The focus remains primarily on preventive measures and methods that apply to simulation studies. Although these methods can be applied to monitored data, they are not usually constructed with that in focus. The aim of this paper is, therefore, to compare how two overheating assessment approaches apply to large-scale monitoring datasets and identify their respective strengths, weaknesses, and limitations. The approaches examined are exceedance over thresholds and the overheating assessment method TM59 by CIBSE. Hourly indoor air temperature measurements from 25,788 apartments in Gothenburg from 2018 and 2024 are used in the analysis. The results demonstrate that the approaches studied are highly sensitive to threshold selection, whether static or adaptive, and the method followed. The analysis also highlights the uncertainty in selecting assessment period and occupancy schedule when dealing with measured data. There is a need for benchmarking methods specifically tailored to big data overheating analysis, which can provide reliable indicators of both the extent and patterns of overheating in existing building stocks.