Hotspot analysis can identify areas with a higher concentration of events compared to the expected number in a given random distribution of events. Hotspot analysis has been widely used in various research disciplines, such as public health, crime, and environmental quality-related research. In this study, we applied hotspot analysis and Local Moran’s I indices to pinpoint high and low-temperature clusters within commercial buildings. Our study demonstrates that hotspot analysis is applicable inside a building. Using the Moran's I statistic, we introduced a method to determine the optimal number of neighbors.
Additionally, random forests were employed to identify important placement features influencing hot/cold spots. The vacancy of the place and the presence of windows emerged as crucial factors in our application. In conclusion, our combined application of Moran’s I statistic and hotspot analysis effectively identifies hot and cold areas within a building, when the number of neighbors is accurately identified and there are no outliers or abnormal sensors.
Additionally, random forests were employed to identify important placement features influencing hot/cold spots. The vacancy of the place and the presence of windows emerged as crucial factors in our application. In conclusion, our combined application of Moran’s I statistic and hotspot analysis effectively identifies hot and cold areas within a building, when the number of neighbors is accurately identified and there are no outliers or abnormal sensors.
Date and Time
-
Language of Oral Presentation
English / Anglais
Language of Visual Aids
English / Anglais