The purpose of this study was to explore the relationship between environmental factors and building energy consumption of three Leadership in Energy and Environmental Design (LEED)-certified buildings at the Arizona State University, by establishing the relationships of the outside atmospheric temperature and the energy consumed in the building using real-time data generated from different sources.

K-means clustering analysis is used to calibrate and eliminate unwanted influences or factors from a set of building consumption real-time data. For further statistical analysis, the chi-square is used to verify if the results are ample to prove the findings.

Few studies have addressed building energy consumption real-time data versus LEED Energy and Atmosphere (EA) credits with the data mining technique (k-means clustering) on most of building performance analyses. This study highlighted that the calibrating energy data are a better approach to analyze energy use in buildings and that there is a relationship between LEED credits’ (EA) Optimize Energy Performance scores and building energy efficiency. However, the energy consumption data alone do not yield useful results to establish the cause and effect relationships.

Although there are several previous research studies regarding LEED building energy performance, this research study focused on the LEED building energy performance versus LEED EA credits versus environmental factors using real-time building energy data and various statistical methods (e.g. K-means clustering and chi-square). The findings provide researchers, engineers and architects with valuable references for building energy analysis methods and supplements in LEED standards.