Challenges in implementing data-driven approaches for building life cycle energy assessment: A review.

Over the last few decades, the construction sector's energy consumption has increased tremendously. Buildings consume both embodied energy (EE) and operational energy (OE) during their life cycle. EE is consumed by processes associated with construction, whereas OE is spent operating the building. Studies show that improving the operational efficiency of a building may have serious implications for EE. Building life cycle energy assessments (LCEA) is, therefore, essential to understanding the dichotomy between EE and OE. In recent years, increased availability and accessibility of large-scale data have made data-driven approaches a popular choice for building performance assessments. In this context, numerous review articles have highlighted and tracked current trends in building load prediction methods. While this work is significant, there remains a lack of reviews focusing on data-driven approaches from a building life cycle energy perspective. In this paper, we conduct a systematic review of literature to identify key factors hindering the application of machine learning techniques specifically for building LCEA. They include: (i) issues of data collection, quality, and availability; (ii) lack of standardized methodologies; and (iii) temporal representativeness and granularity of prediction. Finally, we discuss potential solutions, future directions, and research opportunities for data-driven LCEA research. • Building life cycle energy (LCE) consists of embodied energy (EE) and operating energy (OE). • Current research shows that data-driven approaches for building performance assessment are focused only on OE. • This paper identifies key issues hindering the application of machine learning techniques for evaluating building LCE. • This paper discusses potential solutions, and future research opportunities for data-driven LCEA research. [ABSTRACT FROM AUTHOR]