USE OF MACHINE LEARNING (ML) IN GREEN BUILDING MANAGEMENT (GBM)

Abstract

Over the past few years, the Singapore government has emphasized the use of Artificial Intelligence (AI) to drive the nation toward fulfilling its Smart nation aspirations. This is evident from AI initiatives and plans that aim to develop AI capabilities and infrastructure across the nation. Amongst the various applications of AI, Machine Learning has been of great interest due to growth in computational power its ability to exploit the abundance of data generated globally. As part of Singapore’s drive toward a sustainable future, Green Buildings (GBs) have been of keen interest of the Government. As the operational phase occupies majority of a building’s lifecycle, it possesses great potential for reduction in climate change impacts. In addition, modern-day GBs have been infused with smart technologies which generate large amounts of data. Thus, considering the synergistic relationship between large amounts of data and ML, paired with the parallel growth and interest in ML and Green Building Management (GBM) in Singapore – it is probable that ML would be used in GBM. As such, the objective of this dissertation is to explore the various potential risks, drivers, challenges, and enablers of ML in GBM, and the areas of GBM that are most likely to benefit from the use of ML, through conducting in-depth interviews with industry practitioners and academics Findings indicate that data and personnel issues form key challenges and risks, whilst economic benefits, government support, and top-level management commitment are key drivers and enablers, with maintenance management being of keen interest