Project management mode under the concept of low carbon environmental protection and its value in intelligent construction

Abstract Rapid urbanization and climate change are intertwined, making decarbonization of the built environment paramount to stabilizing the future. The commercial and residential sectors generate nearly one-third of carbon emissions. Unexpected fluctuations in operational environments face the flexibility, efficiency, and resilience of building-incorporated energy systems due to climate change and its concerns. Instead, the rapid improvement of Machine Learning (ML) and Artificial Intelligence (AI) has equipped the construction industry with the capability to learn. This paper suggests a Machine learning-based Carbon Footprint Modeling (ML-CFM) to forecast the CO₂ emissions and energy consumption in intelligent constructions. The data has been collected from the World CO 2 Emissions analysis dataset for predicting the carbon emission in residential buildings. A new method based on Deep Neural Networks (DNN) can detect the overall carbon footprint of an intelligent construction design based on the urban layout and building features. A building’s structural characteristics had the most influence on CO 2 emissions and energy consumption, followed by the appropriate micro-climate, socioeconomic conditions, and the provincial climate. The ML-CFM is the most effective forecasting model for predicting carbon emission and energy consumption reduction, which offers building managers a valuable tool to enhance decision-making levels and energy efficiency in smart buildings.