Energy-Saving Solution Options and Energy Consumption Prediction Model Based on RF Algorithm (Random Forest) Combined with BIM Simulation (Design Builder)

By implementing energy efficiency measures, the potential for energy savings in buildings is relatively large, bringing many economic, environmental, and social benefits. However, identifying truly effective solution options is challenging, and this study was conducted to contribute to addressing this issue. The research results identified the most impactful solution options for energy savings in buildings, with those related to selecting high-efficiency cooling systems and optimizing building envelope design ranking at the top. In addition, the study proposed a model for predicting the electricity consumption of buildings, based on the Random Forest (RF) machine learning algorithm combined with Building Information Modeling (BIM) simulation using the DesignBuilder software. The simulation model analyzed the energy performance of a sample building with parameters related to the solution options used in the building design process. The simulation data was then used as input for training and testing the prediction model using the Random Forest (RF) machine learning algorithm. This prediction model can be used in the initial design phase of the project by inputting the building’s technical specifications to predict the electricity consumption, thereby enabling comparison and evaluation of the effectiveness of energy-saving solution options for building design.