Cost premium prediction of certified green buildings: A neural network approach

Built environment has a substantial impact on the economy, society, and the environment. Along with the increasing environmental consideration of the building impacts, the environmental assessment of buildings has gained substantial importance in the construction industry. In this study, an artificial neural network model is built to predict cost premium of LEED certified green buildings based on LEED categories. To verify the viability of the model, multiple regression analysis is used as a benchmarking model. After validating the prediction power of the neural network model, a global sensitivity analysis is utilized to provide a better understanding of possible relationships between input and output variables of the prediction model. Sustainable Sites and Energy & Atmosphere LEED categories were found to have the highest sensitivity in cost premium prediction. In this study, our goal was to reveal the significant relationships between LEED categories and the cost premium, and offer a decision model that can guide owners to estimate cost premiums based on sought LEED credits.