Applying data mining techniques to explore technology adoptions, grades and costs of green building projects

Abstract

Many architects encounter problems during adoption of green building technologies and are unfamiliar with the benefits of green buildings. This study conducted two-stage data mining on 354 green building projects in Taiwan, in order to solve issues in the preliminary design phase of projects, such as technology adoptions, green building grades, and construction costs. This study adopted the association rules to explore the associations between different types and grades of green buildings and technology adoptions. Moreover, a prediction model based on the artificial neural network was constructed to predict the grades and costs of green building projects. The results indicate that different types and grades of green buildings are based on varying green building technologies. In particular, a green building with a high grade places more emphases on the green building technologies of air conditioning, CO2 reduction, and indoor environments. The green building technologies are affected by building types, regulations, costs, climate conditions, or geographic restrictions. This study also found that the accuracy of the artificial neural network in predicting green building grades and costs can reach above 80%. The systematic data mining method constructed herein can effectively assist architects and building owners to reduce preliminary design time and costs, and improve the success rates of green building projects. It is expected that the proposed approach can be adjusted in the future for other regions with different climates or their corresponding green building rating tools, so as to construct more suitable applications.