Energy Benchmarking In A Portfolio Of Educational Buildings In Brazil Using Support Vector Machine And Data Envelopment Analysis
International Journal of Development Research
Energy Benchmarking In A Portfolio Of Educational Buildings In Brazil Using Support Vector Machine And Data Envelopment Analysis
Received 17th December, 2018; Received in revised form 14th January, 2019;Accepted 26th February, 2019; Published online 31st March, 2019
Copyright © 2019, Haroldo Luiz N. da Silva et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Buildings are one of the largest energy consumers in developed countries as well as in Brazil, and no other segment has such a potential for improving energy efficiency. Several policies have been applied for this purpose around the world and energy benchmarking is one of the most used worldwide. Thus, this paper brings a new benchmarking approach that involves not only energy consumption, but it also evaluates managerial issues in energy. For this, Support Vector Machine was used in order to predict the energy consumption using data of vocational schools in the São Paulo state, Brazil, to validate the methodology and Data Envelopment Analysis was used for the elaboration of the efficiency scale. It were considered 92 school buildings for the development of the predictive model and 72 for elaborating the efficiency scale in DEA. The results indicated a great potential for saving energy and financial resources when compared to the best practices.
