Green standard model using machine learning: identifying threats and opportunities facing the implementation of green building in Iran

Environ Sci Pollut Res Int. 2021 Nov;28(44):62796-62808. doi: 10.1007/s11356-021-14991-3. Epub 2021 Jul 2.


Residential buildings consume a major portion of energy resources and hence are seriously involved in environmental pollution. In Iran, fossil fuel consumption is growing, such that it increased by more than 400% from 1990 to 2018. One of the fundamental solutions for reducing fossil fuel consumption and creating a healthy environment inside and outside buildings is implementing and developing green buildings. This study seeks to examine the barriers to and opportunities for developing green buildings and proposes a localized green standard appropriate for the conditions of Iran. To this end, the required parameters were identified using the opinions of experts and the Delphi method. The opinions of 81 building experts, including the employers, consultants, and contractors, were obtained using a three-part questionnaire. Based on the results from the machine learning method, the score of the localized green building in five dimensions, namely, site, water, energy, materials, and quality of the indoor environment was calculated to be 77.2, while the energy dimension was determined to be the most important green standard dimension with a significance coefficient of 0.548. In the ranking analysis of all parameters using the Friedman test, the parameters of energy consumption management, renewable energy usage, and thermal zoning received the highest scores among other factors. Furthermore, a lack of awareness on green buildings (77%) and a high potential for renewable energy production (81%) were respectively identified as the biggest barrier to and opportunity for the implementation of green buildings in Iran.

Keywords: Energy consumption; Environmental pollution; Green standard; Iran; Machine learning.

MeSH terms

  • Environment*
  • Environmental Pollution
  • Fossil Fuels*
  • Iran
  • Machine Learning


  • Fossil Fuels