Developing a Human-Centric AI-based Multi-Criteria Energy Efficiency Advisory Framework for Apartment Housing of Mashhad, Iran
This study proposes and validates a human-centric, AI-based framework that integrates socio-demographic, building, neighborhood, and climatic data to deliver personalized, interpretable energy efficiency recommendations for apartment residents in Mashhad, Iran, achieving high predictive accuracy (R² = 0.86) and demonstrating that occupant characteristics are critical drivers of residential energy performance.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine your home is less like a static brick box and more like a living, breathing organism that interacts with the world around it. For a long time, scientists trying to figure out how to save energy in houses have mostly looked at the "hardware": the thickness of the walls, the type of windows, and the shape of the roof. They treated the building like a machine, assuming that if you built a perfect machine, it would run perfectly. But there's a missing piece in this puzzle: the people living inside. Just like a car drives differently depending on whether a nervous teenager or a calm grandparent is behind the wheel, a house uses energy differently depending on who lives there, how old they are, and what they do all day. This field of study is called "human-centric energy efficiency," and it asks a simple but tricky question: How do we design advice for saving energy that actually fits the specific humans living in the house, rather than just the house itself?
This is exactly what a team of researchers set out to solve in a new study focused on apartment buildings in Mashhad, Iran. They wanted to build a "smart advisor" using Artificial Intelligence (AI) that doesn't just look at the building's blueprints but also listens to the story of the people inside. They gathered data from 50 different households, looking at everything from the age of the building and the amount of green trees nearby to the age of the residents and how many children were in the family. They fed all this information into a computer model—think of it as a super-smart detective that looks for hidden clues in a massive pile of evidence—to see what really drives energy use.
The results were quite revealing. The AI detective found that the most important factors for energy efficiency weren't just the fancy building materials, but a mix of things like natural ventilation (how well the air flows through the house), the age of the building, and even the amount of green cover in the neighborhood. Surprisingly, the age of the people living there and the size of their family also played a huge role. The computer model was able to predict energy performance with a high degree of accuracy, getting it right about 86% of the time.
However, the researchers are careful not to call this a magic wand that solves all energy problems. They describe their work as a "proof of concept," meaning it's a successful test run that shows the idea works, but it needs more testing with bigger groups of people before it can be used everywhere. They explicitly ruled out the old idea that you can just look at a building's physical features and ignore the humans inside; their data showed that ignoring the people leads to a much less accurate picture. Instead, they suggest that the future of energy efficiency lies in a "human-in-the-loop" approach, where technology combines the physical facts of the building with the personal habits and characteristics of the residents.
The study concludes that by using this new AI framework, we can finally give personalized advice. Instead of a generic tip like "install better insulation," the system could say, "Since you have a large family and live in an older building with poor airflow, here are the specific changes that will save you the most energy." While the current study is limited to a small sample of 50 apartments, it lays the groundwork for a future where every apartment dweller gets a custom energy plan that respects their unique lifestyle, helping to make our cities greener and our homes more comfortable without forcing everyone into a one-size-fits-all solution.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.