Designing a building that uses almost no energy while keeping people comfortable is a high-wire act. Architects have to balance dozens of variables — window placement, insulation, air conditioning strategy, occupancy patterns — often without seeing how those choices actually play out until the building is already occupied.
A new tool is changing that. Researchers at Kanazawa University, working with colleagues in China, have created an AI system that lets designers test how a building will perform in real time, while it's still on the drawing board.
The system, called VEEM-ZEB, works like a digital twin of the building. Instead of running a single static simulation and hoping for the best, architects can now adjust a design element — say, moving a workstation or changing how task-ambient air conditioning is configured — and immediately see how that ripples through energy use and comfort levels. The results appear in real-time VR visualization, so designers get instant feedback rather than waiting weeks for computational results.
We're a new kind of news feed.
Regular news is designed to drain you. We're a non-profit built to restore you. Every story we publish is scored for impact, progress, and hope.
Start Your News DetoxWhy this matters for zero-energy buildings
Zero-energy buildings are supposed to generate as much energy as they consume, usually through solar panels or other renewables. But that's only possible if the building itself uses very little energy to begin with. Task-ambient air conditioning (TAAC) systems help with this — they cool individual work areas instead of the whole room, cutting energy waste. The problem until now: designers had no reliable way to test whether a TAAC setup would actually save energy during the design phase.
The new AI model changes that by breaking the building into zones — individual work areas and the wider room — and measuring both comfort and energy consumption simultaneously. It can run through about 48,000 different design scenarios, testing seasonal changes, different numbers of occupants, and how people actually behave in offices. This gives architects real data to work with instead of guesses.
What's particularly useful is that the tool combines rule-based AI (the kind that's transparent and explainable) with an intuitive interface. Architects don't need to be AI specialists to use it. They can compare different cooling strategies, see which layouts are more efficient, and make informed trade-offs between comfort and energy savings before a single brick is laid.
The researchers expect the system to become a standard part of architectural practice for zero-energy buildings — a decision-support tool that helps teams balance competing priorities from the very beginning of a project. That shift from post-construction evaluation to pre-construction testing could accelerate the adoption of truly efficient buildings.
