jAIr (Top 30)

Mission

To reduce the embodied and operational carbon footprint of buildings using artificial intelligence.

The challenge

Currently available (simulation) tools rely on a detailed model of the building to be constructed. As a result, the carbon footprint of designs is usually only considered at the end of the process. This limits the maximum achievable efficiency and leads to an overreliance on more expensive technical solutions to satisfy construction codes such as bouwbesluit.

The solution

A performance informed Deep Reinforcement Learning agent that can collaborate with designers to generate insight into potential design solutions. These designs will match or surpass contemporary designs in climate performance and can be created in a fraction of the time.