While Artificial Intelligence (AI) is the general science of imitating human capabilities, Machine Learning (ML) – or automated learning – trains a machine to learn.
Machine Learning refers to a computer’s ability to ‘learn’ to improve its performance in order to solve problems for which it had not been originally programmed.
ML emerged in 1951, but the general public only became aware of its power in 1997, when the Deep Blue computer defeated chess champion Gary Kasparov. Since then, this science has brought us the self-driving car, voice recognition and spam filters, to name but a few. These technologies are also used by innovative companies to identify security breaches and potential hazards.
While the human mind is only able to process three or four dimensions, AI algorithms have no limits. Since it is not a sector-specific science, the potential applications of ML in the field of construction are vast.
We discuss 2 of them here: generative design and image analysis.
Generative Design makes it possible to design buildings beyond the limits of human creativity. Based on defined constraints (e.g. geometric, thermal and behavioural), algorithms generate all possible solutions that meet the required criteria. Advanced simulations make it possible to anticipate risks while reducing lead times and costs.
A second application that will have a huge impact on construction is image marking and analysis. The technology can be used to identify and analyse hazards, to categorise and label site photographs etc. It can even be used to identify people who are violating safety standards.
Other applications could include processing notifications, identifying missing materials, organising documents, and even piloting drones and operating machines. The possibilities are virtually endless.