Sign In

Communications of the ACM

ACM TechNews

A Genetic Algorithm Predicts the Vertical Growth of Cities


A three-dimensional representation of the Minato Ward of Tokyo, Japan, used for the study.

Spanish researchers have created an evolutionary genetic algorithm that, on the basis of the historical and economic data of an urban area, can predict what its skyline could look like in the coming years.

Credit: Ivan Pazos et al.

Researchers from the University of A Coruña in Spain worked with architect Ivan Pazos have create genetic algorithms that predict how the number of skyscrapers and other buildings in an urban area will increase.

"We operate within evolutionary computation, a branch of artificial intelligence and machine learning that uses the basic rules of genetics and Darwin's natural selection logic to make predictions," says Pazos.

The team developed algorithms that learn urban growth patterns using historical data from the construction sector and economic parameters.

The study focused on the Minato Ward in Tokyo, Japan, one of the neighborhoods with the highest vertical growth worldwide in recent years.

In 2015, the researchers developed maps and three-dimensional representations of Minato to predict the number of buildings that would be erected, and their probable locations, in the 2016-2019 period.

The algorithm's predictions have been very accurate for 2016 and 2017, and initial evaluations indicate the algorithm's accuracy for 2018 and 2019 will be 80%.

Pazo said the study concluded that evolutionary computation can find growth patterns that are not obvious in complex urban systems.

From SINC (Information and Scientific News Service, Spain)
View Full Article

 

Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


 

No entries found