Sitting home guessing the crowd reaction

Want to learn how a crowd reacted when urgent, scientists now do not need to run to the street and shout " fire ", then see how people panic run. A new computer model will help them that.

First, this three-dimensional model contains human movements and behavior, and uses them to simulate a crowd as well as pedestrian habits in certain types of buildings.

"The crowd is vital to the lifeblood of cities," said author Paul Torrens of Arizona State University. But according to him, it would be impractical to set up live experiments of hundreds of thousands of people who are busy running away from the street.

In the Torrens model, he included individuals (with different characteristics of age, gender, size and health) and gave them information about the surrounding context. The combination of unique characteristics will make each individual react in different ways when receiving context information.

In this scene, the crowd runs away from a smoking car, heading to a point to the right of the screen.But the narrow road formed a bottleneck that gathered the crowd.(Photo: LiveScience)

"It's exactly the same way we handle information in the real world , " Torrens said.

The original prototype that Torrens developed was the dispersal of a crowd from a fire that was happening where there was only one exit. However, he applied this method to develop a more primitive model for a situation that spreads through accidental contact.

He intends to create scenarios for extremist crowds. Models can also be helpful in planning for the city to optimize pedestrian space.

Simulation image of people running when having trouble with a smoke car (Photo: LiveScience)

(Photo: LiveScience)

T. An