More and more people are killed in stampedes every year. This problem arises mostly in places where massive number of people gather. Whenever due to some reason one or two people or group of people get or feel threatened, they start panicking and as a result the whole crowd panics causing stampedes and massive casualties. After the Shanghai stampede on New Year’s Eve in 2014 which claimed some 30 lives a team of scientists and researchers from Chinese internet giant Baidu Inc. started working on algorithms to solve this problem. They came up with an algorithm to predict crowd formation.
Baidu’s Big Data Lab published a study which claimed that the location of people doing Map route searches on its search engine when correlated with crowd density of places people searched for can be used to predict future crowd formations of certain places. The senior research scientist, Wu Haishan claims that their algorithm is able to predict how many people will be at a certain place at a certain time in the next two hours. Currently, cameras and sensors with the aid of some advanced algorithms are used to predict crowd density. But that is comparatively costlier and also requires a lot of resources.
Crowd control algorithm could be used to give early warnings to the concerned authorities and local government about huge crowd movement. Places like stadiums and concerts venues could better handle the crowd if they knew beforehand the estimated number of people. Also the best part for those concerned about their privacy is that since it only uses aggregated data, so it has no privacy issues and users can’t be identified howsoever.
The Wonderful World of Algorithms 