Texts and Twitter can help estimate crowd numbers in emergencies
In the event of natural or man-caused disasters and emergencies, first responders often go in without much information as to how many people are in those locations. Having a good estimate could mean better prepared emergency workers with the right amount of resources. Researchers have realized that a tool for providing those estimates rests in most people's pockets -- their smartphones.
Whereas most ideas for gathering crowdsourced information from smartphones requires users to download apps that track them, scientists at England's University of Warwick say that accurate data on crowd sizes could easily be gathered from mobile phone companies and Twitter without phone users having to do anything or give up their privacy.
The scientists have developed a computer model that uses information from mobile phone companies in Milan, Italy and Twitter to estimate crowd size for any given area. It works because all phone users are connected to a grid of relay antennas. When a large group of people is using its cell phones by calling, texting or using Twitter in given area, that corresponds to a spike in activity picked up by the nearest antenna. That spike can be analyzed to estimate how many people are in that area.
The scientists verified the model's accuracy by using it to estimate the number of people in attendance at 10 soccer matches at the San Siro stadium, where the attendance number was already known. After calibrating the software based on those results, the team started estimating the number of people at the city's Linate airport at various times of the day.
Having accurate numbers on how large crowds are could prevent disasters as well as help after they occur.
"Being able to infer the number of people in a specific area is of extreme importance for the avoidance of crowd disasters and to facilitate emergency evacuations," the authors said in a paper published in the journal Royal Society Open Science.