Designing A Program To Predict People’s Irrational Pedestrian Decisions



Keeping pedestrians safe is hard, because they keep having their own opinions about where they should walk. Sometimes that is in front of oncoming vehicles, whose drivers have their own ideas about how fast they should be driving. To design cities that prevent this deadly collision of opinions, scientists have created a simulation that, for the first time, takes into account how people make decisions about where and when they’re going to throw themselves in the path of a moving vehicle.

The SAFEPED program, developed by Tel Aviv University’s Geosimulation Lab at the Department of Geography and the Human Environment and Porter School of Environmental Science, uses behavioral science to program all of its virtual individuals to act as humans do in the real world. What SAFEPED is doing, say its creators, is adding the idea that human
beings see in three dimensions and use that knowledge to predict where
objects will be in the future.

Using SAFEPED, planners have already figured out that if a person thinks a car is going to reach the intersection in 2.5 seconds or less, they won’t try to jaywalk in front of it. If they feel like they have 5.5 seconds or more, they’ll run right out into the street.

With the software, urban
planners don’t need to hope that they know how people will behave in the real world. They can now plug in an intersection and then make changes to traffic
flow or lights and markings and see if that keeps accidents from
happening. Since we’re not going to stop running into the street any time soon, it’s the best option.

[Hat tip: Christie Nicholson at Smarter Planet]

Morgan Clendaniel can be reached by email or on Twitter.