IBM recently announced some groundbreaking new technology designed to not just cut deadlock, but ultimately eliminate it.
We’ve known for years that traffic does not behave randomly and actually behaves closer to water than anything else. Because of that, scientists have been able to predict where and when traffic is likely to be delayed with basic algebra for decades now.
The problem is not so much the math as it is scale. Every time an accident happens or a person puts the breaks on early, the river of traffic responds differently. Add in variables that can shut down multiple roads such as an accident or an emergency vehicle rushing to a hospital, and traffic becomes impossible to predict.
Until now, of course. Thanks to a close partnership with the city government of Lyons, France, IBM has built a system capable of predicting the most likely effects of an accident or other traffic event on the entire city’s traffic grid. The system then makes recommendations to city officials monitoring the grid on how to manage the situation — whether to close roads, open others, change the programing on traffic lights, or dispatch a detour team. Every solution generated is tailored to every problem automatically and, best of all, the system “learns” over time, becoming faster and generating answers that are even more efficient.
The beauty of this system is also that it can be duplicated. If it works in Lyons, there is no reason why transportation grid managers throughout France, Europe, and the world can’t incorporate it into their system as well. Over time, this could wind up shaving a significant amount of time off of your commute, although we’ll still need to wait several years for the system to complete its debut in Lyons.
How do you feel about a computer dictating the flow of traffic? Do you think using technology to control traffic would make commutes faster and easier, or would the device wind up with bugs with could lead to even larger traffic snarls?