Is There a Future Without Braess’ Paradox ?
https://phys.org/news/2010-09-scientist-braess-paradox-high-traffic.html
In class we talked about Braess’ paradox, which shows that the addition of a road doesn’t decrease a commuter’s travel time, but actually increases it. This article explains that new research conducted by University of Massachusetts Amherst Professor Anna Nagurney, proves that Braess’ Paradox isn’t always true. The article first clarifies that Braess’ paradox is not expected to occur for “system-optimizing” behavior in which a controller directs traffic, but it is assumed to occur with “user-optimizing”, or selfish behavior. However, Nagurney saw that at higher demand times, these “selfish” commuters ultimately find and learn to switch to faster routes. In other words, there is a “demand range” for Braess’ Paradox to occur. This can even be observed walking around Cornell campus between classes. Instead of shoving my way through crowds walking on East Ave, to get to Phillips Hall, I find it faster to cut behind Statler.
I found the article to be especially interesting because of its potential application to the future, specifically with the notion of “system-optimizing” and “user-optimizing” behavior. In the past, users were expected to make the “selfish” decision and take the quickest route, ultimately leading to Braess’ paradox. However, today with real-time traffic radio updates and new traffic technologies like Waze, users have more information than they ever had before to help them choose the fastest route to their destinations. Thus, users may now find themselves being “controlled” by these technologies and acting with more “system-optimizing” behaviors rather than with “user-optimizing” behaviors. In the future, if self-driving cars could potentially coordinate with one another, technologies would potentially know the entire system and thus choose the “system-optimizing” path for every commuter and avoid Braess’ paradox all together. The technologies would allow commuters to make decisions, knowing the decisions other commuters around them are making. This could potentially give a Nash Equilibrium that is always optimal for each driver for every travel route.
However, ultimately I do not believe Braess’ paradox will ever be eliminated. In order for there to always be an optimal Nash Equilibrium for every commuter, the drivers will have to work as a system. Therefore, it would only take one driver to “mess up” that system, proving it easily broken. This can be applied to many different fields like health epidemics and sporting events; it might be better to work as a team to gain the most optimal results for everyone, but usually our human “selfish” nature ruins the most favorable outcomes.