Braess's Paradox
resources.mpi-inf.mpg.deI noticed this first hand while playing SimCity 4. I used to make my road network as connected as I could because I thought having more route options would always lead to better traffic. I was surprised one day when I demolished a particularly congested road and my overall traffic actually became better. After some experimentation I realized making the road graph more tree like by reducing pathing options you can force the sims to use higher capacity roads rather than having them all take low capacity backroads to their destination.
I wonder if routing services like apple/google maps consider this when routing drivers. Theoretically they could avoid the paradox by just not routing people the selfish way right? And here that wouldn't make it better than the original, but in the real world where they add helpful roads but they get too congested and slow things down, they could (assuming everyone listened) stop them from getting congested allowing for the benefits to be realized.
Sure, Google Maps could do that, but then the Selfish Maps app (or the Google Maps Express Lane Premium monthly subscription) would be very appealing to the people that can afford it.
Also on wikipedia, for those who don't want to open a PDF.
Related:
Braess Paradox - https://news.ycombinator.com/item?id=31133775 - April 2022 (7 comments)
Braess's Paradox - https://news.ycombinator.com/item?id=27173829 - May 2021 (30 comments)
Braess' Paradox and the Price of Anarchy (2019) - https://news.ycombinator.com/item?id=24231860 - Aug 2020 (12 comments)
Braess’s paradox - https://news.ycombinator.com/item?id=18045164 - Sept 2018 (37 comments)
Braess’ paradox - https://news.ycombinator.com/item?id=13352513 - Jan 2017 (91 comments)
Braess' paradox: adding a new road to a city can slow down traffic - https://news.ycombinator.com/item?id=10397424 - Oct 2015 (61 comments)
Power grid upgrades may cause blackouts, warns Braess's paradox - https://news.ycombinator.com/item?id=4691388 - Oct 2012 (1 comment)
Removing street signs, lights and arrows increases safety and road capacity - https://news.ycombinator.com/item?id=1594478 - Aug 2010 (48 comments)
Braess's paradox: adding roads can increase congestion - https://news.ycombinator.com/item?id=929362 - Nov 2009 (9 comments)
Why the secret to speedier highways might be closing some roads: the Braess paradox - https://news.ycombinator.com/item?id=422152 - Jan 2009 (21 comments)
Fewer Roads, Less Congestion - https://news.ycombinator.com/item?id=353874 - Nov 2008 (5 comments)
There have been other threads on this theme that don't include "Braess" in the title. Anybody want to find some?
https://news.ycombinator.com/item?id=353874 Nov 2008 (5 comments) - dead link, but also mentions SimCity 4
https://news.ycombinator.com/item?id=1594478 Aug 2010 (48 comments) - dead link, "Removing street signs, lights and arrows increases safety and road capacity"
Great finds - I've added them to the list above. Thanks!
Wonder about the relation to prisoners dilemma. Maybe that’s a simple example: offering prisoners opportunity to rat out the other can gives a worse outcome than forcing collaboration.
I might be overly anal, but how is this a paradox at all?
There are multiple types of paradoxes: https://en.m.wikipedia.org/wiki/Paradox#Quine's_classificati...
This would fall under "veridical paradox" - true statement which most people would expect to be false.
It's a paradox because you would think that how can giving drivers more choice worsen their travel times?
If that diagram was electrical resistors, then adding that middle path across the bridge would never increase the resistance from A to B and therefore reduce the current. At worst, the added resistor will do nothing (because the nodes of the Wheatstone bridge are at the same potential). If any current flows at all across the new connection, it improves the flow.
Most people would assume increasing capacity would always mean greater throughput.
I wonder if this has ever been ever observed in real life. The mathematical example makes sense, but the edge weights seem artificial.
FTA
>In Seoul, South Korea, a speeding-up in traffic around the city was seen when a motorway was removed as part of the Cheonggyecheon restoration project.[2] In Stuttgart, Germany after investments into the road network in 1969, the traffic situation did not improve until a section of newly-built road was closed for traffic again.[3] In 1990 the closing of 42nd street in New York City reduced the amount of congestion in the area.[4] In 2008 Youn, Gastner and Jeong demonstrated specific routes in Boston, New York City and London where this might actually occur and pointed out roads that could be closed to reduce predicted travel times.[5]
There are 4 real-life examples quoted at the end of the article under the heading "How rare is Braess's Paradox?"