3 ms·

For the record, see Braess' paradox: https://en.wikipedia.org/wiki/Braess'_paradox https://en.wikipedia.org/wiki/Braess'_paradox "Normal" and intuitive behavi

• mjevans 10y ago
  While this paradox might be a good theoretical tool for theory, I've disagreed that it is a good tool for modeling real situations since first hearing about it. * It is unrealistic that the mid-route would ever be created as described. * It is also unrealistic that the capacity of two parallel routes would also vary so much and have a mid-point where that reverses. * It would be much more realistic to have two routes, one of higher natural capacity, which react differently to increased load (not a static, no matter what, over-simplification). * Finally; it still does not address the issue of systems that are literally unable to cope with peek load; such as modern metro freeways in the US.