I have long held that traffic patterns could be modeled quite closely using principles of fluid dynamics. The roads are the pipes, traffic lights and signs are valves, traffic density is pressure (oddly enough), drivers' reaction times in acceleration and braking are elasticity in the fluid, lane changing is turbulence, and so on.
Driving home, it occurred to me that if fluids can be used to model traffic, then traffic could be used to model fluids just as well. That's when the big realization hit me:
Most atoms are assholes.
If only everyone (drivers) would realize the concept that we are all part of the same dynamic, fluid or otherwise. :/
Posted by: Cruz at October 3, 2006 05:22 PMI don't have any links or anything, but I recall that work on this was done way back, in the 1950s or early 1960s. The analogy works best with the fluid near the supercritical point. But it is more an analogy or metaphor than a really useful method today; with computers it makes more sense to model individual cars with somewhat varying behavior (tailing distances, reaction times, etc.) to the cars ahead of them. The models even came up with the "ripple" effect of braking that we see on almost-congested highways.
Posted by: dwpittelli at October 4, 2006 07:08 AMMost atoms are assholes.
HA! have you seen this mashup yet? they're tracking heat patterns from traffic using google maps -- http://blog.fortiusone.com/2006/10/11/heat-maps-for-google-maps-aka-geoiq-mashup not quit fluid, but a really interesting way of looking at it all the same.
Posted by: cocoricamo at October 12, 2006 11:01 AM