Using statistics to beat traffic
This is great: Brandon Hansen at Omninerd decided to crunch some statistics on how long it takes him to commute to work, to see if there was some hidden way to shorten his drive. Sure enough, he discovered a few secret sweet spots in his local traffic, which are charted above. As he concluded:
Given the above data and analysis, what can be done to improve my commute times? Changing my morning or evening departure time looks promising. The best bet appears to be moving my schedule out a half-hour to 8:30AM and 6:00PM, bringing significant savings (about 7.5 minutes of commute time per day) without getting too far from normal business hours. Spread out over 50 work weeks, that results in a total savings of over 30 hours a year -- the equivalent of about a 38% boost to my existing 80 hours of vacation.
Here's an idea for Detroit's ailing Big Three: Why don't our onboard car computers do this sort of thing automatically? They could spend a few weeks recording when your daily commute begins and when it ends, mix in some GPS telemetry, crunch it with a bit of Mapquest data and presto: A customized drive agenda, produced by automobile A.I.!
(Thanks to the J-Walk blog for this one!)
Posted by Clive Thompson at May 17, 2006 01:56 PM
Trackback Pings
TrackBack URL for this entry: http://www.collisiondetection.net/mt3/mt-tb.cgi/1486
That's a great idea. Something similar occured to me back when I was in high school (the last time I regularly drove anywhere.) With a little data from the city, I could get information about each traffic light on the way to the school, and then have my computer figure out the exact time to leave to maximize green lights. I would probably need an in-car computer with the same data running calculations as I went, in case I got stuck behind some traffic and needed to rework my route. But hey, if Stanley can be built, I'm pretty sure this can too.
Posted by: Peter at May 17, 2006 5:40 PM
Brilliant. I've thought about this too, but before I got around to it I moved someplace where it was less an issue.
In Chicago, my tactic was always to find "secret" roads… there are a few good major streets that don't have much on them that anyone would routinely visit. The ones with the least lights can get you through rush hour at a quick 45 MPH on most days.
But schedule shifting is still the best tactic. Given my weird sleep habits, I'm almost always driving the opposite direction of most people on the road.
Every now and then that freaks me out a little… I start thinking that if everyone is heading *away* from my destination, is that because there's something really scary that just happened there? Yikes!
Posted by: johntunger at May 17, 2006 8:21 PM
A great idea in theory, but it seems like the benefits would evaporate as more and more commuters attempt to identify the perfect commuting schedule. Like the stock market, the very act of people seeking to benefit from statistical anomalies causes them to disappear.
Posted by: Adam Lowe at May 18, 2006 9:31 AM
Peter, yeah, I agree -- these sorts of calculations can't be hard at all to do! Most of 'em could probably be done on a Palm Pilot, actually.
John, I love the "secret" road technique ... like trying to find shortcuts through Liberty City in Grand Theft Auto!
Adam, that is entirely possible. Inject too much intelligence into a massive networked system and the various loosely-joined parts will think themselves into gridlock! It's the same reason "net neutrality" is the most efficient way to run the Internet: If routers had to make sophisticated decisions about what traffic to prioritize, things would get very messy very quickly.
Posted by: Clive at May 18, 2006 1:42 PM
Post a comment
That's a great idea. Something similar occured to me back when I was in high school (the last time I regularly drove anywhere.) With a little data from the city, I could get information about each traffic light on the way to the school, and then have my computer figure out the exact time to leave to maximize green lights. I would probably need an in-car computer with the same data running calculations as I went, in case I got stuck behind some traffic and needed to rework my route. But hey, if Stanley can be built, I'm pretty sure this can too.
Posted by: Peter![[TypeKey Profile Page]](http://www.collisiondetection.net/nav-commenters.gif)
at May 17, 2006 5:40 PM
Brilliant. I've thought about this too, but before I got around to it I moved someplace where it was less an issue.
In Chicago, my tactic was always to find "secret" roads… there are a few good major streets that don't have much on them that anyone would routinely visit. The ones with the least lights can get you through rush hour at a quick 45 MPH on most days.
But schedule shifting is still the best tactic. Given my weird sleep habits, I'm almost always driving the opposite direction of most people on the road.
Every now and then that freaks me out a little… I start thinking that if everyone is heading *away* from my destination, is that because there's something really scary that just happened there? Yikes!
Posted by: johntunger![[TypeKey Profile Page]](http://www.collisiondetection.net/nav-commenters.gif)
at May 17, 2006 8:21 PM
A great idea in theory, but it seems like the benefits would evaporate as more and more commuters attempt to identify the perfect commuting schedule. Like the stock market, the very act of people seeking to benefit from statistical anomalies causes them to disappear.
Posted by: Adam Lowe![[TypeKey Profile Page]](http://www.collisiondetection.net/nav-commenters.gif)
at May 18, 2006 9:31 AM
Peter, yeah, I agree -- these sorts of calculations can't be hard at all to do! Most of 'em could probably be done on a Palm Pilot, actually.
John, I love the "secret" road technique ... like trying to find shortcuts through Liberty City in Grand Theft Auto!
Adam, that is entirely possible. Inject too much intelligence into a massive networked system and the various loosely-joined parts will think themselves into gridlock! It's the same reason "net neutrality" is the most efficient way to run the Internet: If routers had to make sophisticated decisions about what traffic to prioritize, things would get very messy very quickly.
Posted by: Clive![[TypeKey Profile Page]](http://www.collisiondetection.net/nav-commenters.gif)
at May 18, 2006 1:42 PM