Slashdot Mirror

New Scientist has an interesting piece up about the calculable energy costs per mile for various forms of transportation. Despite the headline ("Train can be worse for climate than plane"), the study it describes deals with highway-based vehicles, too: the authors attempted to integrate not just the cost at the tailpipe (or equivalent) for each mode of transport, but also the costs of developing and supporting the associated infrastructure, such as rails, highways and airports. Such comparisons are tricky, though; a few years back, a widely circulated report claimed that the Toyota Prius had a higher per-mile lifetime cost than the Hummer (see that earlier Slashdot post for good reason to be skeptical of the methodology and conclusions). I wonder how the present comparison would be affected by a calculation of (for instance) how much it would cost to move by plane the freight currently carried by trains.

An anonymous reader sends in a story from Central Connecticut State University, claiming that a Prius takes more energy to manufacture than a Hummer — 50% more. In addition, the article claims that the Prius costs $3.25 per mile over its expected lifespan of 100,000 miles compared to $1.95 per mile for the Hummer. The article gets its data from a study by CNW Marketing called Dust to Dust, which is an attempt to account for all the costs of vehicles, from manufacture through operation through repair and disposal. The $3.25/mile cost quoted for the Prius is the 2005 number; for 2006 it is $2.87. This improvement pulled the Prius below the straight industry average — all the other hybrids are still above that average. And the Hummer is not listed at all for 2006. Update: 03/21 00:44 GMT by J : You might want to take those figures with a grain of salt; I don't think anyone's seen the supporting data. Read on for details.

J adds:

The Prius's mediocre cost-per-mile is due mainly to CNW Research assigning the car a short expected lifetime: 109,000 miles. Nobody knows where this number comes from because CNW has not published details about its derivation. If a car will not last very long, then of course its energy cost per mile is high.

Back in July 2006, when CNW's study "Dust to Dust" had just been published (and which remains, unchanged, the original source for today's news), I emailed its president, Art Spinella:

Hello,

I'm with the tech news and discussion site Slashdot.org. One of our readers submitted a story about your Dust to Dust study.

According to Wikipedia, the Prius comes with a 150,000 mile warranty in California and a few other states; 100,000 elsewhere.

On p. 21 and p. 40 of your report I see that you estimate the average Prius will be "removed from the streets... and sent for disposal" at 109,000 miles. Can you explain how you arrived at this figure?

Thank you.

I did not receive a reply.

My question was about the cost-per-mile denominator; here's another critique questioning the numerator.