The founder of Mint.com, Aaron Patzer, has been researching alternative urban transportation under a company called Swift over the past six months, but he has determined that the personal maglev system he had been envisioning is economically not viable for a company to produce. Patzer described all of his findings and development in a blog post (hat tip to Tech Crunch), including the high economics of such a transportation network.
The Swift concept
Swift is a concept to develop a magnetically levitated system using two-person vehicles attached to a track, driven by a linear synchronous motor. Yep, so little pods would move people through the air on an extensive set of tracks built throughout neighborhoods, which could replace a good deal of local car trips.
It’s not as out there as it sounds: Such networks have long been imagined and if ever built could reduce fuel consumption and in really densely populated urban environments provide a useful alternative for car ownership. A project called Shweeb, which Google gave a million dollars to last year, makes a transportation system based on pedal-powered pods that zoom around a monorail track about 20 feet above the ground. SkyTran is another maglev personal transport concept out there, and the Masdar Institute of Science and Technology has built a system of driverless pods that run on a battery pack along tracks guided by magnets under the ground.
One of the really enticing parts of these networks is that moving the maglev pods around could be highly efficient. Patzer says the vehicles could go 134 mph with only a 100 kW motor, and the vehicle-to-passenger weight ratio of Swift is close to 2:1, in contrast to cars, which have a ratio closer to 15:1.
The problem with systems like these is that getting all of that infrastructure built, particularly in the U.S. at any kind of scale, is close to impossible because of high costs and various permitting issues. Patzer came to the conclusion that Swift would not be economical for a company to build after three months of research and three months of coding work on the software for Swift.
Patzer says Swift could cost $8 million per kilometer (or as low as $5 million to $7 million per kilometer) to build, and at that price each station would have to serve 8,000 people. Non-highway roads cost as low as $100,00 to $300,000 per mile on flat terrain, not considering things like land acquisition and tunnels, says Patzer.
At those costs, the population density for such a system would have to be 5,000-plus people per square kilometer. For context, some of the most populous cities in the U.S. have population densities above that: San Francisco has a population density of 6,632 people per square kilometer, and New York City has a density of 10,194 per square kilometer. But many cities in the U.S. don’t.
Then there is the economics of concrete for roads versus steel for tracks. Asphalt costs $17 per ton while steel costs $900 per ton, and copper costs $10,000 per ton, says Patzer. Patzer writes:
Because world population has grown five-fold during the age of cars, we will be stuck with their legacy in roads and low-density metro configurations for at least the next century . . . It is the ubiquity of roads, more than the greatness of cars that is difficult to defeat.
What’s next for Swift?
Patzer says he has abandoned building the project as a company. He calls the failure “fast” and “worthwhile fun” and will be looking for his next venture. He also splits his time as VP of Product Innovation at Intuit, which bought Mint two years ago for $170 million, points out Tech Crunch.
However, Patzer thinks perhaps the traffic engineering algorithms he developed could be used for self-driving cars (like the effort under way at Google) or other technology relating to traffic routing from a central traffic database, or car-to-car wireless signaling (see the research Ford is doing). Patzer says he would be interested in funding or participating in these types of projects with Swift’s software — heads up to other entrepreneurs interested in alternative transport.