Scheduling for failure

Christian Wolmar writes about the romance of the rails in My window on America: From Los Angeles to New Orleans on an epic rail adventure:

“Then in the dark, we broke down, and the lights went off. The conductor said a hose has come undone and that the power had to be switched off while the problem was examined.
They had to send for an engineer, who arrived quickly with a huge lamp that eerily lit up the scene and solved the problem.”

There was more excitement and evictions at San Antonio, which we reached at 9.30pm and where we were scheduled for a two-hour stop. It gave us time for a drink in a bar. A couple of girls on the train, heading for a music festival, had drunk too much, and were too out of it to stop singing on the way back to the train, even after a conductor warned them to stop. The train was delayed for the police to be called. Off they went, still intermittently singing and swearing, into the cooler.

The train meandered around the outskirts of New Orleans, but still arrived an hour early.”

So the train broke down once and had to have the police called, but it was still early. How much slack is built into the system to ensure on-time arrivals?

One taxi Über alles

I am quoted in this article Smartphone apps are changing the way riders hail taxis by Katie Humphrey

“When you go to most cities, there are many different taxi providers and it’s hit-or-miss how reliable any of them are,” said David Levinson, a professor at the University of Minnesota who studies transportation. “It’s easier to download an app than to look up a telephone number.”


Land Developability


Guangqing Chi has a new website devoted to Land Developability:

“Land developability is a measure of land availability for future conversion and development in a geographic entity, such as a state, a county, a city, a census tract, or other geographically aggregated units. The land developability index is generated using spatial overlay methods based on data layers of surface water, wetland, federal/state-owned land, Indian reservation, built-up land, and steep slope, which are all seen as undevelopable. The land developability index can be used for regression modeling when land use and development is a consideration of an analysis, for detecting potentials for land conversion and development, and for predicting the direction of future land use and development.”

This looks like a very useful piece of research data infrastructure.

Welcome to Meteorological Spring

Today I saw one bus unable to get up a hill and one crash, both due to weather conditions (I got some video of the bus after its failure to climb the hill, but none of the crash, which was a minor fender-bender with some grill damage to the offending vehicle with apparently no injuries). Meetings are canceled left and right. My son’s school was canceled. My daughter’s school (a different school in the same building) was not. Welcome to Meteorological Spring.
More on Why we become such bad drivers when it snows at Streets.MN

Lean Machine for the 21st Century

ToyotaiRoadAutoblog tells me about the Toyota i-ROAD :

“According to Toyota, the “i-ROAD takes the company closer to its goal of creating the ultimate range of eco cars.” As you’re surely aware, that range of eco cars includes the enormously successful Prius family, but this new machine is nothing like the hybrid hatchback. And it’s not even a car – Toyota calls the i-ROAD a Personal Mobility Vehicle.
Toyota’s i-ROAD Concept, which debuts at this week’s Geneva Motor Show, is adorned with just three wheels, meaning it’s just as much a motorcycle as it is a car, and the driver and passenger sit in tandem style instead of side-by-side. This arrangement allows for a very thin 850mm width, which is about the same as a large motorcycle. Because the cockpit is enclosed, the occupants don’t need helmets, nor are they open to the elements outside.
Also like a traditional two-wheeler, the i-ROAD tilts through the turns and when driving on uneven surfaces. Toyota says its computer-controlled Active Lean technology automatically balances the vehicle with no input from the driver.


This is of course cool technology, and we have been awaiting skinny cars for a long time (even before GM’s Lean Machine). Even without automation, this could add significant capacity and safety to road networks, as well as providing space conservation and energy reduction. Some videos follow. When will Toyota (or anyone) mass produce this so the costs are below those of passenger cars.



The End of Traffic and the Future of Access: A Roadmap to the New Transport Landscape. By David M. Levinson and Kevin J. Krizek.
The End of Traffic and the Future of Access: A Roadmap to the New Transport Landscape. By David M. Levinson and Kevin J. Krizek.

Sequester and the Lump of Government Mistake

It is March 1, 2013 and apparently a sequester is going to be implemented by the US Federal Government. Most sectors of government are going to be cut by some fixed amount. Much has been written about how stupid this is. The proximate cause is the immediate stupidity of politicians trying to create a Sword of Damocles above their colleagues to get them to do something less stupid. There is a root cause. This is what I call the “Lump of Government Mistake”.

A Political Economy of Access: Infrastructure, Networks, Cities, and Institutions by David M. Levinson and David A. King
A Political Economy of Access: Infrastructure, Networks, Cities, and Institutions by David M. Levinson and David A. King

Almost all agencies of the federal government are on the general budget, paid for from general revenue, with an annual appropriate cycle. This need not be the case.
We should have many separate agencies, each with their own user-based revenue sources, for as many parts of government as possible. For instance Highways already have a Highway Trust Fund (underfunded perhaps, but that is a relatively simple problem if there is an actual desire to govern responsibly).
Air Traffic Control should be handled by a private corporation paid for from some kind of user fee on aircraft movements, like it is in Canada or New Zealand, not part of the Lump of Government.

National Parks should be owned by a Foundation (or better multiple Foundations) that charges admission to users to cover costs, not part of the Lump of Government.
Food Inspection Services should be a Non-Profit Corporation paid for by a small tax on food producers (like the Food Marketing Boards) and administered separately, not part of the Lump of Government.

We can go on and identify many parts of government that can easily be hived off into separable, self-sustaining, non-profit organizations.

Once we did that, the threat of Sequester disrupting the obviously generally useful things that happen to be socialized in the US makes a lot less sense.

Clearly there are exceptions, true public goods like National Defense and Foreign Relations perhaps. Research is another example. Similarly interest (and principal) on the accumulated debt needs to be handled somehow. Everyone receives “defense services” from the Department of Defense (whether you want it or not), so it needs to be paid for from a general revenue source. But this need not be the same general revenue source as used for income redistribution (like Social Security), or health insurance (like Medicaid or Medicare). In fact it is not. Social Security taxes pay for Social Security. Why should not Defense taxes (e.g. a VAT) pay for Defense. If Congress wants more, it raises the VAT rate associated with Defense, if it wants less, it lowers the rate.
The National Science Foundation similarly should not be subject to the vagaries of annual budgets. Like any good foundation, it should have an endowment, and live off the interest.

Every function would have its own associated source of funds and rates, and would stand or fall on its own merits. Horse trading would still exist, but this notion of cutting useful self-sustainable services as collateral damage for reducing the Defense sector would be eliminated.

If this feels familiar, I made basically the same point locally in July 2011 when the State of Minnesota shut down: Lessons from the state shutdown. Avoiding fragility in governance. I wrote a lot, but the main points was:

I propose the lesson to be learned is that, to avoid a total government shutdown, the government should not be totally central.

Brains mapped with network analysis


From KurzweilAI: ‘Rain Man’-like brains mapped with network analysis:

“Agenesis of the corpus callosum can arise if individuals are born missing DNA from chromosome 16 and often leads to autism.
Scientists have long puzzled over what the link is between this disorder and the autistic brain, said co-senior author of the paper Elliott Sherr, MD, PhD, professor of neurology and genetics especially since not all people with this malformation develop autism.
Doctors believe this is because the brain has a rich capacity for rewiring in alternative ways.
Pursuing this question, Mukherjee and Sherr turned to MRI and the mathematical technique of network analysis, which has long supported fields like civil engineering, helping urban planners optimize the timing of traffic lights to speed traffic. This is the first time network analysis has been applied to brain mapping for a genetic cause of autism.
The brain offers a significantly complicated challenge for analysis because, unlike the streets of a given city, the brain has hundreds of billions of neurons, many of which make tens of thousands of connections to each other, making its level of connectivity highly complex.”