Monday, September 29, 2014

Hedgehog Concept

A couple of years back, read ‘Good To Great’, and was really struck by the idea of the hedgehog principle. And I asked myself “What am I the best in the world at?”, and realized it was finding out about things. Not knowing things—other people are better at being experts at things than I am. But give me something no one knows very much about, and I’m golden. One of the reasons I’m in academia right now (rather than consulting) is that it gives me time to ‘Do the reading’. And that’s most of what I do—read on a wide variety of topics, and report back. So, ask anything—even if I can’t answer you today, odds are I can give you an answer shortly.

Friday, September 12, 2014

Downtown Streetcar

Don't know how closely you've been following the downtown streetcar. SLC wants to run it up the hill from downtown to the U, but can't decide how. S. Temple, 100 S., and 200 S. have all been mooted. 

Might be a nice place to recommend a study to them, talking traffic impacts, comparing the 3. 

Part of the deal is they want to talk future expansion to the north part of the U, which is currently a long walk (4+ blocks) from TRAX). Almost certainly along North Campus drive, 200 S. doesn't permit it. The U is currently running an express shuttle along the route.

Wednesday, September 10, 2014

Lyft, Uber, and Taxi

Visiting the Dominican Republic, one of the most impressive things were the share ride taxis. Ever so much more frequent then buses, even if they were more than a bit cramped...always made me wonder 'Why no private transit?'.

However, the furor Lyft and Uber are facing, as they compete with traditional taxis, it suddenly makes sense. In Utah, the taxis are often owned by their drivers, and represent both a considerable investment and their livelihood. I asked a driver, last night, what he thought of it. He considered it a disaster, that making a living driving a taxi was hard enough without the added competition. He noted that to do so "You practically have to live in your car".

But I've also been in a yellow cab, as the dispatcher called out, over open broadcast radio, the names of the next taxi in the queue, due to pick up a ride. Regardless of the location of the taxi, the location of the fare. Every driver gets a fare, in the order which they have declared themselves available. Equitable, but brutally inefficient. For the driver, never a chance to grab a bite to eat, or use the bathroom--their name might be called. There is never any indication of how long the queue is. Nor is it fair to the driver--you get A fare, but their is never any indication of how good a fare it will be. The use of a human dispatcher alone, reading off a clipboard, given the enormous advances made in information technology, was appalling.

And for someone waiting on a pickup, it's misery. You have no idea when your taxi will arrive. Nor that it will pick you up when it does. I've had taxis, unable to find me within the first minute, simply take someone who hailed them. (And in fairness, I've also stepped into taxis that someone else called for).

Part of the reason for the inefficiencies of taxis have to do with the Faustian bargain they have made. Cities need taxis--they are an essential part of transportation infrastructure, especially for business travelers, providing someone who doesn't know how to get around, or where they are going, with a way to do so. In effect, on-demand car hire. But to ensure enough cabs to provide this service, the city must ensure that the cab owners have an incentive to own, maintain, and operate their own cabs. Typically, this is done by limiting the number of cabs, so that all registered cabs must have a 'medallion'. In New York city, where no medallions are being issued, the price for a medallion is over a million dollars--for the right to drive a cab in NYC, you pay a million dollars. And then hope you can pay it off, by driving that cab. 'Gypsy' cabs exist for a reason...

In a more equitable and more humane world, dispatch would be automated, or largely so. Largely, it would be done with a smart phone app, which would map the locations of both cabs and fares.

The second part of the Faustian bargain is the requirement to take any and all fares, regardless of destination. A trip with no possible return fare is known as a 'dead-head'. In effect, the taxi must drive twice the distance--to get the fare to the destination, and then to get back to a central location where fares are more available.

In NYC, there was a furor when gas prices first spiked, as cabs began to refuse fares to the outlying boroughs, which combined long trips with not possibility of a return fare. The price per mile being offered was unequal to the cost per mile required.

The fare cannot vary with costs (which vary with gas prices). It also means that the fair is never perceived as fair. If the rate is too low, the drivers make too little money. If the rate is too high, the public feels they are being gouged.

The third part of the Faustian bargain is the fixed fare--it costs the same to go anyplace. Flag drop plus a mileage rate, with both fixed, regardless of market demand. It puts both a 'floor' and a 'ceiling', on the price. The pricing is static, where dynamic pricing is far more efficient. Nor are different cabs (or cab companies) allowed to charge different rates.

In a nutshell, no one is ever happy with the arrangement--neither drivers nor regulators, nor users. 

In this mess, Lyft and Uber have stepped.  The difference is the lack of the Faustian Bargain. 

They are, in effect, gypsy cabs, with Lyft and Uber providing dispatching services. Pricing is dynamic, set by whatever the driver will accept. As demand rises, so do the fares. And as fares rise, more people are willing to drive, so there is more supply. Ie, a market instead of a Faustian Bargain.

I expect that the 'long-range dead-head' customers are finding themselves paying much more. Which, given that they are receiving much more service, seems like a fair fare.

Monday, September 8, 2014

Big Data comes to transit

Check out Bridj, a new ride sharing service, using multi-passenger vans.

Cell phones have made wealth of geolocation data available. For decades, transit planning has relied on a mix of rules of thumb (population/jobs within a half mile) and empirical experimentation to determine where routes should be established...but now that can change.

Tuesday, September 2, 2014

The Bureaucrats answer vs. the Engineers answer.

This is simple enraging.
It is the bureaucrats answers: "It can't be done because it will require additional resources", rather than the engineers answer: "It can be done, but it will cost..." It would cost. No questions asked. And if the costs can't be paid by the users (through fares) they would have to be paid through other means. For UTA, that means taxes.

"How can we get more money?" is not a bad question. It is a critical question. Because money is the critical input for services. If you want to provide more services, you need to get more money.

Efficiency is a certainly a laudable goal. Asking "How can we provide the most service, the best service, for a given amount of money?" is important for any organization. But it is not necessarily the primary question that a transportation agency should be asking. It is not the question UDOT asks. UDOT asks "What are our needs?" and then goes looking of ways to fund them. It is a growth mentality, a mentality responsive not only to current conditions, but to future conditions.

Saying 'Can't be done' implies that it can never be done. The long range plan is full of projects that are impossible, today. But they are projects that are possible, over time. That's why its a plan--a list of future actions to meet a future objective. Achieving things, over time, that cannot be achieved immediately.