Wednesday, June 24, 2009

Direct Result

A direct result of poor ridership estimation or purposeful underestimation for new lines is not enough capacity to supply service to the people that use the system. Minneapolis has decided to shut down parts of the Hiawatha line this weekend to lengthen the platforms for three car trains. This is going to happen on the South Corridor in Charlotte as well in time. It's unfortunate but this is something we're going to see in San Francisco on the Central Subway as well. Whoever thought cutting platform length to save money now instead of saving money later was a good idea was very very short sighted. These types of cost cutting decisions on the front end really need to stop.

11 comments:

jwb said...

Our train control software is incapable of handling a 3-car rig, so the odds of you seeing a 3-car streetcar in a Muni subway are very near zero.

Glen B said...

Patronage estimation is a tricky exercise (more an art than a science) and dependent on assumptions on a future that may or may not prove to be accurate. A number of rail projects in the 90's were pillored due to over-optimistic ridership estimates, which has led, of course, to more conservative estimates for future projects. Don't be too harsh on the estimators. FTA and localities have been long struggling to find the goldilocks solution to this issue. Wish them luck!!

Jonlin said...

Here in Seattle we built platforms for 4-car trains even though we'll only be running 2-car trains for the next few years.

Pantograph Trolleypole said...

Glen, obviously it is tricky. Though the FTA recently has decided to use the Cost Effectiveness rating to kill projects. The major part of this rating is ridership estimation. There are projects such as that in Columbus Ohio which died by the CE rating that could have been like Charlotte or Minneapolis (who both had medium low CE ratings) and probably overshot their projection rendering the CE rating useless. That is why this stuff matters, because they make it way more important than it should be.

Jarrett at HumanTransit.org said...

It would be good to see some numbers on how much it costs to extend platforms later as opposed to as part of the initial project. You would need to factor out inflation and just looking at the cost of doing the thing in two phases instead of one.

I agree that undersizing platforms on underground lines is a very serious mistake. I'm not so sure that it's all that fatal on surface lines, so long as the initial project secured the right-of-way for platform expansion when it's needed. I'd need to see some numbers on how expensive the two-phase solution really was.

njh said...

I agree, the right strategy in my mind is to implement the basics and have a continual expansion plan. A single track tail on a route is fine, as long as the second track can be added without disruption. This is not hard to arrange, especially with glue in rails rather than excavate and bury.

Cap'n Transit said...

You've gotta be kidding me, Jwb. The cost of upgrading the software to allow three-car trains?

JimS said...

@Cap

Upgrading the software on Muni is actually a huge undertaking.

Alcatel's system does not work off of a data file showing track layout, trainset configuration, routes, etc.

Instead, all of these things are hardcoded subroutines. Each implementation is a custom-built program that needs to be rewritten if you do so much as add a NAME.

This is why Muni T trains are always labeled as Ks. They'd have to rewrite it to add the letter T.

Dead serious.

JimS said...

@Cap

Upgrading the software on Muni is actually a huge undertaking.

Alcatel's system does not work off of a data file showing track layout, trainset configuration, routes, etc.

Instead, all of these things are hardcoded subroutines. Each implementation is a custom-built program that needs to be rewritten if you do so much as add a NAME.

This is why Muni T trains are always labeled as Ks. They'd have to rewrite it to add the letter T.

Dead serious.

arcady said...

I think it might be time to admit that Seltrac was a mistake, and start looking for a replacement. It has too many stupid limitations: no 3-car trains, no new routes, no double-berthing, a fairly long delay on entering the system. And besides, how much longer can they keep running a system based on OS/2?

Bob Davis said...

Sounds familiar; here in the Los Angeles area we had a major upgrade on the LA-Long Beach Metro Blue Line (which mostly follows the old Pacific Electric route) when Metro found that the 2-car trains for which its platforms had been built in 1990 were starting to look like Tokyo subway trains. At considerable expense most of the platforms had to be extended. Now the line runs 3-car trains all day long, and even during non-rush hours, there are often standing loads.