To be fair, this is something most of us knew already. But it's nice having data to back up the empirical (and Twitter) observations...

by Froggie on Feb 2, 2011 10:15 am • link • report

(Sorry for cross posting, but this is probably a better place than the piece about fare boxes.)

Metro is perhaps the best regional system that exemplifies this quotation from Atul Gawande that really grabbed me recently.

Here then is our situation at the start of the 21st century: We have accumulated stupendous know-hows. We have put it in the hands of some of the most highly trained, highly skilled, and hardworking people in our society. Nonetheless, that know-how is often unmanageable. Avoidable failures are common and persistent, not to mention demoralizing and frustrating, across many fields – from medicine to finance, business to government. And the reason is increasingly evident: the volume and complexity of what we know has exceeded our individual ability to deliver its benefits correctly, safely, or reliably. Knowledge has both saved us and burdened us.

That means we need a different strategy for overcoming failure…

From The Checklist Manifesto: How to Get Things Right by Atul Gawande MD

by Jazzy on Feb 2, 2011 10:19 am • link • report

300 trains/month; about 1 a day?

Is there a difference between "being removed" from service and breaking down?

Jazzy, your quote is interesting. Health care is expensive because of hospital care; hospitals are especially ill-run. Have you ever met a hospital executive? These guys couldn't get into a MBA program -- even a crappy one. They would fail out of a typical executive training program. I strongly suspect the same is true of WMATA middle management: they are barely capable of managing a McDonalds or Staples.*

One strong reason to bring in a corporate executive w/no experience on transit is they can recruit slightly better quality middle management. Not rock start, but people who at least would cut it in corporate life.

by charlie on Feb 2, 2011 10:30 am • link • report

It would be nice to put this data in the context of how many trains run per month. Does anyone have those numbers? I'm taking a WAG at 40,000. Which would mean .5% per month out of service for mechanical reasons. About .01% per month for door issues. And less than that for managerial reasons.

Am I way off base?

by Josh S on Feb 2, 2011 10:35 am • link • report

Also would be interesting to know if this sort of analysis has ever been performed inside WMATA HQ. Perhaps you could ask your union guy.

This is a depressing statistic in light of the increase in fares. The death spiral is well under way.

by varun on Feb 2, 2011 10:44 am • link • report

@Josh S

Yes, it would be good to know the percentages, but it's still disturbing that these problems have increased so rapidly. I agree with varun that we're at the precipice of the dreaded death spiral.

by Adam L on Feb 2, 2011 10:47 am • link • report

Any one here care to guess where the accompanying picture in this post was taken?

by Sand Box John on Feb 2, 2011 10:52 am • link • report

@Josh: That's actually a pretty good guess. I had 6.9 million railCAR miles per month in May, dividing by 6.25 cars per train and about 25 miles per trip gives 44,000 train trips per month.

in 2008, I calculated about 34,000 miles as the mean distance before failure for a railcar, which was bad compared to 70,000+ for NYC's worst line. In 2010, it's probably worse, like 20,000 miles.

by Michael Perkins on Feb 2, 2011 10:52 am • link • report

There is a risk of taking this out of context and applying amateurish analysis.

Perhaps the greatest risk is failure to control for confounding factors. What was maintenance spending in each relevant period? Could the "trend" reflect cost savings instead of a "true" reduction in service?

Comparing May 2008 to May 2010 might tell us something important. But it might just be a fluke. Maybe May 2010 was a "bad" month for some reason.

I understand the analysis is done this way because there probably isn't month-by-month data publicly available. Hopefully someone at WMATA does have that, and could present it as a line chart (trends over time) not bar chart.

At the end of the day, I hope people don't look at this data as a "gotcha!" moment. This is an important piece of data but we should be aware it raises more questions than answers.

by WRD on Feb 2, 2011 11:01 am • link • report

For a good understanding of some of the root problems at WMATA that lead to the poor performance illustrated above, I would recommend the book called Atlas Shrugged by Ayn Rand. The book is considered by many to be one of the best they have ever read and by coincidence has alot to do with WMATA. What is measured in the data above are only symptoms of those underlying root problems.

by Jim on Feb 2, 2011 11:03 am • link • report

@Sand Box John: It's a blue line train, going to Addison Road. Which is weird, because Flickr says it was taken in March 2008, when the line to Largo existed. Maybe they hadn't updated destination software by then?

Beyond that, I can't tell. The platform doesn't have bumpy domes nor red lights, which could be a number of stations back then. It's on an island platform station, and the wall behind the train isn't a curved/vaulted wall, so it was probably outside. I'd guess somewhere between Franconia-Springfield and National Airport. Addison Road is also possible, because Morgan Boulevard and Largo both have newer designs and different walls.

by Tim on Feb 2, 2011 11:04 am • link • report

"There are two novels that can change a bookish fourteen-year-old's life: The Lord of the Rings and Atlas Shrugged. One is a childish fantasy that often engenders a lifelong obsession with its unbelievable heroes, leading to an emotionally stunted, socially crippled adulthood, unable to deal with the real world. The other, of course, involves orcs."

by Adam L on Feb 2, 2011 11:06 am • link • report

Fire Sarles!

by Redline SOS on Feb 2, 2011 11:15 am • link • report

@Jim-

I generally try not to post two comments so close together, so apologies.

But Jim, I could not disagree more. Reading Atlas Shrugged might be good from a political or philosophical standpoint. It's not going to help here.

There are tons of resources available that DO help understand the problems. The American Institute of Certified Public Accountants, the Committee of Sponsoring Organizations, the Institute of Internal Auditors, the Chartered Institute of Management Accountants, the International Organization for Standardization, and the Institute of Management Accountants all have various Board Governance resources online. Some of them are very good, others. A lot of it is free. The AICPA and CIMA have a pretty good document, and the IIA's website is probably a great place to start. COSO sets internal control standards in the US and they're great, too. Not to mention Accenture, Deloitte, KPMG, PwC, and Ernst & Young all produce governance and management consulting information often free on their websites.

There is a whole industry dedicated to management consulting and a whole publishing sector that pushes books from the lucky managers and the good ones.

Atlas Shrugged is just not going to help here. I strongly suggest starting at the Institute of Internal Auditor's website. They have a few good blogs that cover Board and Enterprise Risk and governance practices. It's free and great reading. At least, if you're as boring as I am.

by WRD on Feb 2, 2011 11:17 am • link • report

If only WMATA would replace their steel rails with Rearden Metal..then we wouldn't have all of these problems.

by Phil on Feb 2, 2011 11:19 am • link • report

I couldn't get through Atlas Shrugged. I tried to read it once. So the main character runs a railroad and someone else is making metal. But then the government passes a bunch of incredibly stupid laws that make no sense for anyone and would never even be considered in real life.

But they don't hire an army of lobbyists to point out how this is stupid, underwrite think tanks to write reports on how this is stupid, start their own media outlets to point how this is stupid, or just donate to candidates directly, all of which are what happens in the real world and therefore the government ends up mostly helping the captains of industry instead of hurting them.

Instead, they all just get in a funk and... well I stopped reading. Whatever.

by David Alpert on Feb 2, 2011 11:24 am • link • report

In addition to the good questions posed above regarding percentage, how many of these "removals from service" are preventative? As in, train operators are being more attentive to problems and are removing the trains from service before it becomes an actual problem?

In other news, GGW should do an article about the new expansions to the Singapore SMRT system. I watched a TV program about it the other night and I was floored by how efficient the system is. They have walls with doors on the station platforms to prevent suicides. (Picture http://en.wikipedia.org/wiki/File:Ns24dhobyghaut.jpg)

The system itself has the same food/drink restrictions as WMATA, yet they have 7-11s or other convenience stores in many stations, without problems. The system runs similar hours as ours (5:30am to 1am) and has almost the same number of stations (80 to our 86).

They also have one of the lowest operating costs per passenger in the world, tickets are about $1.00-$1.50 (USD) and continue to be a profitable system.

Surely, they are doing something right. I'm not a mass transit expert but if they can do it, why can't we? What are the real differences between theirs and ours?

by Sam on Feb 2, 2011 11:30 am • link • report

I know this article is speaking of rolling stock reliability... but what about track/switch reliability, and what about the perceived reliability of the system via "Next Train" signage?

I think people would be much happier about the system if those were reliable in the least... for example, picking up a yellow line train at P. City this morning, the Blue line train was showing arriving (4 minutes after it left), yellow line train 7 mins away, and the next blue line train at 13 mins. Further it showed the infamous "Train" 14 mins out.

The yellow line train arrived when it was saying 4 mins out...

If those signs show accurate, that would be beneficial... ie: if a car is out of service, call the train a 5 car train.

just my $0.02

by Brad K on Feb 2, 2011 11:39 am • link • report

I just want the train operators to stop yelling at people to who are supposedly "blocking the doors." We all know the doors are crap, and that no one is blocking them.

by OX4 on Feb 2, 2011 11:41 am • link • report

@Brad K: The metric was any train delayed (greater than four minutes, I think), taken out of service or not put into service due to a mechanical, door, or management problem of any kind. This included track problems in some cases.

Since Metro doesn't report individual problems with the PIDS boards, it's not included in the data.

The only reliability data Metro reports to the public currently are the raw data I used for this report, MDBF data reported monthly to the CSO committee (and only one and a half years of data is available) and on time performance data based on a crazy metric which allows a weekend train to be up to 15 minutes late (half the scheduled headway for that period) and consider it "on-time".

by Michael Perkins on Feb 2, 2011 11:46 am • link • report

1. Prefacing this post with presenting the facts that Metro is obviously a long way from perfect, and needs some serious investment in both its physical assets and its personnel in order to bring it up to a standard we would all like to have. There is a lot to improve.

@Sam

Any rail system's biggest cost BY FAR is labor. Among the heavy rail systems in the US labor costs (salaries, wages, and fringe benefits) make up 78% of their operating budgets. For Metro it's a little lower - 74% of their operating budget. So the biggest cost to these systems is the money they must pay to have people operate their trains, run their offices, etc.

It just gets somewhat tiring to have people bring up examples of systems in other countries (happens all the time here) and then say "well they must be doing something right, what can we learn from them?" Here's what can be learned from those systems - if Metro could pay its workers the going rate for labor in Calcutta, or Singapore, instead of the going rate in Washington, DC, then they could reduce operating costs significantly enough to have the same low fares and break even.

If you want to look at peer systems in Europe and see how they do it, it is almost universally supported by government assistance.

by MLD on Feb 2, 2011 12:14 pm • link • report

Is car reliability broken down by model/type? It would seem fairly obvious to blame the old 1000-series cars and problematic 4000-series cars for a lot of the issues. NYCT's 70,000 MDBF figure could be due to the fact that they're currently operating the youngest fleet of railcars in that agency's history.

Did reliability plummet when Metro started mixing car types and abandoned ATC?

by andrew on Feb 2, 2011 12:21 pm • link • report

@Sam - I don't know anything about Singapore's system, but I'm willing to bet a day's wages they spend orders of magnitude more money than WMATA does. That they have a government that is committed to the transit system, rather than seeing it as a nuisance or threat to cars/the American way of life if they think about it at all.

by Ed Hoover on Feb 2, 2011 1:02 pm • link • report

Just to continue on my point, I had a look at the Singapore Transit annual report:
http://www.smrt.com.sg/investors/documents/annual_reports/2009/pdf/SMRT_Full_Report_09.pdf

Labor costs for them are 40% of operating expenses. If WMATA had the same ratio, Metro would make a huge profit, and the transit system as a whole would almost break even. If you took out paratransit, the system as a whole (bus and Metro) would make a profit.

by MLD on Feb 2, 2011 1:29 pm • link • report

@Ed Hoover,

Singapore's system is run by a private company called SMRT Corporation Ltd (SGX: SMRT) and according to their 2010 Annual Report (http://www.smrt.com.sg/investors/documents/annual_reports/2010/index.htm), last year they had after tax profits of SGD162,885,000 (US$128,029,683.78)

by Jacob on Feb 2, 2011 1:38 pm • link • report

@MLD, one of Singapore's four lines is driverless (North East Line), thus is labour cost is far lower, as well as the fact that drivers have lower wages in Singapore since there's no corresponding union that jacks up the cost.

Remember also that Hong Kong and Singapore run property divisions for property above their respective stations. Without these, they would not be profitable.

by Phil on Feb 2, 2011 2:15 pm • link • report

@Michael Perkins
I guess what i was talking about is perceived reliability...
You are right that the metrics are specific to physical rolling stock... however-

One can assume or even rationalize that if the average rider shows up at a station, and there is going to be a ten minute wait because a train broke, and a no-passenger train has to haul through to get out of the way, that would be annotated on the PIDS boards.

If it was available on the display both in and outside the faregates, that rider may not care (as much) if there is a broken train rolling through... they accepted it when they swiped in to the faregate. However, if a rider sees that their train is rolling in in one minute, they may be inclined to run to make sure they make it... if that is truly the case, the system was reliable to them.

On the other hand, if I run to a train that is "1-2" PIDS minutes away, and i get to the station platform 5 actual minutes before the next train even arrives, I am not going to be a happy camper. That is perceived non-reliability.

While I acknowledge the rolling-stock and track have real problems with their maintenance, and further their reliability, all I am saying is that the perceived information provided by resources such as the PIDS is equally important. I know it would be a financial and systematic nightmare, but even if the PIDS didnt show actual times until the next train, but average times between certain color trains, customers would know if they need to run or if they can afford to wait until the next train.

Resources such as PIDS are "assumed" to be accurate until you actually deal with them on a day-to-day basis, at which time, you realize they are only useful to inform passengers of single tracking, broken trains, special events, or elevator outages.

by Brad K on Feb 2, 2011 2:41 pm • link • report

"Surely, they are doing something right. I'm not a mass transit expert but if they can do it, why can't we? What are the real differences between theirs and ours?"

No unions, an authoritarian government, and competent management paid private-sector salaries.

by Phil on Feb 2, 2011 2:55 pm • link • report

Singapore has very high fuel taxes (retail price of $4.92 per gallon is cited, about $1.50 per gallon more than the US right now), fairly rigorous licensing requirements and requires a payment on the order of about $10,000 just to have the right to own a vehicle for ten years. In addition there are parking charges and road usage charges. With a dense city development and high cost for individual automobile use, transit ridership is high, which is efficient.

by Michael Perkins on Feb 2, 2011 3:34 pm • link • report

@Tim:

The destination software was updated prior to the opening of Largo. The destination being displayed is not really relevant as to where the picture was taken

On closer inspection of the picture you will notice that there are in fact bumpy domes behind the granite edge.

When I first looked at the picture I though I new where the picture was taken, on closer inspection I concluded otherwise.

The picture was taken from the inbound New Carrollton platform in the West Falls Church station, the train is berthed on the center track 3.

by Sand Box John on Feb 2, 2011 10:43 pm • link • report