Schedule padding: it’s a bit more complicated than that

I’m on the road: here’s one I prepared earlier. Don’t expect prompt comment moderation, replies, or corrections of embarrassing mistakes.

Telegraph transport hack David Millward, in a rare departure from the normal churnalism schedule, has written an original piece: “State-run rail company accused of fiddling punctuality figures.”

It’s something of a departure from the established definition of “news”. East Coast revised its timetable way back in the spring. They added a little extra time to some journeys — a dirty trick that allows them to inflate their performance statistics and avoid having to give refunds to passengers on trains that run late, by redefining “late”. At least, that’s how the newspapers would have you believe it.

The railway industry will point out that it’s a bit more complicated than that. In an extremely complicated network with a packed timetable, if anything goes just a little bit wrong, there can be massive knock-on effects and the entire timetable will fall apart. And in a system catering for millions of passengers on thousands of trains, things always will go wrong. Doubly so in a country that’s reluctant to invest in maintaining its infrastructure. The timetable padding is there to allow recovery from tiny problems before the tiny problems bring the whole system down. Think of gaps in a domino rally if that helps.

Timetable padding is a good thing, because reliability is more important than speed — despite our obsession with the latter. People taking an East Coast service between Edinburgh and London shouldn’t care if it’s timetabled for 4:04 instead of 3:59, but they certainly will care if it’s five minutes later than they were expecting and they miss their connection or walk into their important meeting late.

But of course, it’s a bit more complicated than that. Performance statistics are generally counted at the terminus of a route, not at the intermediate stations (though companies do still have some incentive to be on time at the intermediate stations, because passengers can claim refunds for very late arrivals), and the industry’s defence of padding might be more acceptable to passengers if it weren’t for the fact that padding is so often added there, right at the end, where the statistics are counted. South West Trains from the Westcountry idle for five minutes on the approach to Waterloo; intercity trains routinely reach Glasgow with ten minutes spare (but spot the 17 minute late train from King’s Cross — it’s that state-run rail company, East Coast). Thus the reliability defence only applies to those going to the very end of the line. Got a tight connection at Clapham Junction or Carlisle? No such luck. As well as benefiting more passengers, padding that is evenly spread can, in absence of disruption, take the form of slower speeds, reducing energy use, emissions, and wear-and-tear.

But it’s probably more complicated than any of that because clever tinkering always has unintended consequences — side-effects. I don’t know what the side-effects of schedule padding are, because I can’t find any research on them, or any evidence that anybody has ever thought to do any research on them. But I can make some guesses at where might be a good place to look for them. As the consequences of minor delays are reduced, will staff get less strict about departure times — a little less rushed, a little more forgiving to the late passenger running down the platform — and would that be a good or a bad thing? Might such adaptive behaviour completely absorb any benefits that timetable padding ever had? Might train company management, in times of high fuel costs, pressure drivers into adapting speed to reach the terminus on time at the expense of arriving late in low-patronage intermediate stations, where performance is less important to the company? What effect does padding have on people’s desire to use trains over alternative modes? Speed shouldn’t be important, but speed sells — does the extra five or ten minutes advertised time affect people’s modal choice? Does arriving at Glasgow ten minutes early impress people, or are they annoyed that they have to wait around longer for their connection? Will regular passengers adapt and expect to arrive in Glasgow earlier — thus being inconvenienced on the occasions when their train is merely “on time”? What does a five minute wait just outside Waterloo — when everybody has packed their bags, put on their coats, and stood in the aisles anticipating arrival — do for the railway’s image? Our impression of the railways is surely affected by unscheduled delays, but does padding do more good or more harm?

I have no idea, because as far as I can see, nobody has really explored the subject.

What I do know, is that railing at state run East Coast for schedule padding is absurd. They copied the idea from the all the other (privatised) train operating companies, who in turn got the idea from airlines. It would appear that a memo went around the right wing newspapers to find reasons to bash state run East Coast — the caretaker organisation that bailed out its failed privatised predecessor.

Advertisements

5 responses to “Schedule padding: it’s a bit more complicated than that

  1. Timetables have always had recovery time in them: ie the advertised time at the terminus is later than the arrival time in the working timetable. Larger amounts of recovery time were added at the time of John Major’s government because of the introduction of passenger compensation for significant lateness: this was obvious because in some cases the departure time of a train on its next journey was before the advertised arrival time of that train, and because the “padding” suddenly disappeared from the schedule at the end of the rush hour (the compensation was calculated only on the basis of lateness in the rush-hour).

    All train companies seem to have quite large amounts of recovery time in the schedules these days. It has been easier to spot with East Coast because the working timetables were available to the public for consultation last year.

  2. People value reliability over journey speed, so you put in padding, as you say. You put it in at destination, and in public times only, so as to minimise the side-effects (there are also little bits put in as you go along, but they are kept to a minimum).

    The signallers will be using the working times, not the public times, though they will give priority to a departure, if there’s a choice, because the arriving train has time to recover before its scheduled departure. Delays are measured against the working timetable, so the signaller will try to avoid making a train more than a couple of minutes late on its working times. Anything more than a couple of minutes has to be explained (and “it had public-book padding” would count as a signaller error).

    If you are standing outside Waterloo, it’s mainly because there aren’t enough platforms / too many conflicts, not because of the padding.

  3. Anyone who cuts it so fine to get to an important meeting that he can’t afford his train to be 5 minutes late is obviously so out of touch with the realities of life he should probably be sacked anyway.

    The state is notorious for claiming excellence in many of the services it provides despite evidence to the contrary – look at schools and the NHS!
    And look at the miserable, poorly planned cycle lanes local government paints which are of no use to anyone. Has a council ever admitted they are rubbish and a waste of taxpayers money?

    If the state isn’t scrupulously honest then why should anyone else be?
    Anyway, the real reason for London’s transprt problems is that far too many people cram into it for work, when 95% of them could probably do their work anywhere in England or anywhere on the planet in these days of instant communication. Trains to London are crammed with people who do the most mundane jobs or are travelling for meetings of dubious worth which could in any case be done by video-conferncing or even by an old fashioned group phone call. Email that powerpoint to everyone – nobody wants to watch it anyway!

  4. Don’t quite see how you can call it “padding” anyhow. Stuff happens, schedules slip. You cannot pretend that there will be no bad weather, no jams of riders getting on/off the train, etc. At rush hour there’s more passengers, hence more stuff that can happen.

    “Adjusting optimism to reality” seems like a fairer term to me.

  5. In Switzerland, they pad the schedule 7%, uniformly if I’m reading the linked article correctly.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s