Monday, 7 January 2008

Up close and personal

By which I mean crowding, but I thought that 'up close and personal' was a more interesting title. Sit comfortably, this is going to be a long one.

It's a tricky subject as no-one likes to stand on trains, especially for long stretches.

The final commenter on my last post did kinda ruin the surpise, because I'm sure it will be a suprise to some, if not most, of you. That being that the legal limit for the number of passengers on a train is, da, da, da: none. That's right, there is absolutely no limit to the number of passengers that can be crammed on to a train.

That may seem outrageous because planes, buses, taxis and seemingly every other form of transport has capacity limits imposed on it, but not the railway. This is, supposedly, because the railway is intended to be a flexible form of transport.

Now, here's another piece of knowledge that you may not like to hear. Many studies were done for the HSE on the subject of crowding on the railway and then for the ORR when it took oevr the HSE's safety responsbilities. Most notable perhaps, is Dr Bottomley's report of 1999. Largely based on that report, the HSE took the view that standing passengers were not a safety issue. Standing is an issue of passenger comfort and there is no additional risk to standing passengers.

If I may provide you with a few choice pieces from that report:

'For collisions occurring at higher speeds, serious injuries can be sustained by both seated and standing passengers ... There is no evidence to suggest a difference in severity.'

'severe chest injuries leading to asphyxiation are a real possibility in heavily overcrowded situations, but there is no evidence to suggest the degree of overcrowding experienced on trains ever reaches the densities witnessed in extreme circumstances.'

(From: Bottomley Report for HSE, 1999)

This led to this, from the HSE:

'HSE does not believe that there is sufficient evidence or grounds for introducing new legislation to control overcrowding because of health and safety risks.'

'Trains are built to operate safely at and above the maximum passenger loads which it is physically possible to accommodate.'

'Passenger loading does not affect stability, braking performance or structural strength, and doors are designed to withstand crush loading.'

'In general overcrowding relates to passenger comfort and well-being, rather than safety.'

(From: Presentation to HSC Railway Industry Advisory Committee, Birmingham, October 2003)

The (now defunct) SRA position is that:

'There is no practicable means of preventing people from boarding a crowded train if they choose, or of compelling them to leave.'

'Published capacity targets (PIXCs) are for service planning purposes only, and unenforceable as actual ceilings on numbers carried.'

(From: Presentation to HSC Railway Industry Advisory Committee, Birmingham, October 2003)

So, to sum up. The HSE says that crowding is not dangerous and the SRA says that even if there were legal limits on passenger numbers, they would not be enforcable. Why, are trains any different from planes then? Why can't the railways predict numbers?

Planes are slightly different. Once you're on, you're there for the duration. There are no stops part way through with people getting on and off. That means that plane operators can confidently and accurately predict the maximum number of people who are going to want/going to be able to get on a partricular trip and they can avoid crowding because once the seats run out, they stop selling tickets.

As I mentioned, the railway is intended to be a bit more flexible. This is why we sell open tickets. They allow you to get on any train you want and, even if you book onto a specific train, there's no guarentee that you'll catch it. Never mind those people who buy an open ticket because they have no idea what train they actually want to catch.

This is why it bugs me so much when people accuse us of overselling trains to maximise profits. We don't oversell. The proportion of tickets that can only be used on a particular train is tiny compared to even the seating capacity of that train. If a train is crowded, it's because a lot of people want to get on that train and have chosen to do so, and most of them have flexible tickets. I find it ironic that, in my experience, the vast majority of those accusing us of overselling have flexible tickets themselves. I'm including season tickets in this. Yes, most of you are commuters and probably get the same train most days, but can you say you do so every day?

The fact is, that with the prevelance of open tickets, there is no way to accurately predict crowding. We can, and do, look at trends, and this enables us to find the worst affected services, but on a day to day basis, it's a no-go.

So, crowding is bad. What can be done to reduce it? There have been several ideas proposed. Let's see what you think.

Idea 1: No standing passengers at all. Make trains seated only.

A fair point you might think. But how many of you would not be able to get to work if you were not allowed to stand up? This reduction in capacity on trains means that there need to be more trains to accomodate the excess passengers. Over to you Dr Bottomley:

'If the excess passengers are removed and put on an extra train, there might be an increased risk to the total population of rail passengers simply due to the addition of the extra trains required to deal with the excess passengers.'

Of course, he other knotty problem is that, there are no excess carriages now. Hands up who thinks we could get enough to run extra trains. Secondly, hands up who thinks that sufficient paths can be allocated to run the extra trains.

Idea 2: Make everybody reserve on a specific train with no more open tickets. Then crowding can be controlled like the airlines do.

OK, that would probably work. You'd need platform staff at every train door checking that people had reserved on the train they were tring to board and this would delay the trains a bit, but it would work. However, it removes flexibility. You can only get on one specific train. You miss you train, tough rocks pal. You might be able to switch your reservation if there's room, but there's a fee and you may find yourself waiting a while if you're trying to get home in the peak. Want to stay a little longer for the shopping trip in London? No dice. I'm thnking that this would get irritating soon. Crowding would reduce though, because everybody would start driving.

Idea 3: Get more carriages.

Wow, it's so simple. Why didn't we think of that? We have been trying you know. It's no like we enjoy getting leters constantly complaining of crowding. Why do you think we were so narked that we had to give those 158s back? We needed those carriages so we had to get in some beat up 142s because it was never an option not to replace those trains.

I think there are a couple more out there, so now I will proicede to tell you why it doesn't matter. The two most crowded routes on the FGW network are Portsmouth to Cardiff and just about anything that goes from Reading to Paddington.

Lets take the Thames Valley first. Very crowded network, especially at the peak where all the problems are. That means no paths for additional trains so no supplemental 142s. They ussually run as 2 car, although I believe you can have two sets run as a 4 car, and they have a top speed of 75 mph. That's up to 4 cars and 15 mph less than the turbos that you've got at the moment. I won't even make a comparison to the HSTs. Also 142s use different couplers, meaning that they can't reinforce turbo services, as well as having the speed issue. Result, we're back to needing Turbos to strengthen services. If we could get any, rest assured, we would have.

Portsmouth Harbout to Cardiff has more overcrowding complaints than any line on the network (see, told you I'd been reading the reports). I have to admit that it has not been one of FGW's top priorities and, between you, me and the world wide web, I don't think FGW really wanted to run it. South West trains has introduced some extra services in the Salisbury area, but that's not a lot of help for the rest of the line.

So, Portsmouth to Cardiff runs 150s (mostly). It's possible that 142s could run as additional services on the line, but there would probably be platforming conflicts at some of the bigger stations along the way. Temple Meads and Cardiff spring immediately to mind. The problem is that, again, they run as 2/4 car sets, and even with 4 that's only 240 passengers max. The other problem is the speed. Slower trains means less available paths means that, ultimately, you don't achieve much of an improvement, especially once platforming and other considerations are taken into, um, consideration. This means again, that we are back to needing more modern trains to run the route. See end of Thames Valley paragraph.

I am sure that some of you have ideas to reduce crowding. If so, I'm interested in hearing them. I'm sure you also have comments on the idea of crowding not being a safety issue. As always, please fell free to comment, and I 'll try and provide the answers.


chris from nailsea said...

Thanks very much for your excellent summary of these 'considerations', Insider!

One suggestion, albeit 'tongue in cheek' to reduce crowding (and save money): why not amend the current refurbishment to simply removing all tables - and seats - altogether? You'd probably double the capacity of each carriage!

On a slightly more serious - and, sorry, a vested interest, note, why is the 1753 Bristol Temple Meads to Weston super Mare (through Nailsea) service booked as a two car? Every evening it's very crowded, sometimes it leaves passengers behind - and when it's reduced to one car, it's absolute chaos! Surely this prime commuter service should be booked as a three / four car? Who decided to book it as a two car, and why?

(Semi)rant over: I would really appreciate your comments!

Anonymous said...

On the subject of reservations, I normally walk through the train, removing reservations that have not turned up / sat somewhere else / put their coat on the reserved seat and sat in the seat beside it.

Probably a third to a half of all reservations have nobody sitting in the seat.

If I am on an evening train out of london, the first thing I do is go through Standard, removing seat reservations, telling people to shift their bags, and then put out a PA backed up by telling passengers where the empty seats are (usually coach A). Most people standing don't move, they stay standing, usually saying "it's not far to Reading where I'm off", or "I've been sitting all day, I want to stretch". The remainder say "I can't move with all this luggage", looking at me accusingly as if it's my fault they decided to turn up too late for a rush-hour train to sort themselves out, and just pile in the door at the last minute.

Lee Fletcher said...

A couple of things :

I use Portsmouth-Cardiff trains, and rather than 150's, they are mostly run by Class 158 units, which have a top speed of 90 mph.

However, Class 150 units do "stand in" surprisingly regularly. They have a top speed of 75 mph, but run to the same schedule as the Class 158's.

Class 142 units also have a top speed of 75 mph, and this makes me wonder whether speed is as big an issue as you fear on this particular route.

A bigger obstacle would probably be that Pacers are banned south of Warminster.

Insider said...


Perhaps not so much on the Portsmouth route this is true, but delays and displaced services are frequent when the 150s stand in.

Yes, pacers are a problem south or warminster, just like in south devon. The rigid wheelbases don't cope well

Lee Fletcher said...

A couple of other things to take into consideration :

Several sections of the Portsmouth-Cardiff route have a speed limit of 75mph or less.

FGW already utilise additional paths along the Portsmouth-Cardiff route during the peak. Examples include from Westbury to Bristol Parkway at 0718 & 0817, and from Bristol Parkway to Westbury at 1647 & 1747.

These journeys are timed for "slow train" running.

AndyW said...

Came across your blog recently and find it interesting.

I would take issue with the statement “The fact is, that with the prevalence of open tickets, there is no way to accurately predict crowding” – while you don’t know exact figures you know full well which trains will be crowded – they are the same trains day in day out.

If you want more capacity for commuters (your main problem I guess) then why don’t you split some HSTs and run them with a single power car plus modified DVT from the West Coast with 4 or 5 carriages.
A 5 coach (4 standard + 1 first class) would have 352 seats or 380 seats if all standard class. You could probably go to 450 seats if they were configured 3+2 like the Turbo rather than 2+2 although I wouldn’t recommend it. This is a useful increase over the Turbos (or Adelantes for that matter). I presume you would us these on the Paddington – Reading route. This could then release Turbos for other routes.

Tim said...

The logic of the statement;

"'severe chest injuries leading to asphyxiation are a real possibility in heavily overcrowded situations, but there is no evidence to suggest the degree of overcrowding experienced on trains ever reaches the densities witnessed in extreme circumstances.'"

is flawed. It is saying that overcrowding can in theory be very dangerous but up to now it hasn't reached dangerous levels. The correct conclusion to draw from this is not "do no have a limit on passenger numbers" but rather "impose a limit on passenger numbers which is high enough to encompass current levels of overcrowding but will not permit them to go much higher"

Lee Fletcher said...

Quote :

"I am sure that some of you have ideas to reduce crowding. If so, I'm interested in hearing them."

I have given it my best shot, but let's be clear on one thing :

FGW cant do it on their own. They will require help from the DfT, Network Rail and, quite possibly, local authorities as well.

What follows is a three-stage plan, inspired in large part by ideas from members of First Great Western Coffee Shop forum. I have taken every factor that I know of into account, but there will undoubtably be things that I have missed. Therefore, I would appreciate it if you'd bear that in mind when reading it.

The plan is also dependent on FGW successfully resolving issues such as staffing and Swindon control.

STAGE ONE (December 2008-December 2010)

3 extra 2-coach Class 142 units (currently in storage) leased to bolster the capacity (and PR) gains made in Devon.

Key "Cross-Bristol" trains run using loco-hauled stock, thus potentially providing extra carriages on these services. This in turn frees up units for use elsewhere, including a Class 153 unit for the extra Frome/TransWilts services contained in the link below. The loco-hauled services are gradually replaced by Class 150/1 units released by London Midland.

The proposed 40 minute frequency Severn Beach Line timetable is assumed to have been implemented in May 2008, as currently planned.

STAGE TWO (December 2010-Franchise End)

A radical overhaul of the franchise is implemented, as detailed in the link below.

STAGE THREE (Franchise End Or Life After FGW)

New rolling stock all round, with the "Cross-Bristol" specification based on the link below.

Now, I'm not naive enough to think that all of the above could necessarily be implemented exactly as I have set it out, but I do think that something close to it could be acheivable.

Lord Hutton said...

We have the lovely 142s in Devon now. Run in pairs they certainly have solved certain overcrowding problems eg from Torre, but the result is, I never have to pay as the guard stays in the rear set, and do we really want these horrible things associated with rail travel?
And if there are spare trains about, what the heck are the control freaks at DfT doing about it?

Anonymous said...

On the subject of one of the most crowded routes on FGW....Portsmouth to Cardiff, the use of the now redundant 180 sets would not only reintroduce a level of comfort not seen on this route since the withdrawal of the 5 coach loco hauled trains, when passenger loadings were lower than now, but would also surely encourage passenger use of this route. These sets would then be available for other very busy sprinter operated services. Common sense says FGW desperately needs these 5 coach trains, just when they are getting rid of them.

Lee Fletcher said...

Yes, I agree (as did Insider in an earlier post) that the Class 180's could be useful on Portsmouth-Cardiff services. However, Andrew Griffiths (FGW Regional Manager, Severn & Solent area) completely ruled the idea out when I asked him about it.

Also, it appears that other operators, such as Arriva Trains Wales and Hull Trains, could be interested in taking them on, although ATW have denied this.

Anonymous said...

unfortunately the question of 'splitting' HSTs into 4+1 was looked into some years ago and then quietly but firmly shelved!
a) they are not a normal couple/uncouple 'item' and therefore have a high probability of not going back together when needed, ie they werent designed to be regularly split, this was only for maintenance and safety purposes (ie fires!)
b) lots of sets would have to be shunted in to Paddington etc to rejoin, this is very very difficult with current staffing and signalling restrictions -how many times have you waited for a signal at paddington for sinatnce whilst another train crosses over to another platform?
c)FGW simply dont have enough HST sets to run their own HSS services now, let alone supplementing local services by splitting sets, even if possible.
there are loads more technical reasons which i wont bore you with, good question though.

Billyo said...

Don't know how feasable this is, and I know it's horribly simplistic (I am far from railway expert), but couldn't train frequency be reduced during the afternoon and the trains "saved up" for an increased frequency during rush hour?

Lee Fletcher said...

Which routes would you reduce the afternoon service on, and how would you explain it to those affected, billyo?

Billyo said...


The same routes which would then have an increased frequency during peak hours. I work a usual 8-5 day, so I have no idea how many people are on each train between 11-4 between say Bristol and Bath (the journey I make). But I'd imagine these trains are not as full as the trains which leave at 4.30, 5, 5.30 and 6. Could an hourly service not be run for a few hours in the afternoon and peak times bumped up to 3 trains an hour?

I didn't say it was the answer, I don't know if there would be room in the timetable etc. But it makes sense to me to reduce service frequency where the trains aren't full and increase it where trains are full to bursting.

Effectively this already happens in some places (there are more trains per hour in the morning rush between Bath and Bristol than the rest of the day) but there never seem to be extra trains in the evening rush, which is always worse.

Billyo said...

Sorry, that should say between Cardiff and Bath, as the journey I make.

Lee Fletcher said...

As you know, trains dont terminate at Bath, but continue to points such as Westbury and beyond.

You may be interested in the timetable that Jacobs Consultancy proposed when they advised the SRA on the original Greater Western Franchise specification (link below.)

I doubt that you would get the above idea, or your proposal past those on the ground, especially at (say) Freshford.