Return to Archive -by date - by topic.

INFORMED SOURCES August 2004

Be positive! Say ‘No'

Not rejecting irrational requests is costing the railway dear – time to import the ‘Silver Ring' campaign to railways

A career in an hierarchical organisation with a command and control structure focused on round the clock real-time operation predisposes railway managers and engineers to obey orders. At times of service disruption someone decides what is to be done and it is done.

That managers like Chris Green could achieve so much was because they controlled just such a responsive machine. Chris could say ‘Let there be red lampposts' and there were red lampposts'.

Yet, as Chris and the other graduate trainees of the 1960s showed, it was the willingness to take unilateral action, or argue the toss that marked out what the French sports writers used to call ‘les comingmen'.

So British Rail presented an interesting dichotomy: an organisation which required compliance but encouraged and rewarded can-do. Unfortunately that combination applied to internal matters only.

When outsiders in authority – government ministers, official inquiries – told the railway what to do, compliance mode tended to rule. Everyone knew that some of the Fennel inquiry recommendations regarding underground stations following the Kings Cross fire were redundant or irrelevant, but no one had the nerve to stand up and say so.

Hence the nonsense of treating Birmingham New Street in its cutting as an ‘underground station', with fire doors and the alternative escape route in the form of the extra bridge. One safety professional reckoned that the Value per Prevented Fatality (VPF) was £10 billion – or it may have been £100 billion.

Can't say no

Post privatisation compliance became institutionalised and the industry lost its will to say ‘no'. And we are paying heavily for this lack of intellect and intestinal fortitude.

Of course Southall and Ladbroke Grove did not help, but as far as I know only the Institution of Railway Signal Engineers and this column were critical of certain recommendations in the reports by Lord Cullen and Professor Uff. Did you read anywhere else at the time that the timescales proposed for ERTMS fitment were unrealistic?

Anyway, to launch a new Informed Sources campaign, here are some case histories to show what happens when you go along with official policies that you must know are wrong. And, referring back to last month's RSSB Safety Report can I commend Porterbrook and the freighters for kicking against the pricks and getting two standards reverse ALARPed.

Just to remind you that ALARP stands for As Low As Reasonably Practicable. Since ‘reasonable' can only be quantified by the time, effort, money and unintended consequences of a course of action, reverse ALARP means saying that this requirement is too expensive.

This is not popular with the HSE which some time back deprecated the use of reverse ALARP to justify not doing something. But if you believe in the cold numbers, as this column does, then ALARP turns both ways and those two test cases have effectively lifted the ratchet for a couple of clicks. But more is needed.

All the case histories are about safety. This may be coincidental and I suspect the Government's Review of the railways – published even as the presses are rolling – will provide a wealth of opportunities for the industry to be positive – say ‘no'.

Test Case 1 – Crashworthiness

In the June column on energy consumption – which actually got the four letter ‘F' word and ‘lardbutt' into the august leader column of the Daily Telegraph – the high mass per seat of modern high speed multiple units was attributed in part to the floor space wasted by crashworthiness requirements.

This I attributed to the HSE from whence a chum pointing out that some rough calculation, comparing Pendolino and an HST, suggested that only 40% of the loss of seats was actually down to crumple zones, the rest being due to RVAR issues, such as disabled toilets, more luggage space, trolley points and so on.

Skipping over the oxymoron of ‘luggage space' in a Voyager or Pendolino, what about the requirement that passengers must not travel in the front third of driving vehicles of multiple units running at over 100mile/h?

‘Not us guv', says my chum. The HSE did not require this extended crumple zone. Rather, it asked the train builders and operators to demonstrate that a passenger in the leading vehicle of Class 180, Voyager or Pendolino was as safe as one in an HST at 125mile/h or an ordinary multiple unit at 100mile/h.

Thus, asked by HSE to demonstrate that the level of risk to which a passenger was exposed in the new trains was no greater than that of an old train, it was the suppliers' and the train operating companies' decisions to meet this in the form of extended crumple zones.

So, in general, HSE does not set detailed standards. It gives objectives to be achieved, but may then illustrate them with ‘specific ways of achieving those objectives'.

Disingenuous

Now to appreciate just how disingenuous this objective was, and the advanced sophistry of the builders and operators, we need to go into some physics. But fear not – it is reader friendly, thanks to Kurt Bang.

When a moving train hits something, the energy of movement – kinetic energy – has to be dissipated. It is the abrupt release of this energy which causes the damage.

Engineers measure energy in Joules which, unlike a foot or a pound, is not readily comprehended. Enter Herr Bang.

This Austrian wunderkind physicist, born in 1889 the year Joule died, emigrated to the United States with his parents around 1900. Already a professor at 19 he was looking for a measure of kinetic energy that would be more meaningful to his adopted country and came up with the unit that today bears his name.

A Bang is the kinetic energy of a mass of 1 ton moving at 1 mile/h. So a small car, say at 1 mile/h, has 1 Bang of kinetic energy to dissipate.

But kinetic energy is proportional to the mass multiplied by the velocity squared. So at walking pace – say 4mile/h, the same car's kinetic energy is 1 times four squared = 16 Bang.

Easy

I told you it was easy, so let's work out the kinetic energy of the HSE's two reference trains.

An HST weighs about 400 tons so at 125mile/h its kinetic energy is 6,250,000 Bang or 6.25 MegaBang (MBang).

Now consider a four car Voyager at 100mile/h. Say 50 tons per car, total kinetic energy 2 MBang.

Stir up the Cummins and accelerate to 125mile/h and although you are running only 25% faster, the speed squared factor means that the kinetic energy, which is what kills people in an accident, has increased by 56% to 3.15 MBang. And the leading vehicle, which hits first, has to absorb more than its own kinetic energy.

Now the Group Standard for crashworthiness requires 1 MegaJoule (MJ) of energy to be absorbed by a vehicle end. This is achieved using crumple zones and the maximum distance the end is allowed to crumple is 1 metre.

Faced with the HSE's requirement on maintaining the same level of risk at 125mile/h the train builders doubled, or in the case of Coradia and Pendolino, trebled, the energy which could be absorbed by the crumple zone of the cab ends. For example, an Alstom simulation showed that a Class 180 would survive a collision at up to 58.9km/h (36.5mile/h) without damage to the passenger compartment.

In addition it was agreed that passengers would be barred from the front third of the driving vehicles, filling the space with kitchens and train managers' offices. And in the remaining space, all seats would be back to engine.

HSE agreed that this met their requirement. At which point I upset a lot of people by describing the outcome as an intellectually dishonest piece of collusive self delusion by all concerned.

When I have made this gratuitously offensive point while researching this item, I have been reminded that the trains were being built and if the HSE had not been mollified the new stock would have not got a tick in the appropriate box and would not have entered service. So that's all right then?

Massive forces

Why am I so cross about this? All will become clear when I convert Bangs into Joules. A Bang is 101 Joules. So the kinetic energy of a four car DMU at 100mile/h is 200 MJ. Increase the speed to 125mile/h and the energy goes up to 315MJ.

So you can see that even trebling the energy absorption to 3 MJ, while it may offer a fig leaf for the disingenuous HSE, is on a par with an elastoplast on a traumatic amputation in terms of usefulness in a high speed collision.

What about the sterilisation of the front third of the end vehicles? Well, at Great Heck the derailed IC225, weighing 490 tonnes, with the Driving Van Trailer (DVT) leading was travelling at an estimated 88mile/h when it hit the oncoming freight train, total weight 1759 tonnes running at 54 mile/h. The kinetic energies wre 380 MJ for the IC225 and 510 MJ for the freight train.

This is how the accident report describes what happened to the DVT. ‘ As the DVT was the first vehicle to collide with the Class 66 locomotive it suffered massive structural damage, resulting in large sections of the vehicle becoming detached and dispersed over the crash site. The leading end cab structure was sheared off, and a 3 metre length of body section following this totally destroyed. The floor structure beneath the cab also became separated, as did the obstacle deflector. The trailing end vestibule structure (the vestibule is the area adjacent to the corridor connection) was also sheared off at the under-frame, flattened, folded over and impressed into the main central portion of flattened body shell'.

As the report concludes, ‘any crashworthiness features of the DVT were totally compromised by the masses and speeds of the vehicles involved in the collision'. On its own the DVT had around 35MJ of kinetic energy and had to dissipate all this plus a proportion of that of the coaches behind.

Laws of physics

So what price 2, 3 or even 5MJ or energy absorption in a 1 metre crumple zone at 125 mile/h? And as for the requirement that passengers in the new generation 125mile/h MUs should be as safe as in an IC125 – how could someone seriously ask for a reassurance which depended on the laws of physics being suspended.

In a HST you have 69 tonnes of power car in front of the leading coach, which is going to dissipate a lot of kinetic energy. The only way to meet the HSE's requirement in an MU with integrity would be to ballast up the leading vehicles and run them empty.

So when the HSE made its request, the engineers responsible for meeting it should have said ‘No, you are asking for the impossible' and showed them the simple maths. The next question would then be ‘are you banning 125 mile/h multiple units'.

Risk

Where do I stand, or rather, sit, on this? Well railwaymen have always preferred not to sit in the leading or trailing coach. But when I do, or when I footplate a high speed train, I don't give the possibility of a collision a second thought.

While, as Einstein said of quantum theory, God does not play dice', in high energy collisions what happens to the vehicles involved is largely down to chance. The more speed, the more energy and the more risk of a catastrophic outcome.

But that is the price of going faster. Equally, the risk of collision continues to reduce.

I suspect that a take off or landing accident in a Handley Page HP42 bi-plane airliner was more survivable than the same incident in a 737. But the increased risk comes from a six-fold cruising speed. It's called progress

Case History 2

More on Thetford

In the June issue this column covered another area where the railway must be positive and say ‘No' – the case brought by the Disability Rights Commission over access a Thetford station. The platforms are connected by a footbridge and for a wheelchair to travel between platforms requires a journey of about half a mile.

To meet the requirements of the Disability Discrimination Act, Central Trains proposed reinstating the barrow crossing, taken out in 1994, with the modern equivalent known as the Station Foot Crossing (SFC). Since the crossing is defined in Appendix 1 of Railway Group Standard GI/RT7011, it must be assumed that such crossings were intended to be introduced subject to a satisfactory safety assessment in line with the appropriate Group Standard (GI/RT 7611 <Ital>Provision, risk assessment and review of level crossings<Ital>

So Central Trains wrote off to Network Rail asking for formal confirmation that compliance with GI/RT7011 would allow an unsupervised Station Foot crossing to be installed at Thetford subject to the minimum requirements of the RGS.

Cannily the train operator also asked if not why not. Also whether, if the answer was in the affirmative, whether Network Rail would be willing to provide an SFC, either funded as a customer's reasonable requirement or through the SRA's Access for all scheme?

Sadly the answer was negative. But Network Rail would be willing to provide a fully accessible footbridge at Thetford on request from the station facility owner. This was a crafty wheeze since Thetford is Grade 2 listed and anyone who has seen a fully accessible footbridge will know that it would give English Heritage conniptions.

Protected

Since an underbridge or subway were equally impracticable Central pressed on with the SFC. Its Safety Compliance Manager carried out a risk assessment in line with GM/RT7611 and concluded that additional protection would be necessary because of the curvature of the line at that point.

Protection could be provided simply by installing a wicket gates at each side of the crossing with remote locking. This could be release by the signalman when a ‘call for assistance' button was pressed.

Back came Network Rail. After pointing out that the site of the crossing was over track and thus outside the station lease area, the official reply was, ‘We have given your suggestion careful consideration and regret that we are not prepared to agree to the installation on the grounds of safety as well as operational and on-going running costs.

Network Rail's counter suggestion was that since it had never seen a planning application for an accessible footbridge ‘perhaps this is an area that should be explored in the first instance. We certainly would be prepared to write letters of support for any well designed and robust application'.

Now since an accessible footbridge would not leave much change out of a million quid, and would have to be regularly inspected and painted, I reckon the ‘operational and on-going running costs' are a pretty flimsy excuse.

HSE view

As Stephen Williams, Head of Technical Division HMRI said in a letter to Forum last month, Mr John Tilly, HM Principal Inspector of Railways, visited Thetford for a site meeting in August 2003. And as Stephen said, Mr Tilly's letter to Central Trains did indeed say ‘The proposal for the crossing put forward by Central Trains is a reasonable proposal, however it misses several important points'.

But what Mr Williams somehow forgot to mention that this quotation came from Item 5 in the letter. And when you read what went beforew the sweetness and light evaporates.

Mr Tilly wrote:

‘I outline HM Railway Inspectorate's position and my understanding of our discussion.

1)HM Railway Inspectorate does not support the creation of new level crossings unless there are exceptional circumstances – this has been national policy for a considerable number of years.

2) Network Rail does not support the creation of new level crossings

3) Railway Inspectorate does not enforce disability legislation, although we take such matters into consideration when considering new works.

4 It is Central Trains responsibility to meet with disability legislation or show reasonable cause why they cannot do so. Cost is a relevant factor in their decisions.

And here are the ‘several important points' Central Trains missed.

a) ‘Effectively this would be creating a public right of way which in my opinion would require an order made under Section 1 of the Transport & Works Act.

b)Central train's proposal places additional workload on the signaller, not one of their employees, and places all the risks onto Network Rail .

C)The proposed ‘buzzer' system will lead almost certainly to ‘vandal activity' in the sense of ‘pushing the button and running away', etc.

d) No consideration of how one qualifies disabled – the vast majority of people will take the easy route which increases the workload even more.

So you can see that when Mr Williams says ‘This did not close the door on a formal discussion or a formal submission' he was putting a bit of a positive gloss on Mr Tilly's message which roughly went, ‘we won't wear it, Network Rail won't wear it, disability provision is nothing to do with us and here's another obstacle plus three more reasons why we won't wear it.' Slam!.

Angry

Now readers will remember that the angry man in wheelchair demanding an accessible taxi to change platforms was Mr Keith Ready. Through the Thetford & District Access Group he got a copy of Mr Tilly's letter and replied vigorously.

He began by pointing out that the heading on the letter was inaccurate. It read ‘Proposal to reinstate barrow crossing' which was not what was being sought. Next he pointed out that there was no explanation of why HMRI and Network Rail did not support crossings.

What was being proposed at Thetford was far safer than many existing crossings relying on traffic lights, let alone those which are crossed visually or by escort.

Having done his homework Mr Ready moved on to DSS (Distracted Signalman Syndrome). There are 856 crossings in the UK controlled by signalmen, including two at East Harling and Attleborough with the same train frequency as Thetford. Mr Ready failed to see that the workload at Thetford would be any greater than at the other stations on the line.

As for the buzzer problem, the obvious solution would be to use the Radar Key which all disabled people carry to get into dedicated facilities. I should add that this would not prevent other people, such as mums with buggies, from using the crossing; many of them have radar keys to provide access to baby changing facilities often found in disabled access toilets.

Useless

Of course, it gets worse. Keith sent the correspondence to local MP Gillian Shepherd who got the wrong end of the stick and slagged off Central Trains.

She also passed on the information to Transport Secretary Alistair Darling who passed it onto his chocolate firescreen aka Parliamentary Under Secretary of State Tony McNulty. He replied with a letter of staggering uselessness, full of statements of the self evident.

For example, Mr McNulty explained that HMRI, a division of the HSE, ‘is the body responsible for overseeing safety on the railways'. Well thanks for that clarification, Tony.

He also set out the positions of HMRI and the Strategic Rail Authority and concluded that disabled access was a matter for the station operator to pursue with HMRI and Network Rail. As a killer PS, he added that First Group had been appointed preferred bidder for Grater Anglia and that Mr Ready might wish to contact them directly with his aspirations for Thetford – so here's the address.

‘I hope this is helpful' he concluded. At which point I must stop because some readers get upset when I savage our elected representatives.

So how does this sorry saga fit into the new campaign? Well being positive is also about not taking ‘No' for an answer. Can I suggest that instead of harrying the railways – who really have been trying hard on disabled access for the last 30 years – the DRC should take HMRI to judicial review on the unreasonable refusal to consider an SFC Thetford for which there is specific provision in Railway Group Standards

Go for it Keith.

Case History 3

Selective Door Operation

It is obviously less than ideal for trains to be longer than platforms. But something being less than ideal does not make it unacceptable.

This is particularly true in a safety case based industry where the aim is to manage safety so that risk is As Low As Reasonably Practicable (ALARP). But, as we saw with the Thetford crossing, prejudice – ‘Level crossings are dangerous and kill people' - is antipathetic to ALARP - ‘we will manage the risk at this crossing because the benefits are worth the minor residual risk'.

At which point a five car Class 444 Desiro runs into Portchester station where the Platform is 4.5metres shorter than the train and ALARP goes out of the window. Let's start with the HSE's views on short platforms.

Risk

There is a risk, says HSE, in cases where the train is too long for the platform, of passengers leaving a part of the train not at the platform and falling on the track. Which is true.

However, while best practice is that trains and platforms should match, ‘HSE recognises that in some circumstances risks can be properly controlled by using Selective Door Opening (SDO) on trains where they are too long for the platform'.

SDO comes in two flavours, manual and automatic (MSDO/ASDO). In MSDO the guard opens only the doors with a platform outside. With ASDO the train knows which station it is at and inhibits the release of doors without a platform.

MSDO, says HSE, ‘still poses some risks in terms of human error'. It should ‘normally only be seen as an interim solution'. ‘HMRI will seek platform lengthening, or the introduction of ASDO over stated timescales, to replace MSDO solutions to capacity problems'

Not that even ASDO can be taken as read. ‘(It) may be accepted as a long term solution to short platform issues where appropriate and following a robust risk assessment to demonstrate that all risks arising from operation have been reduced as low as is reasonably practicable'. According to HSE, the automatic decision may still need confirmation from the train staff.

Cross-industry

In addition to HSE, an SRA Cross-Industry Working Group was set up to consider SDO. In April it published a Guidance Note on ‘managing the risk associated with the alighting of passengers at stations where the train length exceeds platform length'.

It's all good stuff for safety wonks, but cautions that solutions requiring expenditure will be subject to appraisal criteria agreed with the Secretary of State for Transport under the SRA's Directions and Guidance. Presumably this refers to the Value per Prevented Fatality established by the Department for Transport.

When it comes to making a decision, the Note says: ‘for a control measure to be worthy of implementation it must be reasonably practicable. <bold please>This judgement is for the duty holder to judge and defend<end bold>. As a guide the overall cost of the measure (eg accounting for capital, performance, maintenance costs etc) should not be in gross disproportion to the value of the safety benefit'. Readers should hold this last sentence in mind.

Cold numbers

Now in all this paperwork no mention is made of the scale of risk being addressed. But from the concern expressed by HSE and the Guidance Note's suggestion that complete closure of the station is a valid option, you might expect that a real and present danger is involved. So, as this column always does, let's start with the cold numbers – the RSSB Risk Model

Table 1

Extract of Table 8 of the Risk Profile Bulletin Issue 3 February 2003

Source RSSB

You will note that the Model refers to equivalent fatalities per year. This is to bring injuries into the equation, with 10 major injuries and 200 minor injuries equating to a fatality.

Now in the risk model Hazardous Events list the one we are interested in is HEM-09 ‘Passenger fall/injured while boarding or alighting train' . According to Railway Safety & Standards Board (RSSB), HEM 07 ‘ Passenger falls out of train onto track at station' refers to people opening a door on the wrong side of the train and stepping onto the track

Note that getting ride of Mk 1 stock reduces the HEM-07 toll by three quarters.

Now within HEM-09 there are seven possible events. These are boarding/alighting stationary train, ditto moving train, ditto again under the influence of alcohol and, finally falls onto track at station. It is this latter which we need consider when it comes to SDO.

There was one such incident in 2003/04 and the risk per passenger journey given by the Model is 1.079 x 10 -9 , as near as dammit one in one billion. As we know the railways carried one billion passenger last year so the risk model and reality align.

One in a billion

So now we know the scale of the risk that causes HSE such concern that the possibility of human error in MSDO can be tolerated only as an interim solution and the SRA sends out an eight page guidance note on how to deal with it.

How does 1.079 x 10 -9 relate to the real world ? Well at 12 million passenger journeys a year on the Portsmouth Line we are looking at an event once every 49 years even with some slam door stock still in service. And I forgot to mention that some eight car Mk 1 formations stop at Portchester.

Which station serves 229,000 passengers a year. Now assume that they all arrive at the station, the risk of someone stepping out beyond the platform end and falling is once every 4,700 years.

Table 2

Analysis of falls at stations

Source: RSSB Risk Profile Bulletin

This information is summarised in Table 2. And make sure you read the right hand column.

If the risk is expressed in annual equivalent fatalities a year – where one serious injury equates to one tenth of a fatality and one slight injury equates to one two hundredth of a fatality, the annual risk is one eightieth equivalent fatalities a year.

Thus in the rational world of ALARP, where the VPF is currently £1.36million, it is worth spending £17,000 nationally on SDO to eliminate the risk of falls at short platforms.

Clearly it is time to be positive and say ‘no'. The risk at Portchester is so slight that the added risk of human error in MDSO is irrelevant.

Managing the risk is equally straightforward. Obviously guards should be briefed to warn passengers before a station stop. You could put up notices on the affected ramp saying ‘Don't get out here'. Or guards could even work the doors from the affected coach. Or you could put stickers on the end doors with a list of stations where they are not to be used.

Hassle

All it needs is for someone to mount the ALARP challenge. The sad news is that SWT thinks it less hassle for its passengers to go along with stupidity. Apparently Network Rail's original £750,000 cost of extending the platform at Portchester has come down and the SRA may be able to fund the work.

Now obviously the right thing is to have platforms the right length. But when the industry is under extreme financial pressure – dependent on the taxpayer for 60% of its income, spending money to remove a negligible risk can surely wait.

Oops, I nearly forgot. The ASDO for South Central's Electrostars uses the Global Positioning System (GPS) to inhibit the doors at short platforms.

According to Informed Sources, one failure mode in the safety case justifying the use of ASDO was the possibility of a meteorite knocking out one of the satellites, which the GPS uses to triangulate its position, as the train was pulling into a short platform. Were they pulling my leg or have we the ultimate example of endless means to meaningless ends?

Continues.........

Return to Archive -by date - by topic.