Mr A Maginness (Chairperson)
Mr McFarland (Deputy Chairperson)
Mr Byrne
Mr R Hutchinson
Mr Hay
Mr M Robinson
Mr Savage

Mr Rick Eagar ) Department for Regional Development ) / Arthur D Little Consultants

I am pleased to welcome Mr Rick Eagar, a director at Arthur D Little Consultants. Mr Eagar was involved in compiling Arthur D Little's report on railway safety in Northern Ireland in 2000.

It is a pleasure to give evidence on an important piece of legislation governing rail safety in Northern Ireland. It may be useful to say a few words about my background. I have 22 years' industrial experience, 12 of them being spent in safety and risk-management work with Arthur D Little, an international technology and management consultancy. Well over half of my work is in the railway industry, but I also work on safety in such areas as the oil, gas, nuclear and chemical industries.

I also represent the expertise of my company. About half of its work is on rail safety and risk in the rail sector - not just in the UK but abroad, particularly in Italy, Switzerland and Hong Kong. The company worked on the prototype safety case in Great Britain in the early 1990s.

Since 1999 the company has advised Northern Ireland Railways (NIR) and the Northern Ireland Railways Transport Holding Company on safety issues. That began with the strategic safety review of NIR in 1999, which examined railway safety from top to bottom and resulted in the improvement programme now being implemented.

One of the 120 recommendations in that report was that, in the light of what we felt was a sparse regulatory framework for rail safety in Northern Ireland, NIR should lobby for more rail-specific safety legislation. The recommendation also suggested that, in the meantime, NIR should comply with principles of rail safety legislation developed in Great Britain in so far as those are applicable, given the scale and complexity of Northern Ireland's railway network. We believed that it was important, since it would provide the necessary backbone to underpin the safety improvement efforts of the railway. That is where legislation was mentioned in the report.

Assuming that implementation is timely and appropriate, we consider that the draft Bill, together with its secondary legislation, will have a beneficial and positive effect on safety. We feel that it will address the concern raised in the original report.

The safety-case approach was highlighted in the minutes of the previous evidence session. Fundamentally, it requires the duty-holder to conduct a systematic identification and assessment of safety risks and develop suitable control measures to address them. That assessment is documented, and that document is used as a means to provide acceptance or certification from the regulator. It is also used as a basis against which compliance can be audited.

I have four points to make about the safety-case approach's appropriateness for Northern Ireland Railways. Firstly, as has been said, safety cases are used across all the high-hazard industries, such as the railways, offshore oil and gas facilities, the chemicals industry and nuclear installations. That is part of a coherent "permissioning" regime operated for health and safety reasons in the UK. It is important to realise that that is based on the fundamental philosophy that safety is managed best through proactive measures. For example, safety is best managed if companies anticipate possible risks instead of merely complying with rules and legislation. In that sense, we can conclude that, if railways are considered a high-hazard industry - which I feel is a reasonable assumption - it is consistent to use a safety-case approach for them based on that used in other industries.

The second point relates to the direction in which legislation is moving in the European Union. While my understanding is that the interoperability Directives will not apply to Northern Ireland, it would be worthwhile considering the draft European railway safety Directive and its implications for the countries to which it will apply. The legislation will require countries to have a national railway safety authority. It will require railways to submit annual safety reports, and the authority will provide certification to those companies on that basis. In addition, the legislation is considering common safety methodologies and providing guidance on what those should include. The current guidance specifically includes risk management and the use of a risk-based system. It also includes risk-based targets that will require railway companies to conduct risk assessments. For the companies to secure certification, they will be required to demonstrate that they have assessed risks adequately and have the necessary controls in place. Although those measures may not be called a safety case, they are the fundamentals of one.

I shall give some examples. In Italy, there is already a system whereby long-term certification is provided on the basis of a satisfactory demonstration that risks have been assessed and that controls are in place. Again, that is analogous to the safety-case approach. As you know, the proposals in the Republic of Ireland are also to adopt a safety-case approach, although in that case with a railway safety authority. That would be more appropriate considering the scale of its railways. My point is that the safety-case approach is not unique to Great Britain; it is increasingly being used abroad.

Thirdly, the point was made that the safety-case approach is only relevant if you have a fragmented railway, which is fortunately not the case in Northern Ireland. That is not true. To some extent the railway safety-case regulations became associated with fragmentation, since they were introduced at the time of privatisation and were prompted by it. However - and this is completely hypothetical - even if that fragmentation had not taken place, we should have seen something similar to it in the railways anyway, for it is consistent with "permissioning" regimes in other industries.

My fourth point is that, like everything else, a safety-case approach is not a panacea. It has drawbacks as well as advantages. As with any new piece of legislation, there have been problems over the years with safety cases. In particular, there was much over-quantification of risk in the early stages, which led to a spurious complexity without any real meaning. To a degree there was a problem with the safety case becoming a paper exercise and not really representing what happened in the railways.

The two advantages we now have in Northern Ireland are that we can learn from those lessons and that Northern Ireland Railways is already doing virtually everything which would be required by a safety case. In that respect we are not imposing anything completely new on it. Those were the main points I wished to make by way of introduction.

Thank you for coming. My basic question is a little tongue-in-cheek - does it work? When I look at railways in the rest of the United Kingdom, I need proof that their system is working. I cannot see that happening, for at the moment it is horrendous there.

The evidence we have seen is that it works. There is a - probably incorrect - perception that railway safety has deteriorated significantly in Great Britain over the years. While I am the first person to agree that you can prove anything with statistics, the figures do not bear that out. Let us compare the most basic measure - fatalities per year. In the early 1990s, we were having 80 to 90 fatalities on Britain's railways each year. Even with the recent tragic accidents, that figure is lower than it was. It is currently around 40, and it was 65 in 1999-2000. Even with a 50% increase in traffic and the additional complications caused by the fragmentation of the railways, the figure is lower. Most people would accept that fragmentation has made railway safety more complicated to manage.

Are you saying that some train companies are more compliant with safety standards than others?

Inevitably some companies perform better than others.

Your points were in reply to Mr Rayner's evidence. He focused on the fact that the safety-case system was not used in mainland Europe or the Republic, but you have said that it would be introduced in both.

The railway here is small, and the logic for introducing this legislation is that at some stage in the future you might wish to sell it. If that happens, systems will have to be in place to ensure that the contractors do what they are told to do. At present, if in-house contractors are employed, the legislation, in theory, is not required because the contractors belong to you. If it is decided not to sell the railways but to contract out work, what would be wrong with having a contract that states that the contractor must maintain the railway for the next five years? That would set out the parameters within which they had to operate and keep some control over the subcontracting. Control of subcontracting, or lack of it, seems to have caused the problems in England.

Mr Rayner said that this has become a beanfeast for lawyers. The evidence of safety cases is that, if it does go wrong, lots of lawyers make lots of money. Why should we go in this direction, given that the rail system is small and that, at the moment, I think, there are no plans to sell it?

We must distinguish between a contractor safety case and a railway safety case. The proposed legislation is concerned with the requirement on the duty holder, Northern Ireland Railways (NIR), to prepare and submit a safety case for all its operations. The GB railway safety-case regulations do not say anything specifically about the way in which contractors must be managed. They are concerned only with the duty holder. NIR will have its own safety management system that will include appropriate measures for managing contractor safety, and there are many different approaches to that. If, like NIR, a company is running a risk-based safety management system, the most logical approach to take to managing contractor safety is to ask the contractor to identify the risks associated with his activities and to satisfy you, as the client and the contract holder, that it has the right controls in place, such as safe systems of work, method statements and so on. However, that has nothing to do with the railway safety-case regulations.

It is fair to say that the GB situation is a beanfeast for lawyers. That would not be the case here because there is not the same split of responsibilities, at least with the infrastructure controller and the operating and rolling stock companies et cetera. Clearly there is an interface if you contract out, but it is a matter of having suitable and appropriate measures to control safety across it.

A holding company in two parts runs this, and the job of Translink and NIR is to manage it, which is what we are talking about here. Why can they not just manage it without having an entire Bill? Parts of the Safety Bill are important, but why can the company not just get on and manage - which is what it is paid to do - without all these regulations? I could understand it if the railway was large and split up, but what is the imperative behind pushing this legislation rather than saying to this small railway company: "Get on and manage it"?

One imperative is to bring Northern Ireland's railway legislation into line with current legislation elsewhere, or legislation that will soon exist if it does not already. It is a matter of harmonisation. Also, by introducing any legislation, essentially you are saying that you want a higher degree of assurance that things are going to be done correctly - that is why there is any legislation, and safety-case legislation is just part of that. It is not going to impose a huge complex burden. Indeed, the secret of ensuring that it works and is effective is that it should not be overly complex but fit for the purpose. To reflect the scale of the railway, it should also be relatively simple.

Policing this arose at our earlier meetings. We know that we can have a good piece of legislation in place, but, as in England, there can be a problem with policing it. What do you feel about that?

By policing do you mean the role of an authority such as HMRI (Her Majesty's Railway Inspectorate) ?

Yes, and the role of the Department. Where does all that fit in with what we are trying to do?

There is an additional role here, which is an implication of wanting more assurance. The set-up envisaged is to use HMRI to provide the specialist expertise to do that and the manpower to police it. I know that there are, and have been, problems with resources for HMRI, but I have also seen in the minutes of the previous evidence some of the measures that the Department is trying to introduce through contract with HMRI to make sure that this does not adversely affect the system altogether. Given the scale of the railway here, a new authority would be inappropriate - that would be overkill, and so I have some sympathy for the proposed solution.

With our small railway system there has been a consistent pattern to the fatalities over the last 10 or 15 years. Most seem to result from people crossing or trespassing on the lines at level crossings. Will the Bill only address the resource-management system for ensuring better safety, or will it address the resource implications also?

The safety case part of the Bill could have an impact on resources, because part of the safety case will include details on the resources that have been provided by NIR. However, that would be an indirect impact. Level-crossing safety is a key problem, and lots of work is being done on it. Although it will include level-crossing safety, the safety case will be general, covering all aspects. I am not sure whether that answers the question. The Bill will not impose resource requirements on NIR directly.

Will the Bill specify safety standards which will mean that resources will have to be put into signalling or modern level crossing barriers, for example?

The safety case will include a description of the safety standards that the railway complies with, and by accepting that safety case, one accepts that those standards are correct. In that sense, the Bill will cover standards.

What is the lifespan of a locomotive, and is that gauged it by the hours it has operated or its age?

I am not an expert on rolling stock, but the lifespan of a piece of rolling stock is approximately 30 years. However, that is flexible and can be extended considerably by refurbishment, for example. It is judged to be obsolete when its structural integrity or crashworthiness begins to fail or when it no longer complies with accepted international standards.

In other words, the better a locomotive is looked after, the longer it will last.

Yes. That is true of anything.

It has been said that the Bill is piece of enabling, rather than prescriptive, legislation. Will you comment on that?

I cannot comment on that, because I am not qualified to give opinions on legal matters. As a safety adviser to the railway, my only concern is that any process that is adopted should not cause undue delay. If the process can introduce secondary legislation rapidly, it will be satisfactory.

There are two separate approaches. However, you are happy if the safety matters are expedited?

Yes.

There are no more questions. Thank you, Mr Eagar.

10 April 2002 / Menu / 24 April 2002