Airport organisation

There's a wind of change blowing through the air transport industry. To the outsider it may appear that only the regulatory structures have been altered. But it goes much further - right to the heart of making sure that planes are safe to fly.

Maintenance is critical and there are reams of regulations that have to be followed to ensure the right maintenance is done in the right way at the right time. In 2003 the EU set up the European Aviation Safety Agency to promote the highest common standards of safety and environmental protection. Oversight within the UK is still carried out by the Safety Regulation Group (SRG) of the UK’s Civil Aviation Authority (CAA).

Research is about to start at Cranfield University to assist the oversight process. It will look into the virtues and practicalities of using risk assessment to regulate maintenance, rather than the current approach of imposing a schedule of inspections. ‘This is all happening in the light of the Hampton Review,’ said Dr Simon Place of Cranfield's Safety and Accident Investigation Centre.

The Hampton Review was published by the government during the budget of 2005. It was designed to promote ‘more efficient regulatory inspection and enforcement, without compromising regulatory standards or outcomes’.

So everyone, from the engineer who has an EASA Part 66 licence to work in aircraft maintenance, to the companies with an EASA Part 145 licence for maintenance, repairs and overhauls (MRO) and the airlines with the Air Operators Certificates will probably be affected by the findings of the Hampton Review sooner or later.

‘I have found much that is good, and some excellent, innovative practice. However, the system as a whole is uncoordinated and good practice is not uniform,’ said Philip Hampton about regulation in general across the UK, ‘There are overlaps in regulators' responsibilities and enforcement activities. There are too many forms and too many duplicated information requests.’ That last sentence will strike a chord with every engineering business.

Hampton's solution will also have been appreciated by engineers. ‘Risk assessment – though widely recognised as fundamental to effectiveness – is not implemented as thoroughly and comprehensively as it should be. Risk assessment should be comprehensive, and should be the basis for all regulators' enforcement programmes. Proper analysis of risk directs regulators' efforts at areas where it is most needed, and should enable them to reduce the administrative burden of regulation, while maintaining or even improving regulatory outcomes.’

Dr Place has responded to this clarion call on behalf of the aviation sector by starting the Cranfield project to see what part risk assessment can play in regulating aircraft maintenance. ‘There has been some early work, from the Idaho National Laboratory in the US, in the late 1990s,’ said Place, ‘And there are some unpublished university research from Kingston and Bristol in the UK.’ Now he wants to hire a research engineer for three years to help develop the methods for quantifying the likelihood of maintenance error and the inherent risk, depending on the severity of the failure condition. Such methods may include novel applications of Fault Tree Analysis, Petri nets, or take a different approach based on Probabilistic Risk Assessment (PRA), using Bayesian Belief Networks (BBN) for example. The latter features a causal model with components to model the uncertainty within the model, using either data or the opinion of “experts”.

‘Such models offer the opportunity of estimating the risk of the various maintenance processes, so that control measures may be applied,’ said Place. He draws a parallel with other analyses used by industry to assess how well they are doing. ‘We talk about KPIs (key performance indicators) for business. Are there safety KPIs that can talk about the health and safety of an organisation’s maintenance process on a particular aircraft fleet?’

There has been a lot of work done into maintenance human factors in the UK, USA and Australia (notably by Reason and Hobbs), some of which stems from by the nuclear industry, which has tried to put a numerical value on human errors. The Cranfield project will feed into that strand although Place believes it will have broader application. ‘Human error probability would be hard to apply directly to aviation because there are so many different issues that have to be taken into consideration, such as training, aircraft types and so on,’ said Place.

‘We want to develop new knowledge,’ he said, ‘There's a change in the regulatory environment and risk assessment could lead to streamlined and better regulation of aircraft maintenance.’

The Hampton Review recommended that comprehensive risk assessment should be the foundation of all regulators' enforcement programmes. ‘There should be no inspection without a reason, and data requirements for less risky businesses should be lower than for riskier businesses,’ it said. Considering the scale, value and complexity of the aviation sector, if the Cranfield research leads to better regulation, safety and economic viability should get an enormous boost.

Max Glaskin