Sharing best practice: synergy in rail and aviation safety

Citation

(2001), "Sharing best practice: synergy in rail and aviation safety", Aircraft Engineering and Aerospace Technology, Vol. 73 No. 1. https://doi.org/10.1108/aeat.2001.12773aac.001

Sharing best practice: synergy in rail and aviation safety

Sharing best practice: synergy in rail and aviation safety

Keywords: Conferences, Rail transport, Air transport, Safety

Held at the Royal Aeronautical Society this conference focused on common issues in the two industries. Unfortunately, for a variety of reasons, some of the information intended was not available. Nevertheless, sufficient topics were aired to give a good understanding of the present issues as well as looking to the future.

The scene was set by the first paper by Richard Morris, the Technical Director of Shadow Strategic Rail Authority. He noted that 80 per cent of accidents can be attributed to human factors. Even so, if any group of people was asked whether they would travel a distance on an aircraft without a pilot or a high speed train without a driver, there would be some nervousness. So, on the one hand, there is the recognition that human beings make mistakes and, on the other, that people feel a lot safer if there is a human being "at the front".

One or two actions are suggested that could be taken in order to increase the level of safety as far as human factors are concerned. Recruitment is the first issue. For example, Eurotunnel decided that they were looking for people who displayed: common sense; the ability to debate the various standards and rules by which the operation would be worked and to contribute to their development; and, in this particular case, bilingual (French and English) speakers. As a result the control staff came from various backgrounds and the control centre was an extremely professional outfit.

On the job training is of great importance and the railway industry is catching up on the airline industry in the use of simulation. As a result of incidents and investigations, Eurotunnel decided to build two simulators, one for the Railway Control Centre itself and the other a simulation of the Shuttle and Tunnel in order that emergency exercises could be practised easily and that training could be given safely.

As an offshoot from the introduction of the simulators, the staff became aware of the effort that management were making, and became more motivated and more dedicated. Another follow-up activity was a visit to the nuclear submarine base at Plymouth where the importance of "drill, drill, and drill" was emphasised. As a result Eurotunnel introduced a system of task observation. Certification was also slightly changed as a result of this visit.

In safety terms it is vital that an open and transparent culture exists, which means that all staff will feel that they are able, without threat, to tell of a mistake or error exactly as it happened. An example from Eurotunnel was a safety error which, as a result of its admission, enabled the rules to be altered.

Career progression is of importance, with three types of people apparent working within an organisation. The first is quite content with the job that they are doing and wishes to stay there; the second is the one who wishes to get on, who is upward, thrusting, and wants more challenges; the third is the one who does not really know whether he wants to get on or not. These types should be recognised and regular dialogues between the manager and the employee conducted.

Transferring human factors experience

From the University of Texas, Professor Robert L. Helmrich detailed the nature of teamwork and error in situations where teams interact with technology. Errors, both by individuals and by teams, have roots in human limitations. Three factors have helped aviation deal with threat and error. The first was the development of formal training programmes known as crew resource management (CRM) to address the interpersonal aspects of flight operations. The second was the collection and analysis of data to provide an accurate picture of the strengths and weaknesses of organisations and the system. The third was to develop a safety culture addressing the systemic sources of threat and error in the system and in organisation and their cultures. From the observation of errors in normal operations, a typology of team error has been developed that classifies all the errors seen. Five categories are evident (see Figure 1). They are:

1. 1.

   Procedural – following procedures with wrong execution.

2. 2.

   Communications – missing or incorrect information or misinterpretation.

3. 3.

   Proficiency – error due to lack of knowledge or skill.

4. 4.

   Decision – discretionary crew decisions that unnecessarily increase risk.

5. 5.

   Violations – intentional non-compliance with procedures or rules.

Figure 1 Percentage of each type of error and percentage classified as consequential

The most prevalent type of error is a procedural violation, although proficiency and decision errors, although relatively infrequent, are most likely to be consequential, by which is meant likely to place the aircraft at risk.

Investigations of aviation accidents have shown poor organisational cultures to be precursors of disaster. Two accidents are quoted, one indicating the non-relay of a significant message and, in the other, the investigation found that the crew committed four errors, one procedural and three decision, and there were also errors by flight despatch in the flight plan and in launching the flight with maintenance problems.

The University of Texas has developed a conceptual model of threat and error management based on empirical findings. Organisations use the model as a guide for the analysis of accidents and incidents with the goal of detecting latent threats and assessing the effectiveness of countermeasures designed to avoid error and manage threat, error and undesired aircraft states.

Examination of human factors problems in medicine reveals that they have much in common with aviation. The results support the view that similar approaches may be effective for enhancing performance and error management. Despite risks existing in both the medical and the aviation spheres medicine should be able to use the aviation experience to reach the same goals more rapidly with fewer missteps.

Passenger evacuation and transference

The issue of passenger safety has always been a major priority within the aviation and rail industries and Professor Helen Muir of Cranfield University dealt with the many factors present. These can be identified and information obtained from accidents suggests that fire and smoke are the most serious environmental factors affecting behaviour in accidents. They have the potential to limit the number of available exits. The blockages ensuing greatly decreased the efficiency of the evacuation.

In an accident, behavioural responses can include fear, anxiety, disorientation, depersonalisation, panic, behavioural inaction and affiliative behaviour. Fear is a primary response when survival is threatened. Anxiety is a response experienced by the majority of passengers in an emergency situation. Disorientation can be experienced as a result of reduction in visibility, etc. As far as depersonalisation is concerned, people who have encountered life-threatening events often say that the passage of time slows while mental activity increases. Panic may be defined as uncontrollable and irrational behaviour. It appears that there are many more instances of behavioural inaction than of panic. Affiliative behaviour involves movement towards the familiar, e.g. taking hand luggage. Focused attention and enhanced physical performance are features which may also be present.

Research into behavioural factors includes a greater understanding of the differences in orderly and disorderly evacuations. It suggests that one of the primary reasons must rest with the individual motivation of the passenger. A number of research programmes have focused on this issue. It is natural in an emergency to act competitively in order to survive, and a technique used in research is to offer an incentive payment to subjects to influence motivation and performance.

A test to investigate passenger evacuation in an Mk III rolling stock was conducted some years ago. The results of a 100 per cent loaded coach and overloaded conditions both in competitive and in non-competitive trials identified various deficiencies in the internal design of the coach.

In aviation a Trident aircraft was used parked on an airfield. In the succeeding evacuation, the behaviour and evacuation times of the volunteers were recorded using video cameras mounted inside and around the exits from the cabin. In the competitive evacuations statistical treatment of the data indicated that, as the aperture between the bulkheads was increased, the evacuation rate increased, leading to a reduction in the time for the 30 individuals to evacuate the aircraft. In the non-competitive evacuations the means suggested that increasing the width of the aperture through the bulkhead leads to a small reduction in the evacuation times. The data from the competitive evacuations through the Type III hatch indicated that changes to the seating configuration had a significant effect on the mean evacuation times.

The conclusions include the fact that in the design of future passenger aircraft or rolling stock it is important that the behaviour which can be exhibited by passengers, in the event of an accident, is taken into consideration. Research has demonstrated that it is possible to conduct tests on passenger aircraft or rolling stock to evaluate the ease of evacuation.

CFIT and altitude bust

From Capt. Peter Griffiths of the UK CAA Safety Regulation Group was a paper on how do we change behaviour for CFIT or altitude bust? Two typical controlled flight into terrain (CFIT) accidents are examined and also altitude busts in the same manner.

CFIT has been the cause of major fatal accidents in jet and large turbo-prop aircraft. The causal factors have generally been due to human error rather than technical failure. Considering the Boeing 757 accident at Cali (Colombia); in this accident the crew had lost their situational awareness and flew into the mountain side. They appear to have become reliant on the map display and followed the guidance on the FMC, even though this was, with the benefit of hindsight and sufficient time to analyse the situation, clearly the wrong course of action.

What led the crew to take this course of action? The initial problem was one of workload management. The crew had not been briefed for that approach procedure and the change of clearance they received added to the workload. When things became rushed they chose to follow the demands of the FMC rather than rely on the basic instrument presentation and their own mental picture of where the airport should be. As things began to go wrong they failed to take positive action to correct the situation until it was too late. In the Airbus A300 crash at Kathmandu, Nepal, the aircraft was carrying out a non-precision approach. The crew were following the correct horizontal profile but were 1,000 feet lower than the correct vertical profile and had probably misread the charts. A non-precision approach is one where there is no glide slope guidance and the aircraft descends according to a distance or time from a specified point. In this accident the crew had become confused about the vertical profile of the aircraft and were descending in accordance with profile but at the wrong stage. Again, workload management plays a large part in this type of approach, which is quite demanding to fly.

A further training scenario has since been devised and technology has also moved on one stage further and the latest version of ground proximity warning system (GPWS) is the enhanced GPWS. There has been a trend away from non-precision approaches but many still do exist. Again, technology has brought about an improvement in the form of RNAV approaches. The integrity of FMC data is absolutely essential in these cases, however.

An altitude bust is where an aircraft deviates from its cleared level by more than 300 feet. Fortunately, altitude busts rarely result in a mid-air collision but there have been some near-misses and the potential for disaster is all too real. Workload management is a major factor, because altitude busts usually take place in conditions of high workload such as during climb or descent and in busy airspace including terminal areas. Technology is helping in that the aircraft automatics can be used to fly the aircraft and thus reduce workload in busy airspace. CRM training is being further improved with a system of CRM instructor accreditation.

Public perception of safety

Given by Simon Montague, Transport Correspondent, BBC News, this paper noted that public perceptions of safety are influenced by a range of factors; factual safety records, reporting of accidents, and commercial, regulatory and political activities. The public expects a high level of safety when travelling by rail or air or any other form of transport provided by a commercial or other organisation. Rail and air accidents are rare, but they can mean very substantial loss of life and massive media attention. The reality of safety records can be very different from the public perception, but perception is more important in business terms.

Commercial aviation appears to have a better safety record. Between 1990 and 1998, there were only two years when any passengers were killed in a UK airline or air taxi, fixed wing aircraft in UK airspace. During the same period on the railways, total passenger deaths averaged more than 20 a year. Safety reporting is widely misunderstood. Public perception of the risk from "signals passed at danger" and "airprox" incidents is poor. Other risks such as fatigue and "level busts" are little understood. Reaction to accidents is important – management response, corporate openness and safety actions will affect public perception. Survivor groups are becoming increasingly influential. Perceptions of rail safety are crucially influenced by social regulatory and political factors. Perceptions of air safety are less influenced by external factors.

Recent moves or promises by political parties to shake up rail and air safety regulation in turn create perceptions of what may be deficient about existing systems. Watchdogs also influence perception. Rail and air transport are not necessarily comparable modes, and numerous factors mean that safety levels and safety provision cannot be directly compared. But some lessons, and awareness of external factors, could be shared.