Auditory icons caricatures of everyday sounds have the potential to convey information by non-verbal means quickly and accurately. Two experiments investigated the application of auditory icons as warning signals to the civil aviation cockpit environment. Warning signals that are iconic and that stand in a direct relation to the event being signalled, such as the sound of coughing to signal the presence of carbon monoxide, should convey information about the nature of the critical event as well as alerting the operator that there is a problem. By contrast, signals that are arbitrarily associated with an event, such as a beep to signal the presence of carbon monoxide, provide little information about the nature of the event. Speed and accuracy of recognition in response to these different types of warnings may also be influenced by modality (visual, auditory, auditory + visual) and by task demand (low, high). Experiment 1 investigated effects of signal iconicity (iconic, abstract), modality, and task demand on warning recognition speed and accuracy. One-hundred and seventy-eight participants completed a computer-based training session and test task that involved responding to warnings associated with nine critical events while completing low- and high-demand concurrent tasks. As hypothesized, fewer training trials were required to learn iconic warnings compared with abstract warnings. During the test phase, the effect of iconicity, as hypothesized, was influenced by modality and task demand. Bimodal (auditory + visual) warnings were recognized with the greatest consistency and accuracy. Auditory abstract warnings elicited slow reaction times and poor accuracy. Auditory iconic warnings, under conditions of high demand, evoked levels of accuracy comparable with bimodal warnings. Experiment 2 investigated recognition speed and accuracy in response to four auditory iconic and four abstract warnings in an Advanced Aviation Training Device. As hypothesized, accuracy was greater in response to auditory iconic than abstract warnings and recognition accuracy and reaction time were unaffected by level of flying experience. Reaction times in the Advanced Aviation Training Device were approximately 1 second. These initial experiments suggest that there is potential for the use of auditory iconic warnings and bimodal warnings as the means, not only to alert, but also inform pilots about the nature of a critical incident.

Following the grounding of large numbers of piston-engine aircraft across eastern Australia in early January 2000 as a consequence of using contaminated aviation gasoline (Avgas), the Australian Transport Safety Bureau initiated a major safety deficiency investigation into the circumstances of the contamination. Guidance for the investigation was subsequently provided in the form of Terms of Reference, which stated that the investigation was being widened to examine the following:

1. the existing standards for aviation gasoline;
2. the details of risk analyses undertaken prior to and during the production of aviation gasoline at Mobil's Altona refinery;
3. the adequacy of the production control, distribution control, and recording processes used by Mobil and other refiners;
4. the current arrangements for the oversight of aviation gasoline quality, including the procedures followed by Mobil and other refiners to disclose information with potential aviation safety implications; and
5. any other matter of material relevance to the above.

The ATSB investigation team identified a number of factors related to the manufacture, standards and oversight of Avgas that contributed to the contamination, which are outlined below. The relevance of these factors was also considered in relation to the manufacture of other aviation fuels.

The fitness for purpose of aviation fuels is safety critical, however the systems of manufacture, distribution, supply and use in aircraft were not supported by all the defences that are normally incorporated into other safety critical aviation systems. Despite the safety risk, there were no significant redundant systems to enhance the defences for aviation fuel quality.

The deficiencies that have been identified in relation to the supply of Avgas that was not fit for purpose also have the potential to affect the fitness for purpose of other aviation fuels, like Jet A-1. Aircraft that use Avgas are normally small compared with civilian airliners which normally use aviation turbine fuel such as Jet A-1. If a similar contamination of Jet A-1 had led to similar deficiencies in engine reliability, then the potential for a major accident with large loss of life would have been significant.

A temporary variation in the production process in Mobil's Altona refinery in late 1999 led to an increased dosage of an anti-corrosion chemical being injected into the Avgas process stream, which led to a contamination of Avgas. The anti-corrosion chemical, Neutramine D, contained an active ingredient called ethylene diamine. Ethylene diamine was not completely extracted during Avgas manufacture. Excess ethylene diamine from the injected Neutramine D was expected to be extracted from the process stream in water taken from the deisobutaniser tower during manufacture, however the extraction mechanism was not fully effective. The concentration of ethylene diamine in the final product was small and none of the many quality assurance and specification tests used during manufacture and distribution identified the presence of the ethylene diamine in the final product.

The refiner's knowledge of the process within the alkylation unit was not complete. The manufacturing process for Avgas is very complex, and there are many variables and factors that can affect the process. A lot of information was available to the operating team at the refinery, however not all the activities were fully recorded and available for future reference.

Mobil did not define or clearly document procedures for managing process deviations outside some of the limits for normal operations within the alkylation unit. The refiner aimed to operate the plant within predefined parameters, to effectively control the process and maximise its efficiency. The parameter deviations at which the alkylation unit would be considered to be outside normal operations were not clearly defined in all cases, nor were the initial considerations or actions to be taken in such circumstances clearly laid out.

The processes for monitoring the reliability of plant equipment did not provide the best possible indication of reliability. A number of systems were used for predicting and managing the reliability of various components in the alkylation unit. Some of these systems could have been used more effectively to predict reliability. Systems to assess the adequacy of the reliability prediction systems were also not completely effective.

Management of change at the refinery did not consider the effectiveness of the extraction mechanism for ethylene diamine from the Avgas process stream. Changes within the refinery that might have indicated a variation in the properties of the process stream, and therefore might have influenced the efficiency of the extraction mechanism for ethylene diamine included:

• a decrease in the efficiency of the caustic wash system due to problems with caustic circulation pumps and a system leak; and
• concentrations of sulfates and pH in water from the deisobutaniser tower overheads that were outside their normal ranges, indicating increased acid and alkyl sulfate carryover from the alkylation reactor.

These changes were not considered in the context of their potential to affect the ability of the system to ensure that any ethylene diamine that was injected into the process stream would be effectively extracted.

Mobil did not have an effective process in place to identify the adverse consequences of the cumulative effects of multiple planned and unplanned process changes on the degree of control in the alkylation unit. A number of planned and unplanned changes were taking place in the alkylation unit at the time of the contamination event. Any one of the changes could be effectively managed, however the effect of one change on another change would decrease the ability to manage the potential cumulative effect of all the changes, so that the degree of knowledge, and ability to control the unit to the same level of accuracy would be degraded.

The refiner's procedures were not effective in ensuring that decisions were fully implemented, or that progress with recommendations was regularly reported and reviewed. Following a previous contamination event, a number of recommendations and improvement actions were identified. They were not all acted on and followed through to completion.

The refiner's risk management process considered an overly narrow predefined set of undesirable outcomes. The process did not allow Mobil to identify all the undesirable outcomes (such as hazards to aviation safety) that could prevent them from producing products that were fit for purpose and from achieving their broader organisational objectives.

The refiner had not satisfied itself that all compounds that could be in the process stream during manufacture, (with particular attention to process chemicals that were introduced during the manufacturing process), would not adversely affect the systems in which the final product was intended to be used. The manufacturing process was designed to ensure that all chemicals that were in the process stream that were not desired in the end product would be extracted from the process stream during manufacture. Despite this, process deviations may have reduced the effectiveness of these extraction mechanisms. The refiner did not have procedures in place to rigorously consider the likely consequences of product contamination by any of the chemicals that were introduced into the process stream during manufacture, nor of any of the likely products of reaction of those chemicals.

The refiner did not conduct any specific practical validation of its assumption that ethylene diamine would be extracted during manufacture following the introduction of Neutramine D injection in 1991. Neutramine D was first used in the alkylation unit before the introduction of a formal Management of Change process at the refinery. At the time of the introduction, a number of concerns were addressed, however no practical validation was undertaken to assess the effectiveness of the extraction mechanism to ensure that ethylene diamine was removed from the process stream.

The use of Neutramine D to help manage corrosion in the deisobutaniser tower had been contracted out. The process of contracting out the corrosion control at the Altona refinery alkylation unit was not managed to ensure that the fulfilment of the contractor's objectives would not adversely affect Mobil's broader objectives. The corrosion control contractor was required to control the rate of Neutramine D injection as a result of pH indications taken from water samples from the deisobutaniser distillation tower overheads. This requirement did not address the potential for the objectives of the corrosion control contractor (to meet these requirements) to affect the refiner's broader product quality objective of ensuring that the product was fit for purpose.

The refiner's manufacturing process was accredited to ISO 9002, and has been subsequently reaccredited. The refiner's use of its accredited quality assurance system was not effective in ensuring that Avgas was supplied that was fit for purpose.

Following up a recommendation arising from a previous contamination event could have allowed Mobil the opportunity to identify ethylene diamine contamination. The refinery had experienced a previous contamination event from microbiological contamination. Dead bacteria had been transferred along the delivery path and clogged filters. It was thought that the bacteria had been killed by the unusually alkaline water in the bottom of the Avgas storage tank. While the reason for the alkalinity of that water was never ascertained, ethylene diamine dissolved in water will markedly increase its alkalinity.

A clear understanding did not exist among the manufacturers, regulators and users of aviation fuel that compliance with a fuel standard, by itself, would not provide assurance that fuel would be fit for purpose. When the quality of the supplied Avgas was first suspected, it was immediately re-tested to ensure that it met its specification. Avgas is normally sold on the condition that it meets its specification. The fuel that contaminated the aircraft met its specification as defined by the tests that were used to ensure that the Avgas does meet its specification. Fuel is normally fully tested only once during manufacture and distribution to ensure that it meets its specification. A number of other issues have to be addressed beyond the specification to ensure that Avgas is, and remains, fit for purpose.

Despite aviation fuels being a global commodity, no single global standard existed or was used for each main grade of aviation fuel. Manufacturers of Avgas normally use their own specification for their product that meets or slightly exceeds the major international standards. Each manufacturer's specification is normally slightly different, so the actual standard for this global commodity is not consistent.

It was impossible to comply with the literal interpretation of the major international standards for aviation gasoline because they did not specify maximum permissible concentrations of undesired compounds, either singly or collectively. The major international standards for Avgas implied a zero permissible concentration of undesired compounds in the product. It is not possible to measure zero concentrations, only to measure to the lowest measurable limit (and this is normally impractical and expensive in a production environment). It was therefore not possible to comply exactly with the specifications. If the specifications allowed a permissible small concentration of classes of undesired compounds, then this would have allowed the specification to be met exactly. However, this would have required an understanding of the potential impact of such compounds both by themselves and in combination with other compounds that are, or could be, in the fuel.

Accepted definitions did not exist for all the physical and chemical properties of aviation fuels that were required to ensure that aviation fuels were fit for purpose. A number of properties of Avgas are essential for fitness for purpose which are not defined in the international standards. These properties are known by people and organisations who are responsible for ensuring that they exist, however there are no defined levels for these properties for Avgas. This meant that Avgas could have been supplied that met the international standards and yet the undefined essential physical and chemical properties may have been addressed to a varying extent, or not at all.

Despite the criticality to safety of aviation fuel quality, no regulatory requirements for fuel quality testing existed beyond the requirement to visually assess a sample of fuel drained from an aircraft before the first flight of a day, or after refuelling. Australian law that applied to the operation of civil aircraft did not require any testing of fuel quality, beyond the need for a sample to be drained from the bottom of aircraft fuel tanks before the first flight of a day, and after refuelling. The sample was to be examined to confirm the correct clarity, colour and odour, and tested for water, either with water detecting equipment, or visually. These tests would not have identified the presence of ethylene diamine in a sample of the contaminated fuel.

There was a diffusion of responsibility among the various regulatory bodies that had the potential to oversee aviation fuel manufacture, quality assurance, supply and use. Aviation fuel was manufactured at a workplace which was regulated by relevant occupational health and safety organisations. It was sold in commercial transactions that were covered by the obligations of state and federal trade practices legislation. It was used in aircraft that were regulated by the civil aviation regulator. It was possible for each of these responsibilities to have an influence on aviation fuel during its life from manufacture to consumption, but there was no clear delineation of the roles and responsibilities of the respective regulatory organisations in relation to the quality of aviation fuel.

There was no indication to show that the then Civil Aviation Authority considered the effect on safety when it made a safety related decision to discontinue any oversight of aviation fuel quality. When the Civil Aviation Authority discontinued its oversight of aviation fuel manufacture and distribution in 1991, its reasoning was primarily that the expertise in these areas rested with the manufacturing organisations, and they were therefore considered to be the best people to ensure that the quality of fuel was maintained. A lack of expertise within the Authority was not a relevant justification for a change to regulatory oversight which could affect a safety critical aspect of aviation.

No mechanism existed to ensure that the Civil Aviation Safety Authority was made aware in a timely manner of information relating to the management of situations related to fuel quality that could affect the safety of flight. Following the discontinuation of any form of regulatory oversight by the then Civil Aviation Authority, no formal lines of communication existed between the Authority and manufacturers or distributors of aviation fuel, and hence the initial notification of a fuel quality problem was likely to occur through informal channels, and the timeliness of formal notification was at the behest of the manufacturer or distributor.

The Australian Transport Safety Bureau identified a number of deficiencies in the development of manufacturing processes and the management of those processes within the refinery, the relevance of standards that were used in the manufacture of Avgas, and the oversight of aviation fuels.

These safety deficiencies formed the basis for the development of safety recommendations issued by the Bureau. The recommendations are designed to reinforce the defences that are, or could be, put in place to reduce the probability that the safety of civil aviation could be compromised in the future.

The recommendations fall into three main groupings:

• The first group relate primarily to the management of the processes for the manufacture of Avgas. They are addressed to the refiner, and may be considered as relevant to other manufacturers of aviation fuels, as well as managers of complex, safety critical systems.
• The second group relate to the development and use of international standards for Avgas, including their use in ensuring the fitness for purpose of Avgas used in aircraft.
• The third group relate to the use of regulatory oversight as an effective defence in ensuring that fuel quality as a safety critical aviation system is, and remains, consistently fit for purpose, and the need to eliminate any diffusion of responsibility among regulators who have the potential to regulate aviation fuel quality.

The full text of the recommendations can be found in section 5 of the complete report.

The aim of this study was to investigate the feasibility of forensic DNA-based techniques in identifying species involved in Australian aviation bird strikes. Experimental bird tissues were subjected to severely damaging conditions to determine if DNA could be extracted from these samples. In addition, DNA and feather microscopy databanks were created from the species classified as being the highest risk for strikes to provide reference data to compare against unknown samples. Finally, a DNA sampling kit was created and distributed widely to aerodromes across Australia for collection of material from unknown strikes for DNA analysis. Results of experimental bird tissue experiments showed the most detrimental conditions for DNA were to leave a sample at room temperature for 7+ days. DNA was successfully extracted from all strike samples collected with sampling kits then returned to the laboratory and positive identifications were able to be made to species level in the majority of cases. Interestingly, it was found that attempts at visual species identification were often incorrect and that the putative high-risk species were only responsible for 27 per cent of the unidentified strikes. In general, we found DNA identification of strike species to be a reliable method for identifying the species involved in collisions and conclude that it would be a useful addition to the methods already employed to identify wildlife strikes at civilian aerodromes.

This paper provides a preliminary review of options for making marine pilot transfers safer, as the traditional method of transferring via a pilot ladder has resulted in fatalities and serious injuries. The project included a review of the literature, interviews with pilots and a survey of methods used in other industries for similar tasks. Results indicate that the safety of the current transfer arrangement would be increased through strategies including ladder design improvements, use of fall protection systems and/or use of mechanical personnel elevation systems. Each of these strategies requires further investigation to determine the design criteria best suited to the task, the pilot population and the marine environment. The report also identifies a number of additional initiatives requiring review that contribute to reducing the risks in the pilot transfer task.

The primary aim of this study was the development of a set of normative data that captured the performance of a sample of general aviation pilots during a simulated flight from Wagga Wagga to Bankstown via Canberra, Goulburn and Mittagong. A secondary aim was to consider the impact of pilot qualification on the performance of pilots during the five legs of the flight.

Pilots were issued a completed flight plan and all the relevant documents necessary to complete the flight, including weather information, maps, and an aircraft checklist. A total of 34 pilots were recruited to undertake the flight and the exercise was conducted as it would be expected to occur within the operational environment. The experimenter acted as the Flightwatch operator and air traffic controller where necessary, and recorded the details of the flight.

Data pertaining to in-flight performance were recorded at a number of different levels of analysis, the first of which was pilots' own self-reports of their performance. Pilots' performance was also rated by an observer, and assessments were made on a number of different dimensions including the accuracy with which the aircraft was controlled, the accuracy of the track flown, the accuracy in maintaining the prescribed altitude, the level of fatigue management, and the appropriateness of the communication. The final level of analysis involved objective data that were recorded each second that the simulator was in operation. For each of the five legs of the flight, a set of geographic boundaries were identified and representative data that occurred with these boundaries were summarised using measures of central tendency¹.

In relation to the self-report data, pilots considered their performance in the flight simulator poorer than their performance in general. This may be explained by the difficulties that some pilots perceived in exercising control over the simulated aircraft. Indeed, of the eight dimensions assessed, aircraft control was associated with the lowest rating during the simulation. However, it should also be noted that relatively lower ratings were recorded for other variables including fuel management, fatigue management, scanning, and decision-making.

The observations of pilot performance revealed differences between perceived behaviour during the five legs of the flight. Specifically, performance during leg 5, the last leg, tended to be rated at a level consistently lower than performance during the preceding legs. Comparative analyses using pilot qualification as a between-groups factor failed to explain the basis for this difference in perceived performance.

The differences between the perceived performance of pilots in leg 5 and perceived performance during the preceding legs of the flight were further examined using the data recorded by the flight simulator. While differences were anticipated for variables such as altitude, it appeared that performance deteriorated on a range of variables, including the mean range of the heading and the mean range of the pitch angle of the aircraft. The variability in performance during the final leg of the flight could not be explained on the basis of pilot qualification, and suggests that other factors may be impacting on performance. It was considered that these factors might include the impact of fatigue and/or the impact of the demands in conducting a stepped descent to avoid violations of controlled airspace during the approach to Bankstown airport.

Overall, the data acquired in the present study represent a useful normative dataset against which the performance of pilots can be assessed in the future. As expected, there is a significant level of variability in the performance of pilots who conducted the simulated approach. This variability was most evident during the final stage of the flight when the demands on pilots were most acute and when the impact of fatigue was most likely to occur. This represents an avenue for future research and development.

1. Measures of central tendency are statistical summaries of a set of data. The most common measure of central tendency is the mean (average), followed by the median (the middle score in a series of rank-ordered data), and the mode (the most frequently occurring result).

While improper execution of the flare manoeuvre has been implicated in many landing incidents, very few human factors studies appear to have examined this problem. Our flight simulation study examined three different visual strategies that pilots could use to time the flare. On each trial, non-pilots, student pilots or private pilots were required to judge either:

(i) their time-to-contact with the ground; or

(ii) an idealised time to initiate the flare.

Our data provided some support for the hypothesis that pilots initiate the flare when their perceived time-to-contact with the ground reaches a critical value. Pilot performance was generally superior to non-pilot performance. However, both pilots and non-pilots were found to demonstrate flare timing biases during impoverished visual conditions (i.e. reduced depth cues) - indicating that strategies based on perceptions of environmental distance and/or critical runway angle must also have played a role. Importantly, very accurate timing judgments were possible with richer visual displays (i.e. additional depth cues) that provided performance feedback. Thus, we conclude that entry-level flight simulators can be used for flare timing training if certain minimum visual display conditions have been met.

Aircraft maintenance errors are estimated to be involved in 12% of airline accidents worldwide. Records maintained by the Australian Transport Safety Bureau (ATSB) indicate that 4.5% of Australian aircraft accidents involve maintenance deficiencies. Human error in aircraft maintenance is poorly understood and has not been the subject of previous studies in Australia. In late 1998 the Bureau of Air Safety Investigation (now ATSB) distributed a survey to Licensed Aircraft Maintenance Engineers (LAMEs) in Australia. The survey was designed to identify safety issues in maintenance, with a particular emphasis on the human, or nontechnical aspects of the job.

The survey provided LAMEs with the opportunity to describe occurrences that had the potential to threaten the safety of an aircraft, or the safety of maintenance workers. Six hundred and ten occurrence reports were reported via the survey. In most cases the reported events resulted in relatively minor consequences. The most common form of occurrence was one in which an aircraft system was activated in an unsafe manner during maintenance. The next most common form of occurrence involved the incomplete installation of components. Over 95% of the occurrences involved the actions of people. The most common forms of human error contributing to the events were memory lapses and procedure shortcuts. The contributing factors most frequently listed by survey respondents were time pressure, equipment deficiencies, inadequate training, coordination difficulties and fatigue. There was evidence that the frequency of safety occurrences fluctuated throughout the 24 hour day and that the early hours of the morning were times of particular risk for maintenance occurrences.

Several safety deficiencies were identified in the course of this study. These included: a current lack of programs to limit the extent of fatigue experienced by maintenance workers; a lack of recurrent training for licenced aircraft maintenance engineers; a need for maintenance personnel to be trained in crew resource management skills such as communication and the management of production pressures; a widespread blame culture in aircraft maintenance which discourages personnel from officially reporting incidents; and the simultaneous maintenance of critical multiple redundant systems, which can make the consequences of errors more serious. The report concludes with recommendations directed at these issues.

The purpose of this publication is to provide an overview of level crossing accident fatalities in Australia. The information provided is based on unpublished data obtained from the Australian Bureau of Statistics but responsibility for the analyses presented here rests solely with the ATSB.

Many factors contribute to an airline's safety record, some external to the organisation and others internal. An important internal contribution comes from the manner in which the company's flight operations are managed. This study addresses the organisational factors impinging on an airline's safety outcome that are subject to influence by managers in their flight operations divisions. Particular attention is given to evidence of the concept known as 'institutional resilience'.

Twelve major airlines in Australasia and South East Asia participated in the study. The study used a mixed method approach, incorporating both qualitative data (interviews) and quantitative data (audit). The qualitative approach used in-depth interviews, conducted with 36 senior managers in the twelve airlines. The quantitative approach comprised a self-reported audit of organisational management arrangements within each airline. The audit was conducted by means of a questionnaire sent to one senior manager in each airline. Eleven questionnaires were returned.

This report deals with the analysis of results from the audit.

The scope of the audit was determined by both the framework adopted for the study and by information gained during the preceding 36 interviews. The framework of analysis has six-parts: human factors, culture, safety management systems, benchmarking, and theory of high reliability and institutional resilience.

The results show both significant similarities and important differences between the airlines. Attention is given to differences between domestic and overseas airlines. The similar outcomes are useful as a normative guideline on the way airlines should address their management of safety. The differences provide a guide to further development by both airlines and researchers. The findings are discussed in detail at Section 5 of this Report.

The study identifies three areas suitable for further research. The first relates to further development of reactive and proactive measures that can indicate the state of an airlines' 'safety health'. When used in an appropriate combination, such measures should indicate changes in intrinsic safety levels and facilitate the prioritisation of remedial action. The next area builds on the first by investigating the development of a checklist, similar to the Checklist for Assessing Institutional Resilience (CAIR). A suitable checklist must appeal to the airlines in terms of its practical application. The third area is development of a process to improve the reporting rate of flight crew error.

Cranfield University in the United Kingdom, working in collaboration with Virgin Blue Airlines in Australia, applied to the Australian Transport Safety Bureau for an aviation safety research grant in 2004. The grant was awarded to support a two-phase research project into evacuation commands used by cabin crew in managing passengers during evacuations. The first phase was a best practice forum and survey, supported by members of the Asia Pacific Cabin Safety Working Group of the Australian Society of Air Safety Investigators, to establish the commands, policies and procedures currently in use among Australian and Asian operators. The results of this survey informed the development of the research aims for the second phase of the project.

The second phase involved both survey and experimental work, with members of the public participating as passengers. Four groups of up to 40 members of the public were recruited to take part in this phase of the research; all participants completed questionnaires asking for demographic information. In addition, participants were asked about the commands that they would expect to hear in a range of safety-related and emergency situations. The data from these questionnaires were used to explore passenger expectations and comprehension of emergency commands. The results indicated that participants generally had a low understanding of why they might be required to take certain actions in emergency situations. This suggested that it is important that operators take passenger expectations and comprehension into account when devising evacuation commands.

The same participants then took part in one of four sessions of four evacuation trials at Cranfield University in the United Kingdom. In each test session, two evacuations were from the Boeing 737 cabin simulator, and two from the Large Cabin Evacuation Simulator (LCES). The aim of the experimental tests was to investigate the effectiveness of selected cabin crew commands in managing passengers during evacuations. All trials were video-recorded in order that footage from the trials could be time-coded and analysed.

In the Boeing 737 simulator, the first aim was to investigate the use of active and passive safety briefings. Two groups of participants received an active safety briefing, in which the cabin crew generated a high level of interaction in briefing passengers on safety procedures. On the other two test days, participants received a passive safety briefing. Research in cognitive psychology had suggested that actively-briefed passengers would be better able to recall and act on that information if (or when) the need arose. The Boeing 737 evacuation trials were also used to investigate the extent to which having the cabin crew provide passengers with additional relevant instruction would enhance evacuations in poor visibility conditions. The results indicated that while the active safety briefing did not improve evacuation times, passengers rated this briefing as significantly more useful and helpful, and stated that the active safety briefing significantly improved their confidence in evacuating the cabin.

In the LCES, the first aim was to investigate the influence of crew commands during evacuations from a wide-bodied cabin simulator, given that the exits and slides on such aircraft types are typically rated for a dual-lane flow¹. Dual-lane flows significantly increase evacuation rates, and yet results from Phase I showed that many operators do not require their cabin crew to command passengers to move through exits two at a time.

The second aim of the LCES tests was to manipulate the extent to which participants could see the cabin crew during the early stages of an evacuation. These tests aimed to gauge the effectiveness of gestures, eye contact and other non-verbal communications used by the cabin crew in managing passengers.

The results showed that evacuations without dual-lane flow commands were faster, but more disorganised. With a larger passenger load, dual-lane flow commands could be useful for managing the evacuation in a more orderly and less congested fashion. Visibility had a pronounced effect. The half-height bulkheads meant that cabin crew gestures could actually be seen be passengers, who rated the crew's non-verbal communication as significantly more useful in the high-visibility evacuations. The high visibility also gave passengers something to aim for in the evacuations - they had sight of the exit. Hence, it was significantly easier for them to move out of their seats and along the aisles in the half-height bulkhead conditions.

The results of this research could provide useful input to cabin crew training (initial and recurrent), to the design of best practice evacuation procedures and commands, and also potentially to the design of bulkheads and cabin configurations. In addition, the results could assist in designing safety information which is more closely aligned with passenger expectations, and therefore more likely to be effective.

1. The over-wing exits in a typical wide-bodied jet aircraft are full size doors, which allow for dual-lane flow, rather than the smaller push-out exits typical of narrow-bodied jet aircraft.