ISSUE 100 1991-2015 Fo c u s on COMMERCIAL SAFETY

100th Edition

THE OFFICIAL PUBLICATION OF THE UK SAFETY COMMITTEE

AUTUMN 2015 ISSN 1355-1523 Join the debate and develop your thinking about our industry’s paradigm shift beyond regulatory compliance to safety performance

Find out how , manufacturers, MROs and defence organisations are working to secure the strategic and operational benefits of safety performance.

Be inspired by new ideas on how to move an organisational mindset from compliance to safety performance improvement Benchmark your organisation’s safety management performance against other industry players Improve your understanding of what safety performance is Hear keynote speeches from industry VIPs that showcase safety management performance in action Challenge your thinking about the traditional and accepted approach to management Attend FREE workshops: Safety Leadership, Safety Culture, Safety Risk Management, Safety Assurance and Regulatory Compliance Enter the Performance in Aviation Safety Awards and enjoy an evening of networking at the gala dinner

Find out more and book your places online at www.bainessimmons.com/symposium 100 Contents

The Official Publication of THE FLIGHT SAFETY COMMITTEE ISSN: 1355-1523 AUTUMN 2015

FOCUS is a quarterly subscription journal devoted Editorial 2 to the promotion of best practises in aviation safety. It includes articles, either original or reprinted from other sources, related to safety issues throughout all areas of air transport operations. Besides providing information on safety related Chairman’s Column 4 matters, FOCUS aims to promote debate and improve networking within the industry. It must be emphasised that FOCUS is not intended as a substitute for regulatory information or company HEMS 7 publications and procedures. by Zoe Reeves Editorial Office: The Graham Suite, Fairoaks , Chobham, Woking, Surrey. GU24 8HU Tel: 01276 855193 Fax: 01276 855195 e-mail: [email protected] Bringing Thinking Back into the Automated Flightdeck – Web Site: www.ukfsc.co.uk Training Resilience With Simple Simulation 10 Office Hours: 0900 - 1630 Monday - Friday by Captain Mark Cameron Advertisement Sales Office: UKFSC, The Graham Suite, Fairoaks Airport, Chobham, Woking, Surrey GU24 8HU Tel: 01276 855193 Fax: 01276 855195 Look How Far We’ve Come email: [email protected] Web Site: www.ukfsc.co.uk by Zoe Reeves 14 Office Hours: 0900 - 1630 Monday - Friday

Printed by: Woking Print & Publicity Ltd The Print Works, St. Johns Lye, St. Johns, Preventing Loss of Contact in Flight Woking, Surrey GU21 1RS by Michael Ph.D & Randall J Murman 15 Tel: 01483 884884 Fax: 01483 884880 e-mail: [email protected] Web: www.wokingprint.com

FOCUS is produced solely for the purpose of encounters with weather balloons: risk and mitigations improving flight safety and, unless copyright is by R W Lunnon 21 indicated, articles may be reproduced providing that the source of material is acknowledged.

Opinions expressed by individual authors or in advertisements appearing in FOCUS are those of The Dryden Accident - A driver for chance 25 the author or advertiser and do not necessarily reflect the views and endorsements of this journal, the editor or the UK Flight Safety Committee. Members List 28 While every effort is made to ensure the accuracy of the information contained herein, FOCUS accepts no responsibility for any errors or omissions in the information, or its consequences. Specialist advice should always be sought in relation to any Front Cover: A montage of 2 of UKFSC’s earliest members and TAG’s terminal at Farnborough, particular circumstances. the site of the UK’s first powered flight

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 1 EDITORIAL

100 100 Not Out by Dai Whittingham, Chief Executive UKFSC

s you will have noticed from the with more than 10K visibility, killing 92 of wealthy. Demand seems to be recovering well Ahistorical flavour of our front cover, the 146 people on board. Tuxtla Gutierrez following the economic crash in 2008 and this is the 100th edition of FOCUS in its airport saw an F27 hit trees on the approach the manufacturers appear to have reasonably current form. At 4 issues per year, that with fatal results, and an Alitalia DC9 hit healthy order books. But demographics are means 25 years of providing information the ground 5 miles short of Stadlerberg changing and there are increasing concerns aimed at contributing to the safety of having descended 1400 ft below the ILS being expressed about the ability of the . While it would be glidepath. Glass , better FMW, GPS industry to train enough pilots and aircraft pointless to attempt to quantify the and increasingly effective GPWS systems engineers to support the forecast demand. As FOCUS contribution to progress, it would have helped to generate a 7-fold reduction in one of the unexpected consequences of the be worth reminding ourselves of what the CFIT accidents, but they still occur too often, financial pressures that followed the melt- commercial air transport world was like sometimes when crews have ignored all down of the world banking system, the cost 25 years ago and looking at what might the warnings the automated systems have of training is now being borne principally by have changed. been giving them. The UPS A300 accident those being trained. It cannot be right that the in , Alabama, two years ago primary filter for entry to the profession is not As the first edition of FOCUS was being resulted from what had become a ‘dive and aptitude, or intellect, or medical fitness, but is written, the world-wide web was in its infancy drive’, unstable approach; the crew had the instead the size of a candidate’s wallet. and industry was still absorbing the lessons in sight but did not appear to have of the Kegworth and Dryden accidents of been monitoring either their altitude or Financial pressures have also contributed 1989. The Moshansky investigation on the rate of descent, nor did they recognise the to new business models such as ‘pay to F-28 Dryden accident was still over a year approach was unstable while a -around fly’ and zero-hour contracts. Pay to fly is a away and would change the approach to would still have saved them. CFIT may particular concern as it places real pressure safety for both regulators and operators. well replace LOC-I as the leading cause of on individuals to fly when they are not fit Kegworth brought some changes to fatalities once the current focus on UPRT through illness or , the fatigue often standards, specifically to the works its way through the pilot community. arising from the need to have a second job crash-worthiness of overhead bins, but the Hopefully the initiatives to improve flightpath to control the level of personal indebtedness. recent ATR accident in Taiwan suggests we management skills will have a positive effect A further result of the squeeze has been a have still not all fully learned the wisdom of on both CFIT and LOC-I occurrences. reduction in manpower and more pressure having two people agree on which engine to get the maximum out of crews, which has failed and confirming the appropriate Whilst some themes remain constant, there leads to some rostering practices that are actions for shutting it down if that is deemed is no denying that the aviation world today legal but not perhaps sensible. Flight time necessary. is very different from that of 25 years ago. limitations have become targets rather than The collapse of communism and increasing limits, which means that a large proportion The broad statistics show that the efforts globalisation driven in part by the internet of the pilot and cabin crew communities are to improve safety have been working. The has seen an explosion in the numbers of operating at maximum capacity all the time. western-built accident rate in 1990 was destinations now being served by commercial We are now seeing some very experienced approximately 1.4 per million . Today, aviation. Many of these destinations are pilots opting to work part-time as a defence the global rate is 0.5 per million. That would ex-military airfields, a tangible sign of the against long-term fatigue, which is a warning be seen as a success in any terms, but it so-called peace dividend – despite the sign that should not be ignored. becomes more remarkable when set against fact that the world is arguably less stable a doubling in traffic over the same period. now than it was before the Wall came Aircraft are much more reliable now than And for the 4th-generation types (fly-by- down. So while there is a much increased they were 25 years ago, which is testament wire and envelope protection) the fatal choice of destination, there are certainly to the efforts that the , engine accident rate is only 0.11 per million flights, many more conflict hotspots that add to and manufacturers have made to with a hull-loss rate of 0.23 reflecting the the risk of over-flight, or of operating into improve their products. As part of those ever-present danger of runway excursions. nearby destinations, as MH17 has proven so improvements the industry has seen The trend lines are still descending, though graphically. In the early 90s it was reasonable increasing levels of automation. While there the media view on MH370, MH17 and to assume that heavy weaponry was under is no doubt that automation has contributed Germanwings would of course have us think strong control, but that assumption is no significantly to a reduction in the accident otherwise. longer valid for all areas. rate, there is understandable concern about the dangers of over-reliance on automation CFIT accidents featured regularly in 1990. numbers and the success of low- and on the atrophy of manual flying skills. For example, an A320 landed 2300 ft short cost operators are ample evidence of the fact The younger generation of pilots have been of the runway at Bangalore in day VMC, that flying is no longer the preserve of the immersed in computer technology since

2 100 focus autumn 15 EDITORIAL

100

childhood and trust it implicitly, but those runways, the tendency being to add options of causing disabling eye injuries at 250m attitudes tend to prevail now across the pilot rather than simplify decision making. It is now range; the easy availability of such devices community, a function of the reliability of too easy just to follow the decisions taken has to be a concern to us all, and the UKFSC avionics and coupled with the by the nav system, on the normally correct is working with others on measures to normal human tendency to accept anything basis that the kit is right. Electronic flight control the problem in the UK. The other that makes life easier. Small wonder then bags have become the norm and the use of new threat is from the proliferation of that gross errors generated at the data entry data-links is now routine business for CAT. remotely-piloted air systems (drones) and stage are sometimes not detected until it’s Complexity is rearing its head in other areas their seemingly un-controlled proximity to too late. Similarly, TCAS has worked wonders as well, with the pace of change in regulation other aircraft. Many small drone operators for awareness of traffic and for collision and operating instructions being hard to keep are ignorant of the airspace in which they avoidance, but as another example of an up with. Electronic manuals offer managers fly, of the regulations that pertain to their unintended consequence of change it has the opportunity to make rapid changes to operations, and of the risk they pose to also largely removed the incentive for pilots documents and procedures, but the danger is others when they are airborne. That too has to look out of the flight deck despite ‘see and of a significant change being lost in the noise to change; hopefully the cross-Government avoid’ being the last line of defence. of sometimes daily alterations of a minor strategy now in development will help to nature. Such is the price of progress. control a problem that is already being seen As for other changes, it is now almost in the rising numbers of drone encounters 14 years since the 9/11 attacks brought Passenger behaviour has been a common being reported by flight crews. us the armoured and locked flight deck theme across the last 25 years – it was door along with major changes in security raised in the very first edition of FOCUS. So, a skim across the last 25 years and a arrangements. It will never be possible to It is perhaps becoming more prominent as little look at the future by way of celebrating know how many other terrorist attacks or an issue because of increasing numbers of the first 100 issues of FOCUS. If you would hijacks have been prevented by the presence , the added frictions from security like to contribute to the next 100 issues of the flight deck door, but its impact on processes and constraints on the use of by providing us with an original article, we communications between pilots and cabin personal electronic devices, and a willingness would love to hear from you. crew has been manifest on many occasions. to prosecute the worst offenders. In societal We will also never know how many young terms there is less deference being shown people failed to try for a career in aviation to those in authority, which may go some because they were never inspired by a visit to way to explaining why passengers being a working flight deck; hopefully we will still evacuated in an emergency will still try to be able to attract the half-million pilots and take with them despite being engineers that some forecasts suggest will be told not to, and why some will refuse to required over the next 15 years. comply with legitimate instructions from crew members. It is a problem that will not One of the major changes since 1990 has go away quickly, if ever. been growth in the use of space-based navigation systems, both in the air and Lastly, it is worth remembering that threats on the ground. The proliferation of RNAV change with time as well. Lasers have been approaches seems to be accelerating and yet with us for a while but available power the ATM environment becomes ever more levels are increasing, as are raw numbers of complex and congested. A bewildering array hand-held devices. It is now possible to buy of approaches is now available for most cheaply a 5W green laser that is capable

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 3 CHAIRMAN’S COLUMN

100 Justice or a dark day for flight safety? by Capt Chris Brady, Chairman UKFSC

trust that the pilot community has in So, from the above we can see that, in the safety investigation system thereby , an investigation into any potential undermining any Just Culture that the criminality must occur. industry has tried so hard to engender. 3. The use of CVR/FDR by Police Scotland It appears that the Lord via SARG “to provide an expert opinion Advocate has decided on the performance of the flight crew” that, in the absence of suggests that the focus of any potential an AAIB report, given n 19 June 2015 Lord Jones at the prosecution will be on the pilots, rather that the 3 special OCourt of Sessions in ruled than looking deeper into the reasons bulletins released by that the CVR and FDR for the accident why two highly trained and experienced the AAIB have suggested no technical failure, to Super Puma G-WNSB at professionals may have made an error, (if that the flight crew may have been at fault. Shetland in 23 August 2013, should be they did). Reasons such as human factors, He has therefore instructed Police Scotland released by the AAIB to Police Scotland. training, SOPs, aircraft design, aircraft or to start an investigation along those lines. They will in turn ask CAA SARG to provide airfield equipment shortcomings, ATC, One item of “best evidence” for such an an expert opinion on the performance weather, commercial pressure, company investigation is the CVR/FDR hence the Lord of the flight crew during the accident or national culture etc. could have been Advocates petition to the Court of Sessions flight. To quote the judgement precisely: causal or contributory factors. (the supreme civil court of Scotland) to get 4. This judgement is indicative of the their release from the AAIB. “In my judgment, there is no doubt that wider issue of the criminalisation of air the Lord Advocate’s investigation into the accidents. circumstances of the death of each of those Just Culture who perished in this case is both in the public interest and in the interests of justice. The Why the investigation? The petition was heard by the Honourable voice recording and the flight data Lord Jones, a former RAF Phantom pilot. recording which the Lord Advocate seeks to To try to present this particular case in a BALPA and the crew joined the process as recover will provide relevant, accurate and balanced way it is necessary to give some interested parties, to oppose the application. reliable evidence which will enable SARG to background to the specifics of the case. There were also affidavits from Keith Conradi, provide an expert opinion of value to assist The first question is why is there a police Chief Inspector of Air Accidents at the AAIB, him in his investigation of the circumstances investigation at all? expressing concerns that release of CVR/FDR of the death of the four passengers whose was contrary to Article 14(3) of Regulation lives were lost, and his decision whether and, This accident occurred is Scotland where the EU 996/2010 and also from Rob Bishton, if so, against whom to launch a prosecution. legal system is different to that of CAA Head of Flight Operations within SARG For that reason, the disclosure of the CVFDR and Wales. which included the following: will bring benefits for the purpose of the Lord Advocate’s investigation. It is important to “The Lord Advocate is responsible for the “Mr Bishton expresses the view that a feature stress that the analysis of the recordings in Procurator Fiscal’s investigations into potential which is key to the successful implementation the CVFDR for the purposes of its opinion will criminality and prosecutions. He also has sole of safety regulation is to achieve an open and be carried out by personnel within SARG who responsibility for directing the investigation of honest reporting environment within aviation have the expertise and experience necessary deaths in Scotland. In respect of every matter organisations, regulators and investigation for the performance of these tasks.” of fatality reported to the Procurator Fiscal, authorities. The effectiveness of that reporting the petitioner directs that the Fiscal must culture depends on how organisations This judgement is disturbing for several investigate the full circumstances of the death manage blame and punishment. Only a reasons: and must also consider if criminal proceedings small proportion of human actions that are are appropriate. Such investigations are in the unsafe are deliberate and deserve sanctions of 1. The AAIB investigation is not yet public interest.” appropriate severity. A blanket amnesty on all complete. unsafe acts, however, would lack credibility 2. The release of CVR/FDR to an agency The Lord Advocate states that “Given that and could be seen as contrary to the interests other than the AAIB for anything other the four people who died in the crash died in of justice. What is required, suggests Mr than exceptional circumstances (such as the course of their employment, the deaths Bishton is a system of “just culture”, which say a Germanwings type event) is contrary will be the subject of a mandatory Fatal creates an atmosphere of trust in which to the way that industry understands the Accident Inquiry.” people are encouraged, or even rewarded legislation. This may in turn erode the for providing safety-related information, but in which they are also clear about where the

4 100 focus autumn 15 CHAIRMAN’S COLUMN

100

line is to be drawn between acceptable and circumstances, this is usually taken to mean ICAO Annex 13 of the Chicago Convention unacceptable behaviour.” a similar pilot of similar experience and contains the following: training, often referred to as a substitution The Lord Advocate responded in his petition test, this again is where expertise is required. “3.1 The sole that: objective of the FDM investigation of “Just Culture is not intended to be a culture in an accident or which those involved in are free An interesting side-line to this story is the incident shall be from scrutiny or investigation.” issue of FDM. The Police request was initially the prevention not for the CVR/FDR but for FDM on 22 of accidents and Nobody is suggesting that those involved in January 2014. It appears that CHC choose to incidents. It is civil aviation should be free from scrutiny abide by their FDM agreement with BALPA not the purpose or investigation. This is covered by the that restricts the distribution and use of FDM of this activity to definition of Just Culture which is defined in data. The Police then choose not to pursue apportion blame Commission Regulation (EU) No 691/2010, the FDM data but seek the CVR/FDR. On the or liability.” Article 2, as follows: subject of release of FDM the Lord Advocate stated in his petition that: “5.12 The State “‘Just culture’ means a culture in which front conducting the line operators or others are not punished for “It is averred that disclosure of the material investigation of actions, omissions or decisions taken by them sought in the petition would be contrary to an accident or that are commensurate with their experience the terms of the FDM Agreement and have incident shall not and training, but where gross negligence, an adverse effect on the functioning of the make the following records available for wilful violations and destructive acts are not FDM Programme. It is respectfully submitted purposes other than accident or incident tolerated;” that the FDM Programme and Agreement are investigation, unless the appropriate authority of little or no relevance to the issues before for the administration of justice in that State So the line between acceptable and the Court. The FDM Agreement may, as the determines that their disclosure outweighs the unacceptable behaviour is gross negligence, interested parties aver, be binding in contract adverse domestic and international impact wilful violations and destructive acts. as between BALPA and CHC. However, it does such action may have on that or any future Therefore, the prosecution of pilots should not, and could not, preclude the granting of investigations”. be reserved for such cases rather than the order sought in terms of the petition if the for weaknesses in human performance or Court is satisfied that the relevant statutory EU 996/2010 echoes Annex 13 very closely errors, especially if they are system induced requirements are satisfied”. in Article 14 Protection of sensitive safety and the crew took reasonable decisions information. However it also allows for commensurate with their experience and disclosure of CVR/FDR as follows: training. A problem with having non-experts The Law(s) (ie not the AAIB) investigate this, or any other “3. Notwithstanding paragraphs 1 and 2, case, is that any system-induced human Those of us in the industry are familiar with the administration of justice or the authority performance errors may appear at first sight the applicable laws surrounding accident competent to decide on the disclosure of to be negligence but they will often involve reporting and investigation. However some records according to national law may decide reasonable decisions and understandable may not be aware that they do permit that the benefits of the disclosure of the errors. The test of negligence is an the disclosure of CVR/FDR under certain records referred to in paragraphs 1 and 2 for assessment of what a ‘reasonable person’ circumstances as follows: any other purposes permitted by law outweigh could have been expected to do in the same the adverse domestic and international impact

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 5 CHAIRMAN’S COLUMN

100

that such action may have on that or any The above legislation, that allows under I fully understand that the law requires future safety investigation. Member States court orders for CVR/FDR data to be handed an investigation into any potential may decide to limit the cases in which such over, should only be invoked in exceptional criminality given that four passengers lost a decision of disclosure may be taken, while circumstances, otherwise a precedent is set their lives. However we should leave the respecting the legal acts of the Union.” and it becomes the norm and trust is lost. reading and analysis of CVR and FDR to the I would not consider the lack of evidence acknowledged, impartial, experts in whom Finally there is Regulation of a technical failure for an incomplete we can trust to get to the real root causes of 18 of The Civil Aviation investigation to be an exceptional what are usually incredibly complex chains of (Investigation of Air circumstance. events and who will publish their conclusions Accidents and Incidents) and safety recommendations in line with Regulations 1996, which is I also firmly believe that the timing was international best practices. The judiciary can again similar to EU 996/2010 and Annex 13. premature. The Judge could have reserved then take such a report into consideration. It was under this regulation that the Lord judgment until after the AAIB Report was The alternative is to have non-experts sifting Advocate sought the order. published. The safety investigation could through CVR/FDR to identify an honest be completed without interference and mistake and closing the case as To summarise, these laws do allow for the prosecutor would then have an expert with the potential for litigation against the disclosure if the interests of justice outweigh opinion on the cause of the crash to inform flight crew; with it destroying years of trust any adverse domestic and international impact his decision on criminal activity. between pilots and investigators which has on future accident investigation. This, I would yielded so much valuable safety information suggest, is where it becomes subjective and to make this industry as safe as it is today. where the opinion of the aviation community diverges from the courts.

The Lord Advocate stated that…

“The suggestion that disclosure of the CVR and FDR “would have an adverse impact on future investigations into civil aviation accidents, incidents and occurrences in general” has no evidential basis and is nothing more than speculation (see, in I have tried to present a balanced view of the this connection, Société Air France v NAV judgement but inevitably with such a detailed 2010 ONCA 598, per Goudge JA at case the areas I have chosen to quote may be §29).” [This refers to the Air France A340 seen by some as selective; and of course as overrun at ] a pilot, I may have an unconscious bias. For that reason I recommend that members read Lord Jones has had to decide if release the judgement of Lord Jones themselves. of the CVR/FDR would have any adverse He has done a remarkably thorough job impact; he did not forsee any but I am not of unpicking all of the various overlaying convinced. The criminalisation of accidents, national and international laws and deciding particularly if driven by the compensation upon the merits of the arguments of the Lord culture, is an insidious malaise which drives Advocate, the crew, BALPA, AAIB and the behaviour away from openness towards an CAA. It should also be said that he has also unhealthy defensive attitude which stifles imposed a very strict set of conditions on reporting and hence progress in flight safety. the release of the data including appropriate Whilst canvassing opinion from my peers redaction if any transcript is published. The on the line, one pilot said that “Satisfying a full judgement is available on-line at https:// short term punitive objective at the expense www.scotcourts.gov.uk search under “Court of long term flight safety benefits will, to of Session Judgments” using reference reduce it to its starkest terms, “compensate” P628/14. a few but kill many more.” It is difficult to disagree with this viewpoint.

6 100 focus autumn 15 Helicopter Emergency Medical 100 Service (HEMS): A life-saving operation by Zoe Reeves, BALPA’s Flight Safety Officer

“Fairoaks Information Helimed six zero alpha.”

“Helimed six zero alpha Fairoaks Information, pass your message.”

“Fairoaks Information Helimed six zero alpha, request to transit your ATZ and traffic information en route to a scene along the M25 near junction 10.”

eceiving this kind of call from an air Rambulance would be thought provoking. I would sit in the tower knowing that the scene of the accident was not far away from my ATC base at Fairoaks and I often wondered how everything would turn out for the people involved. expansion, life-saving innovation, and heart- flying, so crews also have rapid response rending personal stories. Now there are more to maximise their availability. The air ambulance would be deployed to an than 30 air ambulances flying nearly 20,000 incident site of a serious accident where the missions every year. The air ambulance Each aircraft is crewed by at least one area is not easily accessible to vehicles or charities in the UK generate over £45 experienced pilot and a minimum of one when time is of priority to save a life. million a year, mostly through donations and doctor and one paramedic, who are trained in sponsorship by local people and businesses. advanced pre-hospital care, which gives them You never know when you may need the help the knowledge and skills necessary to assess of the air ambulance. I tend to hear about it Emergency Aircraft and stabilise critically ill and injured adults and going to help injured mountain bikers or children. This means that specialist clinical horse riders quite frequently, being involved Last July I had the pleasure of having a procedures that are normally only available in both sports myself. These activities tend tour of the Kent, Surrey and Sussex Air in the resuscitation area of an emergency to be in remote areas and when things go Ambulance at its Redhill base in Surrey. department can be delivered to patients wrong in the biking or equine world they can at the scene, such as general anaesthesia, go very wrong, very quickly. The Kent, Surrey and Sussex Air Ambulance advanced pain relief and, for some patients, Trust’s (KSSAAT) state-of-the-art helicopter surgical interventions. Effectively, what they Life-Saving Innovation emergency medical service aircraft operate aim to do, as far as possible, is to bring the 365 days a year out of the bases of Marden, emergency department to the patient and The ambulance in the UK flew into Kent and Redhill, Surrey. They are capable then take them quickly and directly by land operation in 1987. Since then, the story of delivering their crews anywhere within or air to the most appropriate hospital best of helicopter emergency medical services the region in under 20 minutes’ flying time. able to treat their injuries. (HEMS) in the UK has been one of sustained Occasionally weather prevents the aircraft

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 7 100

type of operation however, the main one being that there is only one major trauma centre, Southampton, which is within the aircraft’s flying zone that is equipped to accept at night.

There are two major trauma centres in that have helipads, St George’s Hospital and the Royal London Hospital. Neither of these is available for night-time operations. To get around this problem, if the air ambulance is deployed at night it will land at the scene of the accident and blue light the patient to the most appropriate trauma unit along with the medical team that treated them from the aircraft to maintain continuity for the patient. The flight crew will then take the aircraft back to its base at Redhill. King’s Hospital in London is in the process of raising funds to build a helipad, which should be in place KSSAAT works very closely with the South East NOTAR system. This means that the by spring/summer 2015. It is not yet known, Coast Ambulance Service NHS Foundation helicopter does not have a tail rotor. Instead it however, if it will be night equipped. Trust (SECAmb), responding to over 2,000 has an enclosed fan which directs air through emergency calls a year. The normal operating the tail boom to the thruster and out of slots, Fuel is also an issue, as away from the area of the charity is therefore defined by using the Coanda effect for yaw control. The helicopter’s base, the fuelling infrastructure the region that SECAmb covers and includes benefits of this system are increased safety, is challenging. Kent, Surrey, Sussex, and a small area of lower noise levels, better performance and north Hampshire; this region has a resident controllability enhancements. The medical team at KSSAAT has long been population of 4.5 million people. faced with the difficulties of stabilising The helicopter has twin engines and patients at the scene of an accident who The air ambulance is deployed by HEMS at speeds of up to 150mph, essential for a have suffered catastrophic bleeding. paramedics working on the HEMS desk at the rapid response to serious medical traumas. control centre of SECAmb, who screen all 999 The 902 is still one of the most advanced During the early months of 2012, KSSAAT emergency calls coming into the ambulance aircraft available for air ambulance work, and committed to exploring the possibilities of service to establish if the air ambulance the combination of this and the skills and carrying blood on board their aircraft for the would be of benefit to the patient. experience of the pilots allows the helicopter benefit of these patients. to land in some of the most challenging MD902 Explorer locations to respond to medical emergencies. The blood safety and quality regulations (2005) are incorporated into UK law and are upheld The charity has been flying a MD902 Explorer Night Operations by the Medicines and Healthcare Products helicopter in Kent since the beginning of Regulatory Agency (MHRA). The regulations the century, registration G-KAAT. When the Since September 2013 KSSAAT has been stipulate that blood products must be fully service was expanded into Surrey and Sussex operating at night. To enable the crew to traceable from donor to recipient and that the obvious choice for the second helicopter do this it operates its night capable aircraft they are stored at a constant temperature, was another MD902, registration G-KSSH. G-KSSA along with night vision goggles between two and six degrees centigrade. If This has since been replaced with a night- (NVG). The visibility using the goggles KSSAAT was to fulfil its aspirations to carry capable MD902, registration G-KSSA. greatly improves the safety of the operation blood on board its aircraft, then compliance but at £17,000 a pair they are not cheap! with the regulations would have to be ensured. The MD Explorer helicopter was one of For KSSAAT there was the need to source a the first designs to incorporate the unique There are some hurdles that do restrict this

8 100 focus autumn 15 100

As I left the Redhill base that day I felt very lucky to have such services here in the UK. To see that most of the operations around the country are funded mainly by public donations and run by charities is amazing. If you can help support your local air ambulance charity I would encourage you to do so as you just never know if you or a loved one may need their help one day.

For further information on Kent, Surrey and Sussex Air Ambulance visit www.kssairambulance.org.uk.

Original article by Zoe Reeves published in BALPA’S The Log Magazine Winter 2014 edition. supply of the blood product and a transport and safe administration. The Credo EMT solution that would ensure a consistent ‘Golden Hour Box’ is an insulated transport supply to its bases. box that can maintain the required cold temperatures for up to 72 hours, and the Service by Emergency Rider Volunteers Belmont Buddy Lite fluid warmer is a light, (SERV) is a charity that supplies an out-of- portable, battery-powered warmer that hours blood transport service to the NHS would efficiently warm the cold blood to which is free of charge. They agreed to near normal body temperatures. A supply provide a transport service for KSSAAT, of the boxes was made available to the servicing both the Redhill and the Marden haematology teams at both hospitals. bases 365 days a year. Extensive training was provided to all involved, and on the evening of the 3rd Critical Care February 2013, two Credo boxes packed with four units of O negative blood, were Arrangements were made for blood to be delivered by SERV to a secure drop-off point supplied by two neighbouring hospitals. at each of the KSSAAT bases, and blood East Surrey Hospital would provide blood was available to the medical teams for the products to the Redhill base, and the first time from 7am on the 4th February William Harvey Hospital would do the 2014. The equipment used to carry out this same for the Marden base. Two new pieces service was funded by the Henry Surtees of equipment were introduced to support 2015 Foundation, founded by legendary the regulations around temperature control motor racing champion John Surtees OBE.

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 9 Bringing Thinking Back into the Automated Flightdeck100 – Training Resilience With Simple Simulation by Captain Mark Cameron

ince the 1940’s, the emergence Sof procedural instructions for the operation of increasingly complex aircraft has made a substantial contribution in reducing the accident rate. With multi-crew operations, the formalization of issuing Standard Operating Procedures (SOPs) has had a further beneficial effect, one pilot is more able to predict what the other will do, if the predicted action is missed or omitted, the Pilot Monitoring (PM) is able to intervene. However there are signs that some pilots are having difficulties making sense of events outside of the routine, there are also concerns that the reassurance of procedural compliance trumps resilience.

As an active line pilot for the past 35 Figure 1. “What’s the problem?” years, I have been primarily interested in towards ‘increasing realism’ to make their flight, however with the increasing levels human behavior and the cognitive processes micro environments even more plausible. of safety that results there is the attached that form our interaction with the aircraft, As someone who used to train pilots in a paradox of compliance versus resilience. To especially mine. ‘live’ helicopter, having the ability to safely be provocative it could be suggested that the exercise differing failure modes in a simulator difference between resilience and violation is Starting with the Bell 47 helicopter, the tiny is a huge improvement; it becomes possible simply the outcome. Airlines would like their instrument pedestal surrounded by a large to present realistic failure scenarios without pilots to follow procedures, the measured Plexiglas bubble hinted at how the designer having the safe flight trajectory occupying a operational risk drops as a result, however intended flight operations to be conducted. considerable part of the instructional process, since procedures cannot address every Nonetheless, with a power-on engine RPM never mind the increased risk and exposure. possible eventuality, there comes a time band of 3000 ±100, there was a tendency where a novel event occurs for which there for the novice student to spend what felt However, there are limitations to the current is no procedure. There is little doubt that like 90% of their time fixated on this single simulator iteration. For example there is airlines would also like their crews to insulate parameter. Eventually, within a short time, still the difficulty of creating a cognitive flight operations from unpredicted hazards. other senses came to the rescue. If the RPM replication of a naturalistic line flight. When was drifting high or low, it became possible I do any simulator training positioned in to hear the change. From those slightly naïve a virtual London Heathrow or New York Simulation and Compliance days of visual operations, I now operate JFK, there is the immediate and detectable an aircraft where the thrust levers do not implausibility of being the only aircraft in the Conversion and recurrent training tend to move with power changes while operating in sky. The only exception to this is when the have footprints; this is a natural extension instrument conditions. regulator requires the training or checking from the contention made by the airframe of a TCAS manoeuver. As a second example, manufacturers of the familial structure in the scenario in Figure 1 is a case in point; it their systems design. Common Type Ratings High Fidelity Simulation? remains beyond current simulator recreation. expose the trainees to ‘differences’ from the This event was taken directly from a line archetype. This proscribed strategy tends An invaluable component of training pilots flight and it illustrates the difficulty that the to generate lists of events that have to be to manage their craft has been the simulator. crew is about to face. Even without Airbus trained or checked to satisfy regulatory From the early days of the , experience the reader can try and deduce compliance. As a result the “44 items in pilots have practiced and exercised their what the problem is. An explanation will be four hours” process can emerge, with a skills at translating the indications on their found at the end of this article. simulator system reset in between each instruments into a situational mental model. (nearly) completed item. In the past Line With increasing powerful and cheaper A further influence of the simulator process is Orientated Simulation was held up to digital computers the capability of rendering the training and examination of compliance fulfill the promise of simulated naturalistic high fidelity external imagery arrived. The with procedures. As indicated before, these training, but regulatory compliance inhibited simulator manufacturers continue to strive SOPs are central to the safe conduct of this more free-ranging strategy.

10 100 focus autumn 15 100

The subject then discovers the meaning of a ‘dampened phugiod’ as initially nothing happens since the still increasing temperature in the coldstore has to be suppressed and then reduced. The size of the system that is being controlled is not always apparent to the subjects; none of my experimental group had worked in a supermarket in a previous life and so the expertise was beyond direct experience. With the initial apparent lack of response to the initial input, the next stage usually followed a typical path of a more adventurous reduction on the control wheel closely followed by genuine surprise as Figure 2. Coldstore Simulation Interface the powerful cooling system overcame the temperature inertia. The system temperature Industry is offering Evidence Based Training Atlantic. This event had such a strong impact decreases rapidly and the control panel emits (EBT) as a way of reassuring regulators on me that I was inspired to use one of these complaints of freezing stock (too cold is as that training and checking would provide simulations as part of my Masters research. bad as too warm with dairy products). The a competence regime ‘at least as safe as’ Dr. Dahlström introduced me to Dr. Stephan subject then typically readjusts the control previously existed, while moving away from Strohschneider who, as part of the Dörner wheel in a fashion that will be familiar to the ‘box ticking’ routine that still pervades. group, kindly gave me access to a simple pilot trainers and trainees in the first few The difficulty with this move is exemplified simulation called “Coldstore”. career minutes of attempting straight and by the differing views between the LOSA level flight. Collaborative compared to those held by The premise of the “Coldstore” simulation individuals such as Sidney Dekker; the former is very simple. The automatic temperature offering event data collection as a way of controller of a supermarket coldstore has Simple But Clever assessing operational risk, while the latter failed; there is a manual system that can worries that ‘counting stuff’ does not offer be used in this eventuality. This mechanism For such a simple, single axis simulation, any predictive indications of future hazard consists of a slider or control wheel marked it remains remarkably nuanced. The interactions. from 0 to 200; the indexes are not correlated temperature of the coldstore is not to the temperature of the coldstore (Figure 2). contiguous; a temperature snapshot is At the beginning of the exercise, the coldstore released every eight seconds, the subject Other Simulations temperature is +18°C and rising, it should has to wait to see what the response is. This be +4°C. The system is already issuing dire waiting for rationed data can produce some As part of our CRM departmental development, warnings of decomposing stock. The task of interesting effects. The other remarkable our manager, Dr. Nicklas Dahlström the subject is to find a setting on the control aspect to the simulation is that it can introduced us to a simple simulation created wheel that will yield a stable temperature of generate data. A small text file is created by a German academic group under Professor +4°C within a limited period of time. after each simulation run that can be used to Dietrich Dörner. This scenario is designed graph time against control wheel input and to exercise the team dynamics that emerge The typical response of the subject is to resultant coldstore temperature. This data when a group is given the task of running a intervene in the immediate crisis by adjusting capture was useful for producing statistical small cruise somewhere in the North the temperature wheel to a lower setting.

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 11 100

Other distinctive patterns emerged; one of them seemed to occur frequently and was always associated with a pilot subject. This approach could be called the “Tentative Pilot” strategy. The initial temperature recovery is largely dampened but then there is a slow and progressive trimming towards the final stable temperature, but from only one direction (usually downwards). What was notable about this tactic was that the subject frequently failed to complete the exercise in time (Figure 5).

Figure 3. Coldstore Simulation Record as well as a unique visual illustration of the Other Solving Strategies subject’s interaction with the simulation, as can be seen from Figure 3. Occasionally, with the inertia of the system, Figure 5: Tentative the subject will increase the temperature demand on the control wheel and the system, Making Comparisons Observing Mental Processes still in a downward cycle will generate a reduction of temperature that completely Gathering the quantitative data and mapping Within the graphical dataset where there collapses the still fragile mental construct. correlations offered further insights. Figure 6 appears to be three distinct phases: chaos, Several time cycles can pass before there is shows a plot of solve time versus subject age. learning and mastery. In the initial stages of confidence to return to the original thesis or The signal that emerged was unexpected. The the simulation, the subject is uncertain of another theory is embarked upon. negative correlation between age and solve what to do except drop the temperature. The time is statistically significant. The reason for time pressure to solve the problem combined There is another subset of subjects that this is speculative and further work would be with uncertainty about the relationship could not rationalize the system in any way, required to make any definitive statement, between wheel position and temperature they constantly chased the temperature up however theories could be made involving lead to rapid but incoherent wheel inputs. and down the scale; in his book “The Logic compliance and conditioning. It may be worth noting that the green triangle data point As the simulation progresses a solidification of Failure” Dörner labeled this activity as was a result from a female Asian who of mental model seems to occur; the subject ‘Garlanding’ (Figure 4). develops their “Theory of System” and then out performed all but one of the pilots; the experiments with the control wheel to either slowest were experienced line commanders, many of which are also CRM trainers. validate or disprove that idea. Typically the subject arrives at a tentative but valid theory Giving pilots a venue and experience to think and after some time will achieve the required problems through can only have a positive parameters. Once the theory seems to impact on operational safety. hold, the adjustments to the control wheel become more carefully modulated with one or two index unit adjustment to trim the system to the final target temperature.

Figure 4: Garlanding

12 100 focus autumn 15 100

Figure 6: Graph of age versus solve time

Figure 1: “What’s the problem?”

For someone with expertise, it is very simple to see that the white Top-Of-Descent arrow is somehow wrong. The aircraft is 141 nm from the VOR ‘JDW’ that is situated on the airfield and the Flight Management System is suggesting that the TOD will be in 40 nm. This would leave 100 nm to descent 40,000 ft. While this is possible to do with pilot intervention, it is less than desirable, especially with the slower traffic below (our aircraft is flying at M0.85). The high rates of descent that need to be achieved to recover the profile may have later implications.

The crew executed a Direct-To the Final Approach Fix for the approach they were going to fly, as cleared by Air Traffic. Somehow the Flight Management calculations became corrupted and miscomputed the TOD.

The real issue here is that there is only ONE data point amongst the other fifty-five data points that indicates another ONE data point is incorrect in a usually highly reliable system. This is part of the daily pilot experience, where only experience gives the pilot enough expertise to notice the discreet signaling that the system emits that either validates or disproves the mental construct the crew collectively possess about their current and future situation; thinking pilots will see the discrepancy.

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 13 100 Look How Far We’ve Come... by Flight Safety Officer Zoe Reeves on the rise in helicopter automation within offshore operations, the issues and the future.

the industry faces’, it was discussed how to tackle the issues of complex automation in helicopters and how we best learn from the mistakes the airlines have made, and how we train our crews better to allow them to understand the technology they are using daily.

Below are some areas of improvement raised by leaders in the industry:

n Pilots need to reduce ‘automation surprise’ by being educated properly in the use of the automation

n Looking at the benefits of humancentric design within the cockpit modern helicopter. Perhaps a redesign of the cockpit layout with a greater focus on n The benefits of reducing the complexity of colours and graphical displays which display human-machine interface design (HMI). information more intuitively, SOPs that mirror hen I think of automation I think This is already happening with the use of the lessons learnt from the world, of fixed , I don’t know why, touch screen devices but looking forward W ATQP and EBT all have their role to play. perhaps it’s because the helicopters I to colours, shapes and graphs to simplify used to fly were so basic that having an complex operations One fact remains however, commercial artificial horizon indicator was pure luxury. pressure and the fast declining resource of n Standardising across operators, standard highly experienced military personnel are Fixed wing aircraft are much more stable than operating procedures (SOPs) and learning making it very difficult for the helicopter helicopters and thus lend themselves better from the airline industry’s mistakes operators to recruit the right level of to automation. Helicopters are unstable experienced pilots and keep it that way. by nature. Maintaining the aircraft altitude n Understanding OEM (original equipment This should not have a detrimental effect requires constant inputs by the pilot on all of manufacturer) design philosophy, sharing on safety but is the time and money there the controls (cyclic, collective and anti-torque lead customer experience, agreeing to train the pilots in a way the airlines do? pedals) generating a high workload. For this common procedures and incorporating A recent BALPA survey asked our helicopter reason, autopilots for helicopters were rapidly these procedures for all training providers pilot members whether they felt the taken up. Recently, functions have and mandating those philosophies outsourcing of training to aviation training expanded to help the pilots to control not organisations (ATOs) would lead to a rise only the aircraft altitude but also its trajectory n Regulated training – generate training in safety standards and 87 per cent said no. in three dimensions, and sometimes in four that meets basic safety, reflects role, Perhaps we need to look at incorporating dimensions including time constraints. environment and types but can also adapt training back in-house again, which is easier to include global and local evidence said than done. In the wake of recent helicopter incidents, the safety of offshore helicopter operations n The introduction of ATQP – alternative Automation is the way of the modern has been comprehensively reviewed by the training and qualification programme helicopter and the requirement to stay ahead British Government, European regulators, of the drag curve is vital if we are going to manufacturers, operators and the oil and n Learning from the positive – EBT reduce errors in the future. gas industry. Loss of control associated (evidence-based training). with the sophistication and automation of If you have any comments or feedback modern helicopters, the training that the At the end of the conference it was asked if please contact: pilot receives and the rise of commercial the advent of digital avionics and complex Zoe Reeves, Flight Safety Officer, pressure have been identified as issues automation in the modern helicopter cockpit [email protected]. requiring attention. changed the way we train. It was also asked: Tel: +44 (0) 20 8476 4039. Does current training prepare a pilot for How do we go about doing this you ask? operational flying or is it a matter of ticking Original article by Zoe Reeves published We get together and discuss a different the regulatory boxes? The question we were in BALPA’s The Log Magazine Autumn approach… all asked was: Could there be a better way 2014 edition. to train? The consensus is that offshore At a two-day conference presented by the helicopter training needs to offer a better Royal Aeronautical Society on ‘Automation platform for pilots to understand the greater of offshore helicopters and the challenges automated design and procedures of the

14 100 focus autumn 15 100 Preventing Loss of Control in Flight

Boeing, as part of the Commercial Aviation Safety Team, recently completed a multiyear effort to analyze loss-of- control–in-flight events and generate feasible solutions in areas of training, operations, and design. These safety enhancements have now been adopted by the Commercial Aviation Safety Team for implementation in the and are being advocated for worldwide adoption. by Michael Snow, Ph.D., Associate Technical Fellow, Human Performance, Aviation Safety, and Randall J. Mumaw, Ph.D., Associate Technical Fellow, Human Factors, Flight Deck Design Center, Flight Crew Operations Integration

n the last decade, loss of control– Iin-flight (LOC-I) has become the leading cause of fatalities in commercial aviation worldwide. A subcategory, flight crew loss of airplane state awareness, has risen as a causal factor in these accidents.

This article explains safety enhancements that were recently adopted by the Commercial Aviation Safety Team (see “What is the Commercial Aviation Safety Team?” on page 19) and the process that drove the development of the enhancements. Implementation of the resulting training, operations, and airplane design safety enhancements is estimated to reduce the risk of future airplane state awareness events approximately 70 percent by 2018 and 80 percent by 2025.

A large, complex problem

Accident rates and fatalities in commercial aviation are at historic lows in recent years, even as air traffic has climbed. However, Boeing continues to work with industry and government partners to improve safety for the traveling public. In August 2010, the Commercial Aviation Safety Team chartered the Airplane State Awareness Joint Safety Analysis Team as a follow-on activity to previous work done by a LOC-I Joint Safety Analysis Team in 2000. The primary purpose of the Airplane State Awareness Joint Safety Analysis Team was to analyze a representative set of LOC-I accidents and incidents in which the flight crew lost Proposed loss-of-control–in-flight interventions cover a broad spectrum of potential awareness of the airplane’s state, defined as: solutions, including flight simulator training.

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 15 100

Figure 1: Worldwide jet transport fatal accidents, 2003–2012 The loss of airplane state awareness has been a major factor in worldwide jet transport fatal accidents during the last 10 years. n Attitude (pitch or bank angle) or operations, based on information available in airplane state awareness dataset, which may the Aviation Safety Information Analysis and be representative of common issues present n Energy (the combination of airspeed, Sharing database. in similar events (see fig. 3). Note that no altitude, vertical speed, thrust, and single factor causes an accident or incident. configuration control surfaces). In these events, it took a combination of Studying loss of control-in-flight at least six themes to result in a hazardous A review of worldwide transport airplane situation. The Airplane State Awareness Joint accidents during the period from 2003 to Nine of the events analyzed involved loss of Safety Analysis Team did not assign a ranking 2012 revealed that more than half of all attitude awareness and nine involved loss of to these themes and notes that higher LOC-I accidents and resulting fatalities energy awareness (see fig. 2). The objective frequency of occurrence (i.e., appearance in involved flight crew loss of airplane state of the analysis was to identify underlying more events) should not necessarily imply awareness (see fig. 1). problems that contributed to the accidents and greater importance. incidents analyzed. In the course of this analysis, The Airplane State Awareness Joint Safety the teams identified 161 distinct problems, of n Lack of external visual references. In Analysis Team was co-chaired by Boeing which 117 were common with those identified 17 of the 18 events, the event airplane and the U.S. Federal Aviation Administration by previous Joint Safety Analysis Teams and was flying at night, in instrument and staffed with subject matter experts from 44 were newly developed by the Airplane meteorological conditions, or in a major airplane manufacturers and suppliers, State Awareness Joint Safety Analysis Team. combination of night and instrument pilot unions, airlines, research organizations, The analysis teams then identified a total of meteorological conditions, sometimes data mining organizations, and government 274 intervention strategies to address these at high altitude or over dark land or aviation safety departments and agencies. problems, of which 181 had been documented water. As a result, the crew had to rely Two analysis teams studied 18 events, previously and 93 were newly developed. on instrumentation to establish and identified problems and major themes, and maintain orientation. developed intervention strategies. A data team complemented the work of the analysis Common themes among loss of n Flight crew impairment. In seven of teams by assessing the presence, frequency, control-in-flight the 18 events, at least one member of and characteristics of airplane state awareness the flight crew was affected by fatigue, precursors (conditions commonly leading The Airplane State Awareness Joint Safety illness, or alcohol consumption, and in to these events, such as warnings Analysis Team discovered 12 major themes some cases by a combination of factors. or extreme bank angles) in U.S. Part 121 that appeared across the events in the

16 100 focus autumn 15 100

Figure 2: Airplane State Awareness Joint Safety Analysis Team event dataset Of the 18 events studied by the Airplane State Awareness Joint Safety Analysis Team, nine involved loss of attitude awareness and nine involved loss of energy awareness. n Training. In nine of the 18 events, flight maintenance was performed, but it did not pairing — particularly the pairing of pilots crew training played a role. In some cases, directly address the actual problem or was with low time in type — was also an the crew had not received training that performed on the wrong system. issue (see the section on crew resource is generally considered industry standard management). and is widely available. In other cases, n Safety culture. Safety culture played a role the training had taken place but was not in 12 of the 18 events. In some cases, the n Invalid source data. In five of the 18 recalled properly or did not address the operator had a poor safety record, extending events, invalid source data from the air scenario encountered. In some instances, back for months or years. Many of the data system sensors or probes, inertial the Joint Safety Analysis Team considered flights operated with compromised safety, or rate gyro systems, angle-of-attack the training that the crew had received such as with less than fully functioning vanes or sensors, or other signals were counterproductive or negative. systems or with a poorly defined flight plan. used as input to primary flight displays, In several events, the coordination and the autoflight system, or the navigation n Airplane maintenance. Airplane interaction with the air traffic management, systems with little or no indication the maintenance was an issue in six of the 18 both in and during the flight, data were invalid. events. In some cases, maintenance was was poor. Schedule pressure was prevalent, not performed in a timely manner, allowing resulting in crews pressing on with flights n Distraction. Distraction played a role in all problems to persist until they became or other activities despite warning signals 18 events and manifested itself in two ways. factors in the accident chain. In other cases, that the situation was deteriorating. Crew

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 17 100

Figure 3: Summary of significant themes across all events

First, a flight crew would make a decision number of events, the pilot monitoring awareness, understanding, and behavior based on faulty information or incorrect failed to properly perform the monitoring in the manner intended. It is important reasoning (sometimes when task-saturated) function.Crews also failed in some to note that alerting effectiveness is not and would be distracted by pursuit of instances to manage their workload solely the result of airplane design: it is also actions or thought processes associated properly. In a few events, an authority significantly affected by flight crew training, with that decision, a phenomenon known gradient between the captain and first communication, attention, and other factors as confirmation bias. Second, the flight crew officer likely played a role in preventing in the flight deck environment. would become focused on one instrument the from taking control of the or one response to the exclusion of all other airplane from the captain, even when the n Inappropriate control inputs. In 12 of relevant inputs, comments, or alerts and captain was clearly failing to correct a the 18 events, the flight crew responded to would essentially block out any information hazardous airplane state. hazardous airplane states and conditions that may have led them to fully understand with control inputs that were opposite to the problem they faced, a phenomenon n Automation confusion/awareness. In what was necessary to recover the airplane. known as channelized attention. 14 of the 18 events, the flight crew was The term “inappropriate” is intended to either confused about the state (i.e., on/ convey only that the control inputs were n Systems knowledge. In seven of off) or mode of the autoflight system not correct for the purpose of recovering the 18 events, the flight crew lacked or else was unaware of trim or control the airplane and should not be construed to understanding of how major airplane inputs made by the autoflight system. automatically imply pilot error. subsystems — such as autoflight, air data measurement, navigation, and inertial n Ineffective alerting. In all 18 events, alerting systems — interact and how information was an issue. The intended function of a flight Preventing loss of control-in-flight from one system influences another. deck alert is not simply to go off: rather, it is to raise flight crew awareness to a potential Hundreds of intervention strategies were n . In 16 of the hazard, assist the crew in understanding identified by the Airplane State Awareness 18 events, crew resource management the hazard, and (where possible) provide Joint Safety Analysis Team to mitigate was not practiced effectively. Specifically, guidance to avoid or recover from the the problems observed in the 18 Airplane flight crews failed to communicate hazard. The term “ineffective” in this State Awareness Joint Safety Analysis effectively or work together to understand context is meant to convey only that the Team events, and they were grouped into and resolve problems or confusion. In a alert, if present, failed to impact flight crew categories, based on how, and by whom, they

18 100 focus autumn 15 100

What is the Commercial Aviation Safety Team?

The Commercial Aviation Safety Team is a voluntary collaboration between U.S. government and industry that was founded in 1998. Its goal is to reduce fatality risk 50 percent in airline operations by 2025. It operates by consensus, deciding as a group which problems represent the greatest threats to aviation safety, chartering teams (e.g., Joint Safety Analysis Teams) to analyze those problems and underlying issues, determining feasibility of potential solutions (via Joint Safety Implementation Teams), and then tracking the implementation and effectiveness of adopted solutions (i.e., safety enhancements).

would be implemented. These categories n Reemphasized/expanded crew not receive an overall effectiveness score. include airplane design, flight crew training, resource management. Ranking of research interventions for priority maintenance, and safety data and research. n Flight crew proficiency. was based on which research interventions n Flight simulator fidelity. addressed the highest number of high-scoring Airplane design. These interventions problems. The top research interventions, called for action on the part of airplane Airline operations and maintenance. based on this methodology, fell into these manufacturers or suppliers related to the These interventions called for action on the general areas: design of current and future . part of operators or air traffic management The highest-rated interventions related to to improve and expand operating policies n . airplane design fell into these general areas: or procedures. The interventions related to Displays to prevent spatial airline operations, including disorientation. n Flight envelope protection. issues and airplane maintenance, fell into Alerting of spatial disorientation n Improved alerting. these general areas: conditions. n Flight path/control guidance on displays. n Maintaining flight crew awareness in n Source data integrity. n Maintenance procedures. high-workload environments. n “Day-visual meteorological conditions” n Flight crew qualifications. n Automatic systems for error detection, display systems. n Nonstandard flight operations. prevention, and recovery. n Automation design. n Reemphasis and rationale for standard n Human performance benefits of poststall n Energy management display/ operating procedures. recovery training using advanced flight prediction systems. n Flight crew impairment. simulator aerodynamic models. n Safety culture. Flight crew training. These interventions called for updates to current flight crew Safety data. These interventions called for Developing safety enhancements training curricula, standards, additional expanded data mining and sharing programs training, and improvements to flight simulator and safety management principles. The After the Airplane State Awareness fidelity. The highest-rated interventions interventions related to safety data fell into Joint Safety Awareness Team identified related to flight crew training fell into these these general areas: intervention strategies, the Commercial general areas: Aviation Safety Team chartered the Airplane n Sharing of safety-related data (e.g., the State Awareness Joint Safety Implementation n Aviation Safety Information Analysis and Revised approach-to-stall training. Team to review them; assess them for n Sharing Program). Expanded upset prevention and recovery technical, financial, operational, schedule, n Operator safety management systems. training. regulatory, and social feasibility; and develop n Sharing of service difficulty reports. Scenario-based situations. new safety enhancements. The team then Stall recognition and recovery. Research. Research interventions based on developed detailed implementation plans Spatial disorientation recognition based on the approved safety enhancement and recovery. the Joint Safety Analysis Team process do HOW MANY DO YOU REMEMBER?

32277®Flight Safety Iss 67 17/8/15 14:00 Page 1

SUMMER 2007

1

100 focus autumn 15 19 100

concepts. The proposed training and partners in research and development to operations, and design safety enhancements, operations safety enhancements focus bring these systems to application readiness. and it recommends these enhancements primarily on: be communicated to international aviation safety communities for their review and n Revisions and improvements to existing implementation where applicable. The flight crew training in upset prevention Commercial Aviation Safety Team and its and recovery, including revised approach- members have now officially adopted and to-stall training. published these safety enhancements as part n Revisions to go-around training. of the Commercial Aviation Safety Team n Policies and training for prioritizing Safety Enhancement Plan and are working with controlled flight in non-normal situations. the International Civil Aviation Organization n Training verification and validation. and the international safety community to n Enhancement of crew resource increase adoption worldwide. The plan can be management training to further define found at http://www.skybrary.aero/index.php/ and practice the duties of the pilot Portal:CAST_SE_Plan. monitoring. n Monitoring and understanding of habitual noncompliance to standard operating “Roll Right!” Summary procedures and improvements to Figure 4: Bank angle alerting with standard operating procedures. recovery guidance Loss of airplane state awareness plays a n Policies for conducting nonstandard, significant role in at least half of all LOC-I nonrevenue flights. Boeing is implementing auditory and visual category events. bank angle alerting with recovery guidance in the 737 MAX and the Next-Generation 737. In addition to training and operations An industry analysis of a representative safety enhancements, the team generated set of events identified specific problems three airplane design safety enhancements The airplane state awareness safety and major themes and resulted in proposed that the Commercial Aviation Safety enhancements are integrated into a interventions that cover a broad spectrum of Team has adopted and that Boeing and coordinated safety plan with a goal of potential solutions in the areas of airplane other Commercial Aviation Safety Team– balancing short-term tactical mitigations design, flight crew training, airline operations represented airplane manufacturers have provided by operational and training and maintenance, and safety data. committed to implementing on their next programs with longer term, more strategic all-new type designs: solutions resulting from improved design. The Commercial Aviation Safety Team has n Flight envelope protection. This safety now officially adopted the resulting safety enhancement has already been implemented The airplane state awareness safety enhancements and is working to implement by Boeing on its latest fly‑by-wire commercial enhancement portfolio was constructed by them in the United States and worldwide. airplanes, the 777 and the 787. the Airplane State Awareness Joint Safety Implementation Team to provide both near- Credit: Copyright, Boeing. n Bank angle alerting with recovery and far-term solutions that reinforce each guidance. Boeing is now working to other and provide a balanced, redundant implement this safety enhancement in approach to addressing the issue of flight the 737 MAX and the Next-Generation crew loss of airplane state awareness. Like 737 (see fig. 4). the underlying problem being solved, the solution set is complex and addresses n Virtual day-visual meteorological multiple issues. The analysis estimates that conditions displays. Boeing’s commitment implementation of the training, operations, is contingent on successful completion of and airplane design safety enhancements relevant research and development and would reduce the risk of future airplane state supporting industry standards. Boeing awareness events approximately 70 percent recently demonstrated these displays, also by 2018 and 80 percent by 2025. referred to as synthetic vision systems, in the 787 EcoDemonstrator. Because these The Airplane State Awareness Joint Safety displays are effective at supporting flight Implementation Team recommended crew attitude awareness, Boeing continues adoption by all U.S. Commercial Aviation to engage with government and industry Safety Team members of the training,

20 100 focus autumn 15 Aircraft encounters with weather 100 balloons: risks and mitigations by R W Lunnon: Royal Meteorological Society

1. Background There have been a number of incidents stemming from aircraft encounters with airborne objects similar to balloons, where the pitot systems on the aircraft have been affected. As far as is known, none of these encounters have been with radiosonde balloons, and it is not clear, given that a radiosonde balloon is designed to burst, that such a balloon poses a threat to the pitot system and other measurement systems on aircraft. This study considers the threat from radiosonde balloons and mitigations: one of the mitigations applies to all balloons and other causes of problems with pitot systems (although this mitigation can be considered to be a stand alone topic).

2. Current use of radiosondes in the British Isles and elsewhere Radiosonde stations in the British Isles fall into 3 categories. Reference stations (Camborne, Lerwick, Valentia) release radiosondes twice Figure 1. Shows current Met Office Radiosonde launch sites daily, at 2315 GMT and 1115 GMT. Automatic stations (Herstmonceux, Watnall, Albermarle and Castor Bay) release radiosondes daily, at The nominal ascent rate of radiosonde 2.1 Movement of small balloon in 2315 GMT. MOD stations (Larkhill, Aberporth balloons is 1000 feet/minute (between 5 and proximity of an aircraft and South Uist) release radiosondes as/when 5.5 m/s). This figure can be used to quantify It is noted that while there have been needed to support trials (e.g. artillery at the risk of an encounter at a particular time a number of “near misses” between Larkhill). Thus there are no regular releases at a particular flight level. weather balloons and aircraft, there of radiosondes along the SSE/NNW axis have been no reports of collisions. It of Britain except at night when domestic Use of radiosondes in other parts of the world is of importance to understand why passenger flights are minimal. follows a similar pattern to that in the British this might be. Immediately ahead of Isles. Radiosondes are rather expensive and an aircraft in flight there is a “nose” of Information on the web, for example, http:// it is much more cost effective to obtain air where the pressure is higher than badc.nerc.ac.uk/data/radiosglobe/. wind, temperature and is possible humidity would otherwise be the case. This nose html (which as advertised in an FSB article information from commercial transport serves to deflect the air well ahead of on In Flight Impacts) imply that there are 30 aircraft. Therefore the use of radiosondes in the aircraft round the fuselage of the launch sites in the UK, whereas in fact there areas where there is dense commercial air aircraft. Figure 2 shows the trajectories are 9 as described above. That web link states traffic will tend to be avoided at the times of of (a) air (in blue), (b) a hypothetical that there are 200 sites across Europe: this day when air traffic density is at its highest. heavier than air object (such as a UAV) figure is almost certainly too high. (in red) and (c) a hypothetical lighter than air object (such as a small helium

HOW MANY DO YOU REMEMBER? 09:19 Page 1 ght Safety iss 71 18/8/15 35096®Fli

ISSUE 71 SUMMER 08 SUMMER

ON COMMERCIAL AVIATION SAFETY

ISSN 1355-1523

The official publication of the United Kingdom Flight Safety Committee

100 focus autumn 15 21 100

a dedicated App would be very helpful in implementing this mitigation.

4. Mitigation 2 – diagnosis of position of radiosondes Radiosondes routinely broadcast their position (along with other met data such as temperature) and do so in one of only two frequencies – 403Mhz or 1680MHz. There is nothing in principle to prevent a suitably equipped aircraft “listening in” to the transmissions of any radiosondes within radio range. The position information could then be fed into a system such as TCAS which could then provide advisories (and other warnings) to the pilot recommending changes of flight path which would enable the aircraft to avoid the radiosonde. It is noted however that there are considerable cost implications in equipping aircraft to Figure 2. Shows aircraft and fuselage (lower part of figure) together with receive these frequencies. trajectories (relative to the aircraft) of various objects which the aircraft is approaching. The blue lines indicate trajectories of the free air, the red line indicates the trajectory When this material was presented (at SIE of a small heavier than air object, and the green line indicates the trajectory of a small meeting in January 2015) it was suggested lighter than air object. that radiosondes could be fitted with transponders so that TCAS systems could balloon) (in green). As can be seen 3. Mitigation 1 – prediction of position interrogate them. It is understood that the lighter than air object swerves well of radiosondes fitting radiosondes with transponders is not out of the way of the aircraft, whereas Radiosondes are released from well defined technically very difficult. However it would the heavier than air object follows points at predictable times. Assuming they require ICAO to make an approach to the a trajectory which is much straighter are filled with a pre-set quantity of helium, World Meteorological Organisation (WMO). than that of the air, and the probability their ascent rate is predictable. Therefore The transponder approach is the solution of a heavier than air object hitting a the trajectory (in 4 dimensions) of the preferred by some in the aviation community. sensor such as a pitot system would radiosonde is largely predictable – it depends be relatively high. The figure does on the wind at levels from the surface to the 5. Mitigation 3 – reduced reliance on not tell us much about a large lighter level of interest. In principle an airline with information than air object, such as a balloon at, access to forecast winds generated by the As mentioned in the introduction, this say, 35000 feet, but it is clear that Met Office could predict the trajectory of mitigation can be seen as something of a such an object would distort and divert any radiosonde anywhere in the world. standalone topic compared to the previous so that the probability of a collision The involvement of Air Traffic Management mitigations but is included in this paper for would be much lower than observations service providers in the provision of completeness. It is helpful to bear in mind that of such an object from the cockpit predictions of radiosonde predictions is the most important issue is to recognise when of an approaching aircraft would recommended. One possible scenario is pitot systems are not performing nominally suggest. Note that the combination that individual Met Services who release and thereafter the emphasis should surely be of a helium balloon and radiosonde radiosondes provide predictions of their on immediate achievement of safe flight. instrumentation is significantly lighter positions to ATM providers controlling the than air: at aircraft cruising altitudes airspace through which the radiosondes Radiosonde balloons, and other similar the combination is still ascending at are expected to pass (this would take into objects, pose a threat because of the risk of approximately 5m/s which would not account the three-dimensional structure of affecting the determination of airspeed using be possible if the combination had the airspace). The ATM providers would then pitot systems on aircraft. same overall density as air. vector aircraft round any radiosondes in their airspace. The use of new software such as

22 100 focus autumn 15 100

5.1 Other threats to measurements by 5.3 Accuracy of upper level wind 5.4 Indicated airspeed and true airspeed Pitot systems forecasts In most contexts the critical quantity A number of mechanisms can affect the Upper level winds are the most accurate that a pilot will refer to is indicated performance of Pitot systems. These forecasts the Met Office produces airspeed rather than true airspeed. In include (compared to natural variability) and order to convert between the two it is (a) Icing, as affected flight AF447 RMS errors have approximately halved in necessary to make reference to outside (although note that icing can be a the last 20 years. Statistics on accuracy air temperature and barometric pressure. temporary problem as in the case of are available on the Met Office website. Although it does not follow that if AF447, and that the solution can be Currently 24 hour forecast winds at FL390 the pitot system was not performing simply to carry on flying more or less have a RMS vector error of 3m/s for the nominally anomalous measurements as before) zone north of 20oN. This figure applies to would be made by the air temperature average wind over 10-20km: for shorter sensor and/or the static pressure sensor, (b) Volcanic ash distances there will be larger errors. it is certainly true that air temperature Shorter range forecasts have smaller sensors are prone to icing problems and (c) Bird strikes errors. The 3 m/s figure is for vector error: foreign objects could affect any sensor. for a single component the RMS error will However forecast information is available (d) Foreign objects be 3/√2 which is approximately 2m/s. on both temperature and the geometric height of flight levels. Temperatures have (e) Balloons and other airborne objects It is clear that in the absence of any of a RMS error of 0.7 degrees which would made of rubber, e.g. banners the effects (a) to (e) above, airspeed is give rise to a true airspeed error of less more accurately determined from the than 1m/s. The forecast true heights of If a mitigation can be developed which pitot system. However, in the presence or flight levels are also broadcast as part works through reducing dependency on suspicion of any of the effects (a) to (e) of the services provided by World Area pitot systems, then this can be applied above, use of forecast wind data coupled Forecast Centres. It is possible to combine to the other causes of pitot unreliability. with ground velocity information from the forecast heights with the geometric on-board sources can significantly reduce aircraft height derived either from the 5.2 Accuracy of components of the uncertainty. For example, if the two pitot IRS or GPS to derive the flight level of the “wind triangle” systems give different figures for airspeed, aircraft without reference to the static In the absence of any of the effects in many cases it should be possible to pressure. It follows that if all relevant (a) to (e) above, airspeed has a typical decide which of the two systems is more forecast information was available on the RMS error of ~1m/s. The accuracy of accurate using forecast wind information. flight deck, an aircraft could fly without the ground velocity vector is also very This was a noted aspect of flight AF447. pitot systems, outside air temperature good, using a combination of Inertial For the period between 2:10:04 and sensors or static pressure sensors. Reference Systems (IRS) and Satellite 2:10:26 the two computed airspeeds Navigation Systems (typically GPS) were significantly different 40% of the 5.5 Practical use of forecast wind, giving a typical RMS error in either of time; for the period between 2:10:26 temperature and geometric height the components of the vector of ~1m/s. and 2:10:50 the two computed airspeeds information Aircraft heading is a significant source were significantly different 70% of the If there was a sudden malfunction of the of error in determining the wind vector time; for the period between 2:10:50 pitot system giving rise to anomalous as derived from airspeed and ground and 2:11:46 the two computed airspeeds airspeed readings, it is unlikely that a velocity, and this has a typical RMS error were significantly different 30% of the pilot who had never made use of forecast in either of the components of ~1.5m/s. time. (See figures 26 to 28 of the BEA wind information on the flight deck final report). would be able to solve wind triangles

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 23 100

and derive the aircraft’s true airspeed. 5.7 Specific recommendations on the 5.10 Appendix 1: statistics Therefore it is recommended that pilots use of forecast data Earlier a RMS vector error of 3 m/s practise accessing the required data and If errors are to be kept to a minimum, it is was quoted. In this section data are performing the requisite calculations in essential to use the scientifically correct provided which make it easier to order to fly the aircraft safely. In an era approach to utilising the forecast data. interpret this statistic. In general, errors of highly automated aircraft, a significant In general upper air forecast data are in forecast wind components satisfy a role for the pilot is understanding provided on a 4-dimensional grid and normal distribution. This enables us to anomalous indications and taking it is necessary to apply 4-dimensional quantify the risk (probability) of a wind appropriate action – “debugging the interpolation to obtain the correct component with an error exceeding a aircraft”. This is made much easier if the forecast value at the current position specified threshold. Specifically we can pilot has a good appreciation of plausible and time of the aircraft. Data used in say that the probability of a wind error values of relevant parameters – in this flight plan calculations often assume exceeding three standard deviations case the sides of the wind triangle, and a specific take-off time and a specific (6.3 m/s) is 0.001. Clearly lower if necessary, both indicated and true trajectory in 4 dimensions. Therefore probabilities apply to larger errors. The airspeed and the relationship between if wind data are only available for probability of an error in excess of 50 geometric height and Flight Level along the flight planned route, these may knots is less than 10-12. the expected trajectory of the aircraft. well be inadequate in the event of a Clearly there is a role for Electronic pitot malfunction if the aircraft has 6. Appendix 2 – extension of mitigation Flight Bags here, enabling some of the departed from the planned route in 3 to Angle of Attack sensor problems more challenging calculations to be any way. Therefore it is essential to The following comments are provided in the performed and providing plausible limits have available on the flight deck wind context of an incident to a Lufthansa A321 for unfamiliar parameters. information for a range of latitudes, near Bilbao on 5/11//2014. In the incident longitudes, altitudes and times covering both AoA sensors experienced icing and both Note that applying some common sense both the expected route and a range of transmitted incorrect, but very similar, values rules about Power/Attitude and altitude plausible reroutes. to the on-board computers. As a result flying could be as effective as application of the aircraft was made very difficult and a the wind triangle. 5.8 Training for pilots on flight with 4000 foot loss of altitude was experienced. unreliable airspeed indication 5.6 Altitude considerations when flying There is considerable reference to this in AoA sensors when operating nominally with imperfect airspeed information the report on the accident to AF 447. In measure the direction of flow of air relative to Generally speaking at a specific gross particular there are three appendices: the fuselage (in the relevant plane). If it were weight and altitude, there is a range of Appendix 5: Air France “Vol avec IAS possible to predict the vertical motion of the airspeeds at which an aircraft can safely douteuse” procedure atmosphere and it was possible to determine fly. If the pitot system is performing Appendix 6: Airbus “Unreliable speed the pitch angle of the aircraft (in addition to all nominally, an aircraft can fly safely close indication” procedure the quantities needed to determine airspeed) to the ceiling altitude appropriate to the Appendix 7: Extracts from Air France then it would be possible to diagnose AoA current gross weight. In the event of briefing brochure (“IAS douteuse” from that information. The Met Office (and using forecast wind vector information exercise) other meteorological centres) predicts the to determine airspeed, it is probable It was noted that all three pilots had vertical motion of the atmosphere. However, that an aircraft should fly at a lower undertaken simulator training on IAS the predictions do not form part of the service altitude so that the actual airspeed flown douteuse. However, the pilots apparently provided as a World Area Forecast Centre by the aircraft lies within safe limits did not apply common sense rules for aviation. Additionally, as there are no even though there are errors in the about Power/Attitude and altitude, as routine measurements of vertical motion of diagnosed airspeed arising from the use recommended earlier. the atmosphere there is uncertainty as to how of the forecast wind. The calculations accurate these predictions are. performed by the pilot in “practice 5.9 Comment on necessity for good mode” as described in the previous measurements of airspeed section should include consideration of Current accuracy of forecasts of upper Acknowledgement any altitude changes required in the level winds from the Met Office event of pitot malfunction. depends critically on the availability of This paper has been improved as a result of accurate measurements of wind vector, review by Alex Fisher. particularly automated reports from aircraft. These in turn depend critically on accurate airspeed measurements.

24 100 focus autumn 15 100 The Dryden Accident – A Driver for Chance by Dai Whittingham, Chief Executive UKFSC

n 10 March 1989, C-FONF, a Fokker inquiry also acted as an inquest on behalf of as the day progressed. Terminal weather OF28-1000 Fellowship operated the Coroner’s Office. forecasts for and were by Air took off from Dryden available to the crew before their departure. Municipal Airport, an intermediate stop The facts These forecasts indicated conditions that on its journey from Thunder Bay to could potentially deteriorate to below the Winnipeg. The aircraft crashed after only Air Ontario had been formed 2 years earlier captain’s limits at their scheduled 15 seconds of flight. Of the 65 passengers from a merger between 2 other operators, one arrival times.” Fuel requirements for the and 4 crew, 44 passengers and one flight running a commuter operation in southern alternate and an upload of 10 passengers attendant survived; the FDR and CVR data Canada and the other effectively a bush meant that, most unusually, refuelling at were destroyed in the post-impact fire. operation in the northern parts. The northern Dryden would be required. The subsequent investigation uncovered a routes had been sold off to another entity catalogue of human error, organisational but a commercial arrangement remained When the crew checked in at Winnipeg, they and regulatory shortcomings, and for scheduling purposes. Air Ontario was discovered that their aircraft had no APU; generated 191 recommendations, many operating essentially as a feeder for Air the equipment had been malfunctioning for of which were addressed to the operator Canada’s national network. The F28 captain the previous week and rectification work and . was a product of the commuter operation had proved unsuccessful. A management but his experience was mainly piston/turbo, decision had been taken the previous The investigation was unusual in that it was though he had also flown the Gulfstream day deferring the defect until the aircraft conducted by a juducial commission led by II. The first officer was from the northern returned to Toronto later on the night of the the Hon Virgil Moshansky, a Justice in the operator and his experience was also mainly accident. Unfortunately, Dryden lacked the Alberta courts. Judge Moshansky had specialist piston/turbo, albeit with some Cessna GSE needed for a ground start, which meant accident investigators as part of his team, but Citation time. The F28 was the largest jet the crew would need to keep an engine his inquiry also considered the role of Air aircraft either had flown in commercial running throughout the refuelling process. Ontario and Transport Canada in some detail. service and, at the time of the accident, As a further consequence of the decision the Pertinent to the current debate about release their combined experience on type was less aircraft could not be legitimately de-iced or otherwise of the Sumburgh accident FDR than 150 hours. The lack of type experience at Dryden because a proscription had been and CVR data, Moshansky was also faced with was the subject of 2 recommendations to published in both a Fokker aircraft winter questions about disclosure of confidential Transport Canada. operations bulletin and an Air Ontario safety and other data in support of his operational directive against de-icing the investigation; he determined that the public Air Ontario Flight 1362/1363 was scheduled F-28 aircraft with one or both engine(s) interest demanded the release of information to fly a return Winnipeg to Thunder Bay, running. Although could for accident investigation purposes even if it with intermediate stops at Dryden. The same be expected, the captain did not ask for had been provided in confidence for safety crew were then to operate as Air Ontario a waiver to allow C-FONF to be de-iced. and accident prevention work. Unusually for Flight 1364/1365, a Winnipeg - Thunder Bay However, the captain did require the aircraft an air accident investigation, and perhaps return with no intermediate stop, a duty day to be de-iced before departing Winnipeg. driven by the different nature of judicial of less than 10 hours. Moshansky reports The weather at Dryden was within the inquiries, Moshansky’s report includes the that “The area weather forecasts for the captain’s limits for the first leg but at that names and positions of all those involved day’s operations showed generally unsettled stage Thunder Bay was unsuitable. in the accident and every witness who gave and deteriorating weather, including lowering evidence to the inquiry commission. The cloud ceilings and freezing precipitation

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 25 100

Commercial pressure 170” at 4 knots.” The weather reports for with and the other a captain Thunder Bay and the alternate also included from another fleet within Air Ontario. Both The first leg was unremarkable, though an freezing rain throughout the period, and the described ¼ to ½ inch of snow on the wings, engine remained running throughout the investigation concluded both the operations one describing the snowflakes melting and time on the ground at Dryden. The second control system and the crew should have adhering to the wing surface. Both presumed leg to Thunder Bay went ahead after a been aware of the potential exposure to the captain would have the aircraft de-iced short delay on the basis of an improving airframe icing on the ground. before take-off and neither brought their forecast which subsequently proved to be concerns to the attention of a crew member. correct, arriving 20 minutes late. After The hot refuelling was conducted by However, the surviving also refuelling at Thunder Bay the captain was personnel inexperienced with the F28 and noticed the snow and said nothing. This informed that he would be carrying an with passengers still embarked; moreover, was also the subject of a recommendation, extra 10 passengers and that the flight was the fuel was Jet-B which is flammable at namely that captains should have a duty therefore overweight. He decided to offload +1C. This practice was condemned as unsafe to check, or have checked, wing surface the additional passengers but was over-ruled by Moshansky and his early recommendation conditions in the event of a report from a by the duty manager, who decided that the for a prohibition was immediately accepted crew member; this also implies that crew weight would be adjusted by defuelling, by Transport Canada. There were also members should have an understanding of imposing a further 35 minute delay. At this recommendations relating to training for the dangers of ice and snow contamination stage the crew were coming under pressure refuelling personnel. and this was duly recommended as a matter from passengers concerned about their of urgency in the interim report. connecting flights. There was evidence given at the inquiry to suggest that the various Pre-take off phone calls in connection with the additional The take-off passengers and the overweight situation The captain initially remained on the aircraft had left the captain angry and frustrated; during the refuel before heading to the As the aircraft began its take-off roll it was there had been a noticeable change in his terminal to call the ops controller. He was snowing heavily. The off-duty Air Ontario demeanour during the Thunder Bay stop. seen to walk quickly back to the recalled seeing about 10-20% of the but it was observed by one survivor that he snow blowing off the surface as the aircraft The flight release for the return leg to “rather looked disgusted ... just not a happy gathered speed, with the rest changing colour Dryden anticipated an engine-running refuel expression”. Neither of the pilots was seen and texture. The other pilot recalled seeing (hot refuel). There had also been a new TAF carrying out a walk-round inspection of the the snow crystallising into ice and observing issued for Dryden which forecast freezing aircraft. Before the aircraft taxied it was about half an inch of slush on the runway. precipitation, whereas the previous forecast snowing quite heavily and its movement Witnesses on the ground reported the aircraft had not; it is not known whether the crew was delayed briefly by the arrival of a C172 as being slow to accelerate but it lifted off read the revised forecast, though it was on a recreational flight, the pilot reporting near the 5700ft point of the 6000ft runway. certainly available to them. The aircraft that he was having severe difficulties in The initial rotation produced buffet (observed arrived at Dryden about 1 hour behind the snow. Radio transmissions from the by the Air Canada pilot) and the attitude schedule. On approach in VMC, the runways aircraft further reflected frustration on the was reduced before a second rotation; it were reported as being bare and dry, though part of the captain. At this stage the snow is probable that the aircraft was airborne light snow grains had been observed to the was reported to the investigators by the briefly after the first rotation but settled back west. It began to snow lightly soon after Cessna pilot as being ‘heavy and wet’. The onto the runway. C-FONF gained very little the aircraft touched down. During the 30 first officer, responding to the instruction altitude, and some minor roll excursions were minutes the aircraft was on the ground at to hold position, acknowledged but advised seen before it hit trees and crashed 3000ft Dryden conditions deteriorated significantly; that vis was “down to half a mile in snow”. beyond the end of the runway. 21 minutes prior to its last take-off the actual The Cessna pilot also observed that the first report was “Sky partially obscured, estimated portion of the runway was contaminated ceiling 4000 feet overcast, visibility 2% with around half an inch of slush. The aftermath miles in light snow, wind 260”T at 3 knots” whereas only 18 minutes later the conditions A number of passengers later reported seeing The Dryden Crash Rescue and Fire (CRF) were: “Precipitation ceiling 300 feet, sky snow and ice on the F28 wings prior to take- service responded but one of the vehicles obscured, visibility 3/8 mile in snow, wind off, including 2 off-duty pilots, one a captain was being refilled having been used to wash

26 100 focus autumn 15 100

down a small fuel spill that occurred during The first 171 recommendations in the report who are responsible for or associated with the hot refuelling of C-FONF. When the were all addressed to Transport Canada. They aviation programs are aware of the basis for initial personnel arrived near the crash site range from the hiring of suitable experienced and the requirement to support policies that a few minutes after the crash, the first staff to the need for adequate oversight affect aviation safety”. survivors were struggling through deep snow of commercial operators, and include the towards the airport. Having ascertained that need to review and revise its operator Transport Canada is a very different the chance of extracting survivors from the audit process. One recommendation was: organisation from the one that existed at now-burning wreckage was zero, the CRF “That Transport Canada establish a policy the time of the Dryden accident. Air Ontario initially concentrated on dealing with the that identifies surveillance of existing air also underwent significant change and after injured. The inquiry found that the failure carriers as a non-discretionary task.” A large a period under direct Air Canada control is to deploy hand-lines to the accident had no number of recommendations pertained to now a part of . The full Moshansky report bearing on the survival prospects of those Air Ontario and the F28 programme, where can be viewed at http://publications.gc.ca/ who died in the accident. However, it became shortcomings in training and equipment, collections/collection_2014/bcp-pco/CP32- clear that the accident was actually in the coupled with inappropriately qualified or 55-1-1992-1-eng.pdf area for which the Dryden municipality had experienced managers, had led to a situation responsibility, and there were command and where commercial pressures had over-ridden control issues that would not have occurred good safety practice. had realistic training exercise been held; this was again the subject of a recommendation. Two recommendations sum up the tenor of the report:

The regulator “Transport Canada put in place a policy directive that if resource levels are This accident was notable for the number of insufficient to support a regulatory or recommendations made of Transport Canada other program having a direct bearing on as the regulator. In ascribing probable cause, aviation safety, the resource shortfall and Moshansky stated: “…the pilot-in-command, its impact be communicated without delay must bear responsibility for the decision to to successively higher levels of Transport land and take off in Dryden on the day in Canada management until the problem is question. However, it is equally clear that resolved or until it is communicated to the the air transportation system failed him Minister of Transport”. by allowing him to be placed in a situation where he did not have all the necessary tools “Transport Canada establish a mandatory that should have supported him in making education program to ensure that senior the proper decision.” managers and officials of the department

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 27 100 Members List

Members of The United Kingdom Flight Safety Committee

FULL MEMBERS Baines Simmons Ltd Chaz Counter Rachel Cook Chairman easyJet BALPA Dubai Air Wing Capt. Chris Brady Zoe Reeves Rory Fagan

Vice-Chairman Belfast Intl. Airport UK Ltd Cranfield University Alan Whiteside Ian Smith Cenqiz Turkoglu bmi regional easyJet Treasurer Martyn Sisson Capt. Chris Brady Flight Data Services Ltd Derek Murphy Bond Offshore Helicopters Flight Data Services Ltd Richard Warren Derek Murphy Executive Board CAA Rep Rob Bishton . Capt. Adrian Bateman Capt. Natalie Bush Non Executive Board Member AIG Europe Ltd British International Anthony O’Keefe GAMA Aviation Jonathan Woodrow Nick Mirehouse Acropolis Aviation CAE Oxford Aviation Academies Chris Seal GATCO John Morley Sian Ryan A|D|S Capital Air Services Gatwick Airport Ltd Mick Sanders John Hill Jerry Barkley Aegean Airlines CargoLux Airlines Capt. Stavros Christeas Mattias Pak GE Aviation Roger Katuszka Aer Lingus Cathay Pacific Airways Capt. Conor Nolan Rick Howell Gulf Air Co Capt. Khalil Radhi AIG Europe Ltd Jonathan Woodrow Stephen Morris Ideagen Gael Ltd Steven Cespedes Airbus S.A.S Charles Taylor Adjusting Harry Nelson David Harvey Independent Pilots Association Capt. James West Air Mauritius CHC Scotia Capt. Francois Marion Mark Brosnan Irish Aviation Authority Capt. Dermot McCarthy ALAE CityJet Ian Tovey John Kirke Jet2.com Keith Irving Ascent Flight Training City University London Stewart Wallace TBA L-3 CTC Ltd Capt. Phillip Woodley ASL Airlines Ireland Cobham Aviation Services Jack Durcan Scott Carmichael LHR Ltd Ian Witter AVISA University Dr. Mike Bromfield Phil Stuckle Neil Heron BA Cityflyer Cranfield Safety & Alan Taylor Accident Investigation Centre Dr. Simon Place London’s Air Ambulance Dave Rolfe BAE SYSTEMS Reg. A/C Alistair Scott DHL Air Shayne Broad London City Airport Gary Hodgetts

28 100 focus autumn 15 100

Manchester Airport plc TAG Aviation (UK) Ltd QinetiQ Eng. Rory McLoughlin Malcolm Rusby Rupert Lusty

McLarens Aviation Teledyne Controls RAeS John Bayley Mark Collishaw Maurice Knowles The Honourable Company of Air Pilots RAeS Eng. David Copse Capt. Alex Fisher John Eagles National Police Air Service (NPAS) UK Ltd TAM Brazilian Airlines David Taylor Terry Spandley Capt. Mauricio de Faro Pires Navtech Thomson Airways TAM Executiva Harriet Dalby-Quenet Dimuthu Adikari Capt. Castro Panasonic Avionics CO-OPTED ADVISERS Bob Jeffrey Dominic Perrin Pen Avia UTC Aerospace Systems AAIB Capt. John O’Connell Gary Clinton Capt. Margaret Dean PrivatAir Vistair CAA Julie Biringer Stuart Mckie-Smith Brian Watt - Flight Operations RTI VLM Airlines NV CHIRP Steve Hull Karen Poels Air Cdre. Ian Dugmore

Rolls-Royce Plc West Atlantic UK GASCo Phillip O’Dell Alex Wood Mike O’Donoghue RVL Group Legal Advisor GROUP MEMBERS Michael Standen Edward Spencer Holman Fenwick Willan LLP Ryanair Air Tanker Services Ltd Jim Lynott Dale Grassby NATS Robert Luxton Karen Bolton Seaflight Aviation Ltd Dimitris Kolias MOD Representatives Royal Met. Society Capt. Jerry Boddington RN - MAA Deputy Robert Lunnon Shell Aircraft Intl. Head Analysis & Plan Keith Curtis Wg Cdr Andrew “Errol” Flynn - MAA UK Airprox Board Engineering Oversight & Assurance Air Cdre. Steve Forward SMS Aero Ltd Cdr Ben Franklin - Royal Navy Ian Chapman Lt Col Gavin Spink - Army Gp Capt Brian James - RAF Stobart Air Capt. Clive Martin QinetiQ Flt Lt Tim Burgess Superstructure Group TBA

HOW MANY DO YOU REMEMBER?

100 focus autumn 15 29 Printing in Woking since 1977

BUILDING ON TRADITIONAL VALUES

Have you heard the phrase? “If you want something done, give it to a busy person!”

At Woking Print we dealt with over 6,000 enquiries for our clients last year, most of them urgent. Although busy we delivered a quality job, on time, every time.

Woking Print wish to congratulate UKFSC on their 100th Issue of FOCUS.

Tel: 01483 884884 [email protected] www.wokingprint.com

LITHO | DIGITAL | DESIGN | FULFILMENT BROCHURES | STATIONERY | FORMS | LEAFLETS