Formal
CRM training started sometime in the 70s; the first being KLM that introduced a
human factors training programme, based on the Edwards SHEL model and the
trans-cockpit authority gradient. Accidents involving fully airworthy aircraft
were perplexing the aviation community – the most notable example being the
collision of two B-747s in 1977, while on the runway at Tenerife. Accidents
like these were pointing towards the breakdown of crew co-ordination and
communications between the cockpit crew. It was finally the NASA workshop of
1979 where the role of human factors in aviation accidents came in to sharp
focus. Participants at this conference were convinced that formal training in
crew co-ordination was required to reduce accidents due to human error. The
origin of CRM (initially called Cockpit
Resource Management and later changed to Crew Resource Management in
recognition of the complex nature of aviation operations - where human errors,
anywhere in the chain, can lay dormant for extended periods until an active
failure at the hands of pilot leads to an accident) training is now
universally traced to this NASA workshop of 1979. As more knowledge was gained
on the subject the training has undergone many generational changes, from the
first generation CRM to the present day sixth generation CRM, also called
Threat and Error Management. CRM training is crucial for flight crew (besides
for others too who are connected with aviation) as they are aviation’s last of
line of defence to prevent an accident from happening.
Amongst the flight crew, the Commander of the
aircraft has been given the final authority with regards to the safety of the
aircraft, and its contents, from the time he/ she assumes command until he/ she
relinquishes the command of the flight. This is elaborated in Rule 141 of the Aircraft Rules, which states
that “The Pilot-in-command (PIC) shall have final authority as to the
disposition of the aircraft while he is in command.” The rule also states that
the PIC “shall supervise and direct the other members of the crew in the proper
discharge of their duties in the flight operations.” It further clarifies that
“In addition to being responsible for the operation and safety of the aircraft
during flight time, the Pilot-in-command shall be responsible for the safety of
the passengers and cargo carried and for the maintenance of flight discipline
and safety of the members of the crew.” This is indeed a very onerous task for
any one individual, especially one who is flying above mother earth, in the
third dimension. Why?
It is
because the PIC is human, and has all the capabilities and limitations like any
other human being. These capabilities and limitations have been studied by
experts in the respective fields. These are required to be known by pilots to
get any aviation licence, and pilots are also being tested for the same, before
issue of the licence. ‘Human factors’ is the term designated by the ICAO for
this important area of study by pilots (and lately by others like ATCOs,
Maintenance too). This article is not going to go into details of the
physiological factors like hypoxia, hyperventilation, decompression sickness,
fatigue, sleep, etc. or the factors like illusions that most of the pilots are
familiar with. The author has participated/ facilitated CRM training and always
felt that the training mainly focussed on the ‘What’ and ‘How’ of CRM, after
briefly telling the participants about ‘Why’ it is important for the aviation
industry to have an even lower accident rate. It is the author’s view that if
the CRM training focussed more on convincing the participants of the ‘Why’ it
is important for the participants, then the person would be more receptive, and
self motivated to listen to the ‘how’ and ‘what’.
This
article will attempt to make a strong case for ‘Why’ CRM is important for any
crew, and especially so for the Commander because it is the Commander who
finally signs on to the task of
undertaking a safe flight from A to B, with the help of the crew, of course. It
seems simple and every Commander intellectually understands its importance, but
this fact needs to be internalised and acted upon if we want to prevent
accidents like the one at Mangalore, and others around the world, where in fully
serviceable aircraft met with fatal accidents.
It is a fact that with the new generation of
highly reliable aircraft and engines, the major cause of accident is due to
factors that are termed as human error. When an accident does[J1] take place due to human error, it was
found that the accident happened due to the active failure in the hands of the
pilot, but a number of passive errors had happened at various stages in the
life cycle of the aircraft viz., the design, manufacturing, loading, servicing
stage, etc. or have been caused due to an error by the regulator, company
management, ATC, dispatch or by the meteorologist. The other fact is that in
some cases the PIC was the final person who could have prevented the accident
from happening. A human error accident in the hands of the pilot is an
acceptance by industry professionals that the accident could have been
prevented; thus a human error (aircrew) is more appropriately a ‘pilot
preventable’ accident. This brings us to larger questions – Firstly, Do human
beings like to err? And the answer is ‘No’. Secondly, do humans err? The answer is ‘Yes’. Think of the number of
times one tries to open the lock with the wrong key, or punches in the wrong
password. This implies that although no human likes to err, ‘to err is human’. This also implies
that PICs, being human, are also prone to errors. Why?
Research
in to the human brain has brought out a large number of limitations of the
brain. The following are some of the important memory (storage and retrieval of
information) problems that most human beings, including pilots, suffer from: -
- Absent mindedness – Forgetting due to
lack of attention.
- Blocking – Temporarily forgetting – “did
he clear us to land?”
- Transience – Forgetting information
with time – “What is the Approach frequency…?
- Misattribution – Forgetting the source
of the information.
- Suggestibility – Developing a false
memory because of new information received during retrieval.
- Bias – Unconscious reshaping of memory
due to personal beliefs or mood.
- Persistence – Negative distortion of a
memory of a traumatic event.
- Memory changes – Memory changes from
person to person, and also within the same person due to reasons like
physical & emotional health, stress, quality and quantity of sleep,
diet and age.
- Inattention blindness - Attention
resources are very limited and it is a known fact that things to which we
are not paying attention to are not perceived; more importantly if we are
devoting attention to one task, then we may not have adequate attention
resources for other maybe more important primary tasks – fixation – a
cause factor in many aviation accidents. Inattention blindness is affected
by the following factors:
- Conspicuity – All warnings in the
cockpit are designed keeping this in mind.
- Mental workload and task interference
- Low workload and the effects of
automation – Low arousal; low performance.
- Limited processing capability of the
brain as compared to large acquiring capability of the five senses. In
aviation eyes, ears (hearing and vestibular apparatus) and seat of the
pants are the most relevant sense organs. For example in the human visual
system, the amount of information coming down the optic nerve is estimated
to be in the range of 10 million bits per second. This far exceeds what
the brain is capable of fully processing and assimilating; about 16 bits/
second (max 40 bits/ second). Guess why two people looking at the same
scene would pick up different images – this is primarily based on what is
being paid attention to by each of them.
- Limitations (in terms of time, and
capacity) of the working memory (like the RAM in the computer) in humans.
These severe limitations of the human
information processing mechanism would convince any human about what he is up
to when he/ she is interacting with the outside world and trying to make sense
of what is going on, or in aviation parlance when he is trying to get and
remain situationally aware.
This already difficult situation becomes
even worse when the same set of information is to be processed under time constraints, like in a non-normal or
unanticipated situation in the air. Being situationally aware at all times is
very important for a crew primarily because without being situationally aware
it would not be possible to take the right decisions every time. Poor
situational awareness leads to bad decisions; a situation that is detrimental
to the task at hand – of ‘flying the
aircraft safely and efficiently from one place to another’ – the primary
task of the PIC. How do we ensure that we are situationally aware at all times,
or regain situational awareness at the earliest, if it is lost due to any
reason?
This requires the application of crew
resource management – “the effective use of all available
resources: human resources, hardware, and information.” The PIC is the final decision maker in the
air, but knowing his/ her limitations, he should be the one who makes all
efforts to use all available resources to become situationally aware, and there
after go on to making the right decision. This sounds simple and logical but a
glance through the history of aviation accidents/ incidents would be able to
convince any pilot that this is not so. The reasons are many. The reasons could
be attributed to the way in which an average human grows up in this world – our
formative years in school where we are taught to excel as individuals and
compete with each other to stand first in class/ sports etc. The same pattern
continues through college, and life there after – our individual successes are
celebrated. When we enter aviation, our first solo is the most celebrated event,
and solo flights are cherished by ‘real pilots’. ‘The individual is required to be competitive, and is complete’, is
the message that life has given us thus far. Even our traditional checks and
proficiency testing in aviation are done on an individual basis – testing is of
our skills as individual pilot’s (a column for CRM has also been added to the
check proforma to assess the pilot’s CRM behaviour). The early pilot’s with the
leather jackets, white silk scarves and glasses is the macho image of a pilot,
which even today is ingrained in many individual pilot’s minds, some of whom
have never been close to a fighter. The author flew single seat fighter
aircraft for over two decades and can vouch for the fact that he always felt
complete – missed out a few checks here and there; made approaches on the wrong
runway; had ‘action total’ on first solo; miscalculated fuel in the air while
doing low level navigation, etc. Single seat fighter flying accepted a certain
amount of risk due to the nature of the task, but commercial aviation is
required to be safe, and risk free. Commercial pilots have to work with a
different mindset where the aim safe and
efficient flying has to be kept paramount.
Commercial pilot’s need to be convinced
about using CRM techniques, if we wish to better aviation’s accident record.
The rapid growth in aviation, leading to a larger number of departures, in the
coming years would lead to larger number of accidents, if nothing is done to
bring down the accident rate/ million departures. The accident rate in
commercial aviation has been brought down substantially over the years, from
the 1950 – 60s, due to the move from reciprocating engines to the more reliable jet
engines as a means of propulsion, and
due to usage of better technology. However, the accident rate has stagnated at
a low figure over the past few decades – it has refused to reduce any further.
The major cause of accidents is now human factors. Aviation’s greatest
challenge is to tackle this cause factor if we wish to reduce the number of
accidents, with the growth of aviation. The most significant contributions to
reducing human error accidents can come from internalisation of CRM techniques
by the pilot community, because the pilot is the last line of defence of
preventing accidents in the aviation system.
Being situationally aware in terms of our
location, spatial orientation, environment, aircraft systems, time and fuel
requires all inputs that are available to the PIC. It has been found in many
accidents that the accident happened in the hands of a situationally unaware
PIC, when he was the PF, even though other crew members were situationally
aware – the Mangalore accident is our own case in point. What this implies is that
the whole crew should be on the same page and there should be no unresolved
issues in the cockpit. It is not the PF who causes an accident; it is the
entire crew. Accidents happen because either the PIC does not ask for, or
permit (through verbal or non verbal communications) free flow of relevant
information between the crew, or the crew does not share the relevant
information due to reasons that are part of being human. You can be disrespectful/
rude to a machine and it would even then give you the right information if you
have pressed the right buttons and controls, but humans are different – we are
emotional beings. Emotions are facts as far as humans are concerned. Our life
should convince us that emotions always matter. Other reasons could be: a human
may feel that the other is aware; knows every-thing; or the other does not need
to be told as he is so experienced; or a plain ‘I am not OK, you are OK
situation’; or ‘why should I tell him’ he is responsible for his actions. These
may be false assumptions, as has been brought to light in a number of aircraft
accident/ incident investigations.
Similar to situation awareness, the PIC
needs all the inputs to make a correct decision – two heads are always better
than one; time permitting though. There may be times when there is no time
available and the PIC would then need to take a decision, as he deems fit based
on his training and experience. However, when time is available it is best to
solicit opinions on the problem; options; pros and cons of each option before
taking a decision. Each of these should be weighed in terms of ‘what’ is right,
and not ‘who’ is right, under the given set of circumstances. After the
decision has been made; and responsibilities assigned to implement the decision,
it should be reviewed, and the same process of decision making (DODAR) should
continue. This sounds simple but how do we ensure inputs from all sources? This
can only happen by empowering the crew.
It is the PIC’s responsibility to empower
the crew by facilitating the formation of a team at the earliest opportunity
provided. This seems difficult but is actually not so. All crew members on the
line are trained, licensed, and proficient to do their jobs well. However, it
should be remembered that proficient individuals do not always make competent
teams. Our cricket team brings this out clearly. It is the responsibility of
the leader, the Commander, to turn these competent individual crew members in
to a competent team. The team members are technically qualified, trained and proficient
to undertake the flight, and contribute to the task of a safe flight. However,
it is the human aspects that need greater attention by each and every PIC
before, during, and after the flight to ensure optimum performance from every
crew member. A study of past accidents/ incidents points towards a deficiency
in this area.
Just thinking of the co-pilot as a ‘Second
in Command’, and the cabin crew as additional ‘eyes, ears and brains’ that are
available, would help the PIC in finding better ways to making optimum use of
all the available human resources within the aircraft itself. In addition,
there are other human resources available too via the Radio Telephony. This suggested
way of thinking subtly implies that the PIC needs to internalise the fact that
each one of his crew in the aircraft, besides others who are outside the
aircraft, are making vital contributions to the accomplishment of the common
goal of safe flight from A to B,
while being assigned seemingly different roles. The contribution of the crew
towards the achievement of the common goal should be emphasised, valued, and acknowledged
during the first meeting of the crew at the reporting point itself through appropriate
behaviours and communicated through words, tone, and body language, so as to
create an effective leader/ team relationship.
Research has shown that high performance
Captains use three methods to build an effective leader/ team authority
relationship. They establish their capability to assume the legitimate
authority bestowed on them through law, by establishing competence through a
well organised and logical briefing about the specific task at hand – each
flight may follow the same routing, but no two flights can ever be the same.
Having briefed, they balance the leader/ crew relationship by having the crew
members take responsibility for the work of the group as well – this is an
important element to empower the other crew members. One Captain is known to
have made this statement before an extremely effective crew performance in the
simulator: “I just want you guys to
understand that they assign seats in this airplane based on seniority, not on
the basis of competence. So anything you can see or do which will help out, I’d
sure appreciate hearing about it.” Lastly, these Captains interact with humans (emotional beings,
unlike robots) who would be filling in different roles on the flight. Crew were
encouraged to converse and made to feel comfortable, particularly when
conversation was related to the task at hand. Questions and comments were
encouraged by all crew members on any aspects of the briefing/ task. By doing
this, these Captains had set an authority pattern ranging from the authoritative
to consultative, to participative and finally to the democratic. This is what
balances the need of a single authority responsible for the safety of the
flight with the contributions of all crew members to achieve a safe flight. It
is important for the PIC to understand the role of verbal and non verbal
communications in the accomplishment of the task. Effective communications
before (in the form of a briefing), during (inquiring if something is amiss/
not as planned; advocating one’s own professional opinion, with reasons; etc.),
and after the flight (a thorough critique is essential for learning/ team
building and growth) is crucial towards effective team work on the aircraft.
Effective communications is more likely to result in a situation where the
crews are empowered.
The regulator and the company management
empower the crew by laying down rules, SOPs and checklists that are required to
be followed. The PIC can empower them further by providing them the right
leadership, through effective communications, and by communicating, and more
importantly by following all laid down SOPs and checklists. As has already been
brought out, the leadership style utilised by the PIC would vary from Autocratic
(in time critical situations) to Consultative, to Participative, to Democratic.
However, it should always be borne in mind that this does not absolve the PIC
of any of his responsibilities as laid down by the law. The law gives the PIC
the authority, but he needs to earn the respect of his team members by his
behaviour. A PIC who respects the others will be respected by the others. The
formation of the team starts the moment the individual crew members start
assembling at the reporting point.
The Captain must lead by example by
following all checklists, laid down procedures and SOPs so that all crew
members are on the same page, and know exactly what is being done, and what
needs to be done at each stage of the flight. This helps the crew to identify,
and point out deviations from the normal, which could be unintentional.
Following the Captain’s lead, other crew members would be less tempted to resort
to violations or the intentional disregard for laid down procedures. Sterile
cockpit procedures are one such procedure that has been violated on a number of
flights that have met with accidents or incidents. Sterile cockpit needs to be
observed so that our single channel processing brain does not get distracted
from the task at hand during crucial phases of flight. The Captain sets the
tone for this, and is also responsible to ensure its sanctity by disciplining
non conforming crew members.
The workload in the cockpit keeps varying
on different crew members during different times/ flights. Workload management
is one of the key responsibilities of the Captain. He must ensure that the work
load is evenly distributed and that no crew is over or underworked – both
situations lead to a drop in the arousal levels and in turn affect the processing
capabilities of the brain. High workload leads to stress and fatigue, and its
attendant problems, whereas under load leads to boredom, sleepiness, and loss
of attention. Low workload situations are encountered during auto-pilot cruise
on long flights. The Captain must ensure that the crew continue to monitor the
systems for correct functioning. Even when the aircraft is on autopilot, the PF
is responsible to ensure that the flight is on the desired trajectory and the
systems are functioning as designed. The PM (PNF) must monitor any changes,
even when engaged in non critical activities like flight plan, tuning,
communications, etc. The standard dictum of flight prioritisation viz.,
‘Aviate, Navigate and communicate’ should be followed.
High or low workload, it is a known fact that
a highly motivated individual performs better under all conditions. Motivating
pilots is relatively simple because most love their job – it is easier to
motivate an individual who loves his job, as the motivator is in the job
content itself. The Commander should try and provide opportunities for the crew
to grow, of course within the laid down regulations. This is very motivating
for the crew and the performance of the team improves. Mentoring a relatively
inexperienced crew member pays rich dividends in terms of motivation and job
satisfaction for both the mentor and the mentee.
Having gone through this paper, one is tempted
to ask, “So what is CRM?” It is nothing but understanding that aviation is a
complex system; we are human in that our physical, physiological and
psychological performance keeps changing and that our information processing
system has serious limitations (specially under time critical situations); that
humans are emotional beings; that we need all the help possible to fly safely
from A to B; that it is humanly not possible to do things by ourselves and that
other crew members can be a big help, but would need to be led, motivated,
involved to bring out their most optimum performance. Past accidents provide
evidence that technical competence is an important requirement, but not the
only requirement for safe operations; interpersonal and cognitive (information
processing has severe limitations, especially under time critical situations)
functions also have to be integrated in to flight operations to achieve a safe
flight.
Finally, CRM is a concept that recognises
the critical role of human factors in determining the effectiveness of
technically proficient crew in both normal and non normal situations and gives
one a practical approach which can help in an attitudinal/ behavioural change
so that competent individuals can come together to form a competent team.
How does one recognise a competent team
from just a technically proficient team? A competent team is one that can
operate a safe and efficient flight from A to B, and also terminate the flight
with the crews deriving satisfaction from working as valued professional members
of the team, and also with a readiness/ willingness to perform together as a
crew in the future.
3 comments:
Very well written.
I invite you to view and comment on my blog http://far117understanding.wordpress.com/
I understand that in India, the DGAC is also implementing revised FTDL.
Garret
Lately we have had two high profile accidents where economising on crew training was held as one of the causes of the disasters. Stall training was not part of the syllabus or was inadequately covered practically.Further, a fresh problem has been introduced through too much automation, whereby pilots are losing manual flying skills due to lack of use. There have been incidents due to this also where aircrews were reluctant to take controls manually while investigating erring autoflight functions.
Thank you Garret.
Thank you smhusain_1. Yes, the AF 447 accident over the Atlantic Ocean was quite a shocker. In response to that I had put up an article on my blog, which explains the basics of a stall. Once one knows the basics, it is easier to recover from it. Of course, the accident had many other issues too, besides the crew not knowing how to recognise and recover from a stall. Automation is another grey area which needs to be studied.
Post a Comment