Crew Resource Management
The conduct, aims or qualities that characterize or mark a profession or a professional person
One of the best definitions of a professional is given by the U.S. Congress in the Labor Management Relations Act, amended 1947. According to this Act, a "professional employee" is one who is:
In addition to these criteria, other requirements are frequently added
The FAA notes the following in the Aviation Instructor's Handbook
Professionalism...is achieved only after extended training and preparation...is based on study and research...requires the ability to reason logically, accurately, and make good judgmental decisions...cannot limit their decisions to standard patterns and practice...
Finally, a profession is not seen as a stepping-stone to another career, but rather the pinnacle of many years of disciplined study, research and examinations. Professionals are normally expected to rise within the ranks of their peers.
Non-compliance with regulations can subject you to enforcement action. Pilots are responsible to see that all regulations are upheld in flight operations.
A violation of ethics does not bring about punitive action from the government, but rather a formal or informal reproof from others in the profession. Sometimes the informal reproof can be more detrimental to one's career. A reputation earned as a compulsive "risk taker" can come back to haunt that person. The aviation community is small and reputations stick.
Being a professional can be summed up as being completely dedicated to your field and devoting as much time, energy, and care into your endeavor as possible. Fulfilling a checklist of the items above does not automatically make anyone a professional, and the inability to complete one or two of them does not automatically disqualify someone as a professional.
Most importantly, professionalism is found in an attitude.
Safety is an attitude, a frame of mind. It is being aware of your environment and actions at all times
Safety is knowing what is going on, knowing what can injure, knowing how to prevent that injury and then acting to prevent it. It requires intelligence and a reasonable ability to see, hear, smell and think
Safety is not something you can take or leave alone. It is not an activity that you participate in only when being watched or supervised
Safety is not posters, slogans or rules, nor is it movies, meetings, investigations or inspections
Ignoring safety does not indicate bravery, only foolishness. Conducting business in a safe and correct manner is the mark of a wise person, not a timid one
Please review this accident case study prior to the second day of CRM training.
Excerpts from the Transportation Safety Board of Canada (TSB) accident investigation help to compose this case study. Data was obtained from the Cockpit Voice Recorder (CVR), the Flight Data Recorder (FDR) and ATC radar tapes.
AVIATION OCCURRENCE REPORT
On September 26, 1989, a Fairchild Metro III was on a scheduled flight from Vancouver to Terrace, British Columbia with two pilots and five passengers on board. The aircraft crashed one quarter mile to the west of Terrace Airport while the crew was attempting to carry out a missed approach in IFR conditions. The aircraft was destroyed by the impact and a post-crash fire. All seven occupants were fatally injured in the crash.
The accident occurred at 0829 local time during the hours of daylight, an elevation of 713 feet ASL. The aircraft was new to the fleet, receiving its Certificate of Airworthiness just five days prior to the accident.
History of the Flight
Skylink Flight 070 took off at 0651 from Vancouver on a scheduled domestic flight to Terrace, B.C. The flight was scheduled to take-off at 0631; however when taxiing out for take-off, the aircraft returned part way to the ramp for an undetermined reason and then returned to the take-off position. This resulted in 20-minute delay in the departure time.
The Terrace terminal forecast (FT) issued at 0330 and valid at the time of the accident was sky condition partially obscured, visibility two miles in smoke and fog variable to six miles in smoke. The smoke was the result of slash burning in the local area.
Immediately following the occurrence, fog completely covered all of runway 15/33. The helicopter pilot who located the crash site estimated that the fog bank was 300 feet thick.
Commencing the Initial Approach
There were three comments made by the captain regarding to the weather conditions while the aircraft was en route to Terrace. The nature of these comments indicated that, at this stage of the flight, the captain did not appear to be particularly worried about the Terrace weather and that he expected few problems in this regard.
At 0811, the crew requested and received descent clearance to 15,000 feet. At 0812 the captain briefed the co-pilot on the approach into Terrace. The briefing given was deficient in many areas, in particular it did not include a complete description of the approach, the missed approach procedure, the status of the aircraft navigation aids, or the speeds to be flown. In addition, the Minimum Descent Altitude (MDA) was briefed to a Category C aircraft, when Category D would have been more appropriate. The lack of company sops meant that there were no company standards governing when and how such a briefing would be given.
On three occasions, the captain asked the co-pilot for information which had just been passed to them over the radio or for clarification of the situation. At 0816, the captain asked about the surface wind at Terrace even though this wind information had just been passed to them by the Terrace FSS.
At 0818 the flight was cleared to hold at the KITIMAT NDB because a Lear Jet had missed their first approach, and was commencing a second approach. At 0820, the Lear Jet discontinued their approach and landed on runway 15. Flight 070 was then issued their approach clearance.
At 0820, the captain commented "What's going on?" Just after the radio calls between the Lear Jet and the FSS regarding the Lear Jet's missed approach and between the accident aircraft and the FSS regarding the holding clearance. At 0820, twice the captain questioned whether the Lear Jet had landed even though the FSS had just indicated this when they had passed the approach clearance.
In all these cases, the responses from the co-pilot indicated that he was fully aware of the situation. All these queries from the captain illustrate a certain confusion in the mind of the captain over the developing situation at Terrace and an inability to fully assimilate all the information being presented to him. In addition, the fact that the Lear Jet had missed the approach should have been enough to alert the crew to the fact that the weather might present a problem and prompted them to revise their approach briefing.
At 0821, the aircraft crossed the KITIMAT NDB outbound and commenced the descent
For the approach. During the descent, at 0824 at an altitude of 6,800 feet, the landing gear was selected down and the flaps were selected to one half.
At 0822, the crew altered their intended landing runway from 33 to 15 based on information received from the FSS. The circling procedures for runway 15 are significantly different from those for 33, involving a turn through approximately 190 degrees, and yet there was no additional briefing to cover this. Again this is an indication of inadequate training or a lack or standardization or both.
At 0822, the Lear Jet informed Flight 070 that there was a "hole" in the cloud over the localizer transmitter which would allow them to descend and land on runway 15. The crew of Flight 070 had not voiced any concerns or made any preparations which indicated that they had considered any possibilities other than landing at Terrace. This transmission from the crew of the Lear Jet would have reinforced the impression in the minds of the crew that, although the weather conditions were deteriorating, a successful landing was still expected. The fact that the crew still did not brief for the possibility of a missed approach also indicates that the crew was expecting to land.
This segment of the approach was poorly planned by the crew. They commenced the approach at 15,000 feet ASL, and the aircraft was well above the published approach altitudes throughout this segment. Although these published altitudes are minimum altitudes, remarks made by the captain clearly indicate that he was aware that the aircraft was high. He attempted to rectify this problem by selecting flap early and by lowering the landing gear at 6,800 feet, presumably to increase the drag and expedite the descent. As a result, the captain was constantly trying to establish the aircraft on a more acceptable descent profile throughout the approach. He did not succeed in doing this as the aircraft crossed the TERRACE NDB at 3,050 feet rather than the published altitude of 2,200 feet.
Period from the TERRACE NDB to the Initiation of the Missed Approach
At 0825, the captain had to query the co-pilot about a radio transmission from the FSS regarding the approach light setting on runway 15; at 0826 he asked what the reference speed was. In both cases, the copilot responded with the correct information, indicating he was aware of the situation.
There were three occasions during this segment, at 0826, 0827, and 0827 when the copilot uttered exclamations of surprise and concern. Although the weather conditions were not specifically mentioned in these comments, it appears that all three were associated with the
Co-pilot's perception of the weather conditions and that he perceived these conditions to be worse than he had anticipated. There was no response from the captain to any of these comments, and the co-pilot did not pursue them further.
At 0826:34, 0827:19, and 0827:33, the captain called visual contacts with the ground. The second visual call by the captain at 0827:19 was "OK, I got the button here" at approximately one minute and 50 seconds after the aircraft reached MDA. Based on this timing, it is likely that the button referred to was the button of runway 33 or 27.
The aircraft was observed on what corresponded to a downwind leg for runway 15 at about 500 feet AGL (FDR shows 1,100 ASL), flying straight and level over the threshold of runway 27 at 140 knots.
Approximately 30 seconds after crossing the localizer transmitter, the aircraft commenced a descent out of 1,100 feet. At 0827, the captain cautioned the co-pilot to get ready to extend full flap. The aircraft further descended to 900 feet ASL with full flaps on an approximate one and a half mile base leg for runway 15. The aircraft position at this time was likely in the vicinity of the "hole" reported by the Lear Jet crew, and that the captain presumably descended through this area for landing. Once again there were no indications the landing runway was visual and in light of the other visual calls that had been made, it is logical to assume the crew would have called visual with the landing runway. Throughout the approach there were no indications on the CVR that the altitude alert system was being used.
Period Following the Initiation of the Missed Approach
Twenty-six seconds after full flap was selected, a missed approach was initiated. At 0828:49, the captain called "gear up" and then repeated this call three seconds later despite a clear acknowledgement by the co-pilot of "gear's coming up" between the two calls.
Within five seconds of the initiation of the missed approach, the aircraft began to accelerate and a rate of climb of approximately 1,200 feet per minute was established. This rate of climb quickly diminished to zero, followed by the start of a descent which quickly increased to 3,000 feet per minute.
The missed approach procedure was not carried out in accordance with the Canada Air Pilot procedure. There was no verbal indication on the CVR that a positive rate of climb was established prior to the gear-up selection. The FDR indicates that the altitude began to increase six seconds after the initiation of the missed approach and four seconds after the gear-up call. Given the FDR indications, it is unlikely that the crew had established a positive rate of climb prior to the gear-up call.
Another deviation from the correct procedure was a "flaps up" call by the captain rather than a "flaps to one half" call. It is possible that the captain elected to give this call because the airspeed was such that the flaps could be safely retracted to the fullup position. However, the co-pilot's response to the flaps up call was "Flaps coming up to half." It appears that either the co-pilot was complying with the checklist despite the captain's command or had complied with the captain's command, but was indicating that the flaps were beginning to cycle toward the half position.
As the aircraft descended, the co-pilot called "descending" twice. A few seconds later,
The aircraft struck trees on the west side of runway 15 just inside the airport perimeter.
There are continuous trim beeps recorded on the CVR from 0828:54 to 0829:00 (17 beeps) and from 0829:05 until impact (7 beeps). As the trim beeps during both of these intervals are continuous, the trim input in the cockpit would have been continuously applied during each interval and would have been in the same direction throughout each interval.
The first series of beeps commenced just following the gear-up calls by the captain and ended just prior to the first descending call by the co-pilot. This first series of beeps occurred in response to the power and configuration changes initiated during the first seven seconds of the missed approach. As all these changes would have produced a nose-up pitching moment, it is logical to assume that these 17 beeps were associated
With a nose-down trim input. This assumption is supported by the FDR data which indicate an aircraft pitch change appropriate to a nose-down control input and a vertical acceleration of slightly less than one "G".
The second series of trim beeps began as the captain acknowledged the first descending call by the co-pilot. It is likely that, in response to this call, the captain began to apply nose-up trim. This is also supported by the FDR data which indicates increasing positive vertical acceleration during this interval. The fact that the aircraft appeared to have been in a nearly level attitude when it initially struck the trees also tends to support the fact that a nose-up control input was being applied.
During the last five seconds, the co-pilot twice informed the captain that the aircraft was descending. There was some indication of concern in the voice of the co-pilot in the first call, and this concern appeared to increase in the second call. Certainly, the second call was more emphatic than the first. The captain's response to the first call was a calm, level "OK". There is no indication in this response that the captain was aware of the gravity of the situation.
At the time of the first descending call, the aircraft was at an altitude of 940 feet ASL, or approximately 160 feet above the tops of the trees, and descending at approximately 900 to 1,000 fpm. It is possible that a large, aggressively applied elevator input at this time might have prevented the aircraft from striking the trees; however, the captain did not do this and therefore was likely unaware of the developing situation. The only action on the part of the co-pilot at this time which might have prevented the accident would have been to seize control of the aircraft and apply the necessary elevator input. The co-pilot did not take his action and therefore was not fully aware of just how critical the situation was and/or was not assertive enough in reacting to the situation.
The captain and the co-pilot flew together on Sunday 24 September and Monday 25 September. On arrival back in Vancouver at 2010 on the night of the 25th, the captain was described as exhausted by a company technician. In addition, he declined an invitation to join his co-workers for a coffee at a local restaurant, stating that he was tired and that he had a long day ahead of him. The company Operations Manager stated that he spoke with the captain on the morning of the accident around 0600 and that the captain appeared to be in good spirits.
Although it could not be definitely determined, it is likely that the captain left the airport at about 2100 after his flight on the 25th, and, assuming that he went straight home, he would have arrived there around 2200. It was determined that he attended a pre-flight briefing at 0545 on the morning of the accident. He would have left home at about 0445 in order to arrive at the airport at 0545, so that it is likely that he got up at 0415. Based on this schedule, he could have had a maximum of just over six hours rest.
Attitude Reference Systems
The other Metro III aircraft that Skylink had been operating was equipped with a basic attitude indicator (AI) for pilot attitude reference whereas the accident Metro III was equipped with a more complex flight director (FD) system. Neither the crew nor any other of the company pilots had received any formal training in the use of this FD system, and there was no reference material available at the company to allow the pilots to study the system. The crew had flown the accident aircraft on this same scheduled route on the two previous days. These trips would have been the only exposure that the crew would have had to the operation of the FD system.
There were significant differences between the AI and FD systems. The FD system had features which the AI system did not: command bars; an altitude scale tied to the radio altimeter; a glide slope marker; an inclinometer; a decision height light; and a localizer scale and runway symbol. In addition to the above differences between the two Metro III aircraft, the accident aircraft was equipped with an autopilot, a radar altimeter, and an altitude alert system. There was no information or training provided by the company on these systems.
The most significant difference between the two aircraft was in the method of presentation of roll attitude information. On the FD system, the bank reference was a fixed scale located on the outer part of the face of the instrument and the bank pointer rotated on the inside of this scale to indicate the roll attitude. On the AI system, the bank reference was a rotating scale locate on the inside of the instrument and the bank pointer was a fixed index on the outer face of the instrument. The bank reference for the same roll attitude was taken from different locations on the two instruments.
The crew's flying in the past few months had been almost all on the other Metro III aircraft which were equipped with an AI which displayed roll attitude in a different manner from the FD. As a result, the crew would have been used to the AI roll display. There could have been a tendency for the captain to revert to the system with which he was most familiar, specifically the AI system, during the stressful and high workload circumstances of the missed approach.
In the Aircraft Flight Manual (AFM), the pilot flying is required to set the power during a balked landing procedure and also ensure that the engine limitations are not exceeded. The AFM requirement for the captain to set the power during the missed approach procedure provided a distracting influence at a critical stage of flight.
It is possible that, despite the AFM procedure, the co-pilot tuned the engine power after the power levers were advanced by the captain. If this was not the case, the captain not only had to fly the aircraft and initiate the co-pilot's actions, but he also had to ensure that the engine operating limitations were not exceeded while he advanced the power. This would have been a significant distraction at a critical phase of flight and could have diverted or partially diverted the captain's attention from the task of flying the aircraft. This may indicate the captain was so occupied with flying the aircraft at this time that the co-pilot's acknowledgement to "gear-up" did not register.
The company did not employ Standard Operating Procedures (SOP's), nor were they required to do so. It was determined that the company was in the process of drafting a set of sops; however, company personnel were unable to indicate the exact status of these sops. Interviews with company pilots indicated that there was often confusion among the pilots about what operational policies or directives were in place. Pilots indicated that they would often get different direction from different supervisory personnel.
Spatial Disorientation - Somatogravic Illusion
If one considers an aircraft flying straight and level and accelerating along the direction of flight because of an increase in power, for example, then the direction of the inertial force due to the acceleration is to the rear of the aircraft and for the purposes of this discussion can be assumed to be along the longitudinal axis of the aircraft. This inertial force combines with the force of gravity to produce a resultant which is inclined to the rear of the aircraft. If this resultant is then used by the pilot as the vertical reference, then the pilot will incorrectly sense that the aircraft is in a nose-up attitude. If the pilot then trims or eases forward on the control column to correct for this nose-up perception, the nose of the aircraft will drop and the airspeed will increase. This change in attitude will change the direction of the resultant force vector in such a manner as to maintain and perhaps magnify the illusory perception of a nose-up attitude.
As a result of the acceleration and pitch changes of the aircraft which occurred during the initial 12 seconds of the missed approach procedure, the captain likely experienced a somatogravic illusion which resulted in a significant difference between the pitch attitude which he perceived and that of the aircraft. The captain applied pitch trim and possibly elevator based probably on his perception of the aircraft pitch which resulted in a descent toward the terrain. The descent was recognized by the co-pilot, and corrections were made by the captain to arrest the descent; however, these control inputs were not sufficient to prevent the aircraft from striking the trees.
AVIATION OCCURRENCE REPORT
From the information provided, please give your opinion on:
1. What was the cause of this accident?
2. What do you think of the Captain's behavior? What were his influences?
3. What do you think of the First Officer's behavior? Any influences?
4. What lights lit up on the Situational Awareness Master Caution Panel and at what time during the sequence of events? Were they strong enough cues?
5. Was it just the flight crew at fault in this accident?
6. What would you do differently had it been you in either of those seats?
7. Comment on the usage of the 5 CRM Principles in this accident.
Sky-Links in the Chain of Events
Attitude Indicator was different and opposite to the one familiar to the flight crew; other unfamiliar instrumentation with no company training. This was the third time that the flight crew had flown this aircraft. The aircraft was over 12,500 lbs, which was heavier than the other company Metro III. The company was granted "grouping authority" by Transport Canada, which as a result of this accident, can no longer be granted to aircraft in different weight categories. The maneuvering speed of 140 knots would have kept the Metro III in Category C "minimums", however the aircraft was flown at a higher speed during most of the approach, so Category D "minimums" would have been more appropriate.
Delay in Departure
A 20 minute delay was incurred before takeoff. The report gave no reason for this delay, but the aircraft sat for 20 minutes and did not return to the ramp. We are left to speculate on the reason for this. It also placed the crew 20 minutes behind schedule.
The flight crew was familiar with the Terrace airport and approach. Complacency could have been a factor in not assessing the weather as limiting, or in considering the area slash burning as a factor in the visibility limitations.
Lack of SOPs
The approach briefing was initially inadequate, and was not updated with the change of runway. The initial approach began at 15,000 ft, when a shuttle descent to 5,500 ft was indicated on the CAP procedure. The aircraft was 800 feet high over the FAF, 5 miles away from the MAP. The altitude alert system, radar altimeter, or autopilot was not utilized by the crew, however they received no company training on these systems.
Violation of Air Regs
The aircraft flew below the Minimum Descent Altitude (MDA) without having established visible contact with the intended landing runway. In fact, the aircraft was less than 200 feet AGL on a one and a half mile base leg for runway 15.
Company Procedures Inadequate
Skylink did not provide adequate training for the flight crews on the new aircraft systems. There were no sops in place, and no clear definition of cockpit duties. The flight crew's schedule did not allow for adequate rest. The company allowed flight crews to break Air Regs - we can only assume that this was not the first time the captain had broken minimums and it is possible the company culture expected such. There may have been company pressure for the flight crews to keep to schedule. The pilots reported that they would get different direction from different supervisory personnel.
Captain's Confusion and Mindset
On several occasions, the captain did not mentally process radio information, and sought clarification with the First Officer. The Lear Jet that missed their first approach, then got in after abandoning their second approach could have given the captain a false sense of security about his landing possibilities. The captain fell for the "sucker-hole" trick. The reason why the captain made the decision to switch from runway 33 to runway 15 was not reported, but we still must question if it was sound.
The Captain called for gear up twice within a few seconds, therefore he either forgot, or did not register the First Officer's reply. He called for flaps up when the call should have been one half. The Captain most likely was overloaded during the unanticipated missed approach and resetting max power, and possibly confused by the unfamiliar attitude indicator.
The Captain did not establish a positive rate of climb. Most likely, he experienced a somatogravic illusion of excessive nose-up attitude, hence the 17 nose-down trim input just prior to the accident. The Captain did not react appropriately to the First Officer's "descending" call even though the situation was critical.
Fatigue may have affected the captain's behavior and lack of communication skills.
Non-assertive First Officer
The First Officer made weak exclamations about the visibility, but did not communicate his concern - if any - to the captain about their chances of landing. He did not clarify the inadequate approach briefing, or request an updated plan on their circling procedure. It was never verified who would monitor the instruments, and who would be visual, or what their respective roles were in the event of a missed approach. The first officer did not question the captain on his apparent confused state of mind as a possible subtle incapacitation, nor did he compensate for the captain's inadequate climb attitude during the missed approach. The First Officer seemed aware of all proceedings, yet did not act on the critical cues when they presented themselves.
A Conceptual Model of Human Factors
The SHELL Model provides a conceptual framework to help us to understand Human Factors. It illustrates the various constituents and the interface, or points of interaction, which comprise the subject. The study of Human Factors can be broken down into four conceptual categories:
Software - documentation, procedures, etc.
Hardware - machinery, equipment
Environment - both internal and external to the workplace
Liveware - the human element
Interaction between human beings and the other elements of the SHELL model are at the heart of Human Factors which involves interaction between:
Liveware - The Liveware is the hub of the SHELL Model. This is the most valuable as well as the most flexible component in the system. The edges of this block are not simple and straight, so other components of the system must be carefully matched if stress and eventual breakdown in the system are to be avoided.
The Human Factor elements under the Liveware category include those relating to the psychological state and the physical well-being of operational personnel themselves. (The Liveware element (human) should not be confused with the Liveware-Liveware interface which deals with interpersonal contacts.)
Successful Liveware (Human) Skills:
Liveware-Hardware - This Liveware-Hardware interface is concerned with ergonomics, such as seat designs, panel displays, location of switches and controls. This interface provides routing by which energy and information are exchanged between the hardware and the human.
The L-H interface addresses the mismatch in the human-machine design relationship and the source of confusion and error caused by poorly designed or located equipment.
Successful Liveware-Hardware Interface Skills
Liveware-Software - This Liveware-Software interface encompasses the non-physical aspects of the system, such as procedures, manuals, checklist layout, and computer programs. Problems in the L-S interface are often less tangible than those associated with the L-H interface and more difficult to resolve.
The L-S interface addresses issues that cause delay and error while seeking vital information.
Successful Liveware-Software Interface Skills
Liveware-Environment - Encompassed within the Liveware-Environment are the factors which designers of aviation systems have no control, such as temperature, radiation, weather and air pressure. In addition, the aviation system operates within the context of broad political and economic constraints and the system must be designed to survive and prosper within this element.
The L-E interface is associated with environment factors (noise, heat, lighting and vibration) and the disturbance of circadian rhythms. In addition, the L-E interface encompasses effects of the political, social and economic environments and their impact on operational efficiency.
Successful Liveware-Environment Interface Skills
Liveware-Liveware - The Liveware-Liveware interface is concerned with leadership, crew cooperation, teamwork and personality interactions. The L-L interface addresses the crew performance in flight and the characteristics of the individual pilot or crew member. Flight crews function as a group and so group influences can be expected to play a role in determining behavior and performance.
In the L-L interface, the focus is on the interaction among people because this process affects crew effectiveness. This interaction also includes leadership and command. Shortcomings at this interface reduce operational efficiency and cause misunderstandings and errors.
Successful Liveware-Liveware Interface Skills
Mismanagement in the SHELL Concept
Liveware - Other People
Liveware - Pilot:
Copyright © 1996-2005 by Neil C. Krey unless otherwise indicated.