HUMAN FACTORS AND PILOT DECISION-MAKING

HUMAN FACTORS AND PILOT DECISION-MAKING

Tenerife 1

Had you taken Groundschool for a pilot licence twenty years ago, this section in this course would not have appeared. At that time, decisions made by pilots were not considered to be a “technical” component of aviation safety. Twenty years ago, however, commercial aviation—and in particular large commercial “transport” aviation—was just coming to terms with an extremely unfortunate incident that occurred in the Canary Islands, a group of small resort islands located off Morocco; this incident would forever change the way pilots, especially air transport pilots, would be trained. Transport Canada, like all civil aviation regulators world-wide quickly instigated a program to train pilots that sound decision-making skills was fundamental to aviation safety—so much so that human factors and pilot decision-making became a mandatory component for recreational, private, and commercial ground training.

What happened in the Canary Islands off Spain is now legendary in the aviation community. At about 5 p.m. on the afternoon of March 27, 1977, two Boeing 747s collided on a runway at Los Rodeos Airport on the small island of Tenerife. Within seconds, the lives of 583 people were extinguished. The cause—a Senior Captain with KLM Royal Dutch Airlines—with 30-years’ experience with the Airline—was in a hurry.

Airlines served the heavy tourist traffic in and out of the Canary Islands by way of Las Palmas Airport on Grand Canary, but in the early afternoon of March 27th a terrorist bomb exploded in the terminal. Shortly after the explosion a bomb threat was received by the airport authorities regarding a second bomb and Las Palmas Airport was closed, forcing in-bound traffic to divert to the much smaller and ill-equipped Los Rodeos Airport on Tenerife. By the time the KLM Boeing 747 arrived at Los Rodeos, the ramp and apron facilities were crowded and even more so when a Pan American Boeing 747 arrived a half-hour later after being diverted on its trip from New York.

The facilities at Los Rodeos were overtaxed and the pressure began to mount on the shoulders of Captain Jacob van Zanten, KLM’s chief training captain for their 747s. Captain van Zanten’s responsibilities included ensuring that he and his crew’s “duty time”—the time that a flight crewmember remains at work without rest—did not expire. If the duty time reached the allowed limit established by KLM, the crew would not be able to leave Los Rodeos until after an overnight rest. He would have to find short-notice accommodation for his 234 passengers at considerable expense and inconvenience for the airline; Las Palmas Airport might not be re-opened before his crew’s duty time expired at 6 p.m. that afternoon.

The two 747s were parked tightly together on the run-up bay near the threshold of the one runway at Los Rodeos and, when Las Palmas airport did re-open and aircraft began to depart, the positioning of the two 747s was such that PanAm could not taxi for takeoff until after KLM taxied. During the afternoon Captain van Zanten made the decision to refuel his aircraft, anticipating that there would be delays later when backlogged aircraft converged on Las Palmas.

When Las Palmas did open and aircraft began departing, the PanAm crew would have to wait an additional 35 minutes while KLM aircraft was refuelled. To add to the strain situation, the weather at Los Rodeos began to deteriorate as fog began to move in from off-shore.

At 4:30 p.m. the KLM requested a taxi clearance, but by this time the runway was almost entirely obscured by the fog with visibility fluctuating around 900’. Because of the taxiway configuration the PanAm 747 would have to follow the KLM 747 that was backtracking on the active runway. When the PanAm crew contacted the tower controller as instructed, the visibility was such that the tower controller could not see the runway, and the PanAm crew could not see the KLM aircraft which they were instructed to follow. Now language became a factor as no longer could the crews and the controllers co-ordinate with one-another using visual information. The controllers spoke English with heavy Spanish accents and the KLM crew spoke English with a Dutch accent.

While the controller instructed KLM to taxi back to position on the far end of the runway, the PanAm crew was instructed to leave the runway on the third taxiway exit on the left. The PanAm crew, however, was having difficulty identifying the taxiway exits in the fog, not being sure of the number of taxiway exits they had now passed.

At five minutes after 5 p.m. the KLM crew had reached the far end of the runway and had turned to position the aircraft for the takeoff roll. As evidenced in the Cockpit Voice Recorder and the Flight Data Recorder, Captain van Zanten inexplicably advanced the thrust levers as soon as the First Officer completed the pre-takeoff checklist and the KLM aircraft began to move in the direction of the departure. The PanAm 747 was still taxiing up the runway, but was now obscured from the view of the KLM crew owing to the fog. “Wait a minute . . we don’t have an ATC clearance,” responded the KLM First Officer. “No . . I know that,” responded Van Zanten, holding the 747 back with the brakes, “. . go ahead and ask.”

The First Officer contacted the tower, reporting the KLM 747 was ready for takeoff. Recordings indicate the Tower Controller responded by providing the KLM with its IFR routing clearance, but did not clear the KLM for takeoff. But as the First Officer was reading back the routing clearance, van Zanten released the brakes and set the throttles to takeoff power. The KLM was six seconds into its takeoff roll when the First Officer added: “. . we are now at takeoff.” The Tower Controller responded: “Okay . . standby for takeoff . . I will call you.”

The PanAm crew heard the KLM transmission and sent out a desperate transmission— “We are still taxiing down the runway”—but the transmission was blocked and their fate had been sealed. When van Zanten saw the PanAm he made a desperate attempt to rotate prematurely and climb over the PanAm 747, and this effort is thought to have saved the 70 people on board the PanAm 747, but only the nose of the KLM cleared PanAm, and the KLM fuselage and wing cut though the PanAM passenger compartment. All persons on board the KLM were killed. MacArthur Job describes the scene as follows:

. . the PanAm crew caught sight of lights that were materializing through the fog directly ahead. Hazy at first, they seem for a long moment to be stationary. But as they continued to brighten, it quickly became obvious they were approaching fast! (The Captain) stared through the windscreen in stunned disbelief. “There he is . . look at him!” he cried out. “Goddamn . . that son-of-a-bitch is coming!” Desperately pushing all four throttles wide open, he attempted to swing the Boeing 747 off the runway to the left.

“Get off! Get off! Get off!” (the First Officer) . . yelled frantically as he saw the other aircraft’s nose begin to rise into the takeoff attitude.

Aboard the hurling Dutch aircraft, (the First Officer’s) . . eyes were fixed on his steadily rising airspeed indicator. “V1!” he finally called.

At the same moment, van Zanten sighted the PanAm 747 slewing across the runway ahead of them. Instinctively—there was no hope of stopping—he hauled back on the control column to try to lift over the American. But . . . it was too late. Its nose leg cleared the PanAm fuselage, but at 140 knots the main undercarriage slammed into it, slicing off the fuselage top as the No 4 engine demolished the hump just behind the flightdeck, and both aircraft exploded into flames.

For a few seconds more the burning Dutch aircraft remained in the air, then fell back on to the runway, slewing through 90 degrees before coming to rest with its engines torn off, 150 metres further along the runway. None of the fuselage doors were opened before its fuel tanks exploded, enveloping the entire aircraft in a raging fire.

Aboard the PanAm aircraft, the nightmare of impacts, flash fires, smoke and explosions created utter terror and confusion. Momentarily, (the First Officer) . . grabbed for the engine fire shutoff handles above his head. They were not there—the entire top of the fuselage had been carried away!

Moments later the flightdeck floor, together with that of the wrecked upper deck, collapsed, spilling the uninjured flight crew and the traumatized bodies of passengers who had been seated in the upper deck lounge into the main deck First Class cabin below. Together with passengers from this section of the aircraft, the crew escaped down on to the ground through a hole torn in the port side of the fuselage.

For once, those in the nose of the aircraft proved to be the lucky ones. Many of those seated on the starboard side of the main cabin in the areas of the initial impact were killed outright. On the port side, and further back in the main cabin, many other passengers who were initially spared injury were trapped and prevented from escaping by collapsed sections of the starboard fuselage side. They were soon overcome by the rapidly spreading fire.

Despite a fire under the portwing, where the engines were still running down, many managed to escape through the port side overwing exits to jump to the ground, some sustaining broken legs or other injuries. Still others jumped 20 feet to the ground from a rear door. Within one minute, the evacuation of survivors was effectively over. Fire had now overwhelmed the fuselage and starboard wing of the PanAm Boeing, and those who had not already made their escape would never do so (pp. 174-175, Vol. 1).

For all who lost loved ones, or were injured, or who were involved in other ways in the horror at Los Rodeos Airport that fateful Sunday afternoon on the island of Tenerife, there would have been a seemingly endless sense of recrimination as they thought over what had happened. There were so many “ifs”—so many small coincidences that need not have compounded to make the tragedy inevitable.

If the bomb had not gone off at Las Palmas, if the PanAm Boeing had been permitted to hold instead of landing at Los Rodeos, if the KLM crew had not decided to refuel, if the PanAm aircraft could have squeezed past its KLM sistership without having to wait for it to move, if the weather had not deteriorated, if the Pan Am crew had not bypassed the No 3 taxiway, if they had not transmitted at the moment they did when they feared the KLM aircraft was about to takeoff, if the KLM captain had taken more notice of his flight engineer’s doubt . . any of these factors could have altered the whole course of events as they unfolded.

But no amount of speculation could now change even one of them, much less bring back those who were lost.

The best that could be hoped for was that 583 people had not died in vain—that the sheer magnitude of the disaster would forcibly send home to all involved in aviation that, by its very nature, flying abounds with countless opportunities for “little” things to go wrong.

That, to be routinely safe, aviation requires constant and unswerving vigilance on the part of all its professionals—and that those who carry such responsibilities can never afford to take anything for granted (p. 180, Vol. 1).

References :

1 My comments on the Tenerife event are based on the descriptions and analysis made by Macarthur Job in his excellent and highly recommended Air Disaster, Volumes 1 & 2 (1996, Fyshwick: Aerospace Publications Pty Ltd); there were many factors attributed to this infamous event at Tenerife and they are well documented and discussed by Job (see Pp. 164-180 of Vol. 1).

After Tenerife, the trend became evident.

Back in 1970 an Overseas National Airways DC-9 ran out of fuel over the Caribbean and was forced to ditch at night—22 died; the cause was the crew’s miscalculation of fuel consumption.

In 1972 an Eastern Airlines DC-10 descended inadvertently into the Florida Everglades while distracted by a cockpit light bulb that failed to indicate the landing gear was down, killing 99 people, but leaving 77 survivors.

In 1978 a United Airlines DC-8 ran out of fuel while making an approach into Portland International Airport. The crew elected to circle south of the airport for almost one hour attempting to deal with a faulty landing gear problem. Only 10 of the 189 occupants were killed, primarily because the cabin crew was already prepared for the planned emergency landing on the runway.

These are all but a few of the more dramatic cases of faulty pilot decision-making.

What about you?

If you have not previously noticed local aircraft accidents reported on TV news or in newspapers, your decision to pursue a pilot licence will certainly change all that. Stories or reports of missing aircraft or aircraft “crashes” have little meaning for the public (although they certainly grab the public attention—at least initially). For you, however, this will change.

As you hang out at airports, and develop your friendships with other pilots, you will eventually observe that pilots spend a lot of time doing “armchair” analysis of aircraft accidents and incidents; at the root of this is their desire to know about an unfortunate event so as to educate themselves. They seek to ensure by casual, informal analysis that they do not suffer the same fate. They use these occasions of “gossip” and “hearsay” to reaffirm among themselves their safety practices, and to learn about what Donn Richardson2 calls the “gotchas” (hazards) of flying.

It is said by old pilots that, if you hang around aeroplanes long enough, you will see one bent. Anyone who has seen a bent aeroplane knows how oddly “unrecognizable” they look, how they appear as tattered sheet metal, with little semblance of order and structure. So if the old pilots are true, what can we do to stay out of trouble? The solution is simple—learn from the mistakes of others, and never let it happen to you. Is it really possible to ensure it doesn’t happen to you? You bet it is. A safe pilot is a pilot who is in control of every aspect of a flight. He or she is the Pilot-in-command, and the command authority is rooted in the basic rule that the pilot can take whatever action is necessary for reasons of safety. In developing your command authority as student pilot, assert command and control authority. Practise your ability to effectively analyze situations related to flight operations, and develop your sound decision-making skills. Take control aggressively to avoid potentially hazardous situations.

With command authority, there is privilege; but with this privilege there is responsibility. You will learn the safety rules and rituals of flying—always get a weather briefing, always dip your fuel tanks prior to flight, always use a pre-takeoff and pre-landing checklist, set and never violate personal weather minimum, etc.—and these rules are crucial. But the question remains—what happened to van Zanten? And what could have caused all those other professionals to make simple errors in predicting fuel consumption. Were they aware of the dire consequence of their decisions?

2 Donn Richardson is a Flight Instructor, a professional aerobatic pilot, and a former DC-3 pilot; he was also one of Langley Flying School’s three Flight Test Examiners.

The Pilot Decision-making Process

When faced with a non-normal or critical event, effective management by the pilot will require an effective decision-making processthat maximizes the potential for successful resolution. It is a cyclical and logical process that lies at the base of all decision-making process, but is particularly critical when pilots make critical decisions in critical situations.

Here are the phases of effective decision-making process:

Pilot Decision-making Process. Langley Flying School.

Situation

A situation is the set of circumstances that the pilot is faced with. More specifically, the situation—insofar as we are concerned with here—is critical, and therefore the decisions made are critical. The most important aspect of this part of the decision-making process, however, is that the pilot must be first aware of the impending critical situation. Once you are aware of the situation, you can begin to find a remedy—if you are not aware of an unfolding situation, you cannot even get to first base.

The classic example of this is controlled flight into terrain—CFIT. In all of the CFIT accident cases, the pilots were clearly not aware of the critical circumstances with which they were faced. Quite a few years back, a Canadian Armed Forces C-130 flew into terrain in a controlled fashion while on approach into Alert. The incident occurred at night and the pilot, still many miles back from the airport, had visual reference of the distant airport. Surrounding the airport was utter blackness. As the aircraft gradually descended dangerously close to the ice ridges, one of which the aircraft eventually made contact with, there must have been a period of time—albeit short—when the ridges where zipping past just below the aircraft fuselage. The pilot, however, was not aware of the situation.

Perhaps the most classic case of CFIT is the Eastern 401 accident, which is presented below as a case study. Here the transcript from the minutes leading up to ground impact are truly indicative of the factors that can conspire to make pilots unaware of the deadly situation in which they eventually—perhaps just before impact—find themselves. Nevertheless, we must remember that the pilot decision-making process requires that the pilot is first aware of the situation—he or she must have situational awareness. If we are not aware, we cannot begin to resolve matters towards a favourable outcome. If there is situational awareness, the rest of the problem solving is quite straightforward and natural.

Options

Once a problem occurs, and you are aware of it, the most important thing is to give yourself as many options as possible. Two things to note here, the first is that options are—when faced with a critical situation—like gold. The more options you have, the better your chances of a successful outcome. Remember, options are often merely ideas which, when first considered as an option, may or may not be viable. Further brain-time must be spent on them so that the outcome can be predicted. Some options are better than others, so the more options you can think of, the greater the likelihood of success. Sometimes, though, there is only one option and choice is therefore not involved—a classic example of this is an emergency forced approach resulting from a catastrophic engine failure. There is only one option and you must follow it though successfully. But single-option situations are incredibly rare birds. Since they are so rare, you should be sceptical should you encounter one—perhaps more options exist that you have not considered.

A second interesting feature of options is that they are the product of the “higher” brain—the cerebral cortex. The cerebral cortex is designed to shutdown and give way to the lower brain in moments of panic. The cerebral cortex, for example, allows you to predict and reason that the chances of successfully defending yourself against a grizzly bear are in fact quite small—the size, the teeth, the strength, the claws, etc. This is a reasoning process. In contrast, a panic response is a “lower” brain function which shuts down the reasoning process; panic is powered by adrenaline, not reason, and therefore undermines the rational assessment of options.

Choose

Assess risk and choose a course of action. If the “option” phase of the decision-making process is based on creative ideas, the “choose” phase is based on creative assessment. Again, this is a higher brain process, involving reason, and prediction. If panic forms the basis of any choice, the rational assessment process is undermined.

Act

Time is important. A pilot entering airframe-icing conditions without counter-icing equipment has only seconds to react. Even in less critical circumstances, however, time is in fact a scarce resource for all pilots—time is related to fuel consumption. On every flight there is only so much time you can spend in the air. Be sure to take action before time—and fuel—runs out.

Time limits can take other forms. After an alternator or generator failure, there is only so much electrical energy in the battery. While battery energy may not be consequential to a day VFR flight, it would become critical for a night VFR or IFR flight. Daylight is certainly a scarce resource for the float pilot.

Time can also be controlled by the pilot. Consider the decision to proceed on a cross-country flight into mountainous terrain. The pilot who presses on to have “look-see” is suddenly placed in a time-limited environment, while the pilot who takes the time to sort out weather data prior to launch has all the time in the world.

Evaluate

Has the selected action been successful? Once an act is committed, a new situation presents itself requiring new decision-making.

Loss of Situational Awareness

One of the greatest risks a pilot has when faced with a problem is that the pilot is simply not aware a problem exists. This undesired state is referred to as loss of situational awareness. Loss of situational awareness is like the boogieman sneaking up behind you—danger is imminent, but you are pleasantly unaware of it. What are some of the causes of loss of situational awareness that pilots have become victim to?

Loss of situational awareness can be caused by something as simple as inattention. A pilot is not aware of the 12 o’clock, 1-mile target on a mid-air collision course because he or she hasn’t been attentive in maintaining a traffic watch. A pilot not supervising the refuelling of the piston-engine aircraft is not aware that jet fuel has been loaded into the fuel tanks. A pilot inattentive during the pre-flight inspection is not aware that heavy rains and a poor gas cap seal have conspired to put dangerous quantities of water in the fuel tanks.

Loss of situational awareness is certainly a function of experience and training. A pilot from the prairies crosses the Strait of Georgia at 100’ ASL, oblivious of the hazards of having to ditch an aircraft in the water—unaware of how difficult it is to egress a dark, inverted cockpit that is submerged underwater A pilot on a fresh instrument rating is possibly unaware of just how rapidly dangerous amounts of ice can form on a aircraft after entering large cumulus clouds above the freezing level. We can, of course, go on and on.

An even more interesting cause of loss of situational awareness is false assumptions. In every one of the case studies reviewed below in this section, it is clear that the perceptions of every member of flight crews was undermined by false assumptions—the crew of PSA 182 thought they were clear of the Cessna 172, the crew of Eastern 401 thought the autopilot was still controlling the aircraft’s altitude, the crew of United 173 thought they had sufficient fuel, and crew of Palm 90 thought the dangerously low takeoff power settings were correct. Here are some of the factors that are commonly at the root of false assumptions:

Great expectations—e.g., you hear what you want to hear and see what you want to see. Watch for this in PSA 182. We tend to shape our reality to fit our expectation.

Fixation—e.g., you focus on one item while something more significant goes unnoticed—e.g., a warning light causes loss of control. This is the central theme of the Eastern 401 disaster.

Ignoring the bad news—e.g., subconsciously changing bad-news information into messages that are preferred. A major cause in the crash of Palm 90. A classic form of this cause is gethomeitis—the need to get home.

After period of intense concentration—e.g., after fighting bad weather a pilot lands downwind.

Stress

Physical stressors: hunger, temperature, noise, vibration, lack of oxygen, being tired, poor physical fitness.

Mental stressors: death, sickness, demotion, economic, workload.

Capacity to cope: during a normal flight, pilots are exposed to variable demands.

As demands increase, the safety margin between the work required to meet those demands and pilot capacity to perform the work decreases.

Remember, demands can pile up—e.g., an emergency during an approach. Trouble occurs when demands of a task exceed ability to deal with them.

Stress often invokes an arousal response—flight or fight. Insufficient arousal causes boredom, while too much causes panic.

Common responses to stress:

Omission—failing to respond to a signal—a radio call or warning light.
Error—responding incorrectly.
Queuing—delaying certain tasks because of workload.
Approximation—accepting lower standards of accuracy and performance.
Fixation—concentrating on one item while ignoring another.
Regression—reverting to an earlier procedure or action.
Tremor—trembling or shaking caused by increased muscle tension.
Escape—giving up, panicking, and freezing at the controls.

Cockpit stress management Rules of Thumb:

If an emergency does occur, BE CALM—think for a moment, weigh the alternatives, choose one, and then act.
Remember that fear and panic are your greatest enemies during an in-flight emergency.
Don’t hesitate to declare an emergency. Let other people, including passengers, know about your situation.
Don’t delay until it is too late.
If you feel tension mounting, loosen your collar, stretch your arms and legs, open air vents.
Don’t hesitate to ask ATC for help. Lots of ground resources are available.
Experienced passengers can be asked to look for landmarks and traffic.
If you make a mistake that you subsequently correct, forget about it and concentrate on the task at hand.
Focus on the situation, not the emotion.
Always have a “plan” and an “alternate plan,” and leave yourself an “out.”

Risk Management and the Pilot’s Checklist:

Pilots should expand their concept of “being ready” for a flight; here is a wholistic pre-flight checklist for this purpose.

Risk: Do a pre-flight inspection on yourself. Are you healthy? Are you current?
Aircraft: Weight and balance. Takeoff and landing performance. Cross-wind limitations. Navigation equipment. Survival gear.
Environment: Weather; runway length; navigation aids.
Operation: Different operations impose different risks. A pilot on a medevac assesses risk differently from a pleasure flight pilot.
Situation: The above risk areas compose a “situation.” Upon considering them, fly accordingly.

Deal yourself a good hand: begin the above evaluation with a closed fist and raise one finger for each risk element that you believe is safe for the flight. If you end up with less than a full hand, act accordingly.

Hazardous Attitudes

Here are some attitudes to flying you want to avoid:

Anti-authority. Don’t like being told what to do. This leads to rule and regulation violation; rules, regulations and procedures become unnecessary. Antidote: follow rules; they are usually right.
Impulsivity. When faced with a decision-making situation, the need to do something, anything, immediately. There is a lack of careful consideration. Antidote: don’t act so fast; think first.
Invulnerability. “It won’t happen to me.” You are therefore more likely to take chances. Antidote: think that it can happen to you.
Macho. Proving you are better than someone else. Antidote: “Taking chances is foolish.”
Resignation. Good luck versus bad luck. You can’t make a great deal of difference as to what is happening to you—leave the actions to others—for better or for worse. Just be nice and go along with unreasonable requests. Antidote: don’t feel helpless; you can make a difference.

Human Factors and Pilot Error

Human factors denotes the manner in which people relate and interact with their environments. In the case of aviation, the focus is on how pilot performance is influenced or affected by such issues as cockpit design, temperature, altitude, physiology of the body, emotions, interactions, and communications.

Pilot error is defined as the action or decision of the pilot that, if not caught or corrected, could contribute to the occurrence of an accident or incident, including inaction or indecision.

The Transportation Safety Board of Canada is responsible for investigating all transportation occurrences in Canada, including aviation occurrences. The goal of an aviation safety investigation is to prevent recurrence.

An aviation occurrence is any accident or incident associated with the operation of an aircraft.

A reportable aviation accidentis any accident resulting directly from the operation of an aircraft where a person sustains a serious injury or is killed, or an aircraft sustains damage or failure that adversely affects the structural strength, the performance or flight characteristics of an aircraft, resulting in the need for major repair or replacement of any component parts. A missing aircraft is also a reportable accident.

An incident is reportable only if it concerns the operation of an aircraft with a maximum certified takeoff weight of 12500 lbs.

The actions that must be taken in reporting a “reportable” aviation accident are described in the AIM, GEN 3.3.

It is required that no person displace, move, or interfere with an aircraft involved in an accident, except to extricate persons, or to prevent destruction by fire, or to avoid danger to any person or property.

Accident Rates

Statistics demonstrate improvement in safety—at least this is the general trend based on records kept since World War II. In 2001, the Canadian accident rate was 7.6 accidents per 100,000 hours flown, and less than 2 fatal accidents per 100,000 hours flown. Transport Canada reported for 2001 that there were 167 private aeroplane accidents in Canada, with 17 of these accidents involving fatalities. Pilots are generally considered to be the “cause” or “factor” in 84% of all accidents; in fatal accidents, pilot-related causes increase to 90.6%.

Aircraft Accident Rates since 1946, and the residual Human Factor (From Transport Canada's Human Factors). Langley Flying School.

Canadian Accident Rate (TSB), Langley Flying School.

Accordingly, if a private or recreational pilot has a one-hour flight, once a week, he or she will have to fly for 148 years continuously before an accident is experienced. The same person will have to fly continuously for 1584 years before experiencing a fatal accident.

Transportation Safety Board of Canada’s Statistical Summary—Aviation Occurrences, 2001, P. 1.

Ibid., P. 2.

Ibid.

Exposure to Accident Risk

There are two obvious but significant observations here—accidents are most likely to occur during approaches and landings, and it is the landing phase—at the end of the flight—where the workload and fatigue factor are at their maximum.

Flight Risk exposure based on phase of operation (from Canadian Human Factors). Langley Flying School.

Consider the depiction below showing what may be described as the normal decrease in safety margin during the course of an average flight.

Pilot Safety Margins during Flight (Canadian Human Factors). Langley Flying School.

Below is a table showing the cause of accidents that occurred between 1992 and 2001, relative to the phase of flight, and the first event that gave rise to the accident. The numbers further emphasize the increased exposure to risk associated with takeoffs and landings. Note that the landing risks are associated with events not associated with control or power loss, while these causes are prominent with takeoff accidents. With respect to en route causes, note the significant risk from power loss—be cautious with fuel planning, and be sure to stay current in forced-landing (power-off) skills.

Accident First Events. Canadian TSB. Langley Flying School.

The Accident Pilot Profile

The pilot involved in the average accident is likely to have the following profile:

The pilot is likely to be between 35-39 years of age.
The pilot is likely to have between 100 and 500 hours flying experience.
The pilot is likely to be on a VFR personal flight.

Experience and Greatest Accident Risk

The highest accident risk for a pilot is:

50 hours after receiving Private Pilot Licence;
50 to 100 hours after receiving Instrument Rating.

Here are some reasons that might account for these risk times.

At the completion of a training program, students have a high level of skill and confidence, but have very little experience.
Exposure to risk increases rapidly following training, as pilots are no longer in the protective cocoon of the training program where risks are monitored and controlled.
Despite confidence, new initial pilot or newly rated instrument pilots have not yet developed the experience, knowledge, and skill to recognise and manage the increased risks of IFR flight.

Use of Checklists

It is estimated that as much as 70% of all accidents in which pilot error was a primary error could have been prevented with the use of checklists.

Checklists eliminate forgetfulness. Everything you need to know is on the checklist, and the stress of any situation—where normal or non-normal—cannot change what is on the checklist. A cockpit procedure is more likely to be completed safely by using a checklist than by relying solely on memory.

Disorientation—178 Seconds

What is experienced by non-instrument rated pilots who inadvertently fly into cloud? The following is published by Transport Canada in one of its safety-promotions brochures—Take Five for Safety:178 seconds— which describes (quite dramatically, actually) what could happen:

If you’re ever tempted to take off in marginal weather and have no instrument training, read this article before you go. If you decide to go anyway and lose visual contact, start counting down from 178 seconds.

How long can a pilot who has no instrument training expect to live after he or she flies into bad weather and loses visual contact? Researchers at the University of Illinois found the answer to this question. Twenty student “guinea pigs” flew into simulated instrument weather, and all went into graveyard spirals or rollercoasters. The outcome differed in only one respect: the time required until control was lost. The interval ranged from 480 seconds to 20 seconds. The average time was 178 seconds—2 seconds short of 3 minutes.

Here’s the fatal scenario:

The sky is overcast and the visibility poor. That reported five-mile visibility looks more like two, and you can’t judge the height of the overcast. Your altimeter says you’re at 1500 but your map tells you there’s local terrain as high as 1200 ft. There might even be a tower nearby because you’re not sure just how far off course you are. But you’ve flown into worse weather than this, so you press on.

You find yourself unconsciously easing back just a bit on the controls to clear those non-too-imaginary towers. With no warning, you’re in the soup. You peer so hard into the milky white mist that your eyes hurt. You fight the feeling in your stomach. You swallow, only to find your mouth dry. Now you realize you should have waited for better weather. The appointment was important—but not that important. Somewhere, a voice is saying “You’ve had it—it’s all over!”

You now have 178 seconds to live. Your aircraft feels in an even keel but your compass turns slowly. You push a little rudder and add a little pressure on the controls to stop the turn but this feels unnatural and you return the controls to their original position. This feels better but your compass is now turning a little faster and your airspeed is increasing slightly. You scan your instrument panel for help but what you see looks somewhat unfamiliar. You’re sure this is just a bad spot. You’ll break out in a few minutes, but you don’t have much time left.

You now have 100 seconds to live. You glance at your altimeter and are shocked to see it unwinding. You’re already down to 1200 ft. Instinctively, you pull back on the controls but the altimeter still unwinds. The engine is into the red and the airspeed, nearly so.

You have 45 seconds to live. Now you’re sweating and shaking. There must be something wrong with the controls; pulling back only moves that airspeed indicator further into the red. You can hear the wind tearing at the aircraft.

You have 10 seconds to live. Suddenly, you see the ground. The trees rush up at you. You can see the horizon if you turn your head far enough but it’s an unusual angle—you’re almost inverted. You open your mouth to scream but . . . . . . you have no seconds left.

Hypoxia

Hypoxia is when the cells of the body do not receive enough oxygen; a person who suffers from this is said to be hypoxic.

The form of hypoxia most commonly experienced by pilots is when there is not enough oxygen in the lungs, or when the lungs are unable to transfer oxygen in sufficient amounts to the bloodstream.3 Here is a description of the condition:

In all (cases) . . the net effect is the same—reduced oxygen to the body, more importantly to the brain and eyes, causing a reduction in performance capability. As hypoxia increases, you become less and less able to function properly both mentally and physically. Mentally, as less oxygen reaches the brain, your thinking becomes confused and you are less able to make good judgement calls. Physically, your body increases its respiration in an attempt to get more oxygen. You may also start to feel dizzy and nauseous and perhaps get a headache. You also start losing motor-skill co-ordination and, in extreme cases, may pass out completely . . Hypoxia is an insidious problem in aviation; its effects creep up on pilots without their knowing it. Compounding the problem is the fact that one of the symptoms is a feeling of well-being; not only does hypoxia impair your ability to fly well, but it also makes you feel good at the same time. How you perceive your performance may be quite different from how everyone else sees it (Transport Canada’s Human Factors for Aviation—Basic Handbook, Pp. 46-47).

2 This is referred to as hypoxic hypoxia, while a second form of hypoxia—anemic hypoxia—is when there is sufficient oxygen in the lungs, but the blood is unable to distribute it to the body in sufficient quantities. Anemic hypoxia occurs in carbon monoxide poisoning.

Hypoxia is practically measured by the time at which a person can maintain useful consciousness—function with reasonable competence. As you can see in the table below, performance ability decreases rapidly with altitude. Importantly, the more physically active you are at altitude, the shorter the time of useful consciousness.

Altitude	Time of Useful Consciousness
10,000 ft.	Hours
20,000 ft.	5 to 12 minutes
30,000 ft.	45 to 75 seconds
40,000 ft.	13 to 30 seconds
45,000 ft and above	12 to 15 seconds or less

Smokers have shorter useful consciousness time—an altitude of 5000’ the symptoms and effects for a smoker are equivalent to those experience by a non-smoker at 10000’.

Empty-Field Myopia

Case: Two aeroplanes, a Piper and a Cessna, were flying straight and level on a cross-country flight at an altitude of 1500’ AGL in the Toronto Area. Neither aeroplane was under radar contact. Visibility conditions were seven miles in haze. The two aeroplanes collided almost directly head-on. There were no survivors.

Analysis: The haze conditions produced empty-field myopia in both pilots’ eyes. Therefore, each aeroplane appeared smaller and more distant that it actually was. With limited visibility, the danger did not become apparent until it was too late for evasive action. Since the frontal area of the aeroplane profile is small, an aeroplane viewed directly from the front shows little relative movement. Hence, detection by either pilot was difficult (Transport Canada’s Human Factors For Aviation—Basic Handbook, Pp. 77-78).

A pilot who experiences empty-field myopia is a pilot who is unable to see an aircraft in the distance, despite the unrestricted visibility.

To see something, the lens of the eye must be capable of physically focusing light from the object on the retina. To do this, the eye must be stimulated by an image. If the eye lacks this stimulation, the lens shifts to a resting state some three to five feet away.

When the sky is featureless—as is the case with unrestricted visibility, with hazy conditions, or dark night conditions—you effectively become near-sighted when you look out the windows as your eyes tend to resort to their natural resting state.

To counter empty-field myopia, it is a good practice to focus quite frequently on your own aircraft wing tips. Also, when scanning, focus on distant visible objects or outlines at or near the horizon, stimulating the eyes to establish long-distance focal points.

Consider that a target (another aircraft) on a collision course appears fixed and increasing in size to the observer. Changes in size are difficult to perceive, so a pilot who observes any fixed target should first immediately alter course, then assess its direction.

Landing Errors

Situation	Illusion	Outcome
Runway narrower than usual	Too high	Tendency to flare late
Runway wider than usual	Too low	Tendency to flare early
Runway slopes up	Too high	Tendency to make low approach
Runway slopes down	Too low	Tendency to make high approach
Terrain before runway is higher	Too low	Tendency to make high approach
No lights before runway at night (Black-hole)	Too high	Tendency to make low approach
Air very clear at night	Closer to airport than actual fact	Tendency to descend too early
Smoggy or hazy air	Farther from airport than actual fact	Tendency to descend too late

Crew Resource Management

As Crew Resource Management (CRM) is a complex study in and of itself, this section only seeks to introduce some of the prevailing concepts that lie at the centre of CRM training. Perhaps the best summary of the concept of CRM is provided by Captain Rick Zimmerman of AirBC. AirBC is considered by many to be the pioneer of CRM in Canada. In his Air Canada/AirBC Crew Resource Management, Pilot—Initial Training Program Course Manual (1999), Captain Zimmerman writes as follows:

Professional pilots will spend a large portion of their careers undergoing training and constant evaluation. Be it in the flight simulator twice a year, in recurrent or initial groundschool, during line checks, or even the medical checks required to validate the licence held. The fact is the industry, be it the air carrier or the governing body, requires a very high degree of performance at all times from flight crew members. No other profession on earth requires such scrutiny.

While most of the training and checking a pilot will experience during their careers will focus on the technical, a flight crew member must always maintain a level of “awareness” that is not readily identifiable or even quantitative. IFR skills, or aircraft knowledge is easily assessed, as these are technical skills, yet there is more involved in just “flying the airplane.”

The way in which pilots approach their job, the way they interact with other crew members, the way in which they utilize all available resources, and numerous other subtleties related to Human Performance, will have a profound affect on the way in which they “fly the airplane.”

The examination of the Human Factors involved in aviation, without a doubt, is “the most important” area of training a professional pilot will undergo.

Too many accidents related directly to poor human performance have caused the industry to look closely at this type of training, and most recently legislate its inclusion in air carrier training programs. Unfortunately, much of this training is coming too late in a professional pilot’s career, to be effective.

The “law of primacy” dictates: “What you learn first, you learn the best,” and for many active professional pilots “set in their ways,” the concepts do not come easily. For pilots beginning their careers, the adoption of these skills from the beginning will ensure a high degree of performance throughout their entire careers . .

An old saying, in an industry full of old sayings is: “I’ll never live long enough to make all the mistakes there are, so I hope I can learn from others, and not make the same mistakes that they have!”

Here are some of the concepts of CRM.1

Synergy

The concept of synergy implies that the total of the crew or team effort is greater than the sum of individual efforts. This is the target and key of CRM.

Crew Co-ordination

Crew co-ordination means the efficient assignment and monitoring of duties of all assigned crewmembers.

Targets

Goals, standards, and requirements continually established by crewmembers throughout all flight operations—includes airspeeds, altitudes, aircraft configuration, time, task completion, etc.—which establish the parameters of flight operation. Targets may be verbalized or incorporated in standard operating procedures. Targets may also be impromptu.

The failure to set targets and, more importantly, to not “challenge” a situation when targets are not met, is a major factor in pilot-related accidents. Watch for this in the case of United 173 presented below.

Excessive Professional Courtesy

The orientation to cockpit management in which a crew member experiences reluctance to challenge errors or deviations in flight procedures and skills owing to professional courtesy—“Looks like we’re just a little fast here” or “Seems to me that we’re just a tad above the glide path” are potentially ambiguous comments that may be rooted in excessive professional courtesy.

Safety Window

Time period during takeoff and landing sequences which is the most hazardous—typically below 2000’ AAE—and where increased alertness and concentration by the crew is required. This is a sacred zone during which all attention of all crewmembers must be focused only on the tasks at hand. Only dialogue related to the task of the approach or departure should occur in the cockpit.

Ambiguity

Indefinite, vague, or unclear communications related to flying tasks should never occur between crewmembers.

Establishing Situational Awareness

A pilot initially establishes situational awareness through training and applied crew management skills. Situational awareness for a flight also involves setting goals and targets for a specific flight. These goals and targets become the scale to measure situational awareness. Once established, situational awareness can be maintained by searching for clues to its possible loss. Recognition of these clues is critical to maintaining situational awareness (Dr. Jerry Cockerall, AirBC, 1989).

1 These concepts are from Rick Zimmerman’s Air Canada/AirBC Crew Resource Management, Pilot—Initial Training Program Course Manual (1999).

Clues to the Loss of Situational Awareness

Nobody Flying the Aircraft. Delegation of responsibilities is crucial for safe flight operations. The duty and responsibility of crewmembers must be clearly defined and assigned at all times, especially when faced with abnormal or emergency situations.

Distractions. Be alert to the fact that interruptions from normal procedures gives rise to crewmembers missing important situational clues.

Preoccupation. Be equally alert when preoccupied with something that interrupts normal procedures—again, crewmembers are likely to miss important situational clues.

Failure to Meet Planned Targets. Targets are required on every flight, and all targets—e.g., airspeed, climb and descent rates, power settings, and altitudes—must be met. When targets are not met, crewmembers must question why.

Ambiguity. The rule is clear here: any confusing, incomplete or unclear information must be resolved by crewmembers before proceeding.

Procedural Compliance. Situational outcomes cannot be accurately predicted if crewmembers use improper or non-standard procedures.

Intuition—“Gut Feelings.” In fact the “gut” feeling is likely reliable. The body is capable of detecting and reacting to outside stimulus before we can mentally establish the big picture.

Violating Minimums. Minimums—as applied in IFR—provide a solid frame of reference. If these are violated or ignored, the situation loses definition.

Unresolved Discrepancies. When faced with two or more pieces of information that do not agree or are contradictory, crewmembers must utilize other available resources to establish resolution.

Professional Crew Resource Management Guidelines 2

The positive delegation of flying responsibilities is a top priority, but during an abnormal or emergency situation, this is absolutely essential.
The positive delegation of monitoring responsibilities is another important priority that requires assignment and definition.
The crewmembers responsible for flying the aircraft must not take on secondary tasks that increase workload—these can be assigned to others.
When faced with conflicting interpretations of fact, crewmembers must consult external resources—i.e., charts, manuals, ATC, etc.—to resolve the conflict.
When there is doubt about a clearance, procedure, situation, or decision, crewmembers have the responsibility to clearly express doubt.

Principles of Good Crew Resource Management

Appropriately delegating tasks and assigning responsibilities.
Establishing a logical order of priorities.
Continuously monitoring and cross-checking essential instruments and systems.
Assessing problems carefully and avoiding preoccupation with minor ones.
Using all available data to conduct an operation.
Communicating clearly among crewmembers all plans and intentions.

Pilot-in-command Characteristics 3

The Ineffective Pilot-in-command

conforms with the stereotypes of the “macho pilot” and the “right stuff;”
does not recognize personal limitations of crewmembers who are under stress or who are faced with emergencies;
fails to make use of available crewmember resources—skills, knowledge, and experience;
does not have sensitivity to the problems and reactions of other crewmembers;
is likely to foster a tense flight-deck atmosphere;
is unlikely to foster a flight-deck atmosphere based on team co-ordination among crewmembers.

The Effective Pilot-in-command

recognizes that crewmembers have personal limitations;
recognizes that crewmembers have diminished personal decision-making capabilities in emergencies;
encourages other crewmembers to question decisions and actions;
is sensitive to the fact that personal problems might affect crewmember performance;
personal limitations are openly discussed;
recognizes the need for the pilot flying the aircraft to verbalize planned actions and procedures;
recognizes the role of the captain in training other crewmembers;
recognizes that the flight deck atmosphere must be relaxed and harmonious;
recognizes that management styles must vary with the situation and the make-up of crewmembers;
emphasises that the captain is responsible for co-ordinating crewmember responsibilities.

2 John Lauber, National Transportation Safety Board, as quoted in Rick Zimmerman’s Air Canada/AirBC Crew Resource Management, Pilot—Initial Training Program Course Manual (1999).

3 Transport Canada, Human Factors for Aviation—Advanced Handbook TP 12864 (E), Pp. 93-94.

Letter of Complaint

The following is the actual text of a letter sent by a passenger to Transport Canada in 1992. The letter is apparently used during initial training of staff members, especially pilots. The name of the operator has been changed for publication in this manual, but the remaining text of the letter is identical to the original.

Mr. David Larrigan
Regional Director General of Transportation Canada Aviation
Suite 870 800 Burrard Street, V62J8

14 September 1992

Dear Mr. Larringan,

I am writing this letter to make a formal complaint about how a recent flight was conducted by a Canadian Air Carrier. It is not unlike the one I wrote about on 12 September 1991 to Mr. Rhodes. I will try to include all the facts as I saw them. I will also comment from time to time.

I am a 59 year old retired American Airlines pilot. My flying background includes 10 years of U S Air Force flying which included 3.5 years of Air Sea Rescue flying in Alaska (in effect “bush” operations as I flew mostly helicopters). I flew 23 years for AAL, 14 of them as Captain. My total flying approaches 17,000 hours.

On 6 September 1992, I left Vancouver in a Smith Airline DeHavilland DHC-6, tail lettering C-ABCD. We were the first flight of the day to Rivers Inlet and the Oak Bay Camp. I was seated in the front row, center seat. I was interested in seeing if my letter last year had changed the cruising altitude habits of this airline. I will list the times and happenings of the trip. 0923 levelled at 10,000’ – very large aircraft crossed above, head on-no conflict. 0927 at 10200’ reduced power as if to cruise- the weather had been very nice in climb, Scattered clouds well below and not the slightest sign of choppy air. 0930 reached 10500’ 0935 the copilot started to read The Province Newspaper, he held it in front of himself so he was not watching for traffic and was also blocking some of the view out of the cockpit. 0946 after a very slow climb we reached 10800’ 0958 we reached 11,000’ – copilot was still reading – the clouds were widely scattered and well below, no sign of turbulence, a beautiful day. 1003 the copilot finished the newspaper and began to fly. 1004 the copilot had descended to 10500’ and the Captain started reading the newspaper. 1012 copilot flying and we’re still at 10500’, Heading was 290 degrees +/- 10 degrees most of the time. 1023 copilot still flying and holding 10500’ +/- 75’ (by now I had an idea what the cruise altitude should be, no help on that score from the Captain). 1028 the copilot descended to 10,000’ – Captain still reading. 1030 the Captain stopped reading, took control, and started descending. 1032 stopped the decent at 9500’. 1036 the Captain had climbed back to 10,000’. 1038 now at 10,200’ – a slow descent was started again. 1041 now at 9500’ Captain still flying. 1042 the Captain reduced power – this was the first time he or the copilot had touched the throttles since 0927.

The weather had stayed beautiful.

Our approach at destination was VFR and a surprise to everyone as they all commented strongly about it. We flew over the water straight at the dock, over the camp buildings and into the rapidly rising valley terrain behind it. We stayed very close to the tree tops on the right side of the valley. A very tight turn was executed to the left. The bank angle and subsequent “G” force got everyone’s attention and negative comment. I personally cannot believe he did this! Narrow valley, rapidly rising terrain, tight down wind turn with heavy “G” loading and no way out of this maneuver. The final descent, after the turn, was very rapid to get on the water just in front of the dock. I check the rate of descent after the altimeter went through 200’ and it read 1600’ per minute.

Let’s review: Above 10,00’ 1 hour and 1 minute for no apparent reason. VFR flight while reading a newspaper for 55 minutes. No sign of altitude control by the Captain, not even close. Finally, an approach into a unnecessary situation that any good bush pilot would avoid much less an air carrier pilot with passengers on board.

Please advise me if your regulations allow this kind of flying and behavior.

Sincerely,

Joe Blow,
Cambell, Texas

Transport Canada’s response:

Suite 620
800 Burrard Street
Vancouver, B.C.
V6Z 2J8

September 28, 1992

Smith Airlines
Richmond, B.C.
Attention: Director of Flight Operations

Dear Sir:

The enclosed letters are self-explanatory.

Please provide the undersigned with your explanation of Mr. Doe’s allegations of September 14, 1992, regarding the operation of the September 6, 1992, flight into Rivers Inlet.

Your response is required by October 12, 1992

Yours truly,

N. MacGregor
Regional Director
Air Carrier Operations

encl.

cc: Mr. Doe

Case Study: Pacific Southwest Airlines Boeing 727 (PSA 182), and Cessna 172, Lindbergh Field, San Diego, September 25, 1978 1

B727 Crew:

Captain James McFeron—14,000 hours flying experience, with 12 years on the B727.
First Officer Robert Fox—10,000 hours flying experience.
Flight Engineer Martin Wahne.

Cessna 172 Crew:

Instructor Martin Kazy—5,000 hours flying experience
IFR Student David Boswell—Commercial Pilot Licence

Weather

Clear

Background

The B727 flight originated in Sacramento, with a stop in Los Angeles, and was to terminate in San Diego.
The C172 flight originated from Montgomery Field, located 6 miles north-east of Lindbergh Field, San Diego’s main airport. This training flight included a series of ILS approach at the IFR facilities at Lindbergh Field, whereby the student would orient the aircraft onto the centreline of the runway and track the ILS for a simulated approach. This was done with the student under the hood, with the instructor supervising the student and keeping a lookout for traffic.
It should be noted that the two aircraft are, at the time of impact, on different frequencies. At the start of transcript, the B727 crew is passed off to the Lindbergh Tower from the Approach controller. In contrast, the C172 crew was just handed off by the Lindbergh Tower to the Approach controller. The Approach controller is providing the C172 crew with vectored headings for a repeat practice approach.

The following abbreviations are used:

APP	Approach Controller	P	Cessna Pilot
T	Tower Controller	ODC	Off-duty Captain in jump seat
C	Captain	UID	unidentified person
FO	First Officer

1 Macarthur Job, Air Disaster (Ibid) (see Pp. 23-35 of Vol. 2).

Time to Impact: 120 seconds:
APP:	Cessna 7711G, radar contact, maintain VFR at or below 3500, fly heading 070, vector (for) final approach course.
P:	(to ATC): 7711G on the heading and VFR below 3500.
APP:	PSA182, traffic at 12 o’clock, three miles, out of 1700.
FO:	Got him!
C:	(to ATC): Traffic in sight.
APP:	OK sir, maintain visual separation, contact Lindbergh Tower 133.3. Have a nice day now!
FO:	Flaps two.
C:	(to ATC): Lindbergh, PSA182. Downwind.
T:	PSA182, Lindbergh Tower, traffic 12 o’clock, one mile, a Cessna.
F:	Flaps 5.
C:	(to F): Is that the one we’re looking at?
F:	Yeah—but I don’t see him now.
C:	(to ATC): OK—we had it there a minute ago.
T:	PSA 182, Roger.
C:	(to ATC): I think he’s passed off to our right.
T:	Yeah.
C:	(to F): He was right over here a minute ago.
F:	Yeah.
T:	How far are you going to take your downwind, PSA182? Company traffic is waiting for departure.
C:	(to ATC): Ah, probably about three to four miles.
T:	OK.
T:	PSA182—cleared to land.
C:	(to ATC): PSA 182’s cleared to land.
C:	(to F): Are we clear of that Cessna?
F:	Supposed to be!
C:	I guess.
F:	Flaps 15.
ODC:	(laughing): I hope!
C:	(positively): Yeah—before we turned downwind, I saw him about one o’clock—probably behind us now.
F:	Gear down.
F:	There’s one underneath . . I was looking at that inbound there.
C:	Whoops! (as F pulls up).
F:	Arghhh!
	(Metallic crunching noise.)
ODC:	Oh ####!
C:	Easy baby, easy baby . . what have we got here?
F:	It’s bad.
C:	Eh?
F:	We’re hit man—we are hit!
C:	(to ATC): Tower—we’re going down—this is PSA.
T:	OK—we’ll call the equipment for you!
UID:	Whoo!
	(Sound of stall warning.)
C:	(to ATC): This is it, baby!
UID:	####
C:	(on PA system): Brace yourself!
UID:	Hey baby . .
UID:	Ma, I love you . .
	End of Recording.

Artist rendering of PSA 182 collision (from Wikipedia). Langley Flying School.

PSA 182 photo (Wikipedia). Langley Flying School.

Case Study: Eastern Air Lines Lockheed L-1011 (Eastern 401), Miami International Airport, December 29, 1972 1

Crew

Captain Robert (Bob) Loft—30,000 hours flying experience.
First Officer Albert (Bert) Stockstill—total time unknown, but 300 hours on the L-1011.
Second Officer Don Repo—time unknown.

Weather

Clear evening.

Background

Flight from New York to Miami.
10 flight attendants and 162 passengers on board.

1 Macarthur Job, Air Disaster (Ibid) (see Pp. 98-111 of Vol. 1).

C:	(over PA system): Welcome to Miami. The temperature is in the low seventies, and it’s a beautiful night out there tonight.
ATC:	(Clears Eastern 401 to join the ILS for Runway 09L, and advises the crew to contact the tower.)
C:	(to ATC): Cleared to ILS 09 left, call Miami Tower on 118.3. Eastern 401, so long.
C:	(to ATC): Miami Tower, Eastern 401—just turned on to final.
	(no response)
C:	(to First Officer): Go ahead and throw them (the gear) out.
FO:	(complies)
C:	Miami Tower—do you read Eastern 401? Just turned on to final.
ATC:	Eastern 401, heavy—continue approach to 09 Left.
C:	Continue approach—roger.
C:	(noting that only two of three green gear lights are on, to FO): Bert, is that (undercarriage) handle in.
FO:	No nose gear.
	(Crew recycles gear; indications remain.)
Time to Impact: 8 minutes, 25 seconds.
C:	(to ATC): Well. Tower, this is Eastern 401. It looks like we’re going to have to circle—we don’t have a light on our nose gear.
ATC:	Eastern 401 heavy, roger. Climb straight ahead to 2000—go back to Approach Control on 128.6.
C:	Okay, going up to 2000—128.6.
FO:	(reaches for gear lever)
C:	Put power on first, Bert—leave that damn gear down until we’ve found out what we’ve got.
SO:	(to Captain): Do you want to test the lights.
C:	Yeah—check it.
FO:	Bob, it might just be the light (itself). Could you juggle it?
FO:	We’re up to 2000—you want me to fly it, Bob?
C:	What frequency did they want us on, Bert?
FO:	128.6.
C:	I’ll talk to them.
SO:	It’s right above . . that red one, isn’t it?
C:	Yeah—I can’t get at it from here.
SO:	I can’t make it pull out either.
C:	(to ATC): Approach Control—Eastern 401. We’re right over the airport now, climbing to 2000 feet—in fact we’ve reached 2000—and we’ve got to get a green light on our nose gear.
ATC:	Eastern 401, roger. Turn left, heading 360, maintain 2000—vectors to 09 Left on final.
C:	Left—360.
	(Neither the Captain nor the Second Officer can reach the lights that are on the First Officer’s side of the cockpit.)
C:	Put the damn thing on autopilot. See if you can get it out.
FO:	(complies)
C:	(to the First Officer): Now push the switches just a little bit forward—you’ve got to turn it sideways.
FO:	No, I don’t think it’ll fit.
C:	You’ve got to turn it a quarter-turn to the left.
ATC:	Eastern 401, turn left—heading 300.
	(Captain acknowledges; First officer complies.)
C:	(frustrated, to Second Officer): Hey—get down there and see if that damn nose wheel is down—you better do that.
	(Flight Data Recorder shows a momentary negative vertical acceleration, producing a 200’ per second descent for half a minute.)
FO:	(continuing at removing the light): Got a handkerchief or something so I can get a better grip? Anything I can do it with?
	(A technical officer with Eastern riding in the jump seat now offers assistance.)
TO:	(to the First Officer): . . pull down and turn to the right. Now turn it to your left one time.
FO:	It hangs out and sticks.
TO:	Try it my way.
FO:	(to Captain): It won’t come out, Bob—if I had a pair of pliers, I could cushion it with that Kleenex.
SO:	(about to climb down into the electronics bay): I can give you a pair of pliers. But if you force it, you’ll break it—believe me.
ATC:	Eastern 401, turn left—heading 270.
Altitude: 2000’, Time to Impact: 4 minutes, 24 seconds.
	(Captain acknowledges; First officer complies.)
C:	To hell with it! To hell with this—(to the Second Officer): go down and see if it’s lined up on that red line—that’s all we care! (Laughs) Screwing around with a 20-cent piece of light equipment—on this plane.
	(Others laugh too.)
C:	(to ATC): Eastern 401 will go out west a little further if we can . . see if we can’t get this light to come on.
ATC:	All right. We’ve got you heading westbound now, Eastern 401.
	(Captain and First Officer continue to manipulate the light assembly for an additional two minutes or so.)
FO:	Always something. We could have made it on schedule.
	(The Cockpit Voice Recorder records a C-chord chime at the Second Officer’s panel.)
Altitude: 1750, Time to Impact: 1 minute, 34 seconds.
C:	(to First Officer): Leave it there.
SO:	(returning) I can’t see it down there.
C:	For the nose wheel there’s a place in there where you can look and see if it’s lined up.
SO:	I know—a little like a telescope.
C:	It’s not lined up?
SO:	I can’t see it. It’s pitch dark.
TO:	Wheel well light on?
SO:	Yeah. Wheel well lights are always on if the gear’s down.
	(Second Officer and technical officer descend into electronics bay.)
Altitude: 980’; Time to Impact: 36 seconds.
ATC:	(Controller reads encoded altitude of 900’) Eastern 401—how are things coming along out there?
C:	Okay—we’d like to turn around now and come back in.
FO:	(Observes airspeed increased from 174 to 188 knots, and responds by retarding the thrust levers slightly.)
ATC:	Eastern 401—turn left, heading 180.
	(Captain acknowledges; First officer complies. Controller intended to query altitude, but became preoccupied with other traffic; crew’s response was reassuring.)
FO:	(noting the altitude): We did something to the altitude!
C:	What?
FO:	Were still at 2000—right?
C:	Hey—what’s happening here!
	(Controller notices Eastern 401 has disappeared from radar.)
ATC:	Eastern 401, I’ve lost you on radar—and your transponder. What is your altitude?
Pilot:	Miami Approach, this is National 611. We just saw a big flash—looked like it was out west. Don’t know what it means, but we wanted to let you know.
Pilot:	Lan Chile 451—we saw a big flash—a general flash, like some kind of explosion.

Post Accident Findings

The first record of inadvertent vertical discent—when the Flight Data Recorder shows a momentary negative vertical acceleration, producing a 200’ per second descent for half a minute—is believed to be the result of the Captain inadvertently exerting force on the control wheel as he turn to speak to the Second Officer. This likely disconnected the Altitude Hold function of the autopilot. It is believed that improper wiring of the autopilot indicator lights resulted in the First Officer’s “ALT” (indicating altitude hold function) stayed on, despite the function being inadvertently turned off. 1
It is believed that the C-chord chime that sounded at the Second Officer’s panel (while the Second Officer was in the electronics bay) was in fact the altitude alert system indicating that the aircraft had deviated 250’ from the pre-selected altitude. The chime was not heard by the crews.
The Controller reported that he had noticed the encoder readout for Eastern 401 indicated 900’ only 36 seconds prior to impact. The controller reported that he wanted to see one more radar sweep to verify the deviation in Eastern 401’s altitude, but before this could be done, the controller had to respond to other air traffic.
It is believed that the First Officer’s response to increased airspeed indications—from 174 to 188 knots—was a misinterpretation of the aircraft’s acceleration from descent—the First Officer interpreted this as acceleration in level flight.

1 Post accident investigation found the two computers controlling the autopilot were mismatched with respect to pitch, leading to the misleading altitude hold indications on the FO’s panel.

Artist Rendering (Wikipedia). Langley Flying School.

Eastern 401 Flight Path (US TSB). Langley Flying School.

Case Study: United Airlines DC-8 (United 173), Portland International Airport, December 28, 1978

Crew

Captain Malburn McBroom—28,000 hours flying experience, with 5,500 hours as captain on DC-8s.
First Officer Rodrick Beebe—8,000 hours.
Second Officer Forrest (Frostie) Mendenhall—3,900 hours as flight engineer, over 2,000 of which were on the DC-8.

Weather

Clear evening.
Background
Flight was from New York to Portland via Denver.
On board were181 passengers
The First Officer was flying.
The dialogue begins just after the crew discovers a landing gear problem; the “transit” light did not extinguish, and only the nose gear indicated a “green” light. The extension of the gear was associated with abnormal yaw and a loud thump.

Macarthur Job, Air Disaster (Ibid) (see Pp. 36-46 of Vol. 2).

Just after 5:06 pm:
ATC:		United 173 heavy . . , contact Portland Tower now, 118.7.
C:		Negative, we’ll stay with you—we’ll stay at 5000 (feet). We’ll maintain about 170 knots—we’ve got a gear problem. We’ll let you know.
ATC:		Roger, maintain 5000—turn left, heading 200.
ATC:		United 173 heavy, turn left now, heading 100, and I’ll just orbit you out there until you get your problem right.
C:		(Makes an announcement over the PA system that he is concerned about the condition of the undercarriage and that they would “circle around while the crew did some checking.”)
S:		(Goes back to passenger cabin to examine mechanical gear indicators on wing upper surface—they appear to indicate the gear extended.)
5:38 pm:
		(The Captain contacts United’s Maintenance Control Centre (UMC), and reports the aircraft has 7040 lbs. of fuel, and would hold for another 15 or 20 minutes.)
UMC:		Okay—you estimate you’ll make a landing about five minutes past the hour?
C:	Yes—that’s good ballpark. I’m not going to hurry the girls. We’ve got about 165 people on board and we want to take our time and get everybody ready—and then we’ll go. It’s (the weather) as clear as a bell here and no problem.
5:45 pm:
	(Senior flight attendant, named Dory (FA), reports to flightdeck regarding progress.)
C:	How are you doing, Dory?
FA:	We’re ready for your announcement—you have the signal for protective position? That’s the only thing I need from you right now.
C:	Okay—what would you do? Have you any suggestions about when to brace? What to do it on the PA system?
FA:	. . I’ll be honest with you . . I’ve never had one of these before . . my first, you know.
C:	All right, what we’ll do is we’ll have Frostie . . a couple of minutes before touchdown, signal for a brace position.
FA:	Okay—he’ll come on the PA? And if you don’t want us to evacuate, what are you going to say?
C:	We’ll either use the PA, or we’ll stand in the door and call.
FA:	Okay—one or the other. We’re re-seating passengers right now and all the cabin lights are full up.
C:	All right.
FA:	We’re ready for you announcement anytime.
5:47pm:
SO:	I can see the red indicators from here, but I can’t tell if there’s anything lined up.
FO:	How much fuel have we got left, Frosty?
SO:	5000 (lbs.)
	(An off-duty Captain (ODC) is in the jumpseat.)
ODC:	Less then three weeks—three weeks to retirement—you better get me out of here!
C:	Don’t worry!
ODC:	If I might make a suggestion, you should put your coats on . . both for your protection and so you’ll be noticed . . so they’ll know who you are.
C:	Oh—that’s okay.
ODC:	But if it gets hot (in case of fire), it sure is nice not to have bare arms.
C:	If anything goes wrong, you just charge back there and get . . off—okay?
ODC:	(vacates seat to move to passenger cabin): I told the girl to put me where she wants me—I think she wants me at a wing exit.
C:	Okay fine, thank you.
5:49 pm:
	(The First Officer asks again about the remaining fuel indicated on the gauges.)
C:	Five (5000 lbs.).
	(The Second Officer comments that the fuel pump lights have illuminated.)
C:	That’s about right—the feed pumps are starting to blink. (The cockpit indications for the inboard fuel feed pump illuminates on the DC-8 when the fuel is 5000 lbs.)
C:	Hey, Frostie!
SO:	Yes, Sir?
C:	Give us a current (landing data) card on with—I figure about another 15 minutes.
SO:	15 minutes?
C:	Yeah—give us 3-4000 on top of zero fuel weight.
SO:	(aside) Not enough! 15 minutes is going to really run us low on fuel here.
5:51 pm, 18 NM south of the airport, turning northeast:
FO:	Maintenance have anything to say, Frostie?
SO:	They said, ‘I think you guys have done every thing you can.’ I said we were reluctant to recycle the gear for fear something was bent or broken, and we wouldn’t be able to get it down again.
FO:	(to Captain) Think we ought to warn these people on the ground?
C:	Yeah—we’ll do that right now. (To the Second Officer): Call the United terminal—give them our passenger count including laps (infants)—tell them we’ll land with about 4000 pounds of fuel and to give that to the fire department. I want United mechanics to check the airplane after we stop—before we taxi.
	(The Second Officer contacts United ground staff, and reports the landing will occur at 6:05pm; the Second Officer checks once more with the Captain.)
SO:	They want to know if we’ll be landing about five after.
C:	Yes.
5:54 pm, 17 NM south of airport, heading north-east:
C:	All done?
SO:	Yes, sir—and ready now for the final approach, final descent check.
	(The First Officer asks the Second Officer again about the remaining fuel; the Second Officer responds 4000 lbs.)
C:	(to the Second Officer) You might just take a walk back through the cabin and see how things are going—Okay? I don’t want to hurry them, but I’d like to do it in another . . 10 minutes or so.
	(Second Officer does so, while the Captain and First Officer remark on the competence of the senior flight attendant, the evacuation procedures, and the use of antiskid.)
6:01 pm:
	(Second Officer returns, reporting the cabin crew will be ready in another two to three minutes.)
C:	Okay—how are the people?
SO:	Well, they’re pretty calm and cool. Some of them are obviously nervous, but for the most part they’re taking it in their stride. I stopped and reassured a couple of them—they seemed a little bit more anxious than some of the others.
C:	Okay—well about two minutes before landing—that will be about four miles out—just pick up the PA mike and announce: ‘Assume the brace position!’
SO:	We’ve got about three (3000 lbs.) on the fuel—and that’s it.
C:	Okay—on touchdown, if the gear folds or something really jumps the track, get those boost pumps off . . you might even get the valves open.
	(The aircraft is now 5 NM south of the airport heading south-westward, beginning another orbit. ATC queries how much longer before the landing.)
FO:	(To ATC) Yeah—we have an indication our gear is abnormal—it’ll be our intention in about five minutes to land on Runway 28 Left. We would like the (fire) equipment standing by. Our indications are the gear is down and locked, but we’ve got our people prepared for an evacuation in the event that should become necessary.
ATC:	Okay—advise when you’d like to begin your approach.
C:	Very well—they’ve about finished in the cabin—I’d guess about another three, four—five minutes.
ATC:	If you could give me souls on board and amount of fuel?
C:	172 and about 4000—well, make it 3000 pounds of fuel. And you can add to that 172, plus six laps—infants.
6:06 pm, 17 NM south of airport, heading south-west:
C:	(to Senior Flight Attendant who has returned to the cockpit): How are you doing?
FA:	Well, I think we’re ready.
C:	Okay.
FA:	We’ve re-seated—they’ve assigned helpers, and shown people how to open exits—they’ve got able-bodied men by the windows. The (off-duty) captain in the first row of coach after the galley—he’s going to take the middle galley door—its not far from the window . .
C:	Okay—We’re going to go in now—should be landing in about five minutes.
FO:	I think you’ve just lost number four . . better get some cross feeds open there or something.
FA:	I’ll go and make the five-minute announcement—I’ll be sitting down now
FO:	We’re going to lose an engine!
C:	Why?
FO:	Fuel!
FO:	(to the Second Officer) Open the cross feeds, man!
15 Seconds later, 19 NM from airport:
FO:	It’s flamed out!
	(Captain immediately calls ATC and requests clearance for Runway 28L; ATC instructs to report runway in sight.)
SO:	We’re going to lose number three in a minute, too—it’s showing zero!
C:	You’ve got a thousand pounds—you’ve got to!
SO:	5000 in there . . but we lost it!
C:	All right.
SO:	Are you getting it back?
FO:	No number four—you’ve got that cross feed open?
SO:	No, I haven’t got it open. Which one?
C:	Open ‘em both—get some fuel in there! Get some fuel pressure?
SO:	Yes, sir.
C:	Rotation! Now she’s coming! Okay—watch one and two. We’re showing down to zero or a thousand.
SO:	Yeah.
C:	On number one?
SO:	Right!
FO:	Still not getting it!
C:	Well, open all four cross feeds.
SO:	All four?
C:	Yeah.
FO:	All right—now it’s coming!
C:	You’ve got to keep ‘em running!
SO:	Yes, sir!
FO:	Get this ### (expletive) on the ground!
SO:	Yeah—it’s showing not very much more fuel . .
C:	(to ATC) United 173 has the field in sight now—we’d like to turn left for 28L.
ATC:	Okay—maintain 5000.
SO:	We’re down to one on the totalizer. Number two is empty!
C:	(to ATC) United 173 is going to turn toward the airport and come on in.
ATC:	Turn left, heading 360—and verify you have the airport in sight.
C:	We have the airport in sight.
ATC:	United 173 is cleared for visual approach—Runway 28L.
C:	(to the Second Officer) Reset that circuit breaker momentarily—see if we get gear lights.
C:	(in response to green gear illumination) Yeah—nose gear is down.
C:	(to ATC) How far do you show us from the field?
ATC:	I’d call it 18 flying miles
SO:	Boy—that fuel sure went to hell all of a sudden—I told you we had four (thousand pounds).
C:	There’s an interstate highway type thing along that bank on the river—in case we’re short.
C:	(a minute later) That’s Troutdale (airport) over there—about six of one and half a dozen of the other. (Troutdale is a general-aviation airport located approximately 5 NM east of Portland International.)
FO:	Let’s take the shortest route to the airport.
C:	(to ATC) What’s our distance now?
ATC:	12 flying miles.
C:	About (another) three minutes—four.
SO:	We’ve lost two engines, guys! We just lost two engines—one and two.
FO:	You’ve got all the pumps on and everything?
SO:	Yep!
ATC:	United 173, contact Portland Tower, 118.7—you’re about eight or niner flying miles from the airport.
C:	They’re all going—we can’t make Troutdale!
FO:	We can’t make anything!
C:	Okay—declare a Mayday!
FO:	(to ATC) Portland Tower, United 173, Mayday! We’re . . the engines are flaming out—we’re going down! We’re not going to be able to make the airport!
6:15 pm:
	Aircraft descends into wooded area 6 NM east-south-east of the airport.

Case Study: Air Florida Boeing 737-200 (Palm 90), Washington National Airport, January 13, 1982

Crew

Captain Larry Wheaton—1100 hours of experience as Pilot-in-command on the B737, and previous to this, another 1200 hours as a first officer on DC-9s.
First Officer Roger Pettit—former military fighter jet pilot with approximately 1000 hours as a first officer on B737s.

Weather

It was a cold winter’s day; at the time of the Palm 90 takeoff, the temperature was -4°C, the ceiling 400’, and the visibility fluctuating between ¼ and ¾ mile.

Background

The flight was scheduled to depart at 2:14 pm, but heavy snow closed the airport at 1:40 pm, awaiting the completion of snow removal.
The de-icing of the aircraft was completed at 3:10 pm and the Captain was eager to join the line for departures—there were 12 aircraft scheduled to depart before Palm 90.
Push-back clearance was received by the Captain at 3:23 pm, and a tug attempted to push the aircraft back. The tug, however, was not equipped with chains and got stuck. The Captain, contrary to company policy, elected to start the engines and use the aircraft’s reverse thrust to back out of the gate; this was not successful. A second tug equipped with chains was brought in and pushed the aircraft out without difficulty.
The engines were restarted at 3:38 pm; an analysis of the Cockpit Voice Recorder revealed that the engine anti-ice system was not turned on during the after-start checklist. Included in the engine anti-ice system is the heating of the compressor inlet pressure probe. (The inlet pressure probe measures the pressure of the airflow at the compressor inlet. This pressure is compared to the pressure of the airflow at the exhaust nozzle. The Engine Pressure Ratio, or EPR—pronounced “eeper”—is used to set the thrust on this model B737, and is the turbine discharge total pressure divided by the compressor inlet total pressure. The Palm 90 crew determined that EPR 2.04 was the proper thrust setting for the planned departure.)
The transcript below begins with the aircraft on the taxiway in the line-up for departure. The crew is attempting to position the B737 behind the hot exhaust for the engines of a DC-9 it is following.

Macarthur Job, Air Disaster (Ibid) (see Pp. 83-95 of Vol. 2).

14 minutes to takeoff (Takeoff was at 3:59 pm.):
C:	Tell you what—my windshield will be de-iced; don’t know about my wing.
F:	Well, all we really need is the inside of the wings anyway, the wingtips are gonna . . shuck all that other stuff . .
C:	Get your wing now. (A reference to the effect of the jet efflux of the DC-9’s engines on the Boeing’s starboard wing.)
F:	Did they get yours? Can you see your wingtip over there?
C:	I got a little on mine.
F:	A little! This one’s got about a quarter to half an inch on it all the way . . (now speaking of another aircraft that had just landed): Look how that ice is just hanging on his back . . see that? It’s impressive that these big old planes get in here with the weather this bad . .
11 minutes to takeoff:
F:	(referring to centre panel engine instruments) See the difference in the left engine and the right one?
C:	Yeah.
F:	Don’t know why it’s different unless it’s his hot air going into that right one. That must be it—from his exhaust. It was doing that on the chocks a while ago, but ah . .
ATC:	Okay, Palm 90—cross Runway 03 if there’s space —then monitor the tower on 119.1—don’t call him, he’ll call you.
F:	Palm 90.
F:	I’m certainly glad there’s people taxiing on the same place I want to go, ‘cause I can’t see the runway—taxiway . .
F:	(referring again to engine instruments) This thing’s settled down a little bit—might’a been his hot air going over it.
ATC:	Now for Palm 90—if you’re with me, you’ll be going out after the red DC-9 Apple (New York Air DC-9).
F:	Palm 90.
7 minutes to takeoff:
F:	Boy, this is a losing battle here on trying to de-ice those things. It give you a false feeling of security, that’s all that does.
C:	That—ah, satisfies the Feds (FAA requirements)!
2 minutes to takeoff:
F:	I think we get to go here in a minute . . EPR all the way to 2.04, indicated airspeed bugs are 138, 140, 144 (knots).
C:	Set!
F:	Takeoff briefing: Air Florida standard. Slushy runway—do you want me to do anything special for this (takeoff), or just go for it?
C:	Unless you got anything special you’d like to do.
F:	Unless . . I just lift the nose wheel off early, like a soft field takeoff or something . . I’ll take the nose wheel off and then we’ll let it fly off . .
ATC:	Palm 90, taxi into position and hold, be ready for an immediate (takeoff).
F:	Bleeds?
C:	They’re off.
F:	Strobes, external lights?
C:	On.
F:	Antiskid?
C:	On.
F:	Transponder?
C:	On.
F:	Takeoff (checklist) is complete.
ATC:	Palm 90—cleared for takeoff . . no delay on departure if you will—traffic is two and a half (miles) out from the runway.
Takeoff:
C:	Your throttles.
F:	Okay.
C:	Say if you need the wipers.
C:	It’s spooled . . really cold here (a reference to instrument indications as aircraft accelerates).
F:	Got ‘em?
C:	Real cold . . real cold . .
F:	God, look at that thing! That don’t seem right, does it?
F:	Ah . . that’s not right.
C:	Yes it is—there’s 80 (knots)!
F:	No . . I don’t think that’s right . .
F:	Ah, maybe it is.
C:	120 (knots).
F:	I don’t know . .
C:	V1!
C:	(as aircraft suddenly becomes airborne steeply nose-up:) Easy!
C:	V2!
	(Two seconds later, sound of the stall warning stick shaker begins, continuing to impact.)
ATC:	Palm 90, contact Departures Control.
C:	Forward! Forward! . . Easy! We only want 500 (feet per minute)!
C:	Come on . . forward. . forward . . just barely climb!
F:	Larry—we’re going down.
C:	I know it!
	Sound of impact.

Post accident findings

The engine anti-ice system was switched off at the time of impact.
The actual EPR during the takeoff was 1.70.
Even with an EPR setting of 1.70, the B737 should have achieved a climb rate of more than 1000 fpm.
The First Officer was the PF and the Captain was PNF, while it might have been expected that the PNF would monitor the engine instruments, it was the First Officer who was noting the discrepancy, and there was no evidence that the engine instrumentation was scanned by the PNF.
The engine instrument cluster included N1 speeds (the speed of the most forward low-pressure compressor driven by the most rearward low pressure turbine), and the N1 indications were consistent with the 1.70 EPR.
In the existing conditions, the aircraft should have accelerated to 145 knots (lift-off speed) in about 30 seconds; it took Palm 90 45 seconds.
At 80 knots the B737 could have stopped within 2000’; at 120 knots, the aircraft could still have been stopped on the runway.
Because of the rapid expansion of Air Florida, the Captain had only two years experience as First Officer, in contrast to the then industry average of 14 years.

Palm 90 Flight Path (US TSB). Langley Flying School.

Vertical Fin of Palm 90 (Wikipedia). Langley Flying School.

Rescue efforts for Palm 90 (Wikipedia). Langley Flying School.