TSB Recommendation A96-12

In accordance with standard investigation practice, accident pilots' decision-making processes are analyzed. The temptation to judge the quality of a pilot's decisions by the outcome, however, must be guarded against. Fairness to the individual and the advancement of transportation safety require that the actions of the pilot be understood within the context in which pilots typically are operating.

Cockpit decisions can be conceived of as having two components: situation assessment and selection of a course of action. Thus, the difficulty involved in decision making depends primarily on the degree of clarity of the cues signifying the problem and the nature of the response options available in the situation.

Cues, or information about the situation, can vary between clear and ambiguous. In situations such as emergency procedures, the cues are so strong that decision options are prescribed and the response is automatic. Sometimes, once a situation is defined or interpreted, the pilot must choose among options, or schedule tasks in the most appropriate order for the conditions. This represents a higher level of complexity. The most demanding decision making involves those situations where there are no pre-determined options. In these cases, pilots must develop their plans by using both their knowledge of the system and their assessment of the situation. The more complex and difficult a decision is, regardless of whether the difficulties are in the situation assessment component or in the response selection component, the greater the likelihood of a decision that is less than ideal. Consequently, decision tasks in the cockpit differ in the degree to which they are well-structured and thus have an agreed-upon “best” solution.

The accident pilot was familiar with the area and the aircraft. Low ceilings and poor visibility are common occurrences in the area and he had flown many hours in similar meteorological conditions. As he approached Campbell River, the ceiling at the airport was low (about 300 feet), but over the water, the clouds were about 1,000 feet above the surface.

The decision making in this occurrence required that the pilot assess the situation and choose between continuing to the airport or landing on the water at Tyee Spit, which would undoubtedly have entailed some inconvenience for the passengers. He may also have considered climbing and requesting an IFR approach, for which he was qualified and equipped.

As the pilot approached the control zone, the weather in his immediate area was above ceiling and visibility minima. The Campbell River airport reported a ceiling of 300 feet and two miles visibility, but the ceiling had been varying between 300 and 500 feet through the day and an aircraft ahead reported sighting the Campbell River runway from 900 feet asl. The pilot had an SVFR clearance which only required a flight visibility of one mile and the aircraft to remain clear of cloud. The fact that the aircraft was instrument equipped, and that the pilot was qualified for instrument flight and experienced in conducting instrument approaches into Campbell River would likely have further contributed to his confidence in continuing to the airport. The course of action selected by this pilot, based upon the existing regulations and his experience, would have been taken by many pilots with similar experience.

Once the decision to try to land at the airport was made and the aircraft turned inland, the clarity of the visual cues deteriorated. At about 2 1/2 miles from the airport, the pilot apparently changed his plan and attempted to set up an approach from the same direction as the aircraft which had reported sighting the runway from 900 feet. It is doubtful that the pilot would modify his plan as he exited from the control zone; in principle, though, his visibility criteria increased from one mile for SVFR to two miles for VFR in designated mountainous terrain as he crossed an imaginary line at the boundary of the control zone. It is likely that his attention was focused on controlling the aircraft and his workload was heavy at this point, so that both the transition to uncontrolled airspace and the significance of it were probably not recognized at a conscious level. It appears as though he did maintain visual reference with the ground during this time.

In the light of the outcome, changing his plan and returning to Tyee Spit would have been more prudent; but the cues available to the pilot were apparently not compelling enough to change his mental model, or assessment of the situation. Once individuals select a particular course of action, it takes very compelling cues to alert them to the advisability of changing their plan. Indeed, there is a tendency to use these cues to confirm the validity of the intended plan of action. This pilot had control of the aircraft, was maintaining visual contact with the ground, and was able to navigate, probably with the aid of instruments. These cues would be sufficient to lead many professional, safety-conscious pilots to continue the approach to the airport.

A recent analysis by the National Transportation Safety Board (NTSB) in the US involving 37 accidents showed that, when faced with Go/No-Go decisions, 66% of the crews continued with their current plan in the face of cues which suggested they should have abandoned it. “However, in many of these cases the cues were ambiguous and it was difficult to assess with great confidence the level of risk inherent in the situation.”

Inexperience, lack of knowledge, imprecise guidelines, and ambiguous cues will continue to make some pilot decision-making circumstances difficult. However, strategies to maximize the quality of decisions can be learned, and include such things as situation assessment, risk assessment, planning, resource management, communicating, and the identification of special skill requirements.

Of note, increasing the SVFR minima could change the nature of decisions to be made in occurrences like the Campbell River accident. If there had been a greater minimum visibility requirement, or rules requiring a combination of ceiling and visibility, the pilot would have been faced with an easier decision. Had the previous SVFR rules been in force, the only option would have been to land at Tyee Spit. Thus, the need to re-assess the adequacy of current VFR weather minima (recommended above) should be considered in the light of normal pilot decision-making processes.

The accident pilot was operating in an environment which accepted as “normal operations” flights into marginal weather conditions, when other options, such as landing at Tyee Spit, were available. Typically, pilots involved in VFR CFIT accidents have not had any special training in decision making. At the same time, pilots flying in small air carrier operations are arguably the most exposed to such ambiguous situations with the least decision-making support. They often operate as single-person crew into a variety of unfamiliar locations with minimal infrastructure and are self-dispatched, and the aircraft do not usually possess sophisticated instrumentation or control systems.

The Board has previously recommended that Transport Canada develop and implement a means of regularly evaluating the practical decision-making skills of commercially employed pilots in small air carrier operations (TSB A90-86). TC's response in part stated:

It has been Transport Canada's position that the benefits of this training were intrinsic in the enhanced performance of the pilot and that a properly planned and executed Pilot Proficiency Check would provide a practical and realistic assessment of a pilot's ability to make reasoned and timely decisions when faced with a simulated emergency situation. We will continue to keep abreast of developments in the field of decision-making training and assessment and will not hesitate to introduce improvements in our present system should they become available.

National and regional carriers have broadly embraced the concepts of Cockpit Resource Management (CRM) and Pilot Decision Making (PDM) training, and under the new Canadian Aviation Regulations, airline operators will be required to complete initial and recurrent CRM training. However, for other commercial operators, formal programs are being introduced on a voluntary basis. The Board believes that, given the natural human limitations in interpreting distances in marginal visibility conditions, natural human tendencies in complex decision making in the presence of changing and ambiguous cues, and the CFIT accident record involving small commercial operators, further counter-measures are required to facilitate safe crew decision making. Therefore, once again, the Board recommends that