Return to site

It’s Not the Tool. It’s the Craftsman (Part 1)

The Tools of Analytical Decision Making and Why They Sometimes Fail

This is the first of a three-part series on decision making. Parts 2 and 3 will be published successively in the next two blogs.


During the preflight inspection, we notice a rather large nick in a propeller blade. Can we safely make our planned flight? The weather is marginal VFR rather than the forecast good VFR. Can we modify our route and safely make our planned flight?

It is certainly not news that sometimes, even with good intentions, pilots make bad decisions. The literature has an abundance of information on decision making. The behavioral and cognitive psychology folks have produced much valuable information gleaned from meaningful research over the past ten years. We need to recognize that our decision making is separated into two distinct branches. Analytical decision making is the branch in which we have enough time to ponder our decision. The second branch is called automatic decision making by the FAA, but I prefer to call it rapid decision making. That branch has a sub-branch that is called, reflexive or naturalistic decision making. I prefer to think of rapid decision making as being concerned with solving a problem that allows us a short, but reasonable amount of time. That would include and engine failure at 5,000 feet above the ground in clear weather and in an airplane that has a reasonable glide capability. We need to make some important decisions, but we have a little time to consider our options. Reflexive decision making would come into play if the engine was to fail at 100 feet above the ground on initial climb. That would require an immediate response, almost like a reflex action.

Our specialized flavor called Aeronautical Decision Making, or ADM, includes several tools to help us make better decisions. For our analytical decision making, the kind in which we have enough time to ponder our decision, we have the Risk Assessment Matrix and the DECIDE Model. The DECIDE Model spans both analytical and rapid decision making, but is not suited situations that require immediate action, such as an engine failure at low altitude during initial climb. We will address that in Part 2 of this series.

Both the Risk Assessment Matrix and the DECIDE Model can be valuable and can help us make better, safer decisions. But we must remember that they are tools and must be used properly or, like any tool, they can be dangerous. Let’s start with the Risk Assessment Matrix.

text here.

broken image

To analyze an issue and determine how much risk is involved, we begin by choosing the likelihood that the undesirable outcome will occur. We then choose the severity of the outcome should it occur. We extend the lines and find the point of intersection and learn if our risk is low, medium, or high. The flaw in this technique is our humanness. We must make critical decisions about the likelihood and severity of an outcome. We may not have complete or accurate information to make the decisions. And these decisions are swayed by our unconscious mind, including our cognitive biases. In the opening example of this piece, the pilot observes a nick in a propeller blade. Does the pilot have the expertise to determine the likelihood that the blade will fail? To what degree is the pilot being influenced by optimism bias? Given the same damaged propeller blade, one pilot might say the likelihood of having the blade fail is “improbable” while another pilot might say that it is “remote.” The difference in that determination makes the difference between “Low” and “Serious” in the risk assessment.

The DECIDE Model has similar flaws. It tells us to Detect a change or hazard, Estimate the need to counter or react to the change, Choose a desirable outcome for the flight, Identify actions that can successfully control the change, Do take the necessary action, and Evaluate the effect of the action.

broken image

All those steps can be influenced by our unconscious mind. In the first step, detect a change, an engine failure or a failed vacuum system is like a slap on the back of the head. It is hard to ignore. But a gradual loss of engine power, a faint odor of smoke, or an oil temperature gage creeping toward the high temperature limit, might be ignored. Our perception can be influenced by optimism bias and a problem might be ignored until that slap on the back of the head tells us we have a big problem. Estimate the need to counter or react to the change is great advice, but our estimate of the severity of the situation can be highly influenced by how important it is to complete the flight. It then tells us to Identify actions that can successfully control the change. If our estimation of the severity of the problem in the previous step told us that the problem is not severe, we might not even get to this step. Assuming we do get here, the choices we identify can again be influenced by the unconscious mind. Inconvenient choices might be quickly overlooked. the next step calls for us to “Do” take the action. That might have us on a dangerous path. The final step telling us to “Evaluate” the effect of the action is highly susceptible to confirmation blindness and optimism bias.

I am not saying that the Risk Assessment Matrix and DECIDE Models have no value. They do and they can be helpful in directing our thoughts and actions toward making an informed decision. But they both require subjective rather than objective decisions which provide plenty of wiggle room for our unconscious minds to steer us wrong.

In Part 2 of this series we will look at rapid and reflexive decision making and critiquing the more tools commonly recommended to help us in our decisions.

In Part 3, we will see some ways that we can effectively improve our decision making even if our tools are sometimes flawed.