Tuesday, July 2, 2013

The Intricacies of the Checklist

Checklists.  Some pilots swear BY them, other pilots swear AT them.  You either love them or you couldn't be bothered to ever look at them and your checklist has been lost under the seats for the last 6 months.  You know if I'm talking to you.  I've been on both sides of the camp at one time or another, but I think a lot depends on whether or not the checklist has been tailored specifically for your own specific operation or not.  A well designed checklist is crucial to whether its useful in the cockpit or viewed as an inefficient formality that eventually gets disregarded.

Our operation, being survey, is vastly different than the majority of operations out there in some very specific ways, and if we were to use a stock off-the-shelf checklist for a Piper Navajo in our operation it would be essentially useless.  Being a survey operation we're an extreme example, but the same holds true in some way or another for probably every other operation out there, whether its a private owner/pilot that flies mostly VFR, IFR commercial charter operation, cargo carrier, or any of the other endless combinations in between.

There's a fine line between having every conceivable item on the checklist so that it can be 100% reliable in every situation, and having SO many items on it that in regular operations it becomes so cumbersome that pilots start to skim over it or disregard it completely.  Take for example VFR flying and IFR flying.  In IFR flying there a lot of additional instrument checks that need to be made before each flight.  If you're going to be navigating by VOR (a radio nav-aid) for example, its imperative that you check your VOR instrument's accuracy before you take-off and start to rely on it to take you to your destination.  If you're flying VFR however and primarily by GPS and pilotage (navigation by looking out the window and comparing with your map), checking your VOR indicator for accuracy is of critical importance.  And if you fly VFR 99.9% of the time, if you had to slog through a checklist designed for IFR, eventually you'd get tired of the inefficient process of going through all the "not applicable" items, and you'd either stop using the checklist altogether, or start skimming through the items, which has arguably more potential for missing critical items then not using it at all.

Another reason against using off-the-shelf checklists that are not tailored to your operation is because they may not contain critical items on them that are specific to your particular operation, and you won't be able to rely on them 100%.  If you can't trust your checklist, what good is it?

Furthermore, checklists have to be structured in such a way that they prove to be an efficient and useful tool in an environment as dynamic as a cockpit.  There was an accident several years ago of a passenger airline, the time and place of which I can't remember anymore, that was part of a Mayday episode.  The airplane blew a tire on take-off which went unnoticed by the flight crew.  The blown tire ended up erupting into a fire, which spread throughout the airplane after they retracted the landing gear into the aircraft's belly.  The airplane ended up crashing, and while it wasn't the main cause, the investigators noted that the airplane's cabin fire checklist was a sequence that could take as long as 30 minutes to complete.  Due to this the flight crew chose to circle for several minutes while they went through and completed the checklist.  Unfortunately the airplane didn't have that long before the fire consumed the aircraft.  When they could have been descending to land and survive, their checklist and standard operating procedures failed them by requiring that they remain airborne until they completed their checklist.

That is an extreme example of course, but good checklist design can even be applied in the case of an engine failure in a small aircraft.  After an engine failure there are several things that need to be accomplished very quickly in order to maximize your chances of survival.  After those are finished you may or may not have bought yourself a little extra time to troubleshoot the situation.  In almost all cases it makes sense to complete those immediate critical items by memory first as a "flow", and then only once you have time, pull out the checklist, verify what you've already did, and then continue on with the troubleshooting using the checklist line-by line.  This is a great example of using a checklist in two different ways (Do & Verify, or Read & Do line by line), and a checklist design can either hinder that or help that by organizing each section of the checklist appropriately for the method that is best used at each time.

Consider even a pre-landing checklist on a small airplane.  Most pre-landing checklists are completed in the traffic circuit, the time in the flight with the highest air traffic density and when the pilot should have most of his attention focused outside to look for other aircraft and flying the airplane.  A poorly designed and unnecessarily long-winded checklist takes the pilot's attention away from looking for airplanes and down to his lap has he checks off each item line by line.  It may also increase the chances that it doesn't get completed if a distraction occurs mid-checklist (like a radio call).

Until I started putting thought into designing checklists its easy to overlook the multitude of human and operational factors that need to be considered when creating checklists.  Its quite the science actually, and there's even been studies and papers done on the topic.  Here's a fantastic paper from NASA on the concepts, design & use of cockpit checklists.  I found it one day when I was overhauling our checklist for the Navajo and was looking on some guidance on how to structure it.

If you're one of the pilots who isn't in the habit of using your checklist religiously for EVERY phase of flight, maybe you should consider why.  Is it designed in such a way that makes using it inefficient or cumbersome, is it not tailored specifically to your type of operation?  Maybe you'd benefit from a checklist overhaul.