| BlueSky Business Aviation News | |||||||||||||||||||||||
| . | |||||||||||||||||||||||
The issue as documented in "TERROR, SECURITY, AND MONEY: BALANCING THE RISKS, BENEFITS, AND COSTS OF HOMELAND SECURITY" by John Mueller and Mark Stewart is that there was no cost/benefit analysis done prior to spending a trillion US dollars. At the core of the issue is the difference between possibilistic thinking and probabilistic thinking. Thinking about this issue is a way to view aircraft design, operation, and maintenance issues. The fundamental difference in these two modes of thought is the difference between "worst case" and "likely case." That is, possibilistic thinking is concerned largely with the worst possible case, no matter how remote. This is a rather narrow focus, as you might imagine. Probabilistic thinking, on the other hand, is concerned with what might happen, and how likely might it be. It encompasses all that possibilistic thinking does, but in a broader scope. Let us see how this works out with airplane design. As we design the aircraft, we look not only at the regulations, but also the accidents that drove those regulations and the performance issues that related to them. Sometimes we use heavier structure to mitigate damage due to bird strikes. Sometimes we add systems like EGPWS, to alert the crew that the ground is dangerously close. Sometimes we change procedures like increasing following distances between aircraft to minimize the impact of turbulence due to wing tip vortices. We usually know the worst case already It's true, we already know the worst case scenario: the aircraft crashes, killing all aboard. And there really is no way to completely avoid that particular scenario short of refusing to design, build, operate, and maintain aircraft, though we do make huge efforts to do minimize disasters, and we have been amazingly successful, really. Possibilistic tunnel vision Possibilistic thinking focusing only on the worst possible scenario has limited utility. There are a great many other outcomes that range from bad to very bad but don't result in the aircraft crashing and killing all aboard. Moreover, possibilistic thinking by its very nature does not deal with those less severe risks. It if isn't worst case, it doesn't matter. This is not especially practical. Bird ingestion, for example For example, let us say that bird ingestion is deemed to be the worst case scenario, as one might we focus on the Lear 24 bird ingestion event at Peachtree Dekalb airport in 1973 (killed all aboard; injured another on the ground). Depending on who is your management, the mandated solution might be something like steel screens over the engine inlets that are strong enough to hit a turkey vulture or a Canada goose without hurting anything. Heavy and horrible for engine performance. Bird strikes in general Alternatively, perhaps a more general concern with bird strikes is determined to be the worst possible case, focusing on the 2008 accident where a Challenger 600 struck a flock of pelicans and bird remains blew through the forward pressure bulkhead and on in to the cockpit. Then we might have two inches of armor plate mandated in front of the pilot. The weight penalty alone might be enough to make the aircraft a loser. Of course, this wouldn't address the issues of airplanes hitting deer and elk on the runway (I'm not making this up; a UAL 737 hit a deer at Chicago O'Hare in 1987, and in 2002 a Lear 36's #1 engine ingested a chunk of an elk). Nor does it address any other hazard to the aircraft. Further down the possibilistic rabbit hole If we decide that loss of an engine is our worst case, then perhaps we are mandated to use engines that never ever fail, no matter what. Since that engine has never been designed or built, we are clearly in glider territory at this point. Same thing if the worst possible case is computer failure, since there have never been a computer that won't fail at some point. Focusing on only one potential case to the exclusion of all else creates a kind of tunnel vision. Variations on a theme Maybe we should discuss the wide variation in "worst possible case" for a moment. For me, as an airplane designer, the worst possible case was spending my life designing an airplane, but the company never puts it on offer, or maybe the worst possible case was getting my design put on offer, but not generating enough interest to launch the program--I can't decide. For my boss, the chief engineer, the worst possible case might be putting the airplane into production and then finding that it had a horrible design flaw. For his boss, the program vice president, it might be that the airplane was put into production, but through poor manufacturing, it did not meet its minimum performance guarantees and the company loses a lot of money. The CEO's nightmare worst case scenario might involve horrible crashes that get the type grounded and the company sued. My point is that "worst possible case" can vary quite a bit depending on who exactly is doing the analysis. Informing airplane design For airplane design, we don't concentrate on the worst case. It's moot anyway. We look at the various ways that airplanes can, and have, failed and we look at the cost of those failures versus the cost of addressing them. Not all of them are as dramatic as crashing and burning, of course. For example, a new, heavier design might require either a much larger, heavier landing gear to spread the weight over a broader area, or it might require the airport authority to pour more concrete. I have done that particular study, and I can tell you, it is cheaper in the long run to pour more concrete. Failed options Does anyone remember the option on the A320 for a double-bogie main landing gear? Airbus introduced the option for aircraft they were selling to India. At that time, the taxiways, ramps, and runways of many Indian airports were particularly thin because they were old and not build with modern airliners in mind. To taxi the airplane about without cracking the ramps and runways, they had to have twice as many wheels and tires as usual. This was such a huge weight penalty that it didn't last very long. Even in India, the airport authorities found it cheaper to pour more concrete than continue to penalize the aircraft. How about the original 777 option for folding wingtips so they could fit into a terminal gate built for a 767? Remember that? Thousands of extra pounds of weight displacing cargo and paying passengers. The cost to benefit ratio didn't pay off; airlines found it far more economical to build larger gates. Not a single folding wing was ever bought. Probabilistic thinking Probabilistic thinking, as contrasted with possibilistic thinking, asks questions like, "How often does this occur? How much does it cost? What do we gain here? Is that reasonable?" In essence, probabilistic thinking points us towards the cost/benefit ratio. I.e., how much goodness are we getting for our dollar? There are no magic wands here; it's just hard work all the way. The design process As every engineer know, design is an iterative process. The designers put something together we think meets the requirements, the analysts tell us how well it performs and how much it costs, then senior management discuss the design with potential customers, feeding back the customers' reactions to the designers as to what needs to be improved, bringing us back to the start of the cycle again with a better understanding of the acceptable cost to benefit ratio. Every good design balances a lot of competing interests. It takes a lot of very open, transparent communication about what is desired, what is required, and how much it can cost. And sometimes there are no good solutions with current technology. Boeing's Sonic Cruiser, for example. Addressing complexity After giving this some thought, I believe that possibilistic thinking is the opposite of what we in aviation should use, which is a more holistic, systems-driven, probabilistic thought process. Possibilistic thinking focuses our attention on a single issue, but probabilistic opens us up to an array of things which require our attention. Ignoring the true complexity of the situation and focusing on a single issue is short sighted, to say the least. Given the actual complexity of our industry, we need a great deal more probabilistic thinking. Terry Drinkard is currently consulting on an aviation start-up. His interests and desire are being involved in cool developments around airplanes and in the aviation industry. Usually working as a contract heavy structures engineer, he has held positions with Boeing and Gulfstream Aerospace and has years of experience in the MRO world. Terry’s areas of specialty are aircraft design, development, manufacturing, maintenance, and modification; lean manufacturing; Six-sigma; worker-directed teams; project management; organization development and start-ups. Terry welcomes your comments, questions or feedback. You may contact him via terry.drinkard@blueskynews.aero Other recent articles by Terry Drinkard:
|
|||||||||||||||||||||||
