A thought about safety engineering of complex systems.

The aerospace industry, the nuclear reactor industry, and other high technology industries have pioneered in developing means to create very safe machines and systems and to reduce accidents.

Think of a space mission. Multistage rockets harness huge forces created by controlled explosions that go on and on for minutes at a time. The vehicles have to be so precisely targeted that they travel for days, weeks or months to unerringly reach a destination unseen at the firing.

This reliability is achieved not only be eliminating the obvious glitches in designs, but by identifying combinations of minor problems which if occurring simultaneously would be dangerous to the entire system.

Commercial aviation is perhaps the leading example of safety management. The manufacturers, essentially the same companies and people who developed space technology, use the entire safety engineering toolkit to make planes as safe as they can. A major manufacturer of commercial jet aircraft produces only a small number of multi-million dollar planes a year. If even a few of these fail, the company is out of business. So too, these companies carefully study the records of maintenance and safety of the planes in operation, and provide improvements to meet problems encountered in service.

Source

The commercial airlines similarly depend on their safety records for their very existence -- the jobs of their personnel, the profits of their investors. They take maintenance very seriously. Moreover, they take extreme measures to prevent human error from causing failures of their flights. They hire experienced flight officers, provide extensive training, and have strict safety checks and rules.

Governments have agencies such as the U.S. Federal Aviation Administration that conduct their own studies of accidents to determine causes and assure corrective action to prevent repetition of those causes. They regulate manufacturers and airlines to assure safety. Airports are closely guarded,

Moreover, airlines and governments take extreme measures to prevent criminal efforts to endanger flights or passengers.

So how come the occasional airplane is lost. First, that happens less often than you think. There is an "availability bias". People think that events like those readily available to memory are frequent. But, as the Malaysian Airlines 370 news coverage demonstrates, we will all be able to remember the loss of that plane for a very long time. The disappearance of a commercial aircraft is so newsworthy just because it is so rare.

A few airplanes have been hijacked or destroyed by terrorists, and I suppose those events were made possible by extremists doing something that the safety engineers had not previously imagined could or would be done.

A few airplanes fail due to human, mechanical, electrical or electronic failures. In many of these cases, a combination of things occurs together that was so unlikely that it was not foreseen and prevented.

It would seem that the loss of Flight 370 was of that nature. And of course, the search is so active (costing huge amounts of time and resources) so that investigators can learn what happened to the plane and take appropriate steps to assure such an failure never happens again.

A Final Thought

Let me also note that different investigators using different tools are looking into Flight 370. A number of agencies are trying to recreate the flight path from the information available to them. When they make a "best guess" based on the data and analysis to date of where the plane might have gone down

• various agencies analyze their satellite data to see if there is debris near that location that might relate to the plane;
• other agencies send aircraft to search the area at low altitude with human observers, also to look for debris fields;
• ships search the areas containing possible debris fields identified by satellite and aircraft to retrieve items, which
• are evaluated by still others to determine if they are indeed from the lost plane.

If a better guess can be made of the flight path and final crash site, then the other search agencies move accordingly.

Time is short to find the debris, and each agency tries to do the best it can with the available information.

If and when the crash site is located, very specialized underwater search vehicles (with very limited search ranges) will be put into operation to try to find the wreckage.

Of course, others must be working. The airline is analyzing its maintenance and personnel records without doubt for clues; police is searching for clues that there might have been criminal action involved; the manufacturer of the plane is no doubt seeking clues in records of that plane and similar planes; other organizations are seeking information from their data (radar, radio signals, etc.). Oceanographers are no doubt checking and improving their information on the movement of surface debris in the ocean to the west of Perth, Australia. Even the news agencies are seeking information, not only from official sources, but also from unofficial sources that may have complementary views.

Time is of the essence since a plane's wreckage at the bottom of the ocean will be very difficult if not impossible to find with current technology if the crash site is not determined. It is not a waste to determine that something that might be true is not actually true! 

"How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?'
Sherlock Holmes in The Sign of the Four