Which brings me to suggest that a multitude of video cameras placed around the airplane, inside and out, might be a good idea, no? There have been many incidents in the last few decades (parts blown off, cargo on fire, etc.) where the inability to view critical places was a serious and sometimes fatal factor. With the cost of video cams near zero and the size/weight insignificant, maybe now is the time to consider a surveillance-type scheme for planes that makes it possible to see the unseen?
With large systems (and especially safety-critical systems) like airliners, the cost of the component is the least of your worries. Consider:
-
Cost of designing and building cameras that can endure years and years of use under harsh conditions.
-
Cost of installing the cameras, and installing wiring to each camera.
-
Extensive testing to make the cameras themselves won’t present any risk to the aircraft: Can they fail and short-circuit the power system? Can they send a spurious signal or power surge and screw up the flight control system? Will the video signals interfere with critical signals going through nearby wires? Can the cameras themselves short-circuit and burn up?
-
Cost of adding a display system to the cockpit, or adding that capability to an existing cockpit display. Which is a significant cost if built to airliner quality standards. And of course it needs to be fully tested for safety.
-
Cost of maintaining the camera system throughout the decades of service.
-
What to do if one of the cameras fail. Do you ignore it and fly, because the camera itself is not a safety-critical system? But we said the camera is a safety system. But if you have to fix it before flight, the delay will cost the airline.
It keep$ coming back to the $ame i$$ue. How “many” incidents are we talking about over the last few decades in which a system of on-board cameras could have saved lives, and (more importantly) how many lives do we think could have been saved by such systems if they were installed fleet-wide? Now compare that to the total fleet-wide cost of designing, FAA-certifying, installing, inspecting, maintaining, and repairing such systems of cameras. If the total cost of such a program is more than $5.8M per expected life saved, then it’s not worth it.
Can you give some examples of pasts incident in which you feel an onboard camera would have provided the crew with critical information that would have changed their course of action?
In addition, if you find you don’t have the ability to make it all the way to Denver, you’ll want to land in the best possible conditions given the tire damage. It was 80-plus degrees, clear and dry in Orlando on Thursday. In Denver it barely got above freezing and vacillated between cloudy and something called “ice fog.” Beyond that, if forced to make an off-runway landing the pilot’s options were icy-foggy mountainous Colorado or sunny, flat-as-a-billiard table Florida.
Southwest’s FLL-DIA route is partially overwater as it crosses over the Gulf, and then travels through a lot of relatively unpopulated areas of the South and West with no airports (or at least none big enough to handle a 737).
Similar objections could be made about any other system on a modern airline, yet with multiple, complex systems, flying is safer than ever.
What I propose could have been done decades ago, but the costs of material would have been much higher. Now, the cost of many components has been significantly reduced and the options are greater.
As always with such proposals, designing a single system for a single plane would be prohibitive. But designing for multiple planes or an entire industry is quite different.
Imagine the cost if a GPS system (highly complex, very expensive) served only one plane. And who needs it? We have other, proven navigation methods! Now contrast with the actual system as currently implemented. I don’t think I’ve heard many complaints about the cost recently.
Not without time-consuming research for exact links which I am not willing to do. But as someone who has watched hundreds of air disaster videos, I can remember (IIRC) events like these, and I recall thinking, “If only they could see what is going on!”:
[ol][li]A pilot wasn’t sure which engine was on fire, and he shut down the wrong engine. He couldn’t see any engine from his cockpit seat. A quick glance at a video might have saved his ass. He crashed.[/li]
[li]A fire in the cargo hold, unknown until the plane’s control systems began to shut down. They couldn’t land in time. Fatal. (I believe this scenario happened more than once – oxygen canisters and burning tires.)[/li]
[li]A fire in the overhead area above the cockpit. Not realized until too late how serious it was. Fatal.[/li]
[li]Uncertainty if the front landing gear was down. Pilots couldn’t see, so one climbed into the wheel well and the other wasn’t watching the instruments as the plane descended. Fatal.[/li]
[/ol]
I was only addressing your statement that cost of cameras is “near zero.” The cameras themselves cost almost nothing, but installing a camera system on airliners is a substantial cost, and therefore need to be weighed carefully against the benefits.
Everything I described applies to designing a camera system into a new plane. It’s an additional subsystem that adds cost to every stage of design, manufacturing, testing, certification and operation of the airliner.
I can’t find the cite, but I think I’ve read about one such incident because the temperature sensors on the engines were miswired. Which could happen to cameras too. And there was another incident where the aircraft manufacturer changed the design, and the pilot didn’t know it. (Pilot thought smoke in cabin meant left engine fire, because the earlier version of the same plane drew cabin air from the left, but the newer one didnt’. Or maybe it was the other way around.)
[quote]
[li]A fire in the cargo hold, unknown until the plane’s control systems began to shut down. They couldn’t land in time. Fatal. (I believe this scenario happened more than once – oxygen canisters and burning tires.)[/li][/quote]
Which is why smoke detectors and/or fire suppression systems are now required on most types of airplanes. (IIRC the requirements are less stringent for cargo planes, but one or the other is still required in the cargo hold.)
[quote]
[li]A fire in the overhead area above the cockpit. Not realized until too late how serious it was. Fatal.[/li][/quote]
I’m not familiar with that one. But if you are talking about the avionics bay, there are smoke detectors there already.
There are sensors that tell the pilot the gear is down and locked.
I think you are describing Eastern Air Lines flight 401. While a burnt out indicator light was a factor, the real cause was poor CRM (crew resource management). You can’t fix poor CRM by removing all possible distractions from an airliner.
And I replied to this.
Smoke detectors, at least the simple ones, are useful, but no substitute for a camera. Detectors use smoke, temperature, or other senses. They don’t provide an image, so are deficient beyond just “it’s hot in here!” They don’t easily show the spread or intensity as well as a camera does. Sure, better sensors can be developed, but the better they are, the more like a camera they become. Human intelligence can determine much from an image that might take multiple sensors to accomplish the same thing. Then the cost of a fancy sensor system might be greater than a camera.
Same response as above.
And they aren’t reliable and aren’t always trusted. Nor should they be. A second opinion would be a good idea.
But you can improve your chances with more data and better observations.
Quick now, an indicator says there’s smoke in cargo bay #1. Is it a false alarm, a cause for concern, a fast-spreading fire, an emergency that requires immediate landing, or can we wait for a while to give time for diagnosis?
This. Planes are already pretty well instrumented in areas where problems are most likely to occur.
Suppose for the sake of discussion that you put cameras in all of the places where you’re worried about fire. Which places would those be? A good starting point would be all of the places on the aircraft where there are fire/smoke detectors already. Now you need to display all of those video feeds in the cockpit. But the instrument panel is already pretty crowded. So maybe you make room for a single video display screen, and the pilot has to select which camera feed to show on the screen. And the user will have to choose, because there won’t be any real-time video analysis that can tell the pilot that something important/bad is happening in any particular area. How will he know to look at the video feed from any given area? How about we use the smoke/fire detectors that are already in those areas to let the pilot know he should take a look? OK, great. But what is the video footage going to tell the pilot that a smoke/fire detector isn’t already telling him? That it’s a small fire and he can safely ignore it?
Whereas a camera can only see flames. It won’t detect a smoldering electrical fire, or a burning Li-Ion battery pack inside a cargo container. Also keep in mind that most internal volumes of airliners are dark.
That’s why there are multiple sensors on each landing gear.
You assume the worst and land immediately.
What are you going to do if a smoke detector goes off, but you don’t see anything on the camera? Is it safe to assume it’s a false alarm? I wouldn’t bet my life on that.
By the way, the flight that crashed because the pilot shut down the wrong engine - that was British Midland Flight 92, aka the Kegworth disaster. The accident report says the cockpit instruments almost certainly displayed the vibration warning for the correct engine, but the pilots either had trouble reading the display, or didn’t make the effort. They were misled by the initial impression formed by their prior (and incorrect) knowledge of the airplane design.