A new car built by my company leaves somewhere traveling at 60 mph. The rear differential locks up. The car crashes and burns with everyone trapped inside. Now, should we initiate a recall? Take the number of vehicles in the field, A, multiply by the probable rate of failure, B, multiply by the average out-of-court settlement, C. A times B times C equals X. If X is less than the cost of a recall, we don’t do one.
I agree. But automation has made air travel safer by an order of magnitude. The problem with the 737 Max debacle was trying to use automation as a band aid to avoid costly recirtification and pilot training. They didn’t inform pilots about the new pitch control system and they didn’t train them on how to deal with runaway trim. Oh and relying on a single sensor to detect ACA was also a bad move. So, many mistakes … for which a lot of people died.
Notwithstanding that, every day hundreds of planes rely on automation to help keep passengers safe.
You’d think aerospace engineers would have it down to reflex that things need to be fail safe. It’s ironic a system designed to make the plane safer actually crashed the plane. That one should get an award for world’s worst engineering.
Like any accident it wasn’t just one thing. The maker implemented a safety system that was not fault tolerant, then the airline neglected to train pilots how to deal with a failure of that system. In fact that particular airline didn’t even know the system had been added to their planes. Bad engineering, communication, and training still happens in the industry, but really it’s pretty amazing how safe these machines are overall.
Pilot error is still the cause of a majority of accidents. A big problem is bad pilots that don’t pass regular exams can slip through the system because of management deficiencies. Like pilots it happens in the medical industry where bad doctors or nurses just get passed on from one hospital to the next. Employers fail to do proper checks on previous job performance.
100%. A failure at every possible level. Shame on Boeing for outsourcing the design of the 737-Max. I believe it was contracted to India?
I’m going to give the pilots a mild pass since I’ve read they’re instructed to ignore their gut and trust the instruments since their instruments are “always right” and their gut can be wrong.
Pretty sure I read while this was new that the design changes were considered minor enough that recertification wasn’t required. So I’d put that on Boeing, too. It’s obvious that airlines aren’t going to recertify on functionally equivalent design, and also obvious that these weren’t equivalent designs.
From my understanding of the situation, it did have two sensors but it would act if only one of them said the angle was too high. One of the fixes they added was that it warns the pilot instead of engaging if the two don’t agree.
I think the only way it can realistically work within the next decade is if all road construction is diligently recorded into a database and correct paths are predetermined and checked.
Thing is, I’m sure there’s literally millions of km of road across a country where the road designs only exist in a cad drawing, or some realllly old drawing, or not at all.
So basically, won’t happen this decade :P
Or ya know, build more autonomous vehicles that use tracks, like trains, or busses on predetermined routes.
Or build more trains that are reliable and come often enough and eliminate the need for cars in many areas and so significantly reduce the area needed to be mapped.
Trains, trams, anything. There are always going to areas, especially in the US, that 100% public transit is a pipe dream, but we can certainly try. In my mountain town I’ve floated the idea of a gondola that runs above the main boulevard the length of the valley (~7-8 miles) and people look at me like I have 3 heads. The majority of our town’s services are along said road, as well as hotels and tourist hotspots, so it makes sense to me.
Problem is you get the old-school types that think “public transit” = “you’re taking away my car!” and it always goes nowhere.
I’m not sure any car company could make me feel comfortable about using a full auto feature.
Error: Angle of Attack too high, pitching nose down.
Error: Stall warning! Pitching nose down. Please do not resist!
spoiler
For Context: https://en.wikipedia.org/wiki/Maneuvering_Characteristics_Augmentation_System#Role_of_MCAS_in_accidents
I knew that was the Boeing 737 Max “issue” without clicking the link. When safety is an add-on cost, capitalism/profit margins negates safety.
From Fight Club:
Reading about the 737 Max disaster should be a must for everyone.
I agree. But automation has made air travel safer by an order of magnitude. The problem with the 737 Max debacle was trying to use automation as a band aid to avoid costly recirtification and pilot training. They didn’t inform pilots about the new pitch control system and they didn’t train them on how to deal with runaway trim. Oh and relying on a single sensor to detect ACA was also a bad move. So, many mistakes … for which a lot of people died.
Notwithstanding that, every day hundreds of planes rely on automation to help keep passengers safe.
The other issue is the only had one of those angle of attack sensors and should’ve had more redundancy
You’d think aerospace engineers would have it down to reflex that things need to be fail safe. It’s ironic a system designed to make the plane safer actually crashed the plane. That one should get an award for world’s worst engineering.
Like any accident it wasn’t just one thing. The maker implemented a safety system that was not fault tolerant, then the airline neglected to train pilots how to deal with a failure of that system. In fact that particular airline didn’t even know the system had been added to their planes. Bad engineering, communication, and training still happens in the industry, but really it’s pretty amazing how safe these machines are overall.
Pilot error is still the cause of a majority of accidents. A big problem is bad pilots that don’t pass regular exams can slip through the system because of management deficiencies. Like pilots it happens in the medical industry where bad doctors or nurses just get passed on from one hospital to the next. Employers fail to do proper checks on previous job performance.
100%. A failure at every possible level. Shame on Boeing for outsourcing the design of the 737-Max. I believe it was contracted to India?
I’m going to give the pilots a mild pass since I’ve read they’re instructed to ignore their gut and trust the instruments since their instruments are “always right” and their gut can be wrong.
Pretty sure I read while this was new that the design changes were considered minor enough that recertification wasn’t required. So I’d put that on Boeing, too. It’s obvious that airlines aren’t going to recertify on functionally equivalent design, and also obvious that these weren’t equivalent designs.
From my understanding of the situation, it did have two sensors but it would act if only one of them said the angle was too high. One of the fixes they added was that it warns the pilot instead of engaging if the two don’t agree.
The had two sensors, but used one.
WOOP, WOOP!
WOOP, WOOP!
Terrain.
Terrain.
Pull up!
Pull up!
https://youtu.be/oCqe751tOAo
Not a pilot, but that sound terrifies me everytime I watch an aircraft incident video.
Here is an alternative Piped link(s): https://piped.video/oCqe751tOAo
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I’m open-source, check me out at GitHub.
Thought that was going to be the opening sequence of Mission Impossible 2, but that was so much better
Trying to keep it light and fun :)
Wasn’t Boeing using outsourced software engineers that were being paid $9/hr for that system?
I think the only way it can realistically work within the next decade is if all road construction is diligently recorded into a database and correct paths are predetermined and checked.
Thing is, I’m sure there’s literally millions of km of road across a country where the road designs only exist in a cad drawing, or some realllly old drawing, or not at all.
So basically, won’t happen this decade :P
Or ya know, build more autonomous vehicles that use tracks, like trains, or busses on predetermined routes.
Or build more trains that are reliable and come often enough and eliminate the need for cars in many areas and so significantly reduce the area needed to be mapped.
Trains, trams, anything. There are always going to areas, especially in the US, that 100% public transit is a pipe dream, but we can certainly try. In my mountain town I’ve floated the idea of a gondola that runs above the main boulevard the length of the valley (~7-8 miles) and people look at me like I have 3 heads. The majority of our town’s services are along said road, as well as hotels and tourist hotspots, so it makes sense to me.
Problem is you get the old-school types that think “public transit” = “you’re taking away my car!” and it always goes nowhere.