4 ms·

You could basically make this statement every year since commercial airliners were introduced. For commercial air transport, the 'bleeding edge' (which I put in

You could basically make this statement every year since commercial airliners were introduced. For commercial air transport, the 'bleeding edge' (which I put in quotes because it's actually quite conservative, relative to the bleeding edge in other fields) has always suffered from minor (and sometimes major and deadly) product quirks. Military aircraft even more so. This is also why there's a very strong industry preference for airframes that are 'tried and true' - e.g. 737 variants up until the MAX, A320 and A330, etc.

Someone 5y ago
And it’s also unavoidable. To really know how an airplane will age during its lifetime, you have to build it and use it for 50+ years. That would mean all ‘in service’ aircraft would be based on 50+ year old designs. That clearly is undesirable, if only because it would be too expensive. So, instead, one uses models for the effects of aging, for example by repeatedly freezing and heating parts while they also are being stressed physically and are under intense UV lighting. Those models can be wrong, but if you’re using new tech, you can’t afford to wait with selling airplanes until you know they are right. (Of course, this isn’t limited to the aircraft industry; it happens in all engineering. For example, hard disk manufacturers don’t run new hard disk technology for decades to measure its MTBF, either) Sometimes, such models need major adjustment, as happened with https://en.wikipedia.org/wiki/De_Havilland_Comet https://en.wikipedia.org/wiki/De_Havilland_Comet.
- ethbr0 5y ago
  I guess GP's point could be summarized as "Does the amount of complexity required to improve planes now exceed our ability to manage?" Curious in responses from that perspective. Are airframes that much more complicated? Or just better due to advances in aerodynamic understanding, but not much more complicated? And same question for engines. Are modern high-bypass turbofans more complex?
  - S_A_P 5y ago
    This. It seems that to some degree our knowledge rests on prior experience and passed down knowledge. As we build abstraction layers on top of technology the bare metal understanding is lost. I’m going to contradict my original question to some degree by stating that I know that air travel is safer than ever. My question is more just wondering out loud if there is a point where we rely on “solved” problems so much that we start to lose comprehension of the factors that are involved. It feels like the MCAS was an example. Agree that you can never “know” how an airframe will age until you build it and doesn’t seem like this is as critical as MCAS issue but it could be that we have to wait 15 years to find out.
    - ethbr0 5y ago
      I think complexity + necessarily large organizations are a huge contributing factor. MCAS would have not been an issue is any one of the layers above engineering had questioned the decision to avoid training requirements (e.g. Boeing management, FAA management). Unfortunately, systems are so complex now that it was possible to summarize the details in a way that hid the necessary information. And when it went up further, even more summarization. It feels like this wouldn't have been as easy with simpler, older systems. I mean, presumably a manager could look at a biplane and say "Hey, that doesn't have enough thrust to weight."

Someone 5y ago

And it’s also unavoidable. To really know how an airplane will age during its lifetime, you have to build it and use it for 50+ years. That would mean all ‘in service’ aircraft would be based on 50+ year old designs. That clearly is undesirable, if only because it would be too expensive. So, instead, one uses models for the effects of aging, for example by repeatedly freezing and heating parts while they also are being stressed physically and are under intense UV lighting. Those models can be wrong, but if you’re using new tech, you can’t afford to wait with selling airplanes until you know they are right. (Of course, this isn’t limited to the aircraft industry; it happens in all engineering. For example, hard disk manufacturers don’t run new hard disk technology for decades to measure its MTBF, either) Sometimes, such models need major adjustment, as happened with https://en.wikipedia.org/wiki/De_Havilland_Comet https://en.wikipedia.org/wiki/De_Havilland_Comet.

ethbr0 5y ago
I guess GP's point could be summarized as "Does the amount of complexity required to improve planes now exceed our ability to manage?" Curious in responses from that perspective. Are airframes that much more complicated? Or just better due to advances in aerodynamic understanding, but not much more complicated? And same question for engines. Are modern high-bypass turbofans more complex?
- S_A_P 5y ago
  This. It seems that to some degree our knowledge rests on prior experience and passed down knowledge. As we build abstraction layers on top of technology the bare metal understanding is lost. I’m going to contradict my original question to some degree by stating that I know that air travel is safer than ever. My question is more just wondering out loud if there is a point where we rely on “solved” problems so much that we start to lose comprehension of the factors that are involved. It feels like the MCAS was an example. Agree that you can never “know” how an airframe will age until you build it and doesn’t seem like this is as critical as MCAS issue but it could be that we have to wait 15 years to find out.
  - ethbr0 5y ago
    I think complexity + necessarily large organizations are a huge contributing factor. MCAS would have not been an issue is any one of the layers above engineering had questioned the decision to avoid training requirements (e.g. Boeing management, FAA management). Unfortunately, systems are so complex now that it was possible to summarize the details in a way that hid the necessary information. And when it went up further, even more summarization. It feels like this wouldn't have been as easy with simpler, older systems. I mean, presumably a manager could look at a biplane and say "Hey, that doesn't have enough thrust to weight."