IMO this isn't even an instruments problem. There have been many crashes stemming from pilots diagnosing the wrong engine as the problem and shutting down a working engine.
Where's the NTSB recommendation on creating an official policy on HOW to diagnose an engine? The abnormal checklists only tell you to "shut down the broken engine" but leave it up to you to ad-hoc a way to diagnose the problem in an emergency, which is the wrong time to be inventing something non-trivial. Boeing/Airbus/GE/Rolls royce, whoever should have an official document describing a procedure to test an engine for failure that pilots are expected to refer to unless it would be unsafe
> As I mentioned in my article on the British Midland crash, the most effective way to prevent this kind of accident is by fitting airplanes with an Engine Indicating and Crew Alerting System, known as EICAS, or similar equipment. These systems automatically monitor engine performance and, should an engine fail, will produce a message informing the pilots which engine is the cause of the problem, dramatically reducing the probability of an incorrect identification. However, retrofitting the systems onto older 737s, which are most at risk, is not required and may be impractical or impossible.
Thanks for copying that snippet! It's probably the most interesting paragraph in the whole article. I'm baffled why it was left to very end. This should be the lead!
The author typically ends their articles with "what changes came about because of this type of crash, or exist today that would have prevented it", after describing all the factors that played a role.
> Following the crash and the publication of the NTSB’s damning final report, substantive changes were finally made to prevent similar accidents from happening again in the future. The FAA launched an industry-wide campaign to root out the myth of “ice bridging,” and it became standard practice to activate the deicing boots immediately upon entering icing conditions. The FAA mandated that all EMB-120s and several other aircraft types be fitted with automatic ice detection systems; Comair increased its minimum speed in icing conditions to 170 knots; and the FAA required all manufacturers to provide clear minimum maneuvering speed information for flight in icing conditions. The FAA also established standardized communication channels between FAA experts and POIs stationed at airlines; created a new database to more carefully track airworthiness matters involving foreign manufacturers; and most importantly, launched major research initiatives designed to ensure that the requirements of FAR Part 25 Appendix C were realistic and covered all ice shapes that were likely to form in flight, including thin layers of sandpaper-type ice. These resulted in a series of revisions to FAR part 25 to reflect state-of-the-art knowledge of aircraft icing, which continued all the way through 2016, including via the addition of further testing criteria not originally identified by the NTSB. As a result, it is now known exactly how every aircraft currently in service will react to all the types of in-flight icing which are likely to occur, and limitations exist to ensure that controllability is not compromised.
Detecting engine failure isn't trivial - a failed engine can still spin pretty fast with airflow going through it.
Modern predictive techniques could easily detect engine failure (ie. "Here is the history of the last 10 seconds of the throttle position, and here is the current RPM, is the engine failed?"). But aircraft designs don't like to use such techniques - they prefer simple thresholds. And with simple thresholds, a bunch of engine failures can get missed, especially partial failures.
I think the is a big issue across aviation. The system just gives up and dumps the problem on the pilot who then has to parse conflicting data quickly. I wonder if it would be better to treat this more like a navigation problem. Fuse together different data with known error levels and show the pilot the result. And include more solid state sensors like GPS, gravity sensors, and cameras in the mix.
It worked in Podracer!