uses fault tree analysis to derive likely sources of failure based on subsystems and their relationships. think these methods are empirically useful when straight out of the factory line, but i think tacking on a sort of joint discovery process (grounded q&a session?) along with classification could probs be a good tool to help folks land in a ballpark
was working on something similar with audio classification + ecu logs and research agents checking against manuals for my project truck
had a little go around with diy microfluidics with ecoflex (though i still have some pdms on hand) and water soluble printed pva channels for bilayer cast+molding but the process to dissolve them was a bit less than joyful ¯\_(ツ)_/¯ also depending on how perf conscious you were as you size down, layer line artifacts and flow tuning esp as you go around curves could be make or break for some applications at decreasing scale. already finicky to print smaller than 0.25mm nozzle, pva's particularly a diva too and on top of that you're at the whim of batch variations
https://www.researchgate.net/publication/376877267_Applicati...
uses fault tree analysis to derive likely sources of failure based on subsystems and their relationships. think these methods are empirically useful when straight out of the factory line, but i think tacking on a sort of joint discovery process (grounded q&a session?) along with classification could probs be a good tool to help folks land in a ballpark
was working on something similar with audio classification + ecu logs and research agents checking against manuals for my project truck