I was just flying for the first time in a decade last week and I saw an oversize tail on another plane while we were taxi-ing around the runway. The proportions struck me as non-obvious. That got me wondering if/how much we've begun to see data-mining/machine-learning algos and computational aero/fluid dynamics influence aircraft design.
E.g., genetic algorithms that vary the tail/wing/nose/etc. size/position and runs virtual designs through a simulator to see if various combinations perform better than others.
My next thought was, "how would you design the fitness function?". I.e., what are the right objective measures you can use to determine that one design is 'better' than another? I.e., How do you determine the optimal balance of speed vs. fuel efficiency vs. comfort/stability/safety vs. production/materials cost etc.?
E.g., genetic algorithms that vary the tail/wing/nose/etc. size/position and runs virtual designs through a simulator to see if various combinations perform better than others.
My next thought was, "how would you design the fitness function?". I.e., what are the right objective measures you can use to determine that one design is 'better' than another? I.e., How do you determine the optimal balance of speed vs. fuel efficiency vs. comfort/stability/safety vs. production/materials cost etc.?