ABSTRACT
Computers have enjoyed the reputation of increasing aerospace design productivity at reduced cost for some time, and analytical techniques have proliferated in parallel with this increased computational capability. However, there is still little reliance on these analytical techniques when it comes to certifying aircraft structures. This is due primarily to lack of confidence in either the analytical method or the fidelity of the model. Hence, the aerospace industry accomplishes structural certification through a design process controlled by safety factors and a resource-intensive regimen of physical tests [1]. The current cost of performing these “building block” structural tests, from small specimens to full flight-testing, for certifying a military airframe is estimated to be around $250 million [2]. Additional costs are, of course, incurred to certify structural repairs and operational modifications during the life of every airframe design. These postproduction costs are particularly difficult to predict during design because most aircraft structures experience undesired structural dynamics behaviors — in operational test and evaluation or in the field — that were either underestimated or unanticipated during design.
