Reliability of Airframe Structures - General Aspects of the Problem and Outline of Methods for Attaining 640578
The terms “reliability” and “reliability design” have emerged as popular words to describe a discipline for minimizing failures or operating difficulties in complex mechanisms or systems. Little recognition has been given to past or traditional engineering practices which have been evolved to achieve this objective along with the primary one of fulfilling a specific operational need. The purpose of the following discussion is to describe in non-academic terms how a rather high degree of reliability has been achieved in developing military airframe structures, the difficulties and inaccuracies associated therewith, and some aspirations for the future. It is not intended for structural designers and fatigue specialists who should be better informed on the technical procedures involved than the writer. It is intended however to emphasize the accomplishments of these specialists who have bolstered public confidence in air travel and have provided military flight vehicles with reliable airborne platforms for some of the most demanding requirements devised for weapon systems.
This discussion focuses attention on the shortcomings of the factor of safety, such as used in civil engineering, as well as the probabilistic approach to airframe reliability. It attempts to draw a distinction between the design of small, mass products, for which reliability figures can be derived by statistical interpretation of actual measurements and the design of large structures or systems to be produced in small quantities. In short, a combined approach is needed to assure structural reliability, - for some structural components and operational needs, a safety- factor approach is adequate and indeed is the only one economically feasible; for others, a safe-life approach is essential. To these approaches is added a third, or fail-safe approach, which is used to the extent permitted by performance considerations and to the extent dictated by a critical appraisal of failure effects on the mission of the flight vehicle. To benefit by these approaches, the designer must not be content with the exercise of his mathematical skill alone but must employ all of the cumulative experience at his command to avoid the most frequent source of structural unreliability, viz, poor detail design.
While having sincere respect for the accomplishments of reliability engineers, the writer suggests that the pressing problem in achieving better structural reliability is to concentrate more on analyses of how and why materials fail, on better definition of the operating environments, on better methods for detecting imminent failures, on more realistic simulation of operating conditions in testing, and on better interpretation of test results, than on evolving impressive statistics on rates of failure.
The writer wishes to acknowledge assistance in formulating the ideas expressed herein, - borrowing freely from Professor A. M. Freudenthal of Columbia University and Mr. A. C. Payne of the Melbourne Aeronautical Research Laboratories (Australia). Constructive suggestions were furnished by several dedicated structural engineers within the Navy’s Bureau of Naval Weapons establishment.
The statements and views expressed in this paper are those of the author and are not to be construed as official or necessarily representing the views of the Bureau of Naval Weapons, Department of the Navy.