FACTORS INFLUENCING THE PERCEIVED RELIABILITY OF FATIGUE LIFE PREDICTION IN AEROSPACE STRUCTURES

Abstract

Purpose: This paper will research on the drivers of the perceived reliability of fatigue life prediction in aerospace structures. The study is particularly interested in the influence of the accuracy of the load spectrum estimation, the quality of material fatigue data, the effectiveness of the coupling between the computational fluid dynamics and the finite element analysis, taking into account of manufacturing defects, and the level of the experience of the analyst. It is also a study that determines the mediating effect of confidence in numerical modelling in explaining the effect that these factors have on the reliability of fatigue life prediction.

Design/Methodology/Approach: The proposed conceptual framework was tested with the help of the quantitative approach of research with Partial Least Squares Structural Equation Modeling. The information was gathered among professionals and engineers that work on aerospace structure analysis, fatigue and reliability tests. Measurement model was evaluated with the help of reliability, convergent validity, and discriminant validity tests and the structural relationships were evaluated with the help of bootstrapping methods to identify the significance of the postulated relationships between variables and mediation.

Findings: The results indicate that, accuracy of load spectrum estimation, quality of material fatigue data, effectiveness of coupling between computational fluid dynamics and finite element analysis, manufacturing defects, and the level of experience of analysts play an important part in perceived reliability of fatigue life prediction. Of these considerations, the impacts of manufacturing flaws and the level of experience of the analysts have the best impacts. The results also reveal that confidence in numerical modeling makes the relationships between the independent variables and the perceived reliability of fatigue life prediction partially. The structural model also illustrates acceptable predictive relevance and explanatory power.

Practical Implications: The implications of the findings include the need to enhance load spectrum estimation approaches, reinforce material fatigue database, combine computational fluid dynamics and finite element analysis computational models, and include manufacturing defect modeling into fatigue prediction models. Furthermore, by increasing the experience and competency of the analysts and improving the trustworthiness on the numbers modeling equipment, the quality of prediction of fatigue life can be greatly enhanced and benefits the safer aerospace structural design and maintenance planning.

Originality/Value: This paper offers a composite framework that integrates both technical modeling considerations and human due diligence in the explanation of predictability of fatigue life in the aerospace structures.