Stochastic Fatigue Models for Efficient Planning Inspections in Service of Aircraft Structures

Authors:
Nicholas Nechval;Konstantin Nechval;Gundars Berzinsh;Maris Purgailis;Uldis Rozevskis
Affiliations:
Mathematical Statistics Department, University of Latvia, Riga, Latvia LV-1050;Applied Mathematics Department, Transport and Telecommunication Institute, Riga, Latvia LV-1019;Mathematical Statistics Department, University of Latvia, Riga, Latvia LV-1050;Mathematical Statistics Department, University of Latvia, Riga, Latvia LV-1050;Mathematical Statistics Department, University of Latvia, Riga, Latvia LV-1050
Venue:
ASMTA '08 Proceedings of the 15th international conference on Analytical and Stochastic Modeling Techniques and Applications
Year:
2008

Citing 0
Cited 3

Optimal Adaptive Inspection Planning Process in Service of Fatigued Aircraft Structures

ASMTA '09 Proceedings of the 16th International Conference on Analytical and Stochastic Modeling Techniques and Applications
Invariant embedding technique and its applications for improvement or optimization of statistical decisions

ASMTA'10 Proceedings of the 17th international conference on Analytical and stochastic modeling techniques and applications
Technique of statistical validation of rival models for fatigue crack growth process and its identification

ASMTA'11 Proceedings of the 18th international conference on Analytical and stochastic modeling techniques and applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

For important fatigue-sensitive structures of aircraft whose breakdowns cause serious accidents, it is required to keep their reliability extremely high. In this paper, we discuss inspection strategies for such important structures against fatigue failure. The focus is on the case when there are fatiguecracks unexpectedly detected in a fleet of aircraft within a warranty period (prior to the first inspection). The paper examines this case and proposes stochastic models for prediction of fatigue-crack growth to determine appropriate inspections intervals. We also do not assume known parameters of the underlying distributions, and the estimation of that is incorporated into the analysis and decision-making. Numerical example is provided to illustrate the procedure.