APPLICATION OF THE MONTE CARLO TECHNIQUE TO FATIGUE-FAILURE ANALYSIS UNDER RANDOM LOADING

A METHOD OF ASSESSING STRUCTURAL SAFETY IN FATIGUE IS PROPOSED IN WHICH A STATISTICAL MODEL FOR THE FATIGUE PROCESS IS USED TO CARRY OUT A RELIABILITY ANALYSIS, ENABLING THE PROBABILITY OF FAILURE TO BE ESTIMATED AT ANY STAGE OF THE LIFE. THE STATISTICAL VARIABILITY IN CRACK-PROPAGATION RATE AND RESIDUAL STRENGTH OF THE CRACKED STRUCTURE IS INCLUDED TOGETHER WITH THE EFFECT OF ANY PRESCRIBED INSPECTION PROCEDURE. THE METHOD IS APPLIED TO A STRUCTURE OF HIGH STRENGTH STEEL TYPIFYING A "SAFE-LIFE" STRUCTURE AND TO A REDUNDANT ALUMINUM ALLOY STRUCTURE REPRESENTATIVE OF THE "FAIL-SAFE" CONSTRUCTION. IT IS CONCLUDED THAT THE RELIABLITY ANALYSIS CAN BE APPLIED TO BOTH FAIL-SAFE AND SAFE-LIFE STRUCTURES AND PROVIDES A QUANTITATIVE BASIS FOR ENSURING THEIR SAFE OPERATION, INCLUDING THE PLANNING OF AN INSPECTION PROCEDURE IF FEASIBLE. IN THIS REGARD THE METHOD REPRESENTS AN ADVANCE ON THE EXISTING PROCEDURES BUT INHERENT IN THE QUANTITATIVE APPROACH IT EMPLOYS IS THE ADOPTION OF AN ACCENTABLE SAFETY LEVEL. THE MOST APPROPRIATE WAY OF DEFINING SAFETY LEVEL IS DISCUSSED AND A SUITABLE MEASURE IS PROPOSED. AN EXTENSIVE AMOUNT OF DATA IS REQUIRED IN APPLYING THE PROCEDURE BUT IT IS SUGGESTED THAT IN THE CASE OF AIRCRAFT STRUCTURES THIS DIFFICULTY CAN BE OVERCOME BY USING RESULTS FROM THE COMPREHENSIVE STRUCTURAL TESTING PROGRAM NORMALLY CARRIED OUT, TOGETHER WITH RELEVANT DATA FROM SIMILAR STRUCTURES. /ASTM/