APPLICATION OF LOAD-STRENGTH INTERFERENCE MODEL IN SYSTEM FAILURE PROBABILITY EXTIMATION

Aiming at modeling and assessing failure probability of general series, parallel, or k-out-of-n redundant system in which no premise of “component failure events being s-independent of each other is required, the present paper investigated the load-strength interference theory and its application in component/system failure probability estimation as well. Through a thoroughly analysis to the load-strength interference theory and the process of calculating component failure probability by this theory, the present paper introduced a “system-level load-strength interference method to system reliability modeling. Examples were presented to show the methodology and its agreement with experimental results. Different from the conventional system reliability estimation method by which system reliability was calculated directly from component reliability under the assumption of independent failure among components, the proposed method was applying load-strength interference relationship at system-level and calculated system failure probability according to load distribution and strength distribution. As a basis of developing the methodology, the phenomenon of information loss in the process of conventional component reliability analysis and modeling was discussed, and the reason that why it is impossible to construct dependent system reliability model simply by means of component reliability parameter was declared. Finally, a parametric model was developed to predict high multiple failure probability according to low multiple failure event data of the associated system.