First Order Importance Sampling Method and its Variance Reduction

by Gongkang Fu, New York State Dept of, Transportation, Albany, United States,

Document Type: Proceeding Paper

Part of: Engineering Mechanics

Abstract: This paper proposes an Importance Sampling method for structural system reliability problems with linearized modal failure surfaces. It has the advantage of locating all samples in the failure region for parallel systems. Its resulting variance is estimated by an analytically derived upper bound, which depends on the ratio of the original distribution to the sampling distribution at design points. Application examples are included for illustration. They show that variance reduction by this method does not depend on the system failure probability to be estimated.

Subject Headings: Structural systems | Structural reliability | System reliability | Failure analysis | Structural analysis | Linear functions | Structural failures | Probability

Services: Buy this book/Buy this article


Return to search