Fuzzy analysis and soft computing methods in reliability assessment and optimization

Abstract

The design of structures is influenced by subjectivities and experience especially in early design stages. This leads to epistemic uncertainty and imprecision in data, parameters and computational models. The epistemic uncertainty has to be considered within all required analyses including reliability assessment and optimization. A variety of mathematical formulations are available for the quantification and processing of rare information. Fuzzy methods and soft computing have attracted particular attention in this context. They provide suitable tools for a proper mapping of epistemic uncertainty and imprecision to structural responses and reliability measures. Several concepts for fuzzy analysis have been developed, which can be applied in conjunction with linear or nonlinear algorithms of structural analysis. Soft computing methods are utilized to improve numerical efficiency, for example, with the aid of artificial neural networks for response surface approximation. Further, concepts of machine learning, pattern recognition, evolutionary and genetic computation, reasoning and prediction under uncertainty, etc., have migrated into structural mechanics recently. They are applied for a wide range of tasks such as parameter identification, response surface approximation, prediction of structural responses, structural health monitoring, structural control, and so forth. This mini-symposium is focused on the most recent developments in the field of fuzzy analysis and soft computing methods with applications in structural analysis, reliability assessment and optimization. This includes new ideas, algorithms and models in form of theoretical developments but also innovative applications for the solution of problems in engineering practice.

Organizers

Jan-Uwe Sickert, Technische Universität Dresden, Dresden, Germany, (jan-uwe.sickert@tu-dresden.de)
Michael Beer, National University of Singapore, Singapore