Keywords: reliability-based optimization, reliability assessment, approximation concepts, meta-model, structural performance, approximate reliability methods, advanced simulation methods.

The basic goal of structural engineering is designing and constructing facilities which fulfill most economically certain predefined performance objectives. In practical situations, several different design solutions can satisfy prescribed performance objectives. Hence, the selection of a particular structural configuration is performed using an appropriate criterion, for example minimization of construction costs, maximization of benefits during the life time of the facility, etc. Thus, the task of designing a structure can be interpreted as an optimization problem, because the objective is determining the solution that best fulfills the aforementioned criterion while enforcing a series of performance objectives. However, it should be noted that in realistic situations, it is impossible to characterize structural performance deterministically, because many parameters (such as loadings, geometrical properties, material properties, etc.) are of an uncertain nature. Therefore, the design task can be treated within the framework of reliability-based optimization (RBO), i.e. an optimization problem where the effects of uncertainty are considered explicitly. The objective of this contribution is to present an overview of different techniques developed in the literature for solving RBO problems.

The review presented in this contribution shows that there has been considerable progress in the field of RBO. While early efforts focused on simplified analysis and explicit formulae, modern approaches are capable of addressing complex problems involving realistic models and several failure criteria most efficiently. The spectrum of approaches for RBO is quite broad. Some approaches are highly specialized and can treat certain classes of problems most efficiently. Other approaches are capable of treating very general problems, involving a large number of random variables and failure criteria considering non linear performance functions. Nonetheless, the price of generality is higher numerical costs. In addition, the type of algorithm used for assessing structural reliability influences, to a large extent, the type of RBO problems that can be solved and is discussed.

The conclusions drawn from this contribution indicate that optimization under uncertainty is not any longer the subject of academic examples, but a well developed methodology that can be applied to realistic engineering problems. It is expected that this tendency will accelerate even more in the near future due to progress in the application of high performance computing.

The full-text of this paper is not yet available.

go to the previous paper
go to the next paper
return to the table of contents
return to Computational Technology Reviews
purchase this volume (price £80 +P&P)