Probabilistic Methods in Structural Design and Analysis

Abstract

This chapter discusses the probability-based load and resistance factor design (LRFD) codes and presents a probabilistic analysis of structural loads. The current generation of probability-based code procedures was spawned by the activities of the ASCE Safety Committee activities in the late 1960s. It was developed to its present state by researchers worldwide. Current LRFD code developments are virtually based on the best estimates of probability distributions and their parameters. With the exception of the EPRI Design Guide Effort, they generally neglect the second-order uncertainty; this is the often-major uncertainty associated with both the form of probability distributions and the values of their parameters at a specific site. Probabilistic analysis of structural load or structural loads engineering is the discipline of preparing descriptions of loads for structural analysis and design purposes. The fundamental characteristic of most interesting loads is that they are time varying. Therefore, they are properly represented by the random process theory rather than the random variable analyses that characterize LRFD. The simple random variable representations fail fundamentally to capture a time-varying process; at a minimum, they must be derived from process theory. The chapter discusses the attempts to find simple, tractable, practical schemes for dealing with these processes with acceptable accuracy.