Influence of probability distribution of loads on optimum reliability

Abstract

Abstract Since the Hyogoken-Nambu earthquake, people have begun to pay attention not only to a building's ultimate limit state, but also to its various limit states before collapse through its lifetime. One method to determine optimum reliability has been proposed on the basis of the minimum total cost concept. Load effects are often discussed regarding modeling for probabilistic analyses. Parameters for the probabilistic model, such as the coefficient of variation, the upper bound limit and the characteristics of tail probability, are thought to be important for probabilistic analyses. The purpose of this study is to clarify the effect of these factors on the optimum reliability index, the life-cycle cost and the design point. Numerical investigations have revealed that these factors sometimes significantly affect the optimum reliability, but not the life-cycle cost or the design point for both single and multiple limit states. The minimum total cost concept provides a stable design point with optimum reliability for different load probability models.