ASD and LRFD: Reliability comparison for designs subjected to wind loads

Abstract

Current design practice in the United States allows engineers to use either Allowable Strength Design (ASD) or Load and Resistance Factor Design (LRFD) method, under the assumption that both methods provide similar levels of economy (available design strengths) and safety (probabilities of failure and reliability indices) regardless of the design conditions. Recent studies have demonstrated that significant differences in economy can arise between ASD- and LRFD-based designs subjected to combinations of gravity and wind loads, suggesting inconsistencies between the safety levels. This study investigates the safety levels provided by the ASD and LRFD methods for such design cases. The safety levels were estimated using a formal reliability analysis with consideration of geometrically non-linear second-order effects. The reliability analysis was performed using Monte Carlo simulations with appropriate probability distributions for the random variables associated with both resistance and loads. The findings of this study indicate that generally accepted uniformity of reliabilities achieved by the ASD and LRFD methods does not hold for design cases governed by wind loads. The safety levels achieved by using the ASD method are less consistent than those achieved by using the LRFD method. The use of the ASD method generally entails the acceptance of failure risks many times greater than for those of the LRFD method, and target safety levels lower than those stipulated by the ASCE 7 Standard. These results highlight the need for a reevaluation of the current design practices to ensure that the required target levels of structural safety are consistently met.