System Reliability of Structural Steel Frames Designed Using Component- and System-Based Design Methods

Abstract

The current design practice for structural steel buildings is largely governed by component-based design methods, which are based on variations of 1^st- or 2^nd-order elastic analyses and ensure strength and reliability on the level of individual components of a structural system. However, with the recent advances and increasing accessibility of structural modeling and analysis tools, combined with the development of affordable desktop computers with the processing and memory capacity required to complete advanced analyses at an acceptable runtime, there has been a growing interest in developing and adopting system-based design methods, which are based on variations of 2^nd-order inelastic analyses and ensure strength and reliability on the level of the entire structural system. The main benefit of the system-based design methods is that they explicitly account for most, if not all, component-level limit states directly in the analysis, simplifying or eliminating separate component-level limit state checks while ensuring the overall stability of a structure and accounting for beneficial inelastic load redistribution effects. As the structural engineering profession moves toward broader adoption of system-based design methods, validation, verification, and calibration studies are essential to quantify desired system-level reliabilities and inform the refinement of system-level provisions within future design codes for structural steel buildings. In this study, we investigate system-level reliabilities achieved by two component-based design methods, the Direct Analysis Method and the Advanced Elastic Analysis Method, and two system-based design methods, the Advanced Inelastic Analysis Method and the Direct Design Method. To investigate the system-level reliabilities achieved by these design methods, a series of benchmark structural steel frames were first designed using a structural design optimization framework. System reliability analyses that included uncertainties in geometric properties, material properties, and applied loads were then performed on the resulting optimal designs using the Importance Sampling technique. The findings of this study indicate that component-based design methods consistently produce system-level reliabilities that exceed target levels of reliability; however, these design methods result in designs that are significantly heavier than those produced by system-based design methods. In contrast, the system-based design methods result in significantly lighter designs, with more consistent levels of reliability that are closer to expected target levels. Based on these findings, recommendations are provided to enhance system-level reliability calibration procedures and to support the effective implementation of system-based design methods in future design codes.