RELIABILITY OF COMPOSITE STEEL BRIDGE BEAMS DESIGNED FOLLOWING AASHTOS LFD AND LRFD SPECIFICATIONS

Composite steel beam bridges designed following AASHTO's Load Factor D esign (LFD) method and the newly developed Load and Resistance Factor Design method (LRFD) are proportioned for different loading conditions . They need to satisfy these conditions at maximum design load, an ove rloading condition, and fatigue load. The ultimate strength of compact composite steel beams is based on the fully plastic stress distributi on. The capacity of non-compact beams is based on the yield or partial ly-plastic moment, depending on whether the beam violates all or few o f the compactness or ductility requirements. In this paper, the reliab ility for the ultimate flexural capacity limit state of composite stee l beams is investigated. The structural reliability is measured in ter ms of the reliability index. The statistical data on strength are gene rated starting from statistics on material properties and using the Mo nte Carlo simulation method. Statistical data on load components are c ompiled from the available literature. The scope of the study covers a wide range of rolled beams, span lengths, beam spacings, and two yiel d strengths. The deterministic study showed that the design of composi te steel beams is usually not governed by the maximum design load comb ination, but rather by the overloading condition. The results of the r eliability study indicate non-uniformity in the safety of steel bridge s that are designed by the AASHTO LFD code. This is also true for AASH TO's LRFD code since it is not calibrated for the overloading conditio n. The value of the reliability index is a function of the compactness classification, method of design, beam spacing, span length, and sect ion size.