Maximum deterministic differential settlement is used in current American Association of State Highway and Transportation Officials Load and Resistance Factor Design specification. However, the deterministic differential foundation settlement is quite different from the actual settlement due to the soil’s large variability. This article investigates the effects of the random differential foundation settlement on the reliability of bridge superstructure. A lognormal distribution with a coefficient of variation of 0.25 of random settlement is used in reliability analysis. Dead and live loads are modeled as random variables with normal and Gumbel Type I distributions, respectively. Then, the reliability of bridge superstructure considering random differential settlement effect is calculated using the first-order reliability method. Considering the regional traffic condition on Missouri roadways, the effect of a live load reduction factor on bridge reliability is investigated. The reliability indices of 14 existed bridges and 31 new designed multi-span girder bridges are calculated. It is demonstrated that small change in differential settlement can significantly reduce the reliability of the superstructure, depending upon the span length and rigidity of the girder. For the 31 new designed bridges without the settlement consideration in design can tolerate an extreme settlement of L/500 without reduced live loads and L/2500 with reduced live loads in the case of a bridge reliability index of 3.5. The reliability analysis results also indicate that without reduced live loads, current American Association of State Highway and Transportation Officials specification with a load factor of 1.0 for the differential settlement has enough safe margins within an allowable settlement of L/250. However, when live loads are reduced, the load factor for settlement has to be amplified. The calibrated load factor curve of differential settlement within allowable settlement L/250 is provided to achieve the target reliability index of 3.5.
Read full abstract