ABSTRACT
Akiyama et al. (2012) presented a computational procedure to establish the probabilistic hazard curve of airborne chlorides by applying the concept of seismic probabilistic hazard assessment, and proposed a method for calculating the probability of occurrence of both steel corrosion and corrosion cracking due to corrosion products in RC structures. In addition, a design criterion and durability design factors for a new RC structure were proposed (Akiyama et al. 2014). Based on the results presented in Akiyama et al. (2014), without performing a complex reliability analysis, the designer can determine the concrete cover and water to cement ratio in order that the reliability index associated with occurrence of steel corrosion initiation will be very close to its target value.
Meanwhile, an existing RC structure designed without adequate durability detailing deteriorates with time. To confirm whether the aging structure still satisfies the serviceability requirements, it is necessary to assess the probabilities of steel corrosion and corrosion cracking initiation within the remaining lifetime. Using inspection results, the epistemic uncertainties associated with the service life reliability prediction of existing RC structures can be reduced compared with new RC structures.
This paper presents a methodology for reliability-based durability assessment of RC bridges using observational data. The proposed formulations for durability assessment of existing RC bridge are follows: () ϕ C T , a ≥ γ C a https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315207681/cd556cd4-4dcf-4efe-8e29-56fc67b8bfbd/content/eq136.tif"/>
where () C a = C 0 , a ( 1 − e r f 0.1 ⋅ c a 2 D c , a ( T s + T r ) ) https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315207681/cd556cd4-4dcf-4efe-8e29-56fc67b8bfbd/content/eq137.tif"/> ϕ and γ are the durability partial factors for capacity and demand, respectively, C T,a is the critical threshold of chloride concentration (kg/m3), Ca is the chloride concentration at reinforcing bar (kg/m3), C 0,a is the chloride concentration of concrete surface, ca is the concrete cover (mm), Ts is the time from construction to inspection (year), Tr is the remaining lifetime (year), erf is the error function, and Dc,a is the coefficient of diffusion of chloride. C 0, a and Dc, a are determined by chloride concentration distribution provided by the coring test.
In the proposed durability assessment, Sequential Monte Carlo Simulation (SMCS) is used in conjunction with time-dependent reliability assessment for existing RC structures (Akiyama et al. 2010). To be consistent with the inspection results, the random variables used in the prediction of chloride concentration can be updated. Once the chloride concentration distribution is provided by the coring test, whether repair action will be needed within the remaining lifetime can be determined using Equation 1 and proposed durability partial factors without reliability computations and updating process. Because of the reduction in uncertainty, the partial factors for durability assessment of existing RC structures are different from those used for durability design.
