No abstract

After the 2018 collapse of Morandi highway bridge in Genova, new Italian guidelines (GL) were issued for prioritization of safety evaluations and retrofit interventions on existing bridges. Among several novelties, GL introduced new traffic load models (TLMs) which can be adopted for existing bridges in case of noncompliant safety checks according to the Italian building code (NTC) provisions. In this study, a class of simply supported, beam-type, prestressed concrete bridge decks built between 1970 and 1980 in Italy is considered. Assuming geometric, material, and load random variables together with consideration of capacity model uncertainty, Monte Carlo sampling technique was implemented in MATLAB in order to randomly generate deck models and to evaluate their traffic-load fragility. Fragility analysis was carried out to compare vulnerability levels of existing Italian bridges under different load patterns provided by GL and NTC. In addition to a prescribed TLM, the sensitivity of fragility to bridge usage limitations, such as reduced distance of external load lane from kerb or reduced number of lanes, was also quantified to support decision-making by road management companies. In the final part of the study, the annual failure probability of selected bridges was estimated using a European weigh-in-motion database and convolution of fragility and hazard.Abbreviations: A sp , prestressing steel area; A s , reinforcing steel area; b w , girder web width; CoV, coefficient of variation; d, distance of reinforcing steel rebar from the top compression side; DCR, demand capacity ratio; f ct , concrete tensile strength; FORM, first order reliability method; GL, Italian technical Guidelines for safety assessment of existing bridges; GVW, gross vehicle weight; IM, intensity measure; LRFD, load-resistance factor design; M r , flexural capacity; M r,exp , experimental flexural strength; M r,th , theoretical flexural strength; n l , number of notional lanes; N sim , number of simulations; NTC, Italian building code; P[C], failure probability; P[CjH], conditional probability of exceeding ultimate limit state given intensity measure; P[H], traffic-related hazard; PC, prestressed concrete; PSD, permissible stress design; Q 1d , design first-lane tandem load; R 2 , coefficient of determination; RC, reinforced concrete; r l , remaining area of the carriageway width; RV, random variable; TLM, traffic load model; t ref , residual service life; ULS, ultimate limit state; V r , shear capacity; WIM, weigh-in-motion; x 0 , clear distance of the first notional lane from the kerb; β, logarithmic standard deviation; β 0 , target reliability; β al , logarithmic standard deviation associated with aleatory uncertainties; β ep , logarithmic standard deviation associated with epistemic uncertainties; μ, mean value; σ cp , average compressive concrete stress; σ sp , residual prestressing action; θ, model error.Discussion on this paper must be submitted within two months of the print publication. The discussion will then be published in print,...

No abstract

No abstract

Fragility curves are a well-established tool to define probabilistically the vulnerability of an asset to a prescribed hazard. The majority of previous studies have investigated the seismic fragility of structures and infrastructures. Nonetheless, recent disasters have highlighted the need to assess the vulnerability of aging bridges to gravity loads. Based on existing data from the literature and novel data from some real case studies, the present work presents analytical fragility curves of Italian bridges to traffic loads. The class of simply supported, beamtype, prestressed concrete bridges constructed between 1960 and 1980 was considered, given that those structures are the most recurrent type of existing Italian bridges. Then, a fragility analysis was performed through a fully automated procedure implemented in MATLAB. Geometric, material and load parameters were defined as random variables considering their probability distribution and correlation through either univariate or multivariate models. Structural fragility was evaluated with respect to the limit state of collapse. The output of this study can be of interest to engineers and decisions makers who must manage existing bridges, considering their probability of collapse under traffic loads and eventually limiting their use before structural retrofitting.