Evaluation of probabilistic and deterministic life-cycle cost analyses for concrete bridges exposed to chlorides

Abstract

Owners and designers are increasingly dealing with infrastructure maintenance to ensure adequate level of performance and functioning required by current standards. Corrosion of reinforcement is one of the major factors that reduces the durability of concrete structures. Fiber reinforced polymers (FRP) have emerged as an effective non-corrosive alternative to traditional steel reinforcement (CS), especially for bridges located in chloride-laden environments. This paper aims to undertake probabilistic and deterministic Life-Cycle Costs (LCC) analyses for: a) an entirely FRP-reinforced concrete bridge; and, b) a conventional reinforced concrete (RC)/prestressed concrete (PC) bridge; assumed to be located in different corrosion intensity locations of Florida. The deterministic LCC analysis is investigated as a function of varying chloride concentrations for different locations in Florida. A probabilistic analysis is then performed to account for uncertainties in the deterministic model based on a selected coefficient of variation value. For each maintenance intervention, the probabilistic model yields a range of potential effects produced on the LCC by the same cause, ultimately producing more realistic results. The probabilistic model is validated through the input of maintenance and repair cost data from actual in-service bridges. In the probabilistic model, the repair cost is evaluated as a smooth curve, and the results are sensitive to the selection of the coefficient of variation adopted. The probabilistic model allows for better estimation and prediction of the range of long-term LCC behavior of both bridge alternatives. The CS-RC/PC alternative is found to be a high-risk design alternative with higher range (cost spread) and increased LCC. Data available to validate the probabilistic model show a promising match and revealed that the CS-RC/PC alternatives may become costlier earlier in time with a higher degree of probability.