Stochastic Analysis and Time-Based Modeling of Concrete Bridge Deck Deterioration

Abstract

Adequate prediction of the concrete bridge deck deterioration rate is necessary for maintenance and rehabilitation decisions. The stochastic deterioration of bridge decks can be most accurately modeled with a time-based probabilistic approach. In this work, a semi-Markov time-based model, based on accelerated failure time (AFT) Weibull fitted-parameters, was used to estimate the transition probabilities and sojourn times for the deterioration of concrete bridge decks. Approximately 3 years of in-service performance data for over 22, bridges in Pennsylvania were used in the model development. The proposed approach attempts to relate deck deterioration rates to various explanatory factors, such as structural system attributes, average daily traffic (ADT), route type, and environmental conditions. The following factors were found to be statistically significant with respect to the rate of bridge deck deterioration: type of rebar protection, continuous versus simply supported spans, overall bridge deck length, number of spans, bridge location, type of overlay, and whether or not the deck was located on interstate routes. Furthermore, the effects of remediation on bridge deck deterioration and service life were also evaluated and quantified, based on in-service performance data.