Search

An accurate reliability model for bridge decks is important for effective long-term bridge deck management. The main objectives of this paper are to identify the most suitable reliability model for bridge decks in Wisconsin, and to utilize that model for detailed analyzes of bridge deck reliability and failure rates. The 2005 National Bridge Inventory (NBI) data for the State of Wisconsin were used. In this paper, the hypertabastic, Weibull, log logistic, and lognormal distributions are investigated. The end of service life is defined as the age of deck when rehabilitation or replacement is required (herein defined as a deck rating of between 4 and 5). The effects on NBI deck rating of average daily traffic (ADT), type of bridge superstructure (steel or concrete), and the deck surface area were considered. Based on the Akaike information criteria, the hypertabastic accelerated failure time model was selected as the most appropriate model for this study. Results show that deck area, type of superstructure and ADT are all important factors in reliability of bridge decks in Wisconsin. [ABSTRACT FROM AUTHOR]