Bridge maintenance management based on routine inspection data: a quantitative approach

Gaowei Xu,Fae Azhari
DOI: https://doi.org/10.31224/osf.io/m3fz6
2021-12-31
Abstract:The United States National Bridge Inventory (NBI) records element-level condition ratings on a scale of 0 to 9, representing failed to excellent conditions. Current bridge management systems apply Markov decision processes to find optimal repair schemes given the condition ratings. The deterioration models used in these approaches fail to consider the effect of structural age. In this study, a condition-based bridge maintenance framework is proposed where the state of a bridge component is defined using a three-dimensional random variable that depicts the working age, condition rating, and initial age. The proportional hazard model with a Weibull baseline hazard function translates the three-dimensional random variable into a single hazard indicator for decision-making. To demonstrate the proposed method, concrete bridge decks were taken as the element of interest. Two optimal hazard criteria help select the repair scheme (essential repair, general repair, or no action) that leads to minimum annual expected life-cycle costs.
What problem does this paper attempt to address?