A stochastic model for estimating the network‐scale deterioration and effect of interventions on bridges

Abstract

Visual inspections provide direct information about the deterioration states of each structural element within a bridge. However, when managing a network of bridges, it becomes necessary to aggregate the deterioration information from the element-level to the bridge-level or even the network-level. Deterioration models that are based on state-space models (SSM) have the capacity to model the deterioration based on visual inspections and structural attributes. Nonetheless, the application of such models has been demonstrated at the element-level only. In this study, the SSM deterioration framework is extended to provide overall estimates for the deterioration states of bridges and of an entire network. The proposed approach also includes a framework for handling missing data and interventions. The framework's capacity is demonstrated using data from individual cases of bridges, with and without interventions, in addition to a network of B ≈ 7000 bridges in the Canadian province of Quebec. The deterioration estimates for bridges can aid in examining the effectiveness of previous interventions, the long-term trend of the network's condition, as well as laying the groundwork for planning future maintenance actions.