Generalized bridge network performance analysis with correlation and time-variant reliability

A novel approach and framework for the analysis of bridge networks is presented. The goal of the analysis is to assess the life-cycle performance of the network and its time-variant reliability. The proposed approach combines three important features that determine its ability to estimate with accuracy and robustness the reliability of a network along its life-cycle.The first one is that the reliability of the individual bridges is modeled as time-variant, due to the deterioration of their structural components. The network performance analysis is repeated for several time instants, with the relative reliability of the bridges. Therefore, the network performance indicators are also time-variant. In this way, the proposed procedure can be used as a basic tool for maintenance planning at the network level.The second asset is that the proposed framework takes into account complex (i.e. “generalized”) network layouts, not necessarily describable using series, parallel or series-parallel models. In fact, when all the possible traffic flows in a network and all the possible trip origins and destinations are considered, it is not feasible, in general, to model the network with a simple scheme. In the present paper, techniques derived from transportation engineering for the traffic flow distribution and assignment are used.The third feature is that the proposed approach considers a correlation structure among the states (in/out of service) of the various bridges of the network. In fact, bridges associated with the same network are likely to share similar characteristics and external loads. Therefore, a correlation structure for the service state of individual bridges is estimated and implemented in the analysis.A case study involving a transportation network with fourteen bridges is presented as a numerical application.

► Performance and reliability assessment of a bridge network along its life cycle.
► Correlation among the service states of the individual bridges of the network.
► Comprehensive traffic flow assignment and distribution model.
► Detrimental factors that reduce the bridge reliability over time.