Resumen
The nation?s transportation systems are complex and are some of the highest valued and largest public assets in the United States. As a result of repeated natural hazards and their significant impact on transportation functionality and the socioeconomic health of communities, transportation resilience has gained increasing attention in recent years. Previous studies on transportation resilience have heavily emphasized network functionality during and/or following a scenario hazard event by implicitly assuming that sufficient knowledge of structural capacity and environmental/service conditions is available at the time of an extreme event. However, such assumptions often fail to consider uncertainties that arise when an extreme hazard event occurs in the future. Thus, it is essential to quantify and reduce uncertainties to better prepare for extreme events and accurately assess transportation resilience. To this end, this paper proposes a dynamic Bayesian network-based resilience assessment model for a large-scale roadway network that can explicitly quantify uncertainties in all phases of the assessment and investigate the role of inspection and monitoring programs in uncertainty reduction. Specifically, the significance of data reliability is investigated through a sensitivity analysis, where various sets of data having different reliabilities are used in updating system resilience. To evaluate the effectiveness of the model, a benchmark problem involving a highway network in South Carolina, USA is utilized, showcasing the systematic quantification and reduction of uncertainties in the proposed model. The benchmark problem result shows that incorporating monitoring and inspection data on important variables could improve the accuracy of predicting the seismic resilience of the network. It also suggests the need to consider equipment reliability when designing monitoring and inspection programs. With the recent development of a wide range of monitoring and inspection techniques, including nondestructive testing, health monitoring equipment, satellite imagery, LiDAR, etc., these findings can be useful in assisting transportation managers in identifying necessary equipment reliability levels and prioritizing inspection and monitoring efforts.