The application of Markov models as deterioration-forecasting tools has been widely documented in the practice of infrastructure management. The Markov chain models employ monitoring data from visual inspection activities over a period of time in order to predict the deterioration progress of infrastructure systems. Monitoring data play a vital part in the managerial framework of infrastructure management. As a matter of course, the accuracy of deterioration prediction and life cycle cost analysis largely depends on the soundness of monitoring data. However, in reality, monitoring data often contain measurement errors and selection biases, which tend to weaken the correctness of estimation results. In this paper, the authors present a hidden Markov model to tackle selection biases in monitoring data. Selection biases are assumed as random variables. Bayesian estimation and Markov Chain Monte Carlo simulation are employed as techniques in tackling the posterior probability distribution, the random generation of condition states, and the model's parameters. An empirical application to the Japanese national road system is presented to demonstrate the applicability of the model. Estimation results highlight the fact that the properties of the Markov transition matrix have greatly improved in comparison with the properties obtained from applying the conventional multi-stage exponential Markov model.

Disruptive events, such as earthquakes, floods, and landslides, may disrupt the service provided by transportation networks on a vast scale, as their occurrence is likely to cause multiple objects to fail simultaneously. The restoration program following a disruptive event should restore service as much, and as fast, as possible. The estimation of risk due to natural hazards must take into consideration the resilience of the network, which requires estimating the restoration program as accurately as possible. In this article, a restoration model using simulated annealing is formulated to determine near‐optimal restoration programs following the occurrence of hazard events. The objective function of the model is to minimize the costs, taking into consideration the direct costs of executing the physical interventions, and the indirect costs that are being incurred due to the inadequate service being provided by the network. The constraints of the model are annual and total budget constraints, annual and total resource constraints, and the specification of the number and type of interventions to be executed within a given time period. The restoration model is demonstrated by using it to determine the near‐optimal restoration program for an example road network in Switzerland following the occurrence of an extreme flood event. The strengths and weaknesses of the restoration model are discussed, and an outlook for future work is given.

No abstract

In many practices of bridge asset management, life cycle costs are estimated by statistical deterioration prediction models based upon monitoring data collected through inspection activities. In many applications, it is, however, often the case that the validity of statistical deterioration prediction models is flawed by an inadequate stock of inspection dates. In this paper, a systematic methodology is presented to provide estimates of the deterioration process for bridge managers based upon empirical judgments at early stages by experts, and whereby revisions may be made as new data are obtained through later inspections. More concretely, Bayesian estimation methodology is developed to improve the estimation of Markov transition probability of the multi-stage exponential Markov model by Markov chain Monte Carlo method using Gibbs sampling. The paper concludes with an empirical example, using the real world monitoring data, to demonstrate the applicability of the model and its Bayesian estimation method in the case of incomplete monitoring data.