Dongkyu Lee

As lifeline networks such as transportation or electricity networks in modern societies are intricately interlocked, a small number of components damaged by natural or man-made disasters can have a great impact on network performance. For this reason, it is essential to assure the capability of rapid assessment of network reliability to make prompt follow-up measures. Despite the rapid development of various algorithms and computing power, the capability is still limited due to computational cost for analyzing the connectivity of a single origin and destination (O/D) node pair in large-scale networks. Therefore, this paper introduces a new algorithm utilizing network centrality, termed “centrality-based selective recursive decomposition algorithm” (CS-RDA). By preferentially decomposing the node which is most likely to belong to the min-cut identified based on the betweenness centrality, the convergence of the bounds on the O/D connectivity can be expedited significantly. This paper also introduces a new multi-scale analysis approach termed “edge-betweenness algorithm.” The algorithm groups components such that its modularity is maximized, by sequentially removing edges that have the highest level of betweenness centrality. As a result, the reliability of large-scale networks can be accurately evaluated in a short time owing to the reduced complexity of the simplified network. The proposed methods are successfully demonstrated by a hypothetical network example, the highway bridge networks in San Jose and San Diego in California, USA.

Lifeline systems such as transportation and water distribution networks may deteriorate with age, raising the risk of system failure or degradation. Thus, system-level sequential decision-making is essential to address the problem cost-effectively while minimizing the potential loss. Researchers have proposed to assess the risk of lifeline systems using Markov decision processes (MDPs) to identify a risk-informed operation and maintenance (O&M) policy. In complex systems with many components, however, it is potentially intractable to find MDP solutions because the numbers of states and action spaces increase exponentially. This paper proposes a multi-agent deep reinforcement learning framework, termed parallelized multi-agent deep Q-network (PM-DQN), to overcome the curse of dimensionality. The proposed method takes a divide-and-conquer strategy, in which multiple subsystems are identified by community detection, and each agent learns to achieve the O&M policy of the corresponding subsystem. The agents establish policies to minimize the decentralized cost of the cluster unit, including the factorized cost. Such learning processes occur simultaneously in several parallel units, and the trained policies are periodically synchronized with the best ones, thereby improving the master policy. Numerical examples demonstrate that the proposed method outperforms baseline policies, including conventional maintenance schemes and the subsystem-level optimal policy.

Various simulation-based and analytical methods have been developed to evaluate the seismic fragilities of individual structures. However, the seismic safety and resilience of a community are substantially affected by network reliability, determined not only by component fragilities but also by network topology and commodity/information flows. However, seismic reliability analyses of networks often encounter significant challenges due to complex network topologies, interdependencies among ground motions, and low failure probabilities. This paper proposes to overcome these challenges by a variance-reduction method for network fragility analysis using subset simulation. The binary network limit-state function in the subset simulation is reformulated into more informative piecewise continuous functions. The proposed limit-state functions quantify the proximity of each sample to a potential network failure domain, thereby enabling the construction of specialized intermediate failure events, which can be utilized in subset simulation and other sequential Monte Carlo approaches. Moreover, by identifying an implicit relationship between intermediate failure events and seismic intensity, we propose a technique to obtain the entire network fragility curve with a single execution of specialized subset simulation. Numerical examples demonstrate that the proposed method can effectively evaluate system-level fragility for large-scale networks.

Modern societies depend on various lifeline networks such as transportation, electricity, and gas distribution systems, which are vulnerable to seismic events. Although numerous analytical and simulation-based methods have been developed for efficient seismic system reliability analysis (SRA), dealing with high-dimensional events arising from large-scale infrastructure networks remains challenging. To address this challenge, this paper proposes a system reliability method that efficiently computes the connectivity of directed graphs. The method employs the dual graph representation of a target system to automate the construction of a Bayesian network (BN). This enables the application of the junction tree algorithm, a well-established BN inference method, to perform reliability analysis and quantify complexity based on a network topology. The paper further tackles SRA challenges associated with fully correlated seismic uncertainties, which typically lead to a significant increase in computational complexity. To this end, we propose to combine a cross entropy-based adaptive importance sampling technique with Rao-Blackwellization. Thereby, sampling methods and exact analytical inference can be effectively combined to improve computational efficiency for seismic SRA of lifeline networks. The proposed methods are demonstrated through three numerical examples.