Battle of centralized and decentralized urban stormwater networks: From redundancy perspective

Recent research underpinned the effectiveness of topological decentralization for urban stormwater networks (USNs) during the planning stage in terms of both capital savings and resilience enhancement. However, how centralized and decentralized USNs' structures with various degrees of redundancy (i.e., redundant water flow pathways) project resilience under functional and structural failure remains an unresolved issue. In this work, we present a systemic and generic framework to investigate the impact of adding redundant flow paths on resilience based on three strategies for optimal centralized versus decentralized USNs. Furthermore, a tailored graph-theory based measure (i.e., eigenvector centrality) is proposed to introduce redundant paths to the critical locations of USNs. The proposed framework is then applied to a real large-scale case study. The results confirm the critical role of layout decentralization under both functional (e.g., extreme precipitation events), and structural failure (e.g., pipe collapse). Moreover, the findings indicate that the implementation of redundant paths could increase resilience performance by up to 8% under functional failure without changing the network's major structural characteristics (i.e., sewer diameters, lengths, and storage capacity), only by leveraging the effective flow redistribution. The scheme proposed in this study can be a fruitful initiative for further improving the USNs' resilience during both planning and rehabilitation stages.

Automated summary

This study investigates the effectiveness of topological decentralization for urban stormwater networks (USNs) during the planning stage. It proposes a framework to understand the impact of adding redundant flow paths on resilience and introduces a tailored graph-theory based measure. The results show that layout decentralization and the implementation of redundant paths can enhance the resilience of USNs without changing the network's major structural characteristics.