Urban Water distribution networks (WDNs) are huge and complex infrastructures comprising of numerous components such as pipes, junctions, reservoirs, pumps, valves, etc. The primary function of any WDN is to convey treated water to the consumers at adequate pressure, in desired quantities. A computer model of the WDN is usually prepared to study its performance under various operating conditions. Computer model of the WDN may also be required for determining the best locations for placement of valves (PRVs and Isolation Valves), Sensors (Pressure, Water Quality, Acoustic, Structural health monitoring), Bulk meters, Booster stations (pressure or chlorination) etc. Further, model is useful to solve the problems related to leakage, contamination, aging of the network infrastructure, pipe bursts, pressure imbalances, poor operation & maintenance. It is desirable in many cases to club the pipes or nodes having similar properties to reduce the size of the problem. Clustering is a technique of data analysis and machine learning that divides the network into sub-networks or clusters by virtue of which the complex network becomes simpler. The aim of this study is to comprehensively review various clustering methods, algorithms, and their applications in the WDNs. The chapter introduces clustering and its significance in WDNs. It also explores several clustering methods and their diverse applications in WDNs. Furthermore, it highlights the different objectives for which clustering has been applied in the design and optimization of facility locations in WDNs.