The complexities in the design of water distribution networks (WDNs) and a known fact of the curse of dimensionality using metaheuristic optimization techniques led to the advent of many nature-inspired algorithms. The study demonstrates the application and performance of a recent swarm intelligence innovation, the krill herd algorithm (KHA), in optimally designing the WDNs. Its operational framework emulates the natural phenomenon of herding of krill swarm. The KHA’s advanced operational tool, the fine-tuned model of KHA (FIT-KHA) is employed for designing WDN. Its application in designing a medium-sized WDN in Hanoi city, the Hanoi network (HN), comprising thirty-four pipes for water supply is demonstrated. Performing the sensitivity analysis, the computational results demonstrate that to design HN, a minimum krill population size (Nkrill) of 1000 is essential for an adequate and efficient search through its complex search space size of 634 (2.865 × 1026). Especially, at Nkrill = 6000 and a maximum iteration size (Is,max) of 500, the FIT-KHA model converged to the lowest possible pipe investment cost of 6,148,884 units. Though the algorithm did not converge to the best cost found to date (6,081,564 units), the results signify the excellent intensified searchability of the FIT-KHA model. Even after a long stretch trap at the same solution, the FIT-KHA model notably exploits the available information and locates the best possible solutions at the following iterations. Nonetheless, to enhance its computational efficiency for solution precision and computational ease, there is a need to strengthen its operational framework with diversified search features.