A study on the reconstruction model for minimizing water supply risk in existing water distribution network

Taeho Choi, Dooyong Choi, Kibum Kim, Jayong Koo

Research output: Contribution to journalArticlepeer-review


This study details the development of a reconstruction model for reducing water supply risk (WSR) in an existing water distribution network (WDN). This development included additional gate valves for reducing the impact of a burst pipe (Imp PB ), additional emergency pipes for adjacency block, and pipe replacement to reduce the probability of a burst pipe (Prob PB ). The developed model was applied to the study area, giving a block WSR in the reconstructed network of 0.471 m 3 /year, with a possible input budget of about KRW 140 million (4.5% of the maximum cost for reconstruction); 1,000 KRW = 0.92 USD = 0.69 EUR, 31 Dec 2012. This constituted a decrease of block WSR by 68.7% compared with that of the existing pipe network. The benefit cost for WSR reduction was defined as shown in Scenarios 1–4 at respective ratios of KRW 40,000/m 3 , KRW 60,000/m 3 , KRW 80,000/m 3 , and KRW 100,000/m 3 . Results show that higher the benefits of reducing the WSR in existing WDNs, more the chance for WSR reduction. On the other hand, if the benefit cost for WSR reduction is low, this indicates that it is not possible to actively reduce WSR. In other words, it is currently difficult to accurately convert the benefits of reduced WSRs. However, if such problems are solved in the future, it should be possible to establish more accurate budgets and plans can for minimizing WSR in existing WDNs.

Original languageEnglish
Pages (from-to)126-136
Number of pages11
JournalDesalination and Water Treatment
StatePublished - Mar 2019


  • Additional gate valve
  • Pipe replacement
  • Reconstruction model
  • Water distribution network
  • Water supply risk


Dive into the research topics of 'A study on the reconstruction model for minimizing water supply risk in existing water distribution network'. Together they form a unique fingerprint.

Cite this