WIT Press

Management And Planning Of Water Pipeline Networks


Free (open access)





Page Range

35 - 47




493 kb

Paper DOI



WIT Press


S. J. Habib, P. N. Marimuthu, S. Kim, Y. Pan & S. X. Yang


Water is a scare resource in the region of Middle East especially in Kuwait, where it has limited rainfall and natural ground water. Thus, Kuwait depends more than 90% on non-conventional water sources such as the seawater desalination, whereby the seawater is converted into fresh water through the desalination process. The desalination plants have been proven to supply sufficient amounts of fresh water throughout the years; however, some recent studies in Kuwait show a large amount of water leakages in the water pipeline network (WPN) and intermediate storage units. The wasted amount of water is very high if we consider the cost of producing water through the desalination process that it is roughly around 0.75 to 1.00 dollar per a cubic meter. Also, the up-to-date layout documentation of WPN is unavailable, where this situation poses a challenging problem in monitoring and inspecting WPN. This paper offers three contributions. First, we present a statistical analysis of the water production and consumption and the current status of WPN in Kuwait. Second, we describe an analytical approach to explore WPN based on the analogy between the computer network discovery and WPN discovery, wherein both the networks are viewed as a graph comprising of links and connectors. Third, we propose an integrated WPN discovery system that it deploys smart devices such as mobile robots to discover and validate the layouts of WPN. The manual detection of leakage is limited to surface pipelines, whereas the traditional techniques for passive detection of underground leakage are labor-intensive and very expensive. Thus, the proposed integrated WPN discovery system offers economical, long-term management and planning benefits over the manual inspection. Keywords: capacity planning, network discovery, robot.


Keywords: capacity planning, network discovery, robot.