Intelligent Water Distribution Systems: Emerging Tools for Performance Improvement

Abstract

Water utilities face challenges from aging infrastructure, operational complexity, and limited resources. Most of their investment is in distribution systems, and the scale of the challenge they pose is large in terms of water main breaks, costs of repairs, and damages to property with many mains past their useful lives. To address the problems with investment constraints, intelligent management tools are becoming available across two primary domains, system operations and asset management. Water utilities face important decisions when implementing these tools, but they cannot review or implement all simultaneously. To address the selection process, this paper applies a prioritization framework to identify the most promising tools by using bibliometric-based publication volume derived from the Web of Science Core Collection (January 2015 to May 2026), with emphasis on uses of machine learning as objective prioritization criteria. Using this approach, pipe break prediction models emerged as the primary tool for focus, supported by the highest publication volume among asset management tools and machine learning as the focus in 20.7% of the articles. The paper includes classification and bibliometric prioritization frameworks and a literature review of pipe break prediction methods, as well as an analysis of models and a discussion of implementation possibilities and development needs.