Descriptive Analysis of National Water Pipeline Infrastructure Systems

摘要

Numerous water sector management practitioners have stated an urgent need to have a unified platform for the nation's water pipeline infrastructure data, information, and knowledge that is universally accessible and useful. Pipeline infrastructure database (PIPEiD) a web-based secured platform is envisioned to provide access to the data sources, tools, and models that enable the analysis and estimation of pipeline useful life based on materials and environmental factors and ability to model risk and life-cycle economic analysis for renewal decisions. PIPEiD a web-based secured platform is envisioned to provide access to the data sources, tools, and models that enable the analysis and estimation of pipeline useful life based on materials and environmental factors and ability to model performance, risk, and life-cycle economic analysis for renewal decisions. The research presents the descriptive data analyses methodology for the water pipe field performance data obtained from around 500 water utilities in the United States. The analyses are categorized into material distribution, failure distribution and trends, and performance analysis. Various exploratory and statistical techniques have been employed to represent the data and analyze deeper relationships within the data and pipeline infrastructure systems. The research team has also conducted extensive literature and practice reviews along with interviews with asset managers in water utilities to formulate a list of hypotheses which are essential to verify and validate the mathematical models. The paper will provide robust data structure, centralized database, and model-driven methodologies to benefit water utility, researchers, and water pipeline infrastructure industry. Also, it will provide extensive capabilities in multi-system and multi-scale data analytics for advanced water pipeline data management techniques, statistical analysis, advanced mathematical methods, and machine learning algorithms for water pipeline performance prediction and estimating useful life.