Muddassir Muhammad, Zayed Tarek, Taiwo Ridwan, Ben Seghier Mohamed El Amine
Department of Building and Real Estate, The Hong Kong Polytechnic University, 11 Yuk Choi Road, Hung Hom, Kowloon, Hong Kong, China.
Department of Built Environment, Oslo Metropolitan University, Oslo, Norway.
Sci Rep. 2024 Aug 19;14(1):19218. doi: 10.1038/s41598-024-69855-w.
The failure of water pipes in Water Distribution Networks (WDNs) is associated with environmental, economic, and social consequences. It is essential to mitigate these failures by analyzing the historical data of WDNs. The extant literature regarding water pipe failure analysis is limited by the absence of a systematic selection of significant factors influencing water pipe failure and eliminating the bias associated with the frequency distribution of the historical data. Hence, this study presents a new framework to address the existing limitations. The framework consists of two algorithms for categorical and numerical factors influencing pipe failure. The algorithms are employed to check the relevance between the pipe's failure and frequency distributions in order to select the most significant factors. The framework is applied to Hong Kong WDN, selecting 10 out of 21 as significant factors influencing water pipe failure. The likelihood feature method and Bayes' theorem are applied to estimate failure probability due to the pipe materials and the factors. The results indicate that galvanized iron and polyethylene pipes are the most susceptible to failure in the WDN. The proposed framework enables decision-makers in the water infrastructure industry to effectively prioritize their networks' most significant failure factors and allocate resources accordingly.
给水管网（WDNs）中的水管故障会带来环境、经济和社会后果。通过分析给水管网的历史数据来减轻这些故障至关重要。现有关于水管故障分析的文献存在局限性，缺乏对影响水管故障的重要因素进行系统选择，且未消除与历史数据频率分布相关的偏差。因此，本研究提出了一个新框架来解决现有局限性。该框架由两种针对影响管道故障的分类和数值因素的算法组成。这些算法用于检查管道故障与频率分布之间的相关性，以便选择最重要的因素。该框架应用于香港给水管网，从21个因素中选出10个作为影响水管故障的重要因素。应用似然特征法和贝叶斯定理来估计由于管材和这些因素导致的故障概率。结果表明，镀锌铁管和聚乙烯管在给水管网中最易发生故障。所提出的框架使水基础设施行业的决策者能够有效地对其管网最重要的故障因素进行优先级排序，并据此分配资源。
