Explainable ensemble models for predicting wall thickness loss of water pipes

被引:9
|
作者
Taiwo, Ridwan [1 ]
Yussif, Abdul-Mugis [1 ]
Ben Seghier, Mohamed El Amine [2 ]
Zayed, Tarek [1 ]
机构
[1] Hong Kong Polytech Univ, Dept Bldg & Real Estate, Hung Hum, Hong Kong, Peoples R China
[2] Oslo Metropolitan OsloMet Univ, Dept Built Environm, Oslo, Norway
关键词
Water pipelines; Wall thickness; Machine learning; Ensemble learning; SHAP; FAILURE;
D O I
10.1016/j.asej.2024.102630
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Water Distribution Networks (WDNs) are susceptible to pipe failures with significant consequences. Predicting wall-thickness loss in pipes is vital for proactive maintenance and asset management. This study develops optimized, explainable machine learning models for this purpose. Data from four WDNs located in Canada and the USA are collected and preprocessed. Decision Tree, Random Forest (RF), XGBoost, LightGBM, and CatBoost are employed, with optimized hyperparameters via Tree-Structured Parzen Estimator. The proposed framework performance is assessed using dissimilarity-based and similarity-based metrics. Hyperparameter optimization substantially enhances predictive performance such that the mean absolute error of RF improved by 20.51%. Based on the evaluation metrics, the Copeland algorithm was employed to rank the models, and CatBoost emerged as the best-performing model with a Copeland score of 4, followed by XGBoost and RF. The Taylor Diagram offers a visual representation of the linear proportionality between observed and predicted values across various models, with CatBoost and XGBoost showing strong alignment. SHAP analysis identifies age, diameter, and length as key contributors. The optimized models proactively identify potential pipe failures, enhancing maintenance and WDN management.
引用
收藏
页数:19
相关论文
共 50 条
  • [41] Power transmission loss in exponential horns and pipes with wall absorption
    Phelps, WD
    JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA, 1940, 12 (01): : 68 - 74
  • [42] A Weighted Likelihood Ensemble Approach for Failure Prediction of Water Pipes
    Beig Zali, Ramiz
    Latifi, Milad
    Javadi, Akbar A.
    Farmani, Raziyeh
    JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT, 2025, 151 (02)
  • [43] Optimize Water Distribution Pipes and Water Loss With Digital Solutions
    Karl, Michael
    Culbertson, Erin
    Abrera, John
    JOURNAL AWWA, 2021, 113 (07): : 69 - 73
  • [44] DIAGNOSIS OF THICKNESS IN PIPES AND HEAT EXCHANGER TUBES - A CASE OF STRAIGHT PIPES IN WATER LINE
    Uehara, Akira
    Inaba, Wataru
    Eino, Jyun-ichi
    Hashizume, Takumi
    PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON POWER ENGINEERING 2009 (ICOPE-09), VOL 2, 2009, : 191 - 196
  • [45] Explainable ensemble learning model for predicting steel section-concrete bond strength
    Wang, Xianlin
    Chen, Airong
    Liu, Yuqing
    CONSTRUCTION AND BUILDING MATERIALS, 2022, 356
  • [46] Explainable ensemble learning model for predicting steel section-concrete bond strength
    Wang, Xianlin
    Chen, Airong
    Liu, Yuqing
    Construction and Building Materials, 2022, 356
  • [47] Predicting pediatric esophageal wall thickness: An EUS study
    Rabinowitz, Simon S.
    Grossman, Evan
    Feng, Lisa
    Ebigbo, Nonyelum
    Lin, Bo
    Gupta, Raavi
    Sklar, Rachel
    Schwarz, Steven M.
    Weedon, Jeremy
    Gress, Frank
    ENDOSCOPIC ULTRASOUND, 2020, 9 (04) : 259 - +
  • [48] Prediction Interpretations of Ensemble Models in Chronic Kidney Disease Using Explainable AI
    Jawad, K. M. Tawsik
    Verma, Anusha
    Amsaad, Fathi
    IEEE NATIONAL AEROSPACE AND ELECTRONICS CONFERENCE, NAECON 2024, 2024, : 391 - 397
  • [49] Explainable Ensemble Learning Models for the Rheological Properties of Self-Compacting Concrete
    Cakiroglu, Celal
    Bekdas, Gebrail
    Kim, Sanghun
    Geem, Zong Woo
    SUSTAINABILITY, 2022, 14 (21)
  • [50] Development of explainable recognition models for ADMFB flow regimes based on ensemble learning
    Li, Zhiqiang
    Geng, Peixian
    Li, Shiyin
    Wang, Baoqiang
    Dai, Wei
    Duan, Chenlong
    Dong, Liang
    POWDER TECHNOLOGY, 2024, 445