Improving pipe failure predictions: Factors affecting pipe failure in drinking water networks

Area of Science:

• Environmental Science
• Data Science
• Civil Engineering

Background:

• Water companies use statistical models for pipe failure prediction.
• Models often lack in-field operational insights, leading to data interpretation errors.
• Inconsistent infrastructure data complicates accurate predictive modeling.

Purpose of the Study:

• To enhance the accuracy of pipe failure predictive models.
• To bridge the gap between data scientists and field operatives.
• To provide a comprehensive overview of factors influencing pipe failure.

Main Methods:

• Summarizing typical failure factors for common pipe materials.
• Analyzing infrastructure, weather, and environmental data influences.
• Reviewing material-specific failure mechanisms for cast iron, ductile iron, steel, asbestos cement, PVC, and PE pipes.

Main Results:

• Identified key factors influencing failure across six common pipe material groups.
• Highlighted potential misinterpretations of infrastructure and environmental data.
• Provided a foundation for improved model input selection.

Conclusions:

• An improved understanding of pipe failure mechanisms is crucial for data scientists.
• Accurate modeling requires addressing the disconnect between data analysis and field experience.
• This work enables more precise predictive models for water infrastructure management.