Machine learning based water pipe failure prediction: The effects of engineering, geology, climate and socio-economic factors

•Studied the safety and reliability analysis of water supply network (WSN) in terms of pipe's break probability prediction.•Developed data fusion framework to integrate multi-sourced datasets, leading to the largest real-field dataset related to WSN.•Applied machine learning algorithms to analyze the aggregated dataset and their performance compared.•Analyzed the effects of engineering, geology, climate and socioeconomic factors on WSN service conditions.•Advanced WSN safety and management for water management agencies and stakeholders.