New Research on Network Failure Prediction: Applying Statistics to Open Data and Continuous Data Acquisition

Baseform's latest paper on water network asset failure prediction is now available via our Publications page. The prediction of pipeline and service line failures, and the prioritization of rehabilitation efforts, play an important role in infrastructure asset management. 

 

Traditional condition assessment methods for water pipes are commonly associated with time- and resource-consuming tasks, that are either expensive, inaccurate, or both. Often, these physical methods cannot be applied to the entire system, so the results cannot inform the decision-making process in a structured manner. Several scientific approaches focused on lifetime assessment and failure forecasting have emerged in recent years aiming to identify the most failure-prone pipes and help prioritize system rehabilitation actions in a non-invasive but highly informed and utility-specific way. Failure Analysis, like other Baseform apps, is designed for real data with common, everyday problems, combining breakthrough statistical tools with continuous utility data acquisition. It is therefore prepared to cope with data gaps and inconsistent information, and it takes full advantage of more data-rich scenarios if and when they become available. 

 

This paper presents some key technological aspects and capabilities available in this application, illustrating them through examples from a case study at a large utility where the software is in daily use.