Browsing by Author "Barton, Neal A."

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    The challenges of predicting pipe failures in clean water networks: a view from current practice
    (IWA, 2021-08-09) Barton, Neal A.; Hallett, Stephen H.; Jude, Simon
    Pipe failure models can aid proactive management decisions and help target pipes in need of preventative repair or replacement. Yet, there are several uncertainties and challenges that arise when developing models, resulting in discord between failure predictions and those observed in the field. This paper aims to raise awareness of the main challenges, uncertainties, and potential advances discussed in key themes, supported by a series of semi-structured interviews undertaken with water professionals. The main discussion topics include data management, data limitations, pre-processing difficulties, model scalability and future opportunities and challenges. Improving data quality and quantity is key in improving pipe failure models. Technological advances in the collection of continuous real-time data from ubiquitous smart networks offer opportunities to improve data collection, whilst machine learning and data analytics methods offer a chance to improve future predictions. In some instances, technological approaches may provide better solutions to tackling short term proactive management. Yet, there remains an opportunity for pipe failure models to provide valuable insights for long-term rehabilitation and replacement planning.
  • Thumbnail Image
    ItemOpen Access
    An evolution of statistical pipe failure models for drinking water networks: a targeted review
    (IWA, 2022-01-19) Barton, Neal A.; Hallett, Stephen; Jude, Simon R.; Tran, Trung Hieu
    The use of statistical models to predict pipe failures has become an important tool for proactive management of drinking water networks. This targeted review provides an overview of the evolution of existing statistical models, grouped into three categories: deterministic, probabilistic and machine learning. The main advantage of deterministic models is simplicity and relative minimal data requirement. Deterministic models predicting failure rates for the network or large groups of pipes performs well and are useful for shorter prediction intervals that describe the influences of seasonality. Probabilistic models can accommodate randomness and are useful for predicting time to failure, interarrival times and the probability of failure. Probability models are useful for individual pipe models. Generally, machine learning describes large complex data more accurately and can improve predictions for individual pipe failure models yet are complex and require expert knowledge. Non-parametric models are better suited to the non-linear relationships between pipe failure variables. Census data and socio-economic data requires further research. The complexity of choosing the most appropriate statistical model requires careful consideration of the type of variables, prediction interval, spatial level, response type and level of inference is required.
  • Thumbnail Image
    ItemOpen Access
    Improving pipe failure predictions: Factors effecting pipe failure in drinking water networks
    (Elsevier, 2019-07-29) Barton, Neal A.; Farewell, Timothy S.; Hallett, Stephen H.; Acland, Timothy F.
    To reduce leakage and improve service levels, water companies are increasingly using statistical models of pipe failure using infrastructure, weather and environmental data. However, these models are often built by environmental data scientists with limited in-field experience of either fixing pipes or recording data about network failures. As infrastructure data can be inconsistent, incomplete and incorrect, this disconnect between model builders and field operatives can lead to logical errors in how datasets are interpreted and used to create predictive models. An improved understanding of pipe failure can facilitate improved selection of model inputs and the modelling approach. To enable data scientists to build more accurate predictive models of pipe failure, this paper summarises typical factors influencing failure for 5 common groups of materials for water pipes: 1) cast and spun iron, 2) ductile iron, 3) steel, 4) asbestos cement, 5) polyvinyl chloride (PVC) and 6) polyethylene (PE) pipes. With an improved understanding of why and how pipes fail, data scientists can avoid misunderstanding and misusing infrastructure and environmental data, and build more accurate models of infrastructure failure.
  • Thumbnail Image
    ItemOpen Access
    Using generalized additive models to investigate the environmental effects on pipe failure in clean water networks
    (Nature Research (part of Springer Nature), 2020-07-01) Barton, Neal A.; Farewell, Timothy S.; Hallett, Stephen H.
    Predicting pipe failures using statistical modelling benefits from detailed knowledge of the conditions and circumstances which influence such failures. Incorporating this knowledge into model building improves failure predictions. In this study, we model weather, soil and hydrogeological variables in a generalized additive model for five common pipe materials separately, using partial dependence plots to understand the partial effects of each variable on pipe failure. We show how severe temperatures are associated with high pipe failure. Cold temperatures and air frost and their interaction with soils represent the key factors for pipe failures during the winter for metal pipes. Warm temperatures, high soil moisture deficit and soil movement results in higher pipe failures in asbestos cement pipes during the summer. Warm temperatures, ground movement and soil wash out, and water demand are key factors for polyvinyl chloride pipe failure during the summer. Frost is a key factor influencing polyethylene pipes during winter. An understanding of the physical principals concerning pipe failures can enable the development of more accurate models, guiding network management plans to help reduce asset leakage through appropriate interventions