A data-driven approach is developed and proven for ranking the risk of low disinfection residual in water distribution storage tanks, 1 month ahead. The forecasting methodology uses water quality data collected from drinking water treatment plants, storage tank outlets, and rainfall data as inputs. This methodology was developed and tested with data from a water utility serving more than 5 million people. Results show high-risk category prediction accuracy of 75%–80%. Using a final year of unseen validation data, more than 90% of the storage tanks ranked in the top 20 by the forecasting methodology experienced low disinfectant residual in the following month. Storage tanks are critical water distribution system infrastructure that are currently managed reactively. The adoption of such readily transferable machine learning approaches enables direct proactive management strategies and efficient interventions that can help ensure drinking water quality.

This paper presents machine learning analysis to understand the factors impacting iron concentrations and discolouration customer contacts in drinking water distribution systems. Fourteen years of network sampling and additional data from a large UK utility were collated, analysed, and interpreted using self-organising maps (SOMs), which include complex network theory (CNT) centrality metrics for the first time, investigating how possible explanatory variables interact. The outputs are used to inform ensemble decision trees for risk estimation of iron exceedance and customer contacts for each of the utility’s DMAs, helping inform proactive maintenance.