Author(s): Akrivi Alexandraki; Grigorios Kyritsakas; Martine Van Den Boomen; Luuk Rietveld
Linked Author(s):
Keywords: Asset management; Detectability; Digital reliability; FMECA; Soft sensors; Water treatment
Abstract: The digitalisation of water utilities introduces new operational risks that conventional asset management frameworks do not fully capture. As utilities increasingly rely on machine-learning-based soft sensors for real-time monitoring and process optimisation, the reliability of these digital assets becomes essential for maintaining safe drinking-water production. This study presents an adapted Failure Modes, Effects and Criticality Analysis (FMECA) tailored to soft sensors, focusing on information-related failure modes such as data loss, model drift and alarm malfunction. A modified Risk Priority Number (RPN) integrates detectability through the reduction factor (1–rD), enabling quantification of the mitigating role of early detection. The method was applied to the CT (Concentration × Time) ozone-exposure soft sensor of Waternet. Ten failure modes were identified across data acquisition, preprocessing, inference and output interpretation. Critical modes included missing SCADA input, model drift and alarm failure due to their impact on disinfection reliability. Incorporating detectability reduced RPN values by up to 80%, demonstrating that improving data visibility has a greater effect on risk reduction than changes in likelihood or severity alone. The framework provides a reproducible foundation for digital-asset reliability assessment and supports ISO 55000-aligned strategies within FIWARE-based decision-support platforms.
Year: 2026