The Devil's In The Details: How Advanced Pattern Recognition Can Predict Equipment Failure

With the proliferation of artificial intelligence, machine learning, and other advanced computing solutions, utilities are looking to take advantage of the massive amounts of data at their disposal by using it to make real-time decisions that improve operational reliability and performance. But when is an organization ready to adopt a data-centered culture? This paper will discuss what’s needed for the successful adoption of advanced data analytics tools and share the results of a pilot project with the goal to transform a wastewater utility’s maintenance strategies from reactive to proactive using a predictive analytics software that leveraged the data collected by an existing process control system. The objectives of the pilot were: 1) to determine whether operational deviations could be detected ahead of conventional process control system alarms, and 2) to provide actionable intelligence prior to equipment/operational failure.