Using data for predictive maintenance
Software applications that connect to more data, along with improved data formatting and insights, will grow in use over the next five years in biopharmaceutical manufacturing, suggests Petter Mörée, managing director for Europe, Middle East, and Africa at Seeq. “These applications significantly improve an organization’s ability to share data, information, and insights across the entire organization in real-time,” he says.
In downstream processing, one use of such data is for predictive maintenance, in which equipment diagnostic and condition data can be monitored in real-time to determine when equipment is trending out of specification and maintenance is needed. These tools can be used, for example, in chromatography, media optimization, and centrifuge equipment, Mörée explains.
Another example is in monitoring the effectiveness of ultrafiltration membranes. In one case, flow rate and pressure data were used to calculate the change of membrane resistance over time, and a model was created to predict membrane failure. Warning and alert limits were then set, so that filters could be changed prior to membrane failure. This predictive maintenance model enabled more consistent batch quality and increased yield, reported Seeq (1).
Seeq, “Filter Membrane Predictive Maintenance,” seeq.com (2022).