A paper manufacturer got more out of its AI sensors with a simple administrative fix (www.businessinsider.com)

A significant advancement in AI application has emerged from Domtar, a paper manufacturer, which improved the functionality of its AI-assisted sensors with an administrative approach rather than just technical upgrades. Faced with overwhelming data from 450 sensors monitoring equipment vibrations, reliability engineer Matthew McLaughlin realized the need for a strategic shift in how the company utilized this wealth of information. By leveraging the expertise of the sensor provider, Waites Sensor Technologies, and shifting to a more collaborative, data-sharing methodology, McLaughlin enhanced the predictive capabilities of the AI system. This change included regular meetings with Waites analysts, integrating additional data sources, and ensuring consistent communication among staff. The implications for the AI/ML community are profound, as it underscores the necessity of combining human expertise with machine learning to unlock true operational potential. Rather than simply deploying AI systems, organizations must foster a culture that prioritizes collaboration and active feedback to optimize these technologies. As a result of these efforts, Domtar has experienced a reduction of 1,546.65 hours in unplanned downtime and is now making multimillion-dollar business decisions based on data-driven insights from their sensors. This case illustrates the transformative power of AI when effectively integrated into existing workflows, highlighting the critical role of reliability engineers in bridging the gap between technology and operations.