Procter & Gamble boosts diaper quality with predictive maintenance

Procter & Gamble (P&G), one of the world's largest consumer products companies, faced high defect rates and significant losses due to damaged diapers during the manufacturing process of Pampers. To address this, P&G launched the Hot Melt Optimization platform, using proprietary IoT sensors on assembly lines and Microsoft Azure's predictive analytics and edge analytics. The system monitors glue temperature and pressure in real time across 11 US manufacturing plants. Machine learning algorithms, supported by an AI rules engine within Azure, identify and correct anomalies before they cause large-scale production issues. This data-driven approach has eliminated 70% of defective diaper output, resulting in seven-figure weekly savings. The project also improved uptime, reduced scrap, and supported higher production capacity. The system was extensively piloted, using an industrial control database and Grafana for monitoring, before rollout across the network. P&G credits collaboration with Microsoft and Rockwell Automation for enabling these reliability, efficiency, and cost improvements. P&G now demonstrates how predictive analytics, IoT, and edge computing can have significant, measurable impacts on consumer goods manufacturing.