Nonwovens production is complex and diverse: polymers are transformed through various  processes into fabrics tailored for a wide range of applications. Each step must be carefully managed to meet desired fabric properties, such as strength, flexibility, and filtration efficiency. This versatility allows nonwoven fabrics to fulfill specific demands in fields such as medical and hygiene products, as well as automotive and construction materials. Each step—from melting and extruding to spinning and bonding—must be carefully managed to meet desired fabric properties, such as strength, flexibility, and filtration efficiency. This versatility allows nonwoven fabrics to fulfill specific demands in fields such as medical and hygiene products, as well as automotive and construction materials.

Processes like spunbond, meltblown, and spunlace each present unique challenges that require precise control and careful parameter management. Spunbond involves continuous filament extrusion, which demands control over fiber uniformity and strength. Meltblown focuses on creating fine fibers for filtration, requiring precise temperature and airflow management. Spunlace, on the other hand, uses high-pressure water jets to bond fibers, necessitating careful regulation of water pressure and distribution. Without strict oversight, disruptions like spitting errors, quality degradation, and other production issues can arise, jeopardizing the consistency and quality of the final product.

To address these challenges, a specific production goal must first be defined. This goal could be to minimize spitting errors, reduce disruptions, or improve any technical or non-technical key performance indicator (KPI) such as product quality, scrap rate, or energy consumption. The only requirement is that the targeted disruption or KPI is identifiable within the historical process data.

Raw historical process data is then fed into the causal AI engine, which powers Process Booster X. The AI performs a root cause analysis for the defined goal, identifying the key influencing factors and their characteristic behaviors. This analysis gives process engineers a deeper understanding of the root causes behind production issues.

If the identified influencing factors include controllable parameters, the AI generates a process control model that predicts and recommends appropriate parameter adjustments in real time, optimizing the production process.

The outcome is one or more process control models operating in real-time within Process Booster X for each nonwoven production line, ensuring continuous optimization and minimizing downtime. Each model predicts the targeted KPI and suggests corrective actions when deviations are detected. Additionally, the insights gained through root cause analysis, such as identifying optimal temperature ranges or pinpointing specific steps causing quality issues, help experts further refine the production process, improving both efficiency and product quality.