The production of nonwovens is as diverse as it is sophisticated: Polymers are transformed through various technological processes into fabrics with unique properties for a wide range of applications. From melting and extruding to spinning and bonding, each step is meticulously controlled to achieve the desired fabric properties, such as strength, flexibility and filtration capacity. This versatility allows non-woven fabrics to meet specific demands in fields ranging from medical and hygiene products to automotive and construction materials.

In particular, processes such as spunbond, meltblown, and spunlace, as well as others, each present unique challenges that require strict process control and careful management of parameters. This is essential to prevent disruptions such as spitting errors, quality degradation, and other problems.

First, a specific goal must be defined. This can be to avoid "spitting errors" and other disruptions, or to improve or stabilize any technical or non-technical KPI such as quality, scrap rate, energy consumption etc. The only requirement is that the disruption or KPI can be found or evaluated within the historical data.

Next, the raw historical process data of the targeted nonwoven line is fed into the AI engine that powers Process Booster, aivis®. The AI performs a root cause analysis for the set goal to unveil the key influencing factors and their characteristic behavior. This enables process engineers to gain a deep understanding of the challenge.

If controllable parameters are among the influencing factors, the AI also creates a process control model to live-predict and recommend appropriate parameter adjustments.

One or more process control models per nonwoven line running live in Process Booster. Each model predicts it's target and facilitates corrective actions in case of imminent deviations. In addition, a better understanding of the challenges through root cause analysis enables experts to further improve the process.