Combined analysis of process data and product property measurements to reduce product variability

Process variation is an important obstacle for obtaining uniform product quality. In today’s paper and board machines, a lot of data is available. The process conditions can be obtained from the DCS system, while a lot of information on the product properties is available from the QCS system and the automatized property measurement devices.

By performing data filtering and statistical calculations on data from the different data sources, custom-made KPIs can be continuously determined. These KPIs allow for the detection of critical process conditions leading to extreme streakiness of the CD profile. KPIs derived from process data and the automatized product property measurements can also be combined. This allows for a better understanding of the relationship between process conditions and properties relationship, which leads to a lower product variability.

Different examples of this type of process optimization work are presented. One example deals with an on-line streak analysis based on the QCS data. Another example, based on the Tappi TIS 1101-01 analysis of property variation, is used to continuously quantify and separate the MD, CD, and residuals of the variation. A third example shows the link between coating grammage CD profile and the achieved surface roughness CD profile.