Register Now


Lost Password

Enter your email to reset your password.

Search for answers in posts, Q&A and documents:

BY Author

Improve mill efficiency and meet quality targets
  • Improve raw material assessment to reduce variability in production

  • Improve product quality through better process control and analysis

  • Optimize processes and reduce energy, water & chemical use

Paper manufacturing is a complex process with many variables that affect product quality. Natural variations in raw materials, interactions between different chemicals, the decisions of individual process operators and changes to process settings between paper grades all complicate the goal of achieving consistent, optimized products.

Camo Analytics has worked with pulp and paper manufacturers for many years. Our advanced multivariate data analysis software can be used to make sense of the complex data and identify underlying patterns from the raw material assessment stage through to real-time process and quality control.

As a capital intensive industry heavily affected by environmental issues and narrow margins, innovation is crucial. Our powerful analytical tools can give you the insights and knowledge needed to produce better quality products at lower cost, improve plant and equipment utilization while minimizing the environmental impact of operations. With your know-how and our world leading data analysis software and expertise, we’ll help bring your data to life.

Real business benefits

Product development

  • Reduce variability in product quality with improved control of raw materials
  • Develop better products using multivariate analysis & Design of Experiments
  • Replace expensive laboratory testing with rapid spectroscopy and advanced multivariate analysis

Process monitoring and optimization

  • Identify underlying variables affecting process performance
  • Optimize processes based on deeper process understanding
  • Predictive alerts and real-time decision making using advanced multivariate models and diagnostics


Equipment condition monitoring

  • Reduce costly breakdowns, unplanned maintenance and repairs
  • Improve operator safety and reduce the risk of environmental damage
  • Improve OEE by applying powerful multivariate models to instrument, DCS and SCADA data

Environmental sustainability

  • Reduce chemical use by running processes closer to limits
  • Minimize costly energy consumption by optimizing machinery use and processes
  • Measure and manage contamination levels in wastewater, sludge and air emissions

Example application of multivariate analysis

Developing a multivariate model for on-line control of product quality

A paper producer monitored the quality of newsprint by applying ink to one side of the paper, then measued the reflectance of light on the reverse side of the paper to get a measure of how visible the ink is on the opposite side. This property, Print Through, is an important quality parameter. The paper is also analyzed with regard to several other product variables and raw material variables.

66 samples with 15 process and product attribute variables were collected from the production line and measured against the response variable, Print Through. The purpose was to establish a model that could be used for quality control and production management.


The objectives were:

  • To predict quality from the process variables and other product variables
  • To rationalize the quality control process by reducing the number of variables measured i.e. build a model that includes a subset of variables without losing the underlying variability




Using The Unscrambler® X multivariate analysis software, a PLS regression model was used which enabled the manufacturer to reduce the model down to only the five variables which were shown to be significant:

  • Weight
  • Brightness
  • Scatter
  • Opacity
  • Filler


The new model was then used for on-line predictions. Using the Unscrambler® X Process Pulse real-time process monitoring software, process operators or engineers can view interactive plots during the prediction stage which give insight into any changes in the process. When a new sample falls outside the critical limit the process operator can simply click on the suspect data point in the plot to immediately see which variables are outside the limits as defined by the calibration.

Once a model has been established the next step may be optimization of the process. In this case, some constraints were given for the five important variables in the final model, while at the same time setting a target range for the response variable, Print Through.

The Multivariate Analysis Advantage

Multivariate analysis is the investigation of many variables simultaneously, in order to understand the relationships that exist between them. While traditional (univariate) statistical approaches serve their purposes for investigating and understanding simple systems, when the relationships between variables are complex, a single variable cannot adequately describe the system.

Although modern mills have sophisticated laboratory and process control systems, they typically rely on traditional statistics which often lack the analytical power to show underlying patterns and make sense of the complex data.

Our advanced multivariate data analysis software enhances statistical process control, DCS, PLC and SCADA systems by applying world-leading analytics to the enormous amount of data collected. This gives you deeper product insights, better process understanding and more robust models for predicting quality.

Exploratory data analysis (data mining), clustering, regression and predictive analysis are typical multivariate tools which help process engineers and scientists better identify and understand the complex relationships and variables in their data.

In many processes, the variables have important interactions affecting the outcome (e.g. final product quality) which cannot be detected by traditional univariate statistical process control charts.

When studying a simple process involving two variables, temperature and pH, the sample appears to be in specification when seen with two separate univariate control charts (temperature control chart and pH control chart) but is actually out of specification when seen with the multivariate view.

  • Only with multivariate analysis can the fault be detected
  • The univariate limits are too wide to detect a multivarite fault
  • The two variables under consideration are not independent
  • The “sweet spot“ is defined by the ellipse


Contact Us

WordPress Lightbox