1 / 15

Hello. Thanks for your interest in CAMO and for letting us tell you about what we do.

C A M O Helping Smart People Get Smarter TM. Æ. Hello. Thanks for your interest in CAMO and for letting us tell you about what we do. Our goal is simply to help you get smarter, for when you get smarter, you can make better decisions.

Download Presentation

Hello. Thanks for your interest in CAMO and for letting us tell you about what we do.

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. CAMOHelping Smart People Get SmarterTM Æ Hello.Thanks for your interest in CAMO and for letting us tell you about what we do. Our goalis simply to help you get smarter, for when you get smarter, you can make better decisions. You canget your answers fast. You can take action. And get great results. And isn’t that what it is all about?Getting great results? www.camo.com CamoAsia@Camo.com

  2. CAMOHelping Smart People Get SmarterTM Prediction of a Paper Quality Parameterfrom Process Settings, using Multivariate Modelling

  3. CAMOHelping Smart People Get SmarterTM Context • Print Through (PT) is an important quality property of paper samples • Print Through is a measurement of how a printed text shows on the other side of a paper sheet. A low print through indicates a paper of high quality. • But PT is costly and time consuming to measure! • Therefore we want to be able to perform ”indirect measurements” of PT, i.e. predict PT from other paper properties and process parameters.

  4. CAMOHelping Smart People Get SmarterTM Brainstorming • Property of interest: • Paper Print Through (PT) • 15 parameters are suspected to be influential on Print Through • Weight, ink, scatter, opacity, roughness, permeability, density, PPS, oil absorbance, ground wood, thermo pulp, waste paper, magenf, filler.

  5. CAMOHelping Smart People Get SmarterTM Data Collection • 103 paper samples are collected from the production line • The 15 parameters are measured/recorded on these samples. • Some of the measurements fail to be recorded on some samples, but all 16 samples will successfully be used anyway in a multivariate model. • Print Through itself is measured from our costly, direct measurement method. • These 103 samples will be used to calibrate and validate a prediction model

  6. CAMOHelping Smart People Get SmarterTM

  7. CAMOHelping Smart People Get SmarterTM Model Calibration • A PLS-regression model is built. • Opacity is strongly negatively correlated to Print Through: The higher the Opacity, the lower the Print Through. • Oil Absorbance is in the middle of the plot, therefore not correlated to PT at all.

  8. CAMOHelping Smart People Get SmarterTM Identification of the Important Parameters • The Regression Coefficients plot shows that 6 of the 15 parameters have a statistically significant role in the prediction model for Print Through. • The other 9 parameters are not related to print Through.

  9. CAMOHelping Smart People Get SmarterTM Simplified Model • Now that the relevant parameters for PT are identified, we can build a simplified model based on these 6 only: weight, brightness, scatter, opacity, density and filler. • These 6 parameters will be enough to predict a reliable PT value.

  10. CAMOHelping Smart People Get SmarterTM Model Validation • The validation calculations indicate that this model will predict Print Through with an accuracy of 2,8 units. This is very good: it is very close to the precision of our direct Print Through measurements.

  11. CAMOHelping Smart People Get SmarterTM Collection of New Samples • 12 new paper samples are produced and we will use our model to predict their Print Through value. • The 6 key prediction parameters are measured on these new samples.

  12. CAMOHelping Smart People Get SmarterTM Prediction of New Samples • Print Through is predicted for each of the new samples • The Deviation values indicate how reliable the prediction is on each of the samples. They reflect how well each of these new samples match the samples first used to build the regression model.

  13. CAMOHelping Smart People Get SmarterTM Conclusions • Cost • 16 usual control measurements on a series of ca. 100 samples were recorded • Outcome • 6 easy-to-measure paper properties are found sufficient to replace a costly direct measurement of Print Through • The model takes only seconds to run and gives very reliable PT predictions. • Optimized settings • Knowing which process parameters directly have an impact on PT, we can now regulate the process settings depending on which paper quality level we want to achieve.

  14. CAMOHelping Smart People Get SmarterTM Value for Users • Find out which process settings have a positive or negative influence on your final product quality • Fully understand your process • Easily optimize your process settings accordingly • Save money by replacing costly measurements with unexpensive measurements and a valid prediction model • Possibility to run On-Line predictions if relevant

  15. CAMOHelping Smart People Get SmarterTM Thank You!We look forward to speaking with you soon Please contact: CamoAsia@Camo.com CAMO Software India Private Limited #14-15, Krishna Reddy Colony, Domlur Layout, Bangalore - 560071 INDIA Phone: +91 - 80 - 5125 4242 Fax: +91 - 80 - 5125 4181

More Related