Online Program Home
My Program

Abstract Details

Activity Number: 493 - New Statistical Methods for Lumber Analytics
Type: Invited
Date/Time: Wednesday, August 1, 2018 : 10:30 AM to 12:20 PM
Sponsor: SSC
Abstract #326600 Presentation
Title: Sparse Functional Partial Least Squares Method for Spectral Analysis
Author(s): Jiguo Cao* and Tianyu Guan and Kevin Groves and Martin Feng
Companies: Simon Fraser University and Simon Fraser University and FPInnovation and FPInnovation
Keywords: functional data analysis; calibration prediction models; oriented strand board

A novel laboratory spectroscopy technique was developed at FPInnovations for quickly determining species identification of oriented strand board (OSB) strands and the relative proportions of rot, bark, and species in OSB fines samples. Currently, mills do not monitor these key raw material constituents that play a major role in production operating efficiency and final product attributes. Periodically monitoring raw material can help mills identify problems associated with rot in logs, debarking inefficiency and species variability. Measurement can also provide data to assist in process adjustments and production planning and budgeting.

It is well documented that Vis/IR spectroscopy measurement techniques can differentiate and quantify several material constituents; however, this has not been used to identify strand species or simultaneously measure species, rot and bark composition in OSB fines. In this talk, I will introduce a sparse functional partial least squares method for spectral analysis and generate calibration prediction models.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program