Optimization of particleboard production using NIR spectroscopy and multivariate techniques

This report discusses the possibilities of facilitating cost reduction for the panel-board industry by reducing the consumption of raw material or increasing production capacity. The results of an evaluation of a process modeling system installed in a particleboard plant will be presented and analyzed. The trial clearly shows the advantages of including quantitative raw material characterization in a process modeling system when it comes to optimization of the process to produce boards with uniform strength properties such as modulus of rupture (MOR) and internal bond (113). The raw material in the process is always changing due to variations in the relative amount of wood species, differences in moisture content, storage effects, etc. Some of these properties are difficult to control but with near infrared spectroscopy and multivariate techniques it is possible to monitor these changes and correlate them to variations in the quality of the final product. The reduction in standard deviation accomplished during this study led to cost savings of 2.8 euros/m(3). With a little more fine-tuning, the system could possibly save even more, while maintaining the lower 5th percentile of the board properties.