Transitioning to data-driven quality control in industrial veneer drying: a case study

Veneer drying is a vital step in manufacturing veneer-based wood composites as both under- and over-dried veneers have poor performances on the sequential gluing process. Therefore, achieving a precise and consistent control over the final moisture content of dried veneers has always been the mission of wood veneer suppliers. Existing studies primarily utilized physics-based models or laboratory data to simulate the drying process and thus provide guidance for quality control. However, given more and more commercial veneer dryers are being equipped with sensors for substantial data collection, the industry has embarked on data-driven quality control, turning their data into operational intelligence. In this study, 5 months of data of an industrial veneer dryer were acquired, processed, and analyzed to derive insights on the current drying quality distribution and opportunities for improvement. Logistic regression classifiers and a random forest classifier were established to predict the quality level of individual veneer sheets, but the classification ability of all models was limited. Nonetheless, the results suggested that a more refined sorting of the initial moisture content of veneers could lessen the number of re-dries and over-dries, hence improving quality turnout. The importance of data quality assessment was also discussed.