Abstract
Although the author is not aware of any formal markets paying a differentiated price for logs with a specified minimum stiffness level, some forest products markets show a clear preference for products with a higher stiffness. Bucking and sorting logs based on stiffness measurements will require changes to the bucking algorithms and work procedures on mechanized harvesters. Acoustic velocity is a surrogate measure for stiffness. The effects of three threshold levels for minimum acoustic velocity for veneer logs and two approaches – one more conservative than the other – for predicting acoustic velocity were evaluated using priority list bucking simulation. Log-type distribution and value recovery were the performance measures used. Real external stem descriptions and acoustic velocity data from six Douglas-fir stands were used in the simulation. k-nearest neighbor (kNN) and regression methods were used to predict the acoustic velocity of the first and subsequent veneer logs, respectively, cut from each stem. The conservative approach for predicting acoustic velocity for the first veneer log used the minimum value from its nearest neighbors. The less conservative approach used the average acoustic velocity. A 5% increase in the threshold minimum acoustic velocity resulted in a 50% reduction in the number of veneer logs cut and a 3–5% reduction in value recovery, depending on the approach used to predict the acoustic velocity of the first log to be cut. Considerable variation in value recovery, and changes to value recovery, were noted between stands.
Acknowledgements
The data sets used in this paper were originally gathered for other purposes while the author was a professor at Oregon State University (OSU). The author acknowledges the contribution of Drs Hamish Marshall and Dzhamal Amishev for the collection and development of these data sets. Both were OSU PhD students. Thanks are also extended to Roseburg Forest Products, OSU College of Forestry, and Atlas Technologies Ltd for providing stands, equipment, tools and personnel to facilitate data collection.