A Comparison of Approaches for Predicting Multiple-Product Yields from Weight-Scaling Data
Alternative systems of equations for predicting individual and aggregate product yields from truckload weight-scaling data were developed and tested using data from Champion International's Building Products Division mill at Newberry, South Carolina. The three systems compared were: (a) a multiple-regression system where aggregate yields are predicted first, (b) a recursive system, and (c) a simultaneously interdependent system of equations. Results showed little difference between the three approaches, but the simultaneously interdependent system was slightly superior. Forest Sci. 30:991-998.
No Supplementary Data
No Article Media