A Neural Network Model for Wood Chip Thickness Distributions


  • Emily B. Schultz
  • Thomas G. Matney
  • Jerry L. Koger


Neural network, wood chips, wood chip thickness, wood chip thickness distributions


Wood chip thickness is an important factor in pulp quality and yield. An artificial neural network model was developed and incorporated into a growth and yield simulator to predict wood chip thickness distributions from stand and tree characteristics. Models based on direct parameter estimation and parameter recovery were also developed for comparison to the neural network. Data were derived from 11,771 individual loblolly pine chip thickness measurements. Four stand ages, five dbh (diameter at breast height) classes, and three stem positions were used to predict the cumulative proportion of chip weight per chip thickness class. Results showed that the neural network model was superior to the two deterministic models on the basis of bias, root mean square error, and index of fit. Sensitivity analyses for the neural network model demonstrated that thicker chips were produced by younger stands and lower stem positions. The neural network was combined with a growth and yield simulator to demonstrate its use as a tool for procurement foresters and mill managers in predicting yields from stands of given characteristics.


Bailey, R. L., and T. R. Dell. 1973. Quantifying diameter distributions with the Weibull function. Forest Sci. 19(2): 97-104.nBecker E. 1992. The effects of chip thickness and kraft cooking conditions of kraft pulp properties. Proc. Tappi Pulping Conference, Book 2: 561-565.nBorlew, P. B., and R. L. Miller. 1970. Chip thickness: A critical dimension in kraft pulping. Tappi 53(11): 2107-2111.nBurden, R. L., J. D. Faires, and A. C. Reynolds. 1981. Numerical analysis. Prindle, Weber, and Schmidt, Boston, MA. 598 pp.nChristie, D. 1987. Chip screening for pulping uniformity. Tappi J. April: 113-117.nDell, T. R., D. P. Feducia, T. E. Campbell, W. F. Mann, Jr., and B. H. Polmer. 1981. Yields of unthinned slash pine plantations on cutover sites in the west gulf region. Res. Pap. SO-147. USDA Forest Serv. 84 pp.nDubois, M. R., W. F. Watson, and F. G. Wagner. 1991. Sawmill chip survey in the U.S. South: An analysis of processes and chip quality. Report presented to the Production and Management Issues Committee, Southeastern Lumber Manufacturer's Association, October 1991. Mississippi State University, Mississippi State, MS.nFarrar, R. M., and T. G. Matney. 1994. A dual simulator for natural even-aged stands of longleaf pine in the South's East Gulf Region. South. J. Appl. For. 18(4): 147-155.nFlowers, R. K., W. F. Watson, B. K. Wharton, M. L. Belli, and B. J. Stokes. 1992. Utilization and yield of chips from loblolly pine in central Arkansas. Proc. Tappi Pulping Conference, Book 2: 467-471.nHassoun, M. H. 1995. Fundamentals of artificial neural networks. The MIT Press, Cambridge, MA. 511 pp.nKendal, M., and A. Stuart. 1977. The advanced theory of statistics, 4th ed., vol. 1. MacMillan Publishing Company. New York, NY. 472 pp.nKoger, J. L. 1994. Distributional properties of individual wood chips from loblolly pine plantations in central Arkansas. Ph.D. dissertation. Mississippi State University, Mississippi State, MS. 152 pp.nKoger, J. L. D. J. Leduc, T. G. Matney, W. F. Watson, A. A. Twaddle, and B. J. Stokes. 1993. Distributional properties of individual wood chips from loblolly pine plantations in central Arkansas. Proc. Tappi Pulping Conference, Book 2: 741-748.nMatney, T. G., and A. D. Sullivan. 1982a. Approximating thinned stand diameter distributions with statistical probability functions. Proc. Second Biennial Southern Silvicultural Research Conference, Atlanta, GA: 315-324.nMatney, T. G., and A. D. Sullivan. 1982b. Compatible stand and stock tables for thinned loblolly pine stands. Forest Sci. 28(1): 161-171.nMatney, T. G., and R. M. Farrar, Jr. 1992. A thinned/unthinned loblolly pine growth and yield simulator for planted cutover site-prepared land in the mid-gulf South. South. J. Appl. For. 16(2): 70-75.nNeural Ware, Inc., Technical Publications Group. 1993. Neural computing: A technology handbook for Professional II/Plus and NeuralWorks Explorer. NeuralWare, Inc., Pittsburgh, PA. 329 pp.nTikka, P., H. Tahkanen, and K. Kovasin. 1993. Chip thickness vs. kraft pulping performance, Part 2: Effect of chip thickness screening on cooking, oxygen delignification, and bleaching of softwood kraft batch pulp. Tappi Pulping Conference Proceedings, Book 2: 833-838.nTwaddle, A. A. 1996. The effect of chip length, wood density, and ring orientation on chip formation variability. Ph.D. dissertation, Mississippi State University, Mississippi State, MS. 310 pp.nTwaddle, A. A. and W. F. Watson. 1990. Survey of disc chippers in the southeastern USA, and their effects on chip quality. Proc. Tappi Pulping Conference, Book 1: 77-86.nUelmen, R. L. 1993. Mill cut quality improvement program. Proc. Tappi Pulping Conference, Book 2: 725-733.nWallace, R., W. B. Stuart, and R. Starling. 1992. The effects of wood temperature on fines and pin chip production. Proc. Tappi Pulping Conference, Book 2: 477-480.nWeiss, S. M., and C. A. Kulikowski. 1991. Computer systems that learn. Morgan Kaufmann, San Mateo, CA. 223 pp.nWood, J. R., and G. L. Gosda. 1992. Chip quality improvement for batch kraft pulping. Proc. Tappi Pulping Conference, Book 2: 567-573.nWorster, H. E., D. L. McCandless, and M. E. Bartels. 1977. Some effects of chip size on pulping of southern pine for linerboard. Tappi 60(2): 101-103.nZarnoch, S. J., D. P. Feducia, V. C. Baldwin, and T. R. Dell. 1991. Growth and yield model predictions for thinned and un-thinned slash pine plantations on cut over sites in the west gulf region. Res. Pap. SO-264. USDA Southern Forestry Experiment Station. 32 pp.n






Research Contributions