ALOG: A spreadsheet-based program for generating artificial logs

esearchers in the Department of Industrial and Manufacturing Systems Engineering at the University of Missouri-Columbia, in collaboration with the Southern Research Station of the USDA Forest Service, recently developed a hardwood log sawing simulator for use by the hardwood sawmill industry. LogCast (Log Computer Aided Sawyer Trainer) was designed to be used as a training tool for sawyers in primary hardwood processing mills (Occeña et al. 2000). The program provides a nondestructive method for sawyers to experiment with different log grades, orientations, and sawing patterns and presents the user with yield and value information resulting fromsawingdecisions.Amajor advantage of the program is that it incorporates both external and internal defect information. One limitation of the program, however, is the small number of logs in the database. Currently, there are only 18 logs available. The external characteristics of the sample logs including the surface defects were determined from actual red oak logs. The size and shape of internal defects were artificially generated using defect information obtained from physically sawn logs. The placement and orientation of the internal defects was hypothetically done around the pith and outwards in a random radial pattern following the upward direction of tree growth or where suggested by surface defects. Although many studies have attempted to model internal defects (Pnevmaticos et al. 1974, Richards et al. 1979, Samson 1993, Chen and Occeña 1996), none have successfully correlated internal and external defect attributes. Since log grading rules are based on a known correspondence between external indicators and the presence of internal defects, the program could be greatly improved by incorporating any significant correlation. This paper describes software developed to generate realistic digital red oak logs with external and internal log defects, where the internal defects are based on external defect characteristics. Creating a computer-generated log is much faster, easier, and less costly than physically describing a real log. Another advantage of using computer-generated random logs is the assurance that every log will be unique. Unlike a database,