CROSSCUT OPTIMIZATION OF BOARDS GIVEN COMPLETE DEFECT INFORMATION

Automatic scanning systems for inspecting wood surfaces can give much more detailed defect information than that obtainable from a manual gr ader. For crosscutting operations in the secondary wood industry, this detailed information gives new opportunities to determine the optimum use of a wood board. In this paper, a model is given that matches the quality requirements of the end products to the quality (defect conte nts) of the raw material. The model can be efficiently implemented on a personal computer (PC). An optimization procedure based on this mode l is proposed, giving an integrated system where defect data are input in one end and final cutting lists are output in the other. The proce dure is guaranteed to produce optimal cutting lists from the given inp ut data. Tests with real data from a Swedish window manufacturer show that the average value of the cuttings is 7 percent higher using this procedure than the value obtained using their existing commercial saw. The procedure is also very efficient. Tests show the procedure create s optimal cutting lists for their product mix within an average of 17 milliseconds on an ordinary PC. The model and the optimization procedu re can be extended to limited two-dimensional problems. Other extensio ns, such as variable-length finger-joint pieces and defect clusters ar e also discussed.