Improvements of the Dynamic Programming Algorithm for tree Bucking

Log bucking is one of the most important operations in the transformation of trees into lumber. A bad decision at this stage can jeopardize the optimal recovery in volume or in value. The problem of optimizing the recovery during the bucking process has been solved using, among other things, dynamic programming. This article describes the main approaches and suggests some improvements to the dynamic programming approach. By introducing certain assumptions into the dynamic programming algorithm formulation, this approach becomes both more realistic and more efficient. The algorithm defined here is used in an integrated bucking-breakdown model. Example simulations demonstrate the computational speed improvements that result from the introduction of the assumptions.