Hardwood is becoming an increasingly important resource for the construction sector due to its high mechanical performance and its growing availability linked to ongoing climate changes. However, the use of wood in structures requires that it be classified according to its strength, and for hardwood, the parameters used in production lines, which are typically designed for softwoods, do not have the same predictive efficiency. This study aims to explore the possibility of optimizing the measurement of local grain direction (GD) using laser scattering technology to predict the mechanical properties of beech wood within automated machine grading lines. For the purposes of this study, the extent of knots in approximately 100 beech boards was assessed visually by an expert operator and with the laser scattering technique. The data obtained were compared with one another and subsequently used to predict the bending strength and elastic modulus of the elasticity, as determined by four-point destructive bending tests. Among the predictive parameters considered, localized GD provided the best results, showing medium-to-high correlations with the mechanical properties. This highlights that this technology is very promising for machine strength grading of beech and likely for other hardwood species as well.