Browsing by Subject "kuivaoksaraja"

Limited pre-information of wood raw material properties complicates customer-oriented production planning in sawmilling. The problem is especially concerning Scots pine (Pinus sylvestris L.) because its quality varies significantly within single trees but also within different forests. In the modern cut-to-length method decision of crosscutting points has to be made with help of stem profile prediction, optimization calculations and external quality properties of wood material. In the tree length method that is minimally used in Finland bucking can be done in a terminal that makes it possible to study whole external profile of the stem before the bucking. However, in crosscutting that aims to producing of value added products it would be important to know the internal quality of stem. One method that provides information on internal quality of wood raw material is x-ray log scanning that is used for the quality sorting at sawmills. The purpose of this study was to find out whether it is possible to predict the internal quality of Scots pine stems with the help of some stem characteristics. Moreover, it was also examined whether it is possible to develop the quality bucking of Scots pine butt logs if bucking is supported by x-ray log scanning. The study material was collected from eight final felling stands. From each stand 30 Scots pine stems were chosen and marked and of these was measured diameter at breast height, dead branch height, live branch height and total height. All marked Scots pine stems were felled and bucked as stems whose length varied between 12 and 20 meters starting from the butt of stem. After that the marked study stems were scanned with x-ray system in order to study their internal properties. Individual quality border points were calculated for each of the stems based on the x-ray data. Dead branch height was the most correlated stem characteristic with quality border point. The relationship of the dead branch height and the quality border point was examined with logistic mixed model. A binary dependent variable was determined for each of the stems. Limit value between binary categories (0/1) was 37 decimetres height of quality border. The dead branch height that was classified into four classes was used as an explanatory variable. By using this method it was possible to predict the correct class for 71,6 % of the stems. A stand number that was used as a random effect variable in the model was not statistically significant. The butt logs bucking that was made at the terminal and the butt logs bucking that was proposed by the harvester drivers was compared to the simulated butt logs bucking that considered also the quality border point of a stem. The total volume of A-class logs increased in a bucking option that considered the quality border compared to two other examined bucking options. Average volume of the butt logs decreased as a main way for increasing the logs quality was shortening of their average length. The results indicate that it is possible to rather well predict the internal quality of Scots pine stems with dead branch height and when crosscutting of stems is supported with x-ray scanning system, it is possible to increase the quality of butt logs. In order to reduce uncertain elements it would be necessary to use larger study material in the future studies. In addition, it would be beneficial to pay more attention to the interpretation of x-ray scanning data.