Process control and production strategies in the sawmill industry

The sawmill process itself is not complicated. What makes it complex is thediversity of the raw material, the logs that are processed in the sawmill, and thedivergent production with many different end products. In the sawmill variousnumber of measurement and scanning equipments are installed. These are usedfor controlling the various processes and for measuring how well they are carriedthrough.The main objective of this thesis is to build knowledge of how we can makethe sawing process, one of the main steps within the sawmill process, more efficientwith respect to both volume yield and value recovery by new equipmentand new production strategies. The intention has been that the conclusions inthis thesis can contribute to a knowledge base that can be of assistance in decisionsregarding control process parameters and production strategies for a sawmill.There is a possible economic saving by increased volume yield for the sawmillsif the saw kerf width can be reduced, but there is a fear that the presence andmagnitude of saw mismatch will be affected by this. Saw mismatch occurs on thesawn boards due to displacement in axial direction of the saw blades in doublearbor saw machines as a consequence of wear, heat or mechanical disturbance. Itis shown in this thesis that it was possible to measure saw mismatch automaticallyduring sawmill operation by laser triangulation and that the measurements werecomparable to manual measurements. It is also suggested how the presence andmagnitude of saw mismatch can be evaluated when measurements are carriedout in a sawmill.Another study addressed in the thesis is the consideration of applying analternative log rotation for each log than the in Scandinavia industrial praxis ofhorns down (log crook faced upwards). This possibility for a greater profit returncomes in question since the development of an industrial computed tomographyscanner makes the internal knot structure of the log available.Log breakdown simulations of about 600 Scots pine (Pinus sylvestris [L.])logs and 800 Norway spruce (Picea abies [L.] Karst.) logs mainly from differentgeographic locations in Sweden showed that there is a potential value increasewhen rotating each log for greatest profit return. The potential value increase wasdependent of the rotational error of the sawing machine and the price differencesbetween quality grades. For the 600 Scots pine logs and the 800 Norway sprucelogs in the study an increased average value increase of 13% was obtained ifapplying the rotation that maximizes the value of each log instead of the horns down position. An introduced rotational error of the sawing machine reducedthe value potential to 6%. There was a weak correlation between the log rotationthat maximizes the value of each log and the outer shape of the logs. This meansthat the outer shape can not be used as an indicator of how the log should berotated for greatest profit return.One subject of discussion in the thesis is also the importance of representativeinput data in order to make as general conclusions as possible. The Swedishstem bank has been an important factor in many studies made in the field ofwood technology. It is a well-documented data set and computed tomographyscanning of logs has made it possible to represent internal wood features in logbreakdown software. Since computed tomography scanning of logs is a time consuming process the number of scanned logs are relatively small. Now thatan industrial computed tomography scanner operating at production speed isentering the market this opens up new possibilities. Hopefully simulation studiesthat are performed on larger industrial data sets coming from logs processed inthe sawmills at a daily basis is not too far away.Godkänd; 2013; 20130325 (bendar); Tillkännagivande licentiatseminarium 2013-04-25 Nedanstående person kommer att hålla licentiatseminarium för avläggande av teknologie licentiatexamen. Namn: Anders Berglund Ämne: Träteknik/Wood Technology Uppsats: Process Control and Production Strategies in the Sawmill Industry Examinator: Professor Anders Grönlund, Institutionen för teknikvetenskap och matematik, Luleå tekniska universitet Diskutant: Teknisk doktor/Gruppchef SP Anders Lycken, SP Trä, Stockholm Tid: Fredag den 17 maj 2013 kl 10.00 Plats: Hörsal A, campus Skellefteå, Luleå tekniska universite