Increasing the share of long-lifecycle wood products by extracting the value of data in the sawmill industry

Sweden’s forest industry plays a key role on the path to a circular bioeconomy, but of the annually harvested forest only about 20% are transformed into products (sawn timber) with a long lifecycle and a high rank in the material cascade.

A main reason is the lack of data refinement and integration in the information flow in the sawmill process, with unknown properties of the felled logs when they arrive at the sawmill.

With the help of X-ray computed tomography (CT) it is possible to gather rich data about the internal features of a log prior to sawing.

If this data can be refined appropriately and integrated into the sawmill process, then the material use can be optimised and waste (short-lived products, e.g. chips or sawdust) can be reduced.

We propose a framework called BATA-VLAI, which combines modern machine learning with wood material science, to refine the collected data from CT scans of sawlogs, such that accurate predictions can be made about the potential products inside the log.

The resulting virtual sawmill process makes it possible to optimise which products will be produced, which quality they will have, and to which customer they will be allocated – already before making the first cut in the log. A full integration in the sawmill process would increase the share of long-lived wood products in the society.

The project team consists of wood scientists, machine-learning researchers and specialists, and industry representatives.