06. November 2019

RISC Software: Optimization Enables Sustainable Wood Processing

In cooperation with FILL Gesellschaft m.b.H., RISC Software GmbH developed an optimization solution for the production of glued laminated timber (glulam). The algorithm developed for this calculates the ideal cutting-to-size parameters and arrangement of the boards in glulam for the most efficient use of the wood.

Plant and machinery installers want to guarantee their customers that the plant or machinery in their workshops makes the best possible use of the raw material in every possible product configuration (batch size 1). Using methods of optimization is therefore an obvious path to take. For FILL Gesellschaft m.b.H., RISC Software GmbH devised and implemented a method of optimizing machine control for the production of glulam. Glulam consists of layered timber bonded together with adhesives. Owing to its specific characteristics it is used in wood construction and interior design. Production of glulam begins with upstream scanning of the feedstock, the boards of timber. This identifies usable and unusable areas of the timber such as chipped edges and knots. The optimization algorithm that has been implemented then calculates the current status of the machinery for these scanned boards and the product configuration that should be chosen to ensure the best possible cutting-to-size of the boards and their use in glulam so that as little usable timber as possible is wasted. This makes it possible to change the order of the boards and insert segments of them as intermediate buffers to be used later on. In this way, the glulam is constructed layer by layer and each layer glued. Processing of the completed glulam then continues in downstream production steps.

The wood processing production plant operates with a preset cycle time of approximately two seconds. Within this time, the next processing steps must be calculated and the commands relayed to the control system.

However, the sheer variety of the feed material and the possible product configurations mean that there are countless possible solutions which must be evaluated in this short space of time. This in turn means that, despite the high computational power of modern computers, the optimization algorithms must be conceived in such a way that they take account of cycle time and quality so that they are capable of calculating the best possible solution for each specific task within the defined period.

The optimization solution developed for FILL Gesellschaft m.b.H. has been successfully implemented in operational processes since the first quarter of 2019.