The planning processes in timber panel construction are often time-consuming and require a variety of manual, iterative work and processing steps in design and dimensioning. This often leads to resource-intensive planning and unsatisfactory results in terms of the most economical and efficient use of materials. In particular, the numerous closely linked manual processes involved in the design of customised timber panels represent a major organisational and engineering challenge.

The aim of the research project is to develop a planning methodology using machine learning (ML) algorithms. This should integrate constructive, static and manufacturing elements in automatable process steps. The development of the ML-supported planning tool is intended to simplify and accelerate existing processes. In particular, the Interlocking Dowel System (IDS) an innovative construction method that involves the use of wood-based panels and angularly interlocked wood-rod dowels without additional metal fasteners. Machine learning (ML) is used for the inference of suitable partial solutions and the assembly of these into desired load-affine structures. The focus is on the question of the extent to which ML algorithms can be integrated into CAD software and parametric geometry models and enable more precise and efficient designs of mono-material wall elements.

The research work is supported by the close collaboration between FLEX (research, teaching and experimentation) and LaBP (Laboratory for Biosignal Processing) driven forward. Both research groups are based at the HTWK Leipzig. LaBP leads the project organisationally and is responsible for the training, development and implementation of machine learning methods as part of the project. FLEX contributes its expertise in digital planning and manufacturing methodology, in particular in linking machine learning in CAD software, robotic manufacturing and parametric geometry models. The collaboration between FLEX and LaBP ensures a comprehensive use of overlapping expertise and resources to effectively achieve the project goal.

The research project is part of the activities of the HolzBauForschungsZentrum Leipzig. This centre has robotic production technology at its disposal to implement the developed concepts in practice. The organisational aspects, such as the coordination of the various work packages and the allocation of resources, are implemented by the interdisciplinary team in trusting cooperation. Regular meetings and progress reports are used to monitor progress, in the sense of a continuous comparison of the work with the defined content and time targets. The organisational framework of the project is provided by the Research and Transfer Centre (FTZ)located at the HTWK Leipzig, provides the organisational framework for the project.

Keywords: Timber construction, sustainability, robotics, machine learning, AI, optimisation, parametrics, simulations, FEM, tests, IDS