Automated Panel Segmentation In Clt Buildings Using A Multi-Objective Genetic Algorithm

Cross-Laminated Timber is being increasingly used as an engineered wood product in construction due to its efficient properties. Therefore, there is a substantial need to optimize and automate design to fabrication processes of CLT buildings. Panel segmentation could be a key feature in decreasing CLT waste. Usually, CLT walls, ceilings or roof elements, are made of one large panel with cut-outs for openings. Therefore there are no joints within the wall, avoiding potential weak points and additional connectors. On the negative side, this method creates large amounts of waste, which could not be afforded in the future, considering increasing demand for timber and limited wood production. This paper integrates AI for a novel automated panel segmentation process. A genetic algorithm is developed to find an optimal segmentation scheme to satisfy several engineering constraints and waste minimization. The results shows good adoption with solutions made by experts, in real-world scenarios.