Intelligent Digital Construction Component Systems Based on Machine Learning Algorithms

Building Information Modeling (BIM) has gained significant attention in traditional construction, and its importance has been amplified by the integration of machine learning algorithms. However, existing component systems struggle to achieve efficient automatic planning with complex assembly tasks and information data. Insufficient interaction among functional modules leads to mismatched assembly information within the system and the requirements of actual construction projects. To address these challenges, we propose an intelligent digital construction component system that uses machine learning algorithms for automatic classification of architectural components during assembly data collection, categorization, and production processes. Our system testing on Bai ethnic wooden house components dataset reveals that the system achieves an average feasible probability of 83.0775% for the classification planning of corridor components and 81.3625% for path planning. These results demonstrate the great potential and operational effectiveness of our system in the preservation of historical heritage.