Abstract
<jats:p>Introduction. In modern construction, the digitalization of the design process is becoming a key factor in increasing efficiency and competitiveness. Building Information Modeling (BIM) technologies provide comprehensive data management at all stages of the object's life cycle, but their potential can be significantly expanded through the integration of artificial intelligence (AI) tools. Problem statement. Despite the widespread implementation of BIM, a significant part of design solutions remains labor-intensive and error-prone due to the need to process large amounts of information manually. Low quality of source data, imperfect coordination between performers, and insufficient automation of analytical operations lead to delays, increased cost, and risks in design. Objective. To investigate the possibilities of using artificial intelligence algorithms to increase the accuracy, speed, and reliability of BIM design, as well as to identify effective approaches to integrating AI into standard design business processes. Materials and methods. The paper uses methods of theoretical generalization and analysis of scientific publications, comparison of practical experience in the use of neural networks in modeling and data processing, as well as a comparative assessment of machine learning tools in terms of their ability to automate routine operations of BIM processes. Cases of AI use in automatic classification of elements, collision search, and prediction of performance indicators of building systems were studied. Results. It was found that deep learning algorithms increase the accuracy of modeling due to automatic error correction, optimization of parametric solutions, and intellectual coordination between different disciplines. The use of AI will reduce the time for performing individual design operations, as well as improve the quality of data used during modeling and analysis.</jats:p>