AI, machine learning and BIM for enhanced property valuation: Integration of cost and market approaches through a hybrid model

Abstract

Accurate property valuation is essential for real estate market stability, housing affordability and financial decision-making. However, traditional valuation methods face key limitations. The market approach, reliant on comparable sales data, is prone to subjectivity and data availability constraints. The income approach relies on stable rental income streams, which are often unavailable for newly built dwellings in volatile rental markets. And, the cost approach, based on the Depreciated Replacement Cost (DRC) method, neglects broader market influences by focusing solely on property characteristics. Despite advancements in Automated Valuation Models (AVMs) using Machine Learning (ML), these models remain sensitive to market fluctuations and lack integration with 3D property characteristics. To address these challenges, this study proposes a hybrid Artificial intelligence (AI) and Building Information Modeling (BIM)-driven property valuation model, integrating the DRC method with market-based valuation adjustments using ML, Natural Language Processing (NLP) and BIM 3D models. The framework consists of several key stages, including mass land valuation using ML techniques, automated construction cost estimation through BIM-based Quantity Take-Off (QTO) and NLP-based cost-matching, dynamic depreciation assessment via BIM-integrated maintenance management, entitlement calculation using optimization techniques, and market impact assessment through ML-driven modeling. The methodology was tested on a high-rise residential building in Melbourne, Australia, and the results demonstrated high accuracy, with estimated property values closely aligning with recent market transactions. The estimated values for one-bedroom and two-bedroom units were 100 % within the range of recent market transactions, and the estimate for the three-bedroom units showed only a 0.057 % deviation from the actual market value. The study advances the digital transformation of property valuation, showcasing how AI, ML and BIM enhance automation, accuracy and efficiency. These findings hold significant implications for the real estate sector, offering a scalable and adaptable framework for industry adoption.