An automated BIM-driven framework for early-stage prediction and influencing factor analysis of construction and demolition waste from masonry infill walls

The rapid growth of construction and demolition waste (CDW) presents a critical challenge to sustainable development in the construction sector, yet effective mitigation remains constrained by the absence of reliable quantification methods at the design stage, where waste prevention is most feasible. Addressing this gap, this study targets masonry infill walls, one of the major contributors to CDW and develops an automated Building Information Modeling (BIM)-driven framework for early-stage prediction of material consumption and waste generation. Through secondary development in Autodesk Revit, the proposed framework integrates three interconnected modules: data acquisition, deduction calculation, and quantity computation, enabling geometry-driven and automated estimation of material use and associated waste directly from design models. Validation using as-built data from a mixed-use complex project in Shenzhen, China, demonstrates a prediction accuracy of 98.79%, confirming the robustness and practical applicability of the approach. To further advance understanding beyond quantification, eight architectural design scenarios were constructed and analyzed using correlation analysis and ridge regression to identify key determinants of waste generation. The results reveal that the number of door and window openings exerts a dominant influence on masonry waste rates, while wall volume primarily governs total waste quantities. By enabling accurate, automated CDW estimation and revealing design-sensitive waste drivers, this study provides a scalable decision-support tool for designers and project stakeholders to implement proactive waste-reduction strategies at source. The findings contribute to advancing BIM-enabled circular construction practices and support the transition toward low-waste and sustainable built environments.