Reevaluating IFC and Comparing USD in the Context of the Built Environment

This study conducts a structured comparative analysis between Industry Foundation Classes (IFC) and Universal Scene Description (USD) to evaluate their respective capacities for supporting next-generation information workflows in the built environment. IFC has long been the cornerstone of open data interoperability in Building Information Modeling (BIM), yet its rigid schema, declarative semantics, and file-based federation model present limitations in addressing emerging needs for dynamic collaboration, granular data extensibility, and AI-integrated operations. In contrast, USD introduces a scenegraph-based architecture with non-destructive composition, layered editing, and schema modularity, originally developed for visual effects pipelines but increasingly recognized for its potential across domains such as robotics, geospatial modeling, and digital twin development. The analysis is framed across five analytical dimensions: data structure and hierarchy, identification and referencing, semantics and extensibility, composition logic and collaboration, and standardization and governance. Findings suggest that while IFC provides foundational stability and regulatory alignment, USD offers transformative capabilities for real-time coordination, semantic fluidity, and multi-author workflows. However, its effective adoption in AECO contexts requires the development of domain-specific schema extensions, quality validation protocols, and integration with existing BIM toolchains. Building on comparative insights between IFC and USD, this study recommends a hybrid strategy in which USD is positioned to complement IFC, enabling more robust interoperability across digital building ecosystems. Future research directions include the institutionalization of schema governance, AI-driven querying over USD-encoded data, and pilot implementations in complex infrastructure environments to assess performance, adaptability, and long-term sustainability.