Advancing interpretable AI–BIM integration, physics-informed surrogate modeling, and resource-frugal computational workflows for sustainable building performance.
Coupling BIM parameters and dynamic EnergyPlus simulations with explainable ML (SHAP, sensitivity analysis, uncertainty bounds).
Designing lean tabular and geometric workflows accessible without hyperscale computing infrastructure.
Automating technical interoperability bottlenecks while preserving architectural authorship and engineering judgment.
Systematic mapping of BIM authoring schemas (IFC, gbXML), automated extraction pipelines, machine learning surrogate architectures, and explainability practices across early-stage building energy modeling.
View Protocol & Repository →Foundational governance framework defining project lifecycle states, reproducibility requirements, CRediT authorship guidelines, and open licensing standards.
Read Governance Repository →