Designing a Large-Scale Immersive Visit in Architecture, Engineering, and Construction
DOI: 10.3390/app13053044
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the technical challenges of creating a large-scale, photorealistic immersive visit for a district under construction, specifically the LaVallée eco-district in Châtenay-Malabry, France. The research is motivated by a gap in existing literature: while Geographic Information System (GIS)-based 3D city models are common for urban planning, they lack the high Level of Detail (LOD) required for architectural ambiance studies. Conversely, Building Information Modeling (BIM) offers high fidelity but is rarely used for large-scale immersive experiences due to technical complexities and a lack of documented workflows for integrating heterogeneous BIM data into realistic virtual environments. The study aims to provide practitioners with a documented pipeline for converting professional BIM files into a cohesive, photorealistic digital twin suitable for Virtual Reality (VR) exploration. The methodology involves a multi-stage workflow using data from over 30 buildings and various infrastructure components (roads, landscaping, utilities) managed via the BIM 360 cloud platform. First, Industry Foundation Classes (IFC) files were parsed using a custom Matlab script to identify coordinate discrepancies, as different architectural firms used varying reference frames. These models were manually aligned in Revit to a common georeferenced system. A preliminary prototype was developed in TwinMotion, which allowed for rapid visualization and VR compatibility but resulted in a "cartoonish" aesthetic lacking photorealism and dynamic capabilities. To address these limitations, the authors developed a pipeline in Unreal Engine. This involved automated tasks via Python scripts for material replacement, mesh substitution for vegetation, and lighting configuration, alongside manual adjustments to enhance visual realism. The findings highlight the specific technical operations required to bridge the gap between industrial BIM data and immersive visualization. The study demonstrates that while TwinMotion is effective for rapid prototyping and functional review, it fails to provide the necessary photorealism for architectural ambiance assessment and lacks scripting capabilities for automation. The Unreal Engine pipeline successfully integrated the heterogeneous BIM data into a unified, high-detail model. The authors document the necessity of manual alignment steps due to inconsistent georeferencing in source BIM files and the importance of automated material and lighting adjustments to achieve a realistic appearance. The resulting model supports VR interaction with HTC Vive headsets, enabling detailed exploration of the district’s future state. The significance of this work lies in its contribution to the Architecture, Engineering, and Construction (AEC) sector by providing a reproducible methodology for creating large-scale digital twins from professional BIM data. It offers practical feedback on the automation and manual steps required to handle industrial-level data, addressing the lack of realism in traditional BIM visualization tools. This approach supports advanced applications such as stakeholder engagement, design review, and the study of architectural ambiance and human perception in virtual environments, thereby advancing the integration of Industry 4.0 technologies in urban development.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | OpenAlex-citations | — | — | 1 | 2026-06-25 |
| archive | success | openalex | — | — | 5 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-25 |
| chunk | success | chunk | — | — | 1 | 2026-06-25 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-25 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-25 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.