Innovation in Industrial Asset Management: From BIM to Digital Twin

Industrial infrastructure maintenance is undergoing a profound digital revolution. Traditionally reliant on subjective and reactive visual inspections, the management of envelopes (facades and roofs) is rapidly moving toward digitalization and automation. This changing landscape is the perfect stage for the DigiMaTRIA project, as detailed in the scientific paper "DEVELOPMENT OF A BIM-BASED DIGITAL TWIN FOR THE MAINTENANCE OF THE BUILDING ENVELOPE OF INDUSTRIAL ASSETS".  

The main goal of this research is to develop a Digital Twin ecosystem tailored for the Operation and Maintenance (O&M) phase, using BIM (Building Information Modeling) as the structural backbone and centralized data repository.

Digimatria Project Flowchart: Data Acquisition; Digital Twin; Reporting and Decision Support.

The As-Is Enrichment and Inspection Methodology

The proposed workflow aims to geometrically and semantically enrich digital models through key interconnected stages:

1. Advanced Data Acquisition: Deploying Unmanned Aerial Vehicles (UAV) for high-resolution photogrammetry combined with terrestrial laser scanning (LiDAR).

2. Digital Processing: Merging and aligning 3D point clouds to create a highly accurate, georeferenced geometric mesh of the asset.

3. Automated Damage Detection: Applying Deep Learning algorithms (specifically the Mask R-CNN framework) to automatically detect and segment defects such as corrosion, mechanical damage, and water accumulation on sandwich panels.

4. BIM Integration: Using customized Dynamo scripts, the identified anomalies are automatically projected and integrated into the Autodesk Revit environment as parametric objects (at LOD 200 and LOD 300 levels).

   
   

IoT Connectivity and Predictive Maintenance

While the immediate focus lies in consolidating this Base Digital Model, the roadmap for DigiMaTRIA includes the seamless integration of ioT sensors. These devices will continuously monitor environmental and structural conditions of the physical asset, feeding AI-driven degradation models to revolutionize Asset Management and establish reliable predictive maintenance.

This methodology was successfully validated in a controlled environment (FORTIS Tower) and on a real-world manufacturing plant at ARCO, Santo Tirso. Join us at ptBIM to discover the full insights behind this project!

🔗 Learn more about the conference and register at: https://ptbim.org/