Construction industry (WP8)
Work Package WP8 focuses on the construction sector and is based on a systematic, multi-stage methodology that integrates advanced data acquisition, a multi-agent AI framework with an MLLM orchestrator, and the on-site implementation of results. The research links Scan-to-BIM processes and building diagnostics with adaptive manufacturing, thereby creating a comprehensive and intelligently managed ecosystem with the potential for long-term development.
The main objective of WP8 is the comprehensive optimisation of construction processes and the automation of structural diagnostics, with a direct link to Building Information Modelling (BIM) and construction schedule management. The work package will deliver advanced methodologies, software tools and proven technologies with a clear practical impact on streamlining and modernising real-world construction practice.
What problem does the WP solve, and why is it important?
01
Structural diagnostics and quality control on site are slow and prone to human error
02
Digitising buildings is labour-intensive, and the data is difficult to integrate with BIM modelling
03
Construction schedule management lacks dynamism and is unable to respond flexibly to changes in real time
04
Field data collection is fragmented and lacks automated orchestration
05
Adaptive production is not sufficiently linked to the actual situation on site
Examples of use / areas of application and benefits
Industrial construction
- Automated real-time quality control
- Dynamic time management of the construction project
Manufacturing and modular construction
- Reducing scrap rates and material consumption
- A contribution to decarbonisation
Building diagnostics and management
- Extending the service life of buildings
- Optimisation of operating costs
Key people

WP Leader
prof. Ing. Bc. Radoslav Sovják, Ph.D., LL.M.
National Centre for Construction 4.0 | CIIRC CTU
Participating institutions




Used technologies and procedures
- Multi-agent AI systems
- Scan-to-BIM
- Transmission and diagnostics
- On-site implementation
- MLLM Orchestrator
- 3D laser scanning (TLS, mobile SLAM)
- Dynamic time management
Achieved and planned results
2027
AI applications for the semantic analysis of point clouds in the construction industry
R – software
2028
ZoomBoxAI – Adaptive bending control in the ZoomBox machine using AI
R – software
In situ monitoring unit for construction processes
Gfunk – working sample
2031
An AI system for monitoring the production processes of building structures
Ztech – proven technology