Industry Collaboration, Engineering Applications, and Intelligent Systems
From problem framing and data governance to model development, validation, and field deployment.
Representative Application Areas
Bridges and Infrastructure
UAV inspection, deterioration recognition, repair cost, structural health, and maintenance decisions.
Construction Safety and Quality
Occlusion-aware worker recognition, temporary-works safety, hazardous behavior, segmentation, and early warning.
Engineering Law and Knowledge
LLM- and RAG-based support for procurement, contracts, disputes, regulations, and legal consultation.
Building Energy and Microgrids
Energy loads, PV, BESS, SoC/SoH, grid import, contract capacity, and AI energy management.
Semiconductor Facilities and Circularity
HVAC degradation, RUL/SLP, lifecycle knowledge gaps, maintenance feedback, and reuse decisions.
Environment and Water
Pipeline leakage, water quality, wastewater, air pollution, environmental compliance, and circular materials.
Materials and Structures
Concrete strength, steel, soil, RC beams, and structural design optimization.
Finance and Project Decisions
Cost, finance, portfolios, hedging, schedules, resource allocation, and MCDM.
Disaster and Failure Investigation
Wind-turbine collapse, earthquakes, liquefaction, slopes, dredging, and disaster-resource allocation.
Recent Collaboration and Application Examples
Six-Stage Path from Research to Deployment
1. Problem and Stakeholders
Clarify management needs, technical limits, regulations, costs, and risks.
2. Data Governance
Design sensing, labeling, cleaning, augmentation, splitting, and quality assurance.
3. Method Development
Create original algorithms, AI models, decision frameworks, and explainability.
4. Validation and Comparison
Use benchmarks, cross-validation, field cases, and statistical tests.
5. Systemization and Deployment
Deliver GUIs, dashboards, warnings, schedules, maintenance, and decision support.
6. Feedback and Diffusion
Use field feedback to refine models and generate publications, technical knowledge, talent, and institutional learning.
Collaboration Models
- Joint research: partners provide real problems, data, and sites; PiM Lab supports research design, modeling, validation, and scholarly publication.
- Technical consulting: AI adoption, data strategy, model selection, risk analysis, and decision architecture.
- Living labs: long-term sensing, forecasting, control, and maintenance validation in campuses and public facilities.
- Talent development: transformation of industrial problems into theses, capstone projects, internships, and joint training.