Overview
Artificial Intelligence is gradually changing how we understand and design materials and structures. Instead of relying only on traditional equations and assumptions, engineers can now use AI to learn directly from experimental and field data. This helps in predicting material behavior, identifying damage early, and modeling complex responses like cracking or softening more realistically.
For students and young engineers, this opens up a new way of thinking—where mechanics and data work together. However, AI is not a replacement for fundamentals; rather, it strengthens them when used carefully and with engineering judgment.
Course Highlight
1. Introduction to AI concepts with direct relevance to structural and material behavior
2. Data-driven prediction of material properties and structural response
3. Modeling complex phenomena like cracking, damage, and softening
4. Basics of structural health monitoring using AI techniques
5. Hands-on exposure to simple tools and real engineering datasets
Our Trainers
Dr. Sajan Malviya
Dr. Sajan Malviya is an academician and researcher specializing in geotechnical engineering, with a focus on soil behavior, contaminant transport, and data-driven analysis in civil engineering systems. He holds a Bachelor’s (MANIT), Master’s (IIT Kharagpur), and Ph.D. in Civil Engineering (IIT Kanpur) with a specialization in geotechnical engineering.
His research interests lie numerical analysis using MATLAB and customized codes, and emerging applications of data-driven methods in geotechnical and material behavior. He has been actively working towards integrating traditional engineering approaches with modern computational techniques to better understand complex subsurface processes.
Dr. Malviya has teaching experience in core civil engineering subjects, including soil mechanics, foundation engineering, and geotechnical analysis. Alongside his academic responsibilities, he is engaged in guiding student projects and developing interdisciplinary research ideas at the intersection of civil engineering and artificial intelligence.
His current academic focus includes exploring AI-assisted modeling of engineering systems, with an emphasis on maintaining strong grounding in mechanics while leveraging data-driven approaches for improved prediction and analysis.
Modules
12-Day Course Plan
Day 1: Introduction to AI in Engineering
- What is AI? Myths vs reality
- Why civil engineering needs AI
- Limitations of traditional methods
Day 2: Basics of Data in Engineering
- Types of engineering data (lab, field, simulation)
- Noise, variability, uncertainty
- Data quality issues
Day 3: Introduction to Machine Learning
- Supervised vs unsupervised learning
- Regression vs classification
- Simple examples from engineering
Day 4: Mathematical Intuition (Very Basic)
- Concept of fitting a model
- Error, loss functions, overfitting
- Why more data ≠ better model always
Day 5: Predicting Material Properties
- Strength prediction (concrete/soil)
- Role of input parameters
- Case-based explanation
Day 6: Structural Response Prediction
- Load–displacement behavior
- AI vs FEM (conceptual comparison)
- Surrogate modeling idea
Day 7: Modeling Nonlinear Behavior
- Cracking, yielding, softening
- Why classical models struggle
- How AI approximates complex behavior
Day 8: Introduction to Structural Health Monitoring
- Concept of SHM
- Sensors and data
- AI for anomaly detection
Day 9: Practical Session – Data Handling
- Introduction to dataset
- Basic preprocessing
- Visualization
Day 10: Practical Session – Simple Model Development
- Linear regression / basic ML model
- Training and testing
- Interpreting results
Day 11: Engineering Interpretation of AI Results
- When AI is wrong
- Importance of domain knowledge
- Case discussions
Day 12: Integration and Future Outlook
- AI + FEM + experiments
- Limitations and ethical concerns
- Future directions in civil engineering
Outcomes
After completing the course, learners will be able to:
- CO1: Explain fundamental AI concepts relevant to civil engineering
- CO2: Apply basic regression/ML models to predict engineering properties
- CO3: Analyze material/structural behavior using data-driven approaches
- CO4: Interpret AI outputs in the context of mechanics and design
- CO5: Demonstrate beginner-level use of computational tools for engineering datasets
FAQs
1. Do I need coding experience?
Not necessarily. Basic exposure helps, but the course starts from scratch and focuses more on understanding than coding depth.
2. Is AI going to replace civil engineers?
Unlikely. AI is a tool—it enhances decision-making but cannot replace engineering judgment, especially in safety-critical systems.
3. How is this different from a standard ML course?
This course is contextualized for civil engineering, with examples from materials, structures, and geotechnics.
4. Will we learn deep learning?
Only at a conceptual level. The focus is on fundamentals and applicability, not advanced architectures.
5. Can I use this in research?
Yes—this course can serve as a starting point for research in:
- Material modeling
- FEM–AI hybrid methods
- SHM applications
6. How practical is the course?
Moderately practical. You will:
- Work with real datasets
- Build simple predictive models
- Interpret results critically
