Artificial Intelligence for Civil Engineers

After participating in this course, you will be able to:

• Understand different types of data and apply suitable data processing and analysis
• Learn different types of machine learning methods and their application in civil engineering
• Familiarize with various machine learning algorithms applicable to civil engineering
• Apply various indicators to evaluate the performance of different models
• Employ advanced techniques for data augmentation
• Develop efficient machine-learning solutions for specific civil engineering problems

Description
This three-day course lies at the intersection of Artificial Intelligence and Civil Engineering and aims to equip participants with essential AI skills applicable to civil engineering. Participants will gain a foundational understanding of various types of data in civil engineering, along with proper data processing and visualization techniques.

The course also covers a wide range of supervised, unsupervised, and semi-supervised learning algorithms and their applications in civil engineering. Additionally, participants will learn how to evaluate the performance of models using suitable metrics and indicators. The course emphasizes the importance of explainability and interpretability in machine learning, along with uncertainty quantification using probabilistic approaches. The course offers a practical introduction to Python coding for machine learning and hands-on civil engineering projects in different civil engineering disciplines, such as concrete mixture design, structural analysis and design, and geotechnical analysis.

Case studies and observations of real-world problems will enhance the learning process, and participants will have the opportunity to discuss real cases they have been exposed to in their professional practice.

Who Should Attend
Designers, Construction & Structural Engineers • Project Managers • Technicians & Technologists • Engineers in Training • Construction Inspectors • Inspection Officials • Facility Managers • Architects

Welcome, Introduction, Workshop Preview, Learning Outcomes, and Assessment Method

Day I

Data and Statistics

• Types of data (numerical, categorical, time series, text data, image data, etc.)
• Data processing (data exploring and visualization, correlation, feature selection, dimensionality reduction, etc.)
• Examples

Machine Learning Algorithms

• Supervised learning (regression and classification: linear and logistic regression, decision trees, tree-based ensembles, k-nearest neighbours, support vector machines, neural networks, etc.)
• Unsupervised learning (clustering and dimensionality reduction: principle component analysis, k-means, isolation forest, etc.)
• Semi-supervised learning (search and optimization, reinforcement learning, transfer learning)
• Examples and case studies in civil engineering

Day II

Performance Metrics and Indicators

• Regression metrics (error, mean error, mean normalized bias, mean absolute error, etc.)
• Classification metrics (accuracy, F1-score, confusion matrix, etc.)
• Clustering metrics
• Functional metrics (energy-based metrics, domain-specific metrics)
• Examples and case studies in civil engineering

Explainability and Interpretability

• Supervised machine learning (feature importance, partial dependence, SHAP, LIME)
• Unsupervised machine learning (explainable trees, polar plots)
• Examples and case studies in civil engineering

Synthetic and Augmented Data

• Synthetic minority over-sampling technique (SMOTE)
• Generative adversarial networks (GAN)
• Examples and case studies in civil engineering

Deterministic and Probabilistic Machine Learning

• Probability basics
• Uncertainty quantification

Day III

Introduction to Python Coding

• Useful packages in machine learning

Hands-on Projects in Civil Engineering

• Project 1: Performance prediction and mixture design of concrete
• Project 2: Analysis and design of structural members
• Project 3: Soil parameters prediction

Open Forum: Questions and Answers, Feedback on Achievement of Learning Outcomes

Concluding Remarks and Final Adjournment