Artificial Engineering Course

Lesson 1 • Defining Artificial Engineering

Clarifies what artificial engineering is, how it differs from classical engineering, and why it matters. Sets the conceptual baseline for all subsequent chapters.

Lesson 2 • Core Principles and Systems Thinking

Introduces systems thinking, feedback loops, and emergent behaviour as foundational lenses. Connects abstract principles to concrete engineering decisions.

Lesson 3 • History and Evolution of the Field

Traces milestones from early automation to modern AI-driven design. Provides historical context that explains current methodologies and tool choices.

Lesson 1 • Statistical Inference and Hypothesis Testing

Teaches estimation, confidence intervals, and significance testing for evaluating engineered systems. Enables rigorous performance validation in later chapters.

Lesson 2 • Probability Theory and Distributions

Establishes probability rules, common distributions, and expectation as tools for modelling uncertainty. Prepares students for probabilistic AI methods.

Lesson 3 • Calculus and Optimisation Basics

Introduces derivatives, gradients, and optimisation concepts central to training AI models. Connects mathematical theory to practical parameter tuning.

Lesson 4 • Information Theory Fundamentals

Introduces entropy, mutual information, and coding concepts that underpin many AI algorithms. Bridges mathematics to model evaluation and feature selection.

Lesson 1 • Data Cleaning and Transformation

Applies imputation, normalisation, encoding, and deduplication to prepare raw data. Directly enables reliable model training in subsequent chapters.

Lesson 2 • Data Sources and Collection Strategies

Surveys sensor networks, APIs, databases, and synthetic generation as data origins. Establishes criteria for selecting appropriate sources for engineering tasks.

Lesson 3 • Feature Engineering and Selection

Creates informative features from raw variables and removes redundant ones. Improves model performance and interpretability across all AI engineering tasks.

Lesson 4 • Data Pipeline Design and Automation

Designs reproducible, automated pipelines from ingestion to model-ready datasets. Introduces versioning and monitoring to maintain pipeline reliability over time.

Lesson 1 • Reinforcement Learning in Engineering

Introduces agents, environments, rewards, and policy optimisation for control and scheduling tasks. Extends supervised learning concepts to sequential decision-making.

Lesson 2 • Unsupervised Learning Techniques

Applies clustering, dimensionality reduction, and anomaly detection to unlabelled engineering data. Enables pattern discovery in sensor streams and operational logs.

Lesson 3 • Model Evaluation and Validation

Applies cross-validation, performance metrics, and calibration to assess model reliability. Connects statistical inference from Chapter 2 to practical model assessment.

Lesson 4 • Supervised Learning Fundamentals

Covers regression and classification algorithms with engineering use cases. Establishes the train-validate-test workflow used throughout the course.

Lesson 1 • Generative Models in Engineering Design

Introduces GANs and variational autoencoders for synthetic data generation and design exploration. Expands the design space available to AI-augmented engineers.

Lesson 2 • Backpropagation and Training Dynamics

Derives backpropagation and connects it to the calculus covered in Chapter 2. Addresses vanishing gradients, learning rate schedules, and batch strategies.

Lesson 3 • Neural Network Fundamentals

Explains perceptrons, activation functions, and forward propagation as building blocks. Provides the conceptual base for all deep architectures covered in this chapter.

Lesson 4 • Convolutional Neural Networks for Engineering

Applies CNNs to image-based inspection, fault detection, and spatial sensor data. Covers convolution, pooling, and transfer learning for engineering contexts.

Lesson 1 • Evolutionary and Swarm Optimisation

Applies genetic algorithms, particle swarm, and differential evolution to multi-objective design. Handles non-convex, discontinuous engineering search spaces.

Lesson 2 • Topology and Structural Optimisation

Uses density-based and level-set methods to optimise material distribution under constraints. Produces lightweight, high-performance structural designs.

Lesson 3 • Constraint Handling and Feasibility

Addresses hard and soft constraints in AI-driven optimisation to ensure physically realisable designs. Connects optimisation theory to practical engineering requirements.

Lesson 4 • Surrogate Modelling and Metamodelling

Replaces expensive simulations with fast surrogate models built from sampled data. Reduces computational cost while preserving design accuracy.

Lesson 1 • Condition Monitoring Fundamentals

Defines health indicators, degradation modes, and monitoring architectures for physical assets. Establishes the domain context for all predictive maintenance methods.

Lesson 2 • Maintenance Strategy and Decision Support

Translates model outputs into maintenance schedules, cost-benefit analyses, and risk rankings. Connects technical predictions to operational and business decisions.

Lesson 3 • Signal Processing for Fault Detection

Applies Fourier transforms, wavelets, and statistical features to extract fault signatures from raw signals. Bridges raw sensor data to model-ready feature vectors.

Lesson 4 • Anomaly Detection and Fault Diagnosis

Deploys isolation forests, autoencoders, and classification models to identify and classify faults. Extends unsupervised and supervised learning from Chapter 4 to maintenance contexts.

Lesson 1 • Production Monitoring and Drift Detection

Implements logging, alerting, and statistical drift tests to maintain model performance over time. Extends data pipeline monitoring from Chapter 3 to live model behaviour.

Lesson 2 • Scalability and Performance Engineering

Applies distributed computing, model compression, and hardware acceleration to scale AI systems. Ensures models meet throughput and latency targets in production.

Lesson 3 • Model Deployment Architectures

Compares batch, real-time, and edge deployment patterns for engineering environments. Guides architecture selection based on latency, resource, and reliability requirements.

Lesson 4 • Continuous Integration and Delivery for AI

Applies CI/CD pipelines, automated testing, and rollback strategies to AI engineering workflows. Enables rapid, safe iteration on deployed models.