This module provides an in-depth exploration of the Industrial Internet of Things (IIoT) and Cyber Physical Systems (sensors, control boards) to enable intelligent industrial systems. Students will work with real-world industrial sensors, IIoT gateways, and edge computing devices to collect, process, and analyse manufacturing data for predictive maintenance, quality control, and process optimisation using data analytics & machine learning.

1. Evaluate the role and impact of IIoT and CPS in modern manufacturing environments.
2. Appraise best practice for real time data acquisition systems in manufacturing environments.
3. Develop and deploy edge and cloud-based IIoT architectures, including communication protocols (e.g., MQTT, OPC UA, 5G) to support connectivity, data flow, and system integration.
4. Propose solutions to integrate Cloud & Edge AI solutions to enhance decision-making at the production level.
5. Analyse and visualise industrial process data using time-series analysis and dashboards to inform real-time decision-making.
6. Select and apply Machine Learning Models on edge devices to optimise manufacturing operations.

This module provides a comprehensive introduction to machine learning (ML) with a strong emphasis on deep neural networks, tailored specifically for advanced manufacturing applications. Using PyTorch as the primary framework, students will learn to develop, train, and deploy ML models for real-world industrial challenges such as predictive maintenance, quality control, and process optimisation. The module covers: Core ML techniques (supervised/unsupervised learning) for manufacturing data.

Deep learning fundamentals, including CNNs for visual inspection and RNNs for time-series forecasting.

Cutting-edge architectures like Transformers (for sequence/vision tasks) and Generative AI (VAEs, GANs) for synthetic data generation.

Deployment strategies (ONNX, edge AI) and ethical considerations (bias, explainability).

Hands-on labs and a project will enable students to apply PyTorch-based solutions to manufacturing datasets (e.g., defect detection, tool wear prediction). By the end, students will be equipped to integrate ML into Industry 4.0 systems.

1. Analyse manufacturing datasets and select appropriate machine learning and deep learning techniques.
2. Implement PyTorch-based models (CNNs, Transformers, GANs) for industrial use cases.
3. Evaluate model performance, addressing challenges with limited or imbalanced data.
4. Deploy lightweight machine learning models for edge/real-time inference.
5. Critically assess the ethical implications of AI in manufacturing.

This module introduces students to virtual manufacturing and the creation of digital twins to simulate, optimise and enhance manufacturing processes. By leveraging advanced digital tools and techniques, students will gain hands-on experience in developing process designs, simulating process behaviour and optimising process workflow. Also, students will learn how to apply Lean principles to evaluate process performance, identify inefficiencies and propose process solutions for continuous improvement.

1. Use CAD tools to model realistic products for integration with the digital simulations.
2. Create digital twins of production lines, machines and robots to simulate their behaviours and optimise production.
3. Utilise virtual simulation tools to optimise product designs, manufacturing processes, efficiency and commissioning costs.
4. Apply process improvement techniques to analyse process performance, identify bottlenecks and reduce waste.
5. Evaluate the performance of virtual models against real-time data from physical systems.
6. Evaluate the effectiveness of digital twin integration in improving manufacturing efficiency and costs.