ARTIFICIAL INTELLIGENCE CONCEPTS AND METHODS - 2027/8

Module Overview

Module Overview:

How can AI identify a failure mechanism before it occurs, or optimise a complex design across thousands of variables? This module demystifies the technical "engine" of Artificial Intelligence specifically for engineers.

You will explore the current landscape of AI, from established machine learning techniques to the latest Generative Design and Physics-Informed models. You will learn to navigate the diverse world of AI methods, identifying which algorithm is best suited to solve technical challenges¿whether you are analysing stress-test data or simulating structural performance. This module provides the theoretical foundation and the architectural mindset needed to select the right tool for the right engineering problem.

Please note that the information detailed within this record is accurate at the time of publishing and may be subject to change. This record contains information for the most up to date version of the programme / module for the 2027/8 academic year.

Module provider

SOL - School of Engineering

Module Leader

HAGEN-ZANKER Alex (Sch of Eng)

Number of Credits: 15

ECTS Credits: 7.5

Framework: FHEQ Level 7

Module cap (Maximum number of students): N/A

Module Availability

Semester 1

Prerequisites / Co-requisites

N/A

Module content

Module content:

This module establishes the theoretical framework for AI integration in engineering, focusing on the mechanics of algorithms and the nature of digital learning.

1. Mechanisms of Artificial Intelligence

• Algorithmic Architectures: Deep dive into Neural Networks (CNNs, RNNs, Transformers) and their specific applications in engineering domains.
• Generative & Physics-Informed Models: Examining the mechanics of GANs and the integration of physical laws into neural network loss functions.

2. The Logic of Machine Learning

• Inductive vs. Deductive Learning: Analysing how models acquire "knowledge"¿moving from pure statistical correlation to physics-constrained reasoning.
• Knowledge Representation: Understanding how engineering constraints are encoded within a model and the implications for result fidelity.

3. Training & Operational Constraints

• Data & Domain Dynamics: Evaluating the impact of data scarcity, noise, and domain-specific requirements on model training and utility.
• Architectural Trade-offs: Balancing computational overhead, training time, and model transparency against the need for predictive accuracy.

Assessment pattern

Alternative Assessment

None

Assessment Strategy

Assessment strategy:The assessment strategy is designed to:

• Demonstrate, in first instance, the technical understanding of the modern AI ecosystem and the mechanics of generative AI and physics-informed models.
• Secondly, demonstrate higher level skills using in the design of AI architecture. Including aspects of model utility, data scarcity, and managing the risks associated with interpretability and safety in a technical context.

• Mid-Semester Technical Test: A structured assessment focusing on the technical mechanics and foundational logic of AI methodologies. This ensures a rigorous grasp of the fundamental algorithms and the epistemological differences between data-driven and physics-informed paradigms. (LO1, LO2)
• Final Examination: A comprehensive written exam requiring critical synthesis and professional judgment. Students will be presented with complex engineering scenarios and must justify specific AI architecture. (LO3, LO4, LO5)

• Online Concept Quizzes: Automated "knowledge checks" on the specific mechanisms of AI algorithms (e.g., neural network layers, loss functions) to provide immediate feedback on technical understanding.
• Seminar "Design Challenges": Collaborative discussions where students are given an engineering problem and must argue for a specific AI architecture. Feedback focuses on the logic of the justification and the ability to identify potential "knowledge" gaps in the model.
• Mock Scenario Analysis: Practice exam questions that provide students with a specific engineering challenge that require an AI solution approach.

Module aims

• Equipping engineers with the technical literacy to evaluate the modern AI ecosystem and select optimal tools for complex technical projects.
• Demystifying the mechanics of how machines acquire knowledge from physical and digital data to predict engineering outcomes.
• Examining high-impact developments like Generative AI and Physics-Informed models to understand how they accelerate the engineering design cycle.
• Developing the critical judgment necessary to justify the use of specific AI architectures based on data availability, physical constraints, and safety requirements.

Learning outcomes

Methods of Teaching / Learning

Methods of teaching/learning:

The learning and teaching strategy is designed to:

• Provide sufficient technical depth about the mechanics of AI methods and tools to critical evaluate their use and limitations
• Develop a system level appreciation of the nature of knowledge and knowledge generation in an engineering context.
• Nurture an ability to critically interrogate engineering challenges and draw links with AI abilities as well as limitations.

The learning and teaching methods include:

• Technical Lectures: These sessions establish the mathematical and conceptual framework of AI engines, focusing on the strengths and limitations of different model architectures.
• Applied Seminars: Collaborative sessions focused on "Model Selection" exercises, where specific engineering problems (e.g., jet engine maintenance or bridge stress analysis) are matched with optimal AI methodologies.
• Independent Technical Research: Targeted study into the latest research papers and software documentation, fostering the ability to stay current in the rapidly evolving AI field.

Indicated Lecture Hours (which may also include seminars, tutorials, workshops and other contact time) are approximate and may include in-class tests where one or more of these are an assessment on the module. In-class tests are scheduled/organised separately to taught content and will be published on to student personal timetables, where they apply to taken modules, as soon as they are finalised by central administration. This will usually be after the initial publication of the teaching timetable for the relevant semester.

Reading list

https://readinglists.surrey.ac.uk
Upon accessing the reading list, please search for the module using the module code: ENGM328

Other information

The School of Engineering is committed to developing graduates with strengths in Employability, Digital Capabilities, Global and Cultural Capabilities, Sustainability, and Resourcefulness and Resilience. This module is designed to allow students to develop knowledge, skills, and capabilities in the following areas:

Employability: This module prepares graduates for specialist roles in AI-driven engineering sectors by focusing on architectural decision-making. Rather than just performing technical tasks, students develop the ability to consult on the selection and justification of AI tools, bridging the gap between data science and professional engineering practice.

Digital Capabilities: Students move beyond basic software literacy to master the fundamental underpinnings of digital technology. This includes a deep understanding of algorithmic structures, the mechanics of machine learning, and the integration of physical laws into digital models, fostering a high level of technical fluency.

Global and Cultural Capabilities: The curriculum addresses the global nature of AI development, examining how algorithmic bias and transparency impact diverse societies. By evaluating the societal implications of "black box" systems, students learn to implement AI solutions that are culturally responsible and globally applicable.

Sustainability: Sustainability is addressed through the lens of computational efficiency. Students learn to evaluate the resource-intensive nature of AI training and are encouraged to promote "Green AI" practices¿minimizing energy consumption by selecting the most efficient models and data resources for a given task.

Resourcefulness and Resilience: This module develops the ability to navigate technical uncertainty. By exploring the limitations and failure modes of AI tools, students build the resilience to adapt their strategies when data is scarce or models underperform. This fosters a resourceful approach to complex problem-solving in unpredictable engineering environments.

Programmes this module appears in