Computational Engineering / AI and machine learning for structural engineers
What it is actually doing
Fitting a function to samples — and what that implies about every claim made for it.
Lesson review: expert verification remains open
This lesson has received a full-context editorial review and a check of its technical explanations. Selected arithmetic has been checked separately. This is not independent engineering sign-off.
Checks needed for project use
- A computational engineering reviewer must check algorithms, convergence claims, statistical assumptions and numerical examples beyond the cases explicitly recorded.
- Reproduce seeded learning experiments and label model-selection data as validation, retaining an independent test procedure.
- Obtain qualified review for legal responsibility statements; this is not a legal opinion.
- Review domain guards and calibration before any real engineering use.
These are educational models and exercises. Check their assumptions before using the results. The full technical content and every diagram have not yet been independently verified.
Concrete and steel calculations retain their stated first-generation Eurocode teaching basis. Use the standards and amendments specified for your project; do not mix generations without an appropriate design basis.
Read the lesson register and outstanding review notesWhy this matters
This module is written to be useful in five years, which rules out describing what any particular system can currently do. Capability claims date faster than a page can be maintained, and the course's currency register says so explicitly.
What does not date is the mechanism. A trained model fits a function to samples, and every strength and every failure mode follows from that one sentence — including the ones that will still be true of whatever replaces today's methods.
By the end of this lesson you should be able to
- State what a supervised model is doing in one sentence
- Distinguish supervised, unsupervised and reinforcement learning by what each requires
- Explain features and labels and why choosing features is engineering
- Say why correlation is what these methods find
What you should already know
- Curve fitting and least squares
- The distinction between a model and the thing it models (Module 1)
Learning from examples
Supervised learning uses labelled examples to fit a predictive relationship. Regression predicts a numerical value; classification predicts a class or class probabilities. Labels may come from measurements, simulations, records or expert assessment. Unsupervised methods explore structure in unlabelled data. Reinforcement learning learns from actions and rewards, including simulated environments; it is not excluded from engineering simply because physical trials are costly.
Features must be available at the time of prediction and appropriate to the task. A utilisation ratio leaks the answer into a model intended to predict pass/fail before that utilisation is calculated. Inspect data provenance and splitting as well as the feature names.
Ordinary predictive fitting does not by itself establish causation. Physical equations, constraints and domain knowledge can nevertheless inform learning methods. Where a direct calculation is practical and suitable, it is generally preferable for that calculation; a validated surrogate may be useful when repeated full analyses are expensive.
Uses and limits
Potential uses include preliminary recommendations, classification, anomaly detection, monitoring interpretation, retrieval and surrogate modelling. Each needs a defined role, representative evaluation and a response to errors or unfamiliar inputs. A task is not automatically low risk because a human receives the output.
For structural decisions, establish what checks remain necessary and who performs them. A learned approximation to a force or capacity calculation may help explore options, but its error and applicability must be assessed and final decisions checked through the agreed engineering process.
Check yourself
A model is trained to predict whether a steel beam passes its design checks, from features including span, load, section properties and the computed utilisation ratio. It scores 99.8 % accuracy. What is wrong?
Practice
A classifier is tested on 240 cases and gets 217 right. What is its accuracy, as a percentage to one decimal place?
Worked example
Reading a confusion matrix nobody has explained
Given
- A classifier sorts 500 model-audit cases into 'has a restraint error' and 'does not'
- Of the 100 cases that really do have a restraint error, it finds 72
- It also flags 40 cases that do not have one
Find
Accuracy, precision and recall — and which of the three is misleading here
Practice
A classifier flags 112 cases as having a defect. Of those, 72 really do. What is its precision, as a percentage?
Practice
The same classifier finds 72 of the 100 cases that really have the defect. How many defective models does it let through?
Practice
A classifier is applied to a population where only 2 % of models have the defect. It flags nothing at all. What accuracy does it report?
Check yourself
A vendor reports 94 % accuracy for a defect classifier. What is the first question to ask?
Check yourself
What does a classifier's 'confidence' of 0.94 actually describe?
Predict first
A classifier trained on steel frames is shown a timber frame. What does it do?
Summary
- Supervised learning fits a function to examples so it can be evaluated somewhere new
- Supervised needs labels, which usually means an engineer looking at each case
- Choosing features is where domain knowledge earns its place
- A feature downstream of the label produces a brilliant, worthless model
- These methods find correlation; where a deterministic method exists, use it