Module 6 · Lesson 6.3
The model-purpose statement
Five lines that make a model reviewable — and the order they matter in.
Why this matters
Everything in this module has been a decision: what to discard, which level, what the model is for. Those decisions live in the modeller's head, and a model whose decisions live in one person's head cannot be checked by anybody else.
The model-purpose statement is five lines that move them onto paper. It is the cheapest piece of assurance in the whole course, and it is the thing that makes Module 13's review process possible at all.
By the end of this lesson you should be able to
- Write the five fields of a model-purpose statement
- Say why the question and the invalidation clause come first
- Recognise a statement that is not yet reviewable
- Connect the statement to the model record of Module 13
The five fields
1. The question. What this model exists to answer, in one sentence. Without it the model has no subject, and any result it produces answers something nobody wrote down.
2. What would make it wrong. The behaviour, assumption or piece of information that, if it turned out otherwise, would invalidate the model. This is the field people leave out and it is the one a reviewer needs most.
3. How the result will be checked. Named before the result exists, so the check is not chosen to agree with the answer.
4. What is deliberately excluded. The behaviours discarded on purpose — shear deformation, composite action, connection flexibility, soil stiffness. Writing them down converts an omission into a decision.
5. What it will be used to decide. Which keeps the model proportionate to the consequence.
Why that order
The order is not arbitrary and it is not the order they are usually written in.
A statement with fields 3, 4 and 5 but no question describes a model of nothing in particular. It can be inspected but not reviewed, because there is no standard to review it against.
A statement with a question but no invalidation clause can be reviewed for internal consistency and not for correctness. A reviewer can check the arithmetic; they cannot check the judgement, because nothing has been offered as falsifiable.
That second point is the connection to Module 13, arriving early: a model that nothing could disprove is not being used to find anything out.
An example, worked
A floor beam in a six-storey office:
Question: How much does the secondary floor beam on grid C deflect under the frequent imposed load?
What would make it wrong: Any change that makes the slab act compositely with the beam, which would roughly double the stiffness and make this model substantially conservative.
Check: 5wL⁴/384EI by hand for one bay, before reading the model.
Excluded: Composite action, connection flexibility, and construction-stage loading.
Decision: Whether a 450 mm beam fits the agreed floor zone.
Five lines. A reviewer who has never seen the model now knows what it is for, what to check, what it left out, and what would make it wrong. None of that is available from the model file itself at any level of detail.
When the statement is hard to write
If the question will not go into one sentence, the model is usually doing two jobs. That is worth knowing early, because a model doing two jobs tends to do the second one badly and silently — the settings that suit the first job stay in place.
And if the invalidation clause cannot be written at all, that is the most useful signal in the whole process: it means the model is being built because building it is the process, rather than because a specific question needs answering.
Try it
Model purpose statement
Five fields. Fill them in and see what a reviewer could and could not do with what you have written.
One sentence. Without it the model has no subject.
The assumption that, if it turned out otherwise, would overturn the result.
Named before the result exists, so it is not chosen to agree.
Comma-separated. Writing an omission down converts it into a decision.
Keeps the model proportionate to the consequence.
- Fields present
- 0 of 5
- Reviewable?
- no
- Next gap to close
- question
Not reviewable, and not yet a model of anything: no question has been stated, so any result it produces answers something nobody wrote down.
Still missing, in the order they matter
- The question this model exists to answer
- What would make this model wrong
- How the result will be checked
- What is deliberately excluded
- What it will be used to decide
What this shows: The gaps are reported in the order they matter: without a question there is no subject, and without an invalidation clause there is nothing to review.
Worked example
Reviewing an incomplete statement
Given
- A model-purpose statement gives the question, the decision, the exclusions and the check
- It does not say what would make the model wrong
Find
What a reviewer can and cannot do
Practice
A model-purpose statement has the question, the check and the exclusions filled in. How many of the five fields are missing?
Practice
A line model at L/d = 3 omits 7.4 % of the deflection. If the deflection limit is span/250 and the model reports span/260, by what percentage is the model under the limit, and does the omitted shear term change the conclusion?
Check yourself
Which field of a model-purpose statement does a reviewer need most, and why?
Summary
- Five fields: question, what would make it wrong, check, exclusions, decision
- The question comes first — without it the model has no subject
- The invalidation clause comes second, and it is the one usually missing
- A model that nothing could disprove is not being used to find anything out
- If the question needs two sentences, the model is doing two jobs
- An exclusion is safe only while the margin exceeds the thing excluded
This is educational material. It uses simplified examples to teach principles, and must not be relied on for real design or safety-critical decisions. Module overview and checkpoint