PRIMARY 5 MATHEMATICS LEARNING GUIDE · BATCH 8 · GUIDE 29
Mathematical modelling is the controlled journey from a real situation into mathematics and back again. A learner first decides what quantities matter, which details can be ignored, what assumptions are being made and which mathematical relationships represent the situation. After calculation, the answer must return to the original context and be checked against physical, logical and practical constraints.
This guide is not about making Primary 5 Mathematics unnecessarily abstract. It is about making word problems more honest. Every word problem already contains a model. The question is whether the learner notices it.
For the official curriculum framework and its emphasis on applications, modelling, reasoning, communication and metacognition, see the MOE Primary Mathematics Syllabus.
Series route: return to the Primary 5 Mathematics Learning Hub. Companion guides: Multiple Solution Routes · Metacognition, Self-Monitoring & Strategy Control · Retrieval, Spacing, Interleaving, Variation & High-Quality Practice Design.
1. A real situation contains more information than the mathematics needs
Imagine a school buying water bottles for 186 students. A bus leaves at 8:30 a.m., the weather is warm, the bottles are blue, and each carton contains 24 bottles. If the question asks how many cartons are needed, the colour and departure time are irrelevant. The useful quantities are 186 students and 24 bottles per carton.
Modelling begins by separating mathematical signal from contextual noise.
2. Define the target before building the model
For the bottle problem, the target is number of cartons required. That target determines the useful relationship:
186 ÷ 24 = 7 remainder 18.
Because cartons are whole objects and 7 cartons hold only 168 bottles, 8 cartons are required.
The model includes a capacity constraint, not ordinary rounding.
3. Identify quantities and units
A model is stronger when every quantity is named:
- students: 186 students;
- carton capacity: 24 bottles/carton;
- required cartons: unknown cartons.
Units expose the intended division: students divided by students-per-carton would be strange because the rate is bottles/carton. The situation also implies one bottle per student, an assumption that should be recognised.
4. State necessary assumptions
The bottle model assumes each student receives one bottle, cartons contain exactly 24 usable bottles, and partial cartons cannot be ordered.
If any assumption changes, the answer may change. If five extra bottles are required for teachers, the model must include 191 bottles instead of 186.
Assumptions are not weaknesses. Hidden assumptions are the weakness.
5. Choose the simplest useful representation
A model may use a bar model, equation, table, timeline, diagram or labelled expression.
For a constant-rate problem, a table may be best. For percentage, a 100%-whole statement may be best. For geometry, a diagram may be essential.
The representation should reduce uncertainty, not merely decorate the solution.
6. Model the relationship, not the story words
A shop sells 35% of 480 bottles. The model is not “shop → sell → bottles”. It is:
whole = 480; sold = 35% of whole.
Sold = 168, remaining = 312.
The mathematical model strips the story down to the controlling relationships.
7. Keep state changes explicit
If 20% of the remaining 312 bottles is then sold, the model must update:
new whole = 312.
Second sale = 62.4 bottles in pure arithmetic, but if bottles are indivisible the context may require different interpretation depending on the original problem design.
The physical object can constrain the mathematical result.
8. Models can be exact or approximate
If a problem states a constant rate of 18 ℓ/min, the model can be exact within the problem. In real life, a pump might fluctuate.
School mathematics often idealises reality deliberately. The learner should know the difference between “exact under the model” and “exact in the physical world”.
9. A model can fail outside its assumptions
Suppose a taxi model is “cost = $4 + $2 per kilometre”. If the real system includes waiting charges or different distance bands, the simple model is incomplete.
Do not extend a model beyond the conditions that justify it.
10. Rate models require constant rate
If a machine makes 30 items/min for 8 minutes, total = 240 items only if the rate remains 30 throughout the interval.
If the machine slows after four minutes, the single-rate model no longer represents the process.
Always ask what must stay constant for the multiplication to remain valid.
11. Percentage models require a reference whole
If a 20% discount is applied to $250, the original $250 is 100%. If a further 10% discount is applied to the reduced price, the new price becomes the new 100% for that second stage.
Model failure often begins when the reference whole is not updated.
12. Geometry models require valid dimensions
A triangle-area model requires a base and its perpendicular height. A slanted side is not automatically a height.
The formula works only when its variables match the geometric definition.
Formulas are models with conditions.
13. Tank models distinguish contents from capacity
A rectangular tank 40 cm × 25 cm × 30 cm has full capacity 30,000 cm³. If water depth is only 18 cm, current contents are 18,000 cm³.
The same physical tank supports two different models depending on the target: current water or maximum capacity.
14. Data models depend on what was measured
A line graph connecting hourly readings does not automatically prove the quantity changed linearly between readings.
The graph is a representation of recorded data. Interpolation is an additional modelling assumption unless the context guarantees it.
15. Simplify without changing the structure
If a complex percentage problem is confusing, replace 480 with 100 temporarily while preserving the sequence. Learn the structure using friendly numbers, then return to the original values.
This is a model of the model: a simpler case retaining the same mathematical skeleton.
16. Use bounds to test a model output
If a tank holds at most 30 ℓ, a computed final volume of 42 ℓ cannot remain in the tank without overflow.
If a percentage below 100% produces a part larger than the whole, the result violates a basic bound.
Constraints are model checks.
17. Return the answer to the original question
If the arithmetic gives 7.75 buses, the model output must be interpreted. Buses are indivisible and capacity must be sufficient, so the answer is 8 buses.
Mathematical output becomes a contextual answer only after interpretation.
18. A correct number can be a wrong decision
If a discount amount is $36 and the question asks for sale price, reporting $36 is not a correct solution even though the calculation itself was correct.
Models include the target quantity, not only the arithmetic relation.
19. Model validation can use an independent route
Suppose 25% of 320 = 80. Check using fraction equivalence: 25% = 1/4, and 320 ÷ 4 = 80.
Two routes agreeing does not prove every assumption, but it increases confidence that the arithmetic relationship was represented correctly.
20. Sensitivity: ask what happens if one input changes
If carton capacity changes from 24 to 30 bottles, the required number of cartons for 186 bottles changes from 8 to 7.
Testing changed inputs reveals which variables control the output and whether the model behaves sensibly.
21. Model comparison
A rate problem might be represented with a table or equation. A comparison problem might use bars or an equation.
Different models can represent the same relationship. The best model is usually the one that makes the uncertain feature most visible.
22. Model choice should be economical
Do not draw a full bar model for 84 − 27 unless the comparison relationship itself is under study. Do draw one when several changing groups and reference quantities make the structure difficult to hold mentally.
Good modelling reduces cost.
23. Assumptions can be tested by counterexample
If a learner assumes “more expensive pack means worse value”, find two packs with different quantities where the larger pack has lower unit cost. The counterexample exposes the weakness of comparing total price alone.
Models should be tested against cases that could break them.
24. Error map
| Visible failure | Likely modelling issue | Repair question |
|---|---|---|
| All numbers used automatically | Relevant quantities not selected | Which details actually affect the target? |
| Second percentage applied to original | State model not updated | What is 100% at this stage? |
| Fractional bus accepted | Physical constraint ignored | Can the object be divided? |
| Rate multiplied across changing process | Constant-rate assumption false | Does the rate stay constant? |
| Answer numerically correct but wrong quantity | Target not returned to | What did the question actually ask? |
25. Practice laboratory
- 186 students each need one bottle. Cartons hold 24. How many cartons are required, and what assumption makes the division meaningful?
- A pump fills at 18 ℓ/min for 12 min. State the key modelling assumption and find volume.
- A $240 item receives 15% discount. Identify the 100% quantity and find sale price.
- A tank is 40 × 25 × 30 cm with water depth 18 cm. Find current water volume and full capacity.
- A graph records values at 9 a.m. and 10 a.m. Explain why a straight joining line does not automatically prove constant change between times.
- A bus calculation gives 4.2 buses. Explain why 5 may be correct in context.
- A model predicts 42 ℓ in a 30 ℓ tank. What model check fails?
- Explain why 6 items for $15 and 10 items for $24 should be compared using unit cost rather than pack price alone.
26. Model answers
1. 186 ÷ 24 = 7 remainder 18, so 8 cartons. Assumption: one usable bottle per student and cartons are ordered whole.
2. Constant rate assumption; 18 × 12 = 216 ℓ.
3. Original $240 = 100%; sale price = 85% × 240 = $204.
4. Current = 40 × 25 × 18 = 18,000 cm³; capacity = 40 × 25 × 30 = 30,000 cm³.
5. Only endpoint measurements are guaranteed; continuous linear behaviour is an added assumption.
6. Buses are whole objects and four buses may not hold everyone; capacity requires rounding up.
7. Physical capacity bound fails; overflow must occur or the model/input is wrong.
8. Pack sizes differ; unit costs $2.50 and $2.40 per item create a fair comparison.
27. Full modelling problem
A school excursion has 214 students and 11 teachers. Each bus holds at most 45 people. The organiser wants one empty seat on every bus for equipment.
Total people = 225. Effective passenger capacity per bus = 44 because one seat is reserved.
225 ÷ 44 = 5 remainder 5, so 6 buses are required.
Check: 5 buses provide only 220 passenger places after reservations, which is insufficient. Six provide 264 passenger places.
The modelling decision was not merely 225 ÷ 45. The equipment-seat condition changed effective capacity.
28. Final checkpoint
A strong Primary 5 modeller can identify relevant quantities, state assumptions, choose an economical representation, preserve changing states and units, respect physical constraints, test the output against bounds and return the result to the original context.
Continue to Primary 5 Mathematics Learning Guide | Multiple Solution Routes: Comparing Methods, Efficiency & Verification.
Editorial approach: Wintour House V1.0 · CivDJ · eduKate Publishing. Extract the minimum mathematical world, declare its assumptions, solve inside it, stress-test its limits, and return only an answer that still makes sense in the real situation.
