How do I tell the objective function from the constraint in an optimisation problem?

The objective function represents the quantity the question asks you to maximise or minimise. The constraint represents a condition that every possible solution must satisfy, such as a fixed perimeter or limited amount of material. Words such as “maximum area” identify the objective, while phrases such as “using 100 metres of fencing” usually identify the constraint. Write the two equations separately before substituting anything. If the objective contains two variables, rearrange the constraint and use it to eliminate one of them.

Why must an optimisation function usually be written in one variable?

Standard derivative optimisation in IB Maths AA normally requires a function of one independent variable. An expression such as $A=xy$ does not by itself account for how $x$ and $y$ are connected. The constraint supplies that connection, allowing one variable to be expressed in terms of the other. Substitution then creates a function such as $A(x)$ that can be differentiated using familiar one-variable methods. HL problems may use implicit differentiation in more advanced settings, but the relationship between the variables must still be incorporated.

Does setting the first derivative equal to zero always find the maximum or minimum?

No, solving $f'(x)=0$ only identifies stationary points within the differentiable part of the domain. A stationary point may be a maximum, a minimum or a stationary point of inflexion. You must classify it using a second derivative test, a first derivative sign test or comparison of objective values. An absolute optimum may also occur at an endpoint where the derivative is not zero. Therefore, always consider the domain and compare every relevant candidate.

Should I check endpoints in IB Maths optimisation questions?

You should check endpoints whenever the objective is being optimised over a closed or contextually bounded interval. An endpoint can produce an absolute maximum or minimum even though it does not satisfy $f'(x)=0$. In geometric questions, endpoints may represent degenerate objects with zero area or volume, but those values still help establish the global result. If the question or context excludes an endpoint, state that restriction clearly. Comparing endpoint and interior candidate values is often the safest justification for an absolute optimum.

What should I do if I cannot form the optimisation equation in an exam?

First, underline the phrase stating what must be largest, smallest, cheapest or fastest because this identifies the objective. Define variables on the diagram and write the standard formula for the target quantity, even if it initially contains two variables. Next, list every fixed number or required relationship to locate the constraint. Try rearranging that constraint for the variable that is easiest to substitute into the objective. Even if you cannot finish, a correct objective, constraint and substitution may demonstrate useful method and create a path to later parts.

How can I improve at derivative optimisation problems without memorising solutions?

Practise the setup separately from the calculus by writing only the variables, objective, constraint, domain and one-variable function for several questions. Use varied contexts such as fencing, containers, distance, revenue and surface area so that you learn the shared structure rather than one diagram. After checking a solution, record the first decision that went wrong instead of writing only “optimisation error.” Reattempt a parallel problem a few days later without notes or prompts. Once the setup becomes reliable, combine it with differentiation, candidate testing and concise contextual conclusions under timed conditions.

Back to all posts

Why Is Optimisation with Derivatives So Hard to Start?

Emma C.

• 30 Jul 2026•13

Why Is Optimisation with Derivatives So Hard to Start?

Join 650k+ students breezing through the IBReady to breeze through the IB?

Why Is Optimisation with Derivatives So Hard to… | RevisionDojo

Back to all posts

Why Is Optimisation with Derivatives So Hard to Start?

Emma C.

• 30 Jul 2026•13

Optimisation with derivatives is hard to start because the first task is not differentiation. It is mathematical modelling: deciding what must be maximised or minimised, identifying the information that restricts the variables, and converting both into equations.

Once the correct one-variable function exists, the calculus is often routine. The real difficulty in optimisation IB Maths questions is constructing that function from an unfamiliar context. This explainer concentrates on that setup stage rather than duplicating the broader coverage in IB Maths AA calculus explained through notes and examples.

What an optimisation problem is really asking

An optimisation problem asks for the greatest or least possible value of a quantity under specified conditions. Typical tasks include maximising area, volume, revenue or profit, or minimising distance, surface area, time or cost.

Two mathematical objects control the entire solution:

The objective function represents the quantity to be maximised or minimised.
The constraint represents a fixed condition or relationship that limits the possible choices.

Suppose a rectangle has a perimeter of 40 cm and its area must be maximised. Its area is the objective:

$A=xy.$

Its fixed perimeter gives the constraint:

$2x+2y=40.$

The derivative cannot yet be used effectively because $A$ contains two variables. The constraint must first be rearranged, for example as $y=20-x$ , and substituted into the objective:

$A(x)=x(20-x)=20x-x^2.$

Only now is there a one-variable function to differentiate. This sequence explains why students who are confident at finding derivatives can still struggle with derivative optimisation problems.

Why setting up the problem is harder than solving it

Differentiation is procedural. Once you see $A(x)=20x-x^2$ , you know how to calculate $A'(x)$ and solve . The opening of an optimisation question is less predictable because the required equations may be hidden in prose, a diagram or several pieces of information.

The setup combines several skills at once:

SkillQuestion the student must answerCommon difficultyInterpretationWhat does “best” mean here?Optimising the wrong quantityVariable choiceWhich dimensions or quantities can change?Using too many variables or inconsistent symbolsFormula recallWhat equation represents area, volume, cost or distance?Choosing a formula that does not match the objectConstraint recognitionWhat information is fixed?Treating the constraint as the objectiveAlgebraHow can one variable be eliminated?Substituting incorrectly or rearranging inefficientlyDomain reasoningWhich values are physically possible?Accepting negative lengths or ignoring endpointsCalculusWhere are the candidates for an optimum?Assuming every solution of $f'(x)=0$ is the answerInterpretationWhat quantity did the question request?Reporting $x$ when the question asks for maximum area

The official IB Mathematics: Analysis and Approaches guide describes calculus as a way to model and analyse change, and explicitly includes maximum and minimum points and optimisation. The assessment objectives also require students to transform realistic contexts into mathematics and interpret conclusions, so modelling is not an optional preliminary step. It is part of what the question assesses.

In the current course structure, testing for maxima and minima and optimisation appears in AA SL 5.8, while AA HL students can also encounter optimisation involving implicit relationships in AHL 5.14. RevisionDojo’s AA SL 5.8 optimisation collection and AHL 5.14 practice area separate these levels of demand.

How to identify the objective function

Start by completing this sentence:

I need to make __________ as large or as small as possible.

The missing quantity is the objective. Words such as maximum, minimum, greatest, least, largest, shortest, cheapest and most profitable usually point towards it.

Examples include:

“Find the dimensions giving the maximum volume.” The objective is volume.
“Determine the minimum cost.” The objective is cost.
“Find the point on the curve closest to $P$ .” The objective is distance, or more conveniently distance squared.
“Determine the production level that maximises profit.” The objective is profit, where $P=R-C$ .

Do not assume that the most visible number in the question belongs to the objective. If a box has a fixed surface area and must have maximum volume, surface area is the constraint and volume is the objective.

A reliable annotation is to write OBJ beside the phrase naming what must be optimised. Then write its formula before doing any calculus. At this stage it is acceptable for the objective to contain two variables, such as $V=\\pi r^2h$ .

How to identify the constraint

The constraint answers a different sentence:

What relationship must remain true for every allowed solution?

Look for fixed totals, geometric relationships, equations of curves and practical restrictions. Common signals include:

a fixed perimeter, area, volume or amount of material
a limited length of fencing or wire
a point that must lie on a stated curve
a fixed total cost or production capacity
dimensions connected by similarity, Pythagoras or trigonometry
lower and upper bounds on a variable

In a cylinder problem, $V=\\pi r^2h$ might be the objective while $2\\pi r^2+2\\pi rh=600$ is the surface-area constraint. In a closest-point problem, the distance formula supplies the objective while the equation of the curve constrains the point’s coordinates.

Some questions provide the constraint directly. Others distribute it across a diagram and a sentence. Label the diagram yourself, even if one is supplied, because this forces you to connect each algebraic symbol to a physical quantity.

The objective-constraint-elimination method

A dependable start for IB Maths AA calculus optimisation is:

State the target. Write what is being maximised or minimised.
Define variables with units. For example, “Let $x$ metres be the width perpendicular to the river.”
Write the objective equation. Keep it separate from the constraint.
Write the constraint equation. Use the fixed information.
Eliminate one variable. Rearrange the constraint and substitute into the objective.
State the feasible domain. Use positivity and any contextual boundaries.
Differentiate the one-variable objective. Find stationary points and any other relevant candidates.
Justify and interpret. Establish whether the result is the required maximum or minimum, then answer in context.

The first five steps are the setup. They deserve more attention than the derivative itself. OpenStax’s applied optimisation strategy follows the same underlying sequence: define variables, identify the target, formulate the constraint, reduce to one variable, identify the domain and then locate the optimum.

Worked example: fencing beside a river

A farmer has 120 m of fencing to enclose a rectangular field beside a straight river. No fence is required along the river. Find the dimensions that maximise the enclosed area.

Identify the objective

Let $x$ be the width perpendicular to the river and $y$ the length parallel to it. The quantity to maximise is area:

$A=xy.$

This is the objective, but it has two variables.

Identify the constraint

There are two fenced widths and one fenced length. Therefore:

$2x+y=120.$

This is the constraint. A frequent setup error is to write $2x+2y=120$ , which incorrectly includes fencing along the river.

Rearrange:

$y=120-2x.$

Substitute into the area equation:

$A(x)=x(120-2x)=120x-2x^2.$

The feasible domain is $0\\leq x\\leq60$ . Values outside this interval would produce a negative width or length.

Apply the derivative

$A'(x)=120-4x.$

At an interior stationary point:

$120-4x=0,$

so $x=30$ . The corresponding length is

$y=120-2(30)=60.$

Since

$A''(x)=-4<0,$

the stationary point is a local maximum. It is also the absolute maximum on the feasible interval: the endpoint areas are zero, while $A(30)=1800$ m².

The required dimensions are therefore 30 m by 60 m. Notice that most of the reasoning occurred before differentiation: interpreting the unfenced river side, choosing variables, separating objective from constraint and establishing the domain.

Why setting the derivative equal to zero is not the first step

Students often remember the rule “for a maximum or minimum, set the derivative equal to zero.” That rule is incomplete. You cannot differentiate until you know which function represents the target quantity.

Moreover, $f'(x)=0$ finds stationary points, not automatically the required optimum. A stationary point could be:

a local maximum
a local minimum
a stationary point of inflexion
irrelevant because it lies outside the feasible domain

An absolute optimum on a closed interval can also occur at an endpoint where the derivative is not zero. For this reason, the full candidate set may include stationary points, points where the derivative does not exist, and relevant endpoints.

You can classify an interior stationary point using one of several valid methods:

Second derivative test: $f''(a)<0$ indicates a local maximum and $f''(a)>0$ a local minimum.

If $f''(a)=0$ , the second derivative test is inconclusive. Use a sign test or direct comparison instead.

Choosing the variable that makes the algebra easier

Either variable may sometimes be eliminated, but the choices are not equally efficient. Prefer the variable that gives a simple domain and avoids unnecessary fractions, radicals or difficult differentiation.

For example, if $2x+y=120$ , solving for $y$ gives $y=120-2x$ , which substitutes cleanly into . Solving for is also valid, but it introduces a fraction.

In circle and cylinder problems, it is often convenient to retain the radius because area and volume formulae are naturally expressed using $r$ . In a point-on-a-curve problem, substitute the curve equation into the squared-distance function to avoid differentiating a square root.

This is an efficiency decision, not a mathematical requirement. If both routes are correct, they must lead to the same optimum.

Common starting mistakes and how to repair them

Differentiating the constraint

A student sees $2x+y=120$ and differentiates it immediately. Unless the variables are changing with time in a related-rates problem, that equation is not the quantity being optimised.

Repair: label the target equation objective and the fixed relationship constraint before differentiating anything.

Leaving two independent variables in the objective

Writing $A=xy$ correctly is not enough for standard one-variable optimisation. Setting a partial-looking derivative equal to zero without eliminating a variable does not use the constraint.

Repair: rearrange the constraint and substitute until the objective contains one independent variable.

Assuming the diagram is to scale

An IB diagram may communicate structure without representing accurate proportions. Estimating the optimum visually can therefore mislead you.

Repair: use the labelled relationships, not apparent lengths or angles.

Ignoring the domain

Algebra may produce a negative radius, a length longer than the available material, or a production value outside the allowed range.

Repair: state the feasible interval before solving the derivative equation. This makes invalid candidates easy to reject.

Answering with the input instead of the requested output

Finding $x=30$ does not answer a request for maximum area. It may also be insufficient when the question asks for both dimensions.

Repair: reread the final instruction and substitute back into the constraint or objective as required.

A practical exam routine when you feel stuck

Use a short setup table before attempting any calculus:

PromptYour noteWhat must be best?Objective quantityWhat remains fixed?ConstraintWhat can change?Variables with unitsWhat formula represents the target?Objective equationHow are the variables connected?Constraint equationWhich variable should remain?Substitution choiceWhich values are possible?Domain

If you cannot fill in the objective row, underline the comparative language in the question. If you cannot fill in the constraint row, inspect every fixed number and every geometric relationship.

Write enough working for the method to be visible. The official AA guide explains that marks can be awarded for method, accuracy, reasoning and interpretation, and that a correct answer without supporting work does not necessarily receive full marks. A labelled objective, valid substitution, derivative equation and contextual conclusion make your reasoning traceable.

For Paper 1, practise carrying out the algebra and differentiation without technology. For Paper 2, a graphing calculator can help solve an equation or verify a graph, but it cannot decide which objective function correctly represents the context. RevisionDojo’s IB Maths AA calculus Questionbank is useful for separating setup errors from calculus errors, while the calculus video collection can show how a complete argument is organised.

How to practise the setup rather than memorising examples

Repeating one rectangle question can create false confidence because a new context may hide the same structure differently. Practise classifying problems before solving them fully.

For each question, spend two minutes writing only:

variable definitions
the objective equation
the constraint equation
the substituted one-variable objective
the feasible domain

Then compare your setup with the solution. Record the first incorrect decision, not merely the final wrong answer. Jojo AI can help you question a particular modelling step, but ask for a hint before requesting a complete solution so that you still practise making the decision yourself.

Use calculus flashcards for derivative facts and classification language, but use exam-style questions for modelling. Once direct optimisation questions are secure, move to mixed calculus practice so that the method is no longer announced in advance. The guide to revising IB Maths AA with past-paper questions explains how to attempt, diagnose and reattempt questions systematically.

Conclusion

Optimisation with derivatives feels difficult to start because the opening stage tests interpretation, modelling and algebra before it tests calculus. The central distinction is simple: the objective is what must become best, while the constraint describes what must remain true.

Build the target function, use the constraint to reduce it to one variable, state the domain, and only then differentiate. Finally, justify the optimum and answer the quantity actually requested. RevisionDojo’s Study Notes, Questionbank and Jojo AI are most useful when you use them to diagnose which setup decision failed rather than simply checking a final number.

Sources and referenced URLs

Add as a preferred source on Google

Continue studying with RevisionDojo

What an optimisation problem is really asking

Two mathematical objects control the entire solution:

The objective function represents the quantity to be maximised or minimised.
The constraint represents a fixed condition or relationship that limits the possible choices.

Suppose a rectangle has a perimeter of 40 cm and its area must be maximised. Its area is the objective:

$A=xy.$

Its fixed perimeter gives the constraint:

$2x+2y=40.$

The derivative cannot yet be used effectively because $A$ contains two variables. The constraint must first be rearranged, for example as $y=20-x$ , and substituted into the objective:

$A(x)=x(20-x)=20x-x^2.$

Only now is there a one-variable function to differentiate. This sequence explains why students who are confident at finding derivatives can still struggle with derivative optimisation problems.