What is the key difference between decision variables and parameters in LP formulation?

Decision variables are the quantities the model chooses, while parameters are fixed numerical inputs taken from the problem context. This distinction matters because only variables can change during optimization, but parameters define how choices are valued or limited. If you treat a parameter as a variable, the model no longer represents the original decision environment.

How does an objective function differ from a constraint?

The objective function defines what is being optimized, such as maximizing benefit or minimizing cost. Constraints define what is allowed, so they shape the feasible region where optimization can occur. A correct model needs both: one gives direction, the other gives boundaries.

When should you use a $\le$ inequality versus a $\ge$ inequality in a formulation?

Use $\le$ when expressing an upper limit, such as available capacity that cannot be exceeded. Use $\ge$ when expressing a minimum requirement that must be met. The choice comes from the meaning of the statement, not from algebraic convenience.

What error occurs if non-negativity constraints are omitted when variables represent physical quantities?

Without non-negativity, the model may allow negative production, negative time, or other impossible values. Even if the solver returns an optimal point, that solution may be meaningless in context. Including domain constraints protects realism as well as mathematical validity.

Why is reversing an inequality sign a serious formulation mistake?

Reversing the sign changes the feasible region, often excluding valid decisions or admitting invalid ones. This can completely alter the computed optimum, even when arithmetic is flawless. The symbol must be chosen from problem semantics before rearranging terms.

What goes wrong when units are inconsistent inside a constraint?

A constraint is meaningful only if all terms are comparable in units, such as hours with hours or kilograms with kilograms. If units are mixed without conversion, coefficients lose interpretation and the inequality no longer models reality. Unit checking is therefore a core validation step in formulation.

What is the canonical matrix form of a linear programming model?

A common form is optimize $z = \mathbf{c}^T\mathbf{x}$ subject to $A\mathbf{x} \le \mathbf{b}$ and $\mathbf{x} \ge 0$. This notation compactly represents many variables and constraints using vectors and matrices. It is useful because it separates objective coefficients, structural constraints, and domain restrictions.

How do you define a decision variable in a high-quality LP formulation?

A strong definition states what the variable counts, its unit, and its time scope, such as per day or per week. This clarity ensures coefficients in objectives and constraints are interpreted correctly. Precise variable definitions also reduce ambiguity for marking and implementation.

What does the non-negativity condition $x_i \ge 0$ represent?

It states that each decision variable cannot take negative values in the modeled domain. This condition is usually required when variables represent quantities, flows, or activity levels. It narrows the feasible set to decisions that are both mathematically and physically admissible.

Why must formulation be completed before attempting graphical or simplex solution steps?

Solution methods assume the model already captures the real decision structure correctly. If formulation is wrong, the solver simply optimizes the wrong problem with high precision. Correct modeling is therefore the causal foundation of correct optimization.

Library Podcasts

Courses

Referral & Rewards

Revision Notes

International A-Level

Formulating a Linear Programming Problem

Summary

Formulating a linear programming problem means translating a real decision situation into a mathematical model with decision variables, a linear objective, and linear constraints. The power of formulation is that it separates what can be controlled from what must be respected, so optimization methods can be applied reliably. A strong formulation is precise about units, signs, and non-negativity, which directly determines whether later graphing or simplex steps produce meaningful decisions.

1. Definition & Core Concepts

What the model represents

Linear programming formulation is the process of expressing a decision problem as a linear objective function plus linear constraints. It works because many planning systems can be approximated by additive effects, where total usage or profit is the sum of per-unit contributions. You use it when choices are continuous or count-based and trade-offs must be optimized under limited resources.

Core terms

Decision variables are the unknown quantities you choose, such as production amounts, shipment levels, or staffing hours. Parameters are fixed coefficients like time per unit, cost per unit, or resource limits, and they should not be treated as variables. The objective function is the linear expression to maximize or minimize, while constraints define the feasible set of allowable decisions.
Non-negativity constraints (for example, $x_i \ge 0$ ) are usually essential because many variables represent quantities that cannot be negative in context. Including them prevents mathematically valid but physically impossible solutions. This condition is part of the model definition, not an optional afterthought.

2. Underlying Principles

Coordinate-plane view of two linear constraints, shaded feasible region, and objective improvement direction.

3. Methods & Techniques

4. Key Distinctions

5. Exam Strategy & Tips

6. Common Pitfalls & Misconceptions

7. Connections & Extensions

Formulating a Linear Programming Problem

Summary

1. Definition & Core Concepts

What the model represents

Linear programming formulation is the process of expressing a decision problem as a linear objective function plus linear constraints. It works because many planning systems can be approximated by additive effects, where total usage or profit is the sum of per-unit contributions. You use it when choices are continuous or count-based and trade-offs must be optimized under limited resources.

Core terms

Decision variables are the unknown quantities you choose, such as production amounts, shipment levels, or staffing hours. Parameters are fixed coefficients like time per unit, cost per unit, or resource limits, and they should not be treated as variables. The objective function is the linear expression to maximize or minimize, while constraints define the feasible set of allowable decisions.
Non-negativity constraints (for example, $x_i \ge 0$ ) are usually essential because many variables represent quantities that cannot be negative in context. Including them prevents mathematically valid but physically impossible solutions. This condition is part of the model definition, not an optional afterthought.

2. Underlying Principles

Why linear structure matters

A formulation is linear when each term has variables to the first power and no products of variables, so expressions look like weighted sums. This linearity creates a convex feasible region, which is why optimization is tractable and global optima are reliably found. When relationships are approximately proportional, LP gives transparent and interpretable decisions.
Standard algebraic structure is often written as:

Maximize/Minimize $z = \mathbf{c}^T \mathbf{x}$
subject to $A\mathbf{x} \le \mathbf{b}$ , $\mathbf{x} \ge 0$ Here, $\mathbf{x}$ is the decision vector, $\mathbf{c}$ contains objective coefficients, $A$ is the constraint coefficient matrix, and $\mathbf{b}$ is the resource-limit vector. This structure clarifies what is controlled, what is limited, and what is optimized.

Feasibility and optimality are conceptually separate: a point can satisfy all constraints but still be suboptimal for the objective. LP methods first enforce feasibility, then search within feasible space for the best objective value. This separation is why careful constraint translation is as important as objective construction.

Coordinate-plane view of two linear constraints, shaded feasible region, and objective improvement direction.

3. Methods & Techniques

Step-by-step formulation workflow

Step 1: define decision variables clearly with names, units, and time horizon (for example, per day or per week). This prevents hidden ambiguity when translating text into algebra. If units are unclear, coefficient errors usually appear later in constraints.
Step 2: translate each verbal limit into an inequality by writing total consumption or total requirement as a linear expression. A practical template is “sum of (per-unit usage × variable) compared with available amount,” then choose $\le$ , $\ge$ , or $=$ based on wording. Simplify algebra without changing meaning so the model is readable and checkable.
Step 3: build the objective function from per-unit contribution and decision variables, then state whether it is maximize or minimize. Step 4: write the full formal model with objective, all constraints, and non-negativity in one place. This final formal statement is the deliverable that can be graphed or solved algorithmically.

4. Key Distinctions

High-value comparisons

Distinguishing similar terms is essential because many exam errors are classification errors, not algebra errors. In LP formulation, confusion between variables, parameters, and bounds causes structurally invalid models even when arithmetic is correct. Use the table to separate roles before writing equations.

Feature	Decision Variable	Parameter	Constraint Bound
Can be chosen?	Yes	No	No
Appears where?	Objective and constraints	Coefficients/constants	Right-hand side limit
Typical meaning	Quantity to decide	Rate or contribution per unit	Resource availability or target

Objective direction and inequality direction are different decisions with different logic. Maximize vs minimize comes from the business goal, while $\le$ vs $\ge$ comes from the meaning of a resource or requirement statement. Mixing these choices creates a model that can be mathematically solvable but conceptually wrong.
A model can be feasible but not realistic if integrality is required in context but not included in the formulation assumptions. Classical LP allows fractional values, which is acceptable for divisible quantities but not for indivisible units like whole vehicles. Recognizing this distinction helps decide whether pure LP or integer programming is the correct modeling framework.

5. Exam Strategy & Tips

What examiners reward

Start with a declaration block such as “Let $x$ = ... , Let $y$ = ...” and include units in words. This signals modeling control and reduces marker uncertainty about interpretation. Even if later algebra has a slip, clear variable definition often preserves method credit.
Write the final answer in a formal layout:

Maximize/Minimize objective
subject to listed inequalities
and non-negativity constraints This structure is fast to mark and makes omission errors visible to you before submission. A clean formal layout also helps you perform a final consistency check.

Perform a 30-second reasonableness audit: verify signs, compare units on both sides of each inequality, and confirm every constraint uses the declared variables only. Check that non-negativity appears explicitly unless the context justifies otherwise. These quick checks catch the most common high-impact mistakes before time is called.