Lagrange multipliers are a method for finding maximum or minimum values of a function when the inputs must satisfy a constraint. Many real problems have limits, such as fixed distance, fixed cost, fixed area, or conservation of energy. Instead of searching every allowed point, the method uses gradients to locate candidates where the best value can occur.
This makes constrained optimization systematic and easier to visualize.
Understanding Calculus: Lagrange Multipliers
The central idea comes from allowed motion. A constraint does not let a point move freely across the whole plane or space. It restricts the point to a curve or surface.
At the best allowed location, a tiny move along that curve cannot improve the target quantity to first order. The direction that increases the target most strongly must therefore point across the constraint rather than along it.
This geometric fact explains why the derivatives of the target and constraint become linked. The multiplier measures the amount needed to match their sizes while their directions line up.
In practice, students begin by naming the quantity to optimize and writing every restriction clearly. Units help at this stage. If a company has a fixed budget, cost must be measured in one consistent unit.
If a box has a fixed amount of material, each area formula must use the same length unit. Then take partial derivatives with respect to each independent input. The resulting equations often look harder than the original problem, but they turn a search over infinitely many points into a finite list of candidates.
Algebra matters here. A common mistake is dividing by an expression that could equal zero. Check that case separately before canceling it.
The multiplier itself can have a useful real meaning. Imagine changing a fixed budget by a very small amount. The multiplier estimates how much the best possible profit changes per extra unit of budget, near the current limit.
In physics, a multiplier connected to a conservation rule can represent the effect of slightly relaxing that rule. This interpretation is called a shadow price in economics. Its units come from target units divided by constraint units.
For example, if the target is profit in dollars and the constraint is material in kilograms, the multiplier has units of dollars per kilogram. Its sign shows whether loosening the limit tends to help or hurt the optimum locally.
The method has conditions and exceptions. It works most cleanly when the constraint has a well-defined direction at the candidate point. If every first derivative of the constraint is zero there, the usual equations may miss an important point or provide too little information.
A constraint with corners, endpoints, or separate pieces needs extra care. For a closed path, compare every candidate value, including any endpoints when they exist. For an open path, a largest or smallest value may not occur at all.
Students should sketch the allowed region whenever possible, even a rough sketch. It catches impossible answers, reveals symmetry, and shows whether a result is likely a maximum, a minimum, or neither. Finally, substitute every valid candidate back into the original target quantity.
Solving the derivative equations only finds possible winners. The final comparison decides the result.
Key Facts
- To optimize f(x, y) subject to g(x, y) = c, solve ∇f = λ∇g and g(x, y) = c.
- In two variables, the equations are f_x = λg_x, f_y = λg_y, and g(x, y) = c.
- In three variables, solve f_x = λg_x, f_y = λg_y, f_z = λg_z, and g(x, y, z) = c.
- The gradient ∇f points in the direction of steepest increase of f.
- At a constrained maximum or minimum, the level curve or surface of f is tangent to the constraint curve or surface.
- After solving for candidate points, evaluate f at each point to identify the maximum, minimum, or both.
Vocabulary
- Objective function
- The function f that you want to maximize or minimize.
- Constraint
- An equation such as g(x, y) = c that limits the points allowed in the problem.
- Gradient
- The vector of partial derivatives that points in the direction of greatest increase of a function.
- Lagrange multiplier
- The scalar λ that relates the gradient of the objective function to the gradient of the constraint.
- Level curve
- A curve made of points where a function has the same constant value.
Common Mistakes to Avoid
- Forgetting the constraint equation is wrong because ∇f = λ∇g alone does not guarantee that the point is allowed.
- Treating λ as the answer is wrong because λ is an auxiliary variable, while the actual optimum is found from the point and the value of f there.
- Assuming every solution is a maximum is wrong because Lagrange multipliers give candidate points that must be tested and compared.
- Using the method when ∇g = 0 without checking is wrong because the standard gradient condition may fail at singular points of the constraint.
Practice Questions
- 1 Find the maximum and minimum of f(x, y) = xy subject to x^2 + y^2 = 8.
- 2 Use Lagrange multipliers to minimize f(x, y, z) = x^2 + y^2 + z^2 subject to x + 2y + 2z = 9.
- 3 Explain why the gradient of the objective function must be parallel to the gradient of the constraint at a smooth constrained maximum or minimum.