The process of finding absolute maximum and minimum values of a function within a defined domain is a fundamental concept in multivariable calculus, known as finding absolute extrema. This problem-solving strategy is essential for determining optimal solutions under constraints, a skill that has applications across various fields, including engineering, economics, and data science. The methodology involves a systematic approach: first identifying critical points within the interior of the domain, and then evaluating the function on the boundary of the domain. This dual-phase investigation ensures that no potential extremum is overlooked, whether it lies within the open region or on its edges.
The core strategy for solving these problems is outlined in standard calculus resources. The first step is to find all critical points of the function within the domain of interest. A critical point occurs where the first-order partial derivatives of the function are simultaneously zero or where the derivatives do not exist. For a function (f(x, y)) of two variables, this involves calculating (fx(x, y)) and (fy(x, y)), setting each equal to zero, and solving the resulting system of equations. The solutions are potential candidates for local extrema, but they must be evaluated to determine if they are absolute extrema on the entire domain.
Once the interior critical points are found, the second step involves finding the extrema on the boundary of the domain. The approach for this step varies depending on the shape and complexity of the boundary. For boundaries composed of straight lines, such as the edges of a rectangle or triangle, the problem can often be reduced to a single-variable calculus problem. For example, on a rectangle defined by (0 \leq x \leq 4) and (0 \leq y \leq 2), the boundary consists of four line segments. On each segment, one variable is held constant while the other varies, allowing the function to be expressed as a function of a single variable. The extrema on each segment are then found using standard single-variable calculus techniques, such as finding the derivative and checking endpoints.
For more complex boundaries, such as a circle defined by (x^2 + y^2 = r^2), parameterization is a powerful tool. The boundary can be parameterized using trigonometric functions, for instance, (x(t) = r \cos t) and (y(t) = r \sin t) for (0 \leq t \leq 2\pi). Substituting these into the original function creates a new function of a single parameter, (t). The extrema of this new function on the interval ([0, 2\pi]) can then be found by taking its derivative with respect to (t), setting it to zero, and solving for (t). The corresponding points on the boundary are then evaluated.
In cases where the boundary is defined by a more complicated curve, such as (g(x, y) = c), the method of Lagrange multipliers is often employed. This technique introduces a new variable (the Lagrange multiplier) to incorporate the constraint directly into the optimization process. The system of equations formed by setting the gradients of the objective function and the constraint function proportional to each other is then solved to find potential extrema on the boundary.
The final step of the strategy is to compare the function values at all candidate points: the critical points found in the interior and all points found on the boundaries (including endpoints and points where the derivative of the parameterized boundary function is zero). The largest of these values is the absolute maximum, and the smallest is the absolute minimum.
Example 1: Rectangular Domain Consider finding the absolute extrema of (f(x, y) = x^2 - 2xy + 4y^2 - 4x - 2y + 24) on the domain (0 \leq x \leq 4) and (0 \leq y \leq 2). 1. Find Critical Points: Calculate partial derivatives: (fx = 2x - 2y - 4) and (fy = -2x + 8y - 2). Setting them to zero gives the system: [ \begin{align} 2x - 2y - 4 &= 0 \ -2x + 8y - 2 &= 0 \end{align} ] Solving yields (x = 3) and (y = 1). The critical point is ((3, 1)), and (f(3, 1) = 21). 2. Find Extrema on the Boundary: The boundary consists of four line segments. Evaluating the function on each segment and at the corners ((0,0)), ((4,0)), ((4,2)), and ((0,2)) yields values of 24, 24, 20, and 36, respectively. 3. Compare Values: Comparing (f(3,1)=21) with the boundary values (24, 24, 20, 36), the absolute maximum is 36 at ((0,2)), and the absolute minimum is 20 at ((4,2)).
Example 2: Circular Domain For (g(x, y) = x^2 + y^2 + 4x - 6y) on the disk (x^2 + y^2 \leq 16): 1. Find Critical Points: (gx = 2x + 4) and (gy = 2y - 6). Setting to zero gives the critical point ((-2, 3)). However, this point lies outside the domain (((-2)^2 + 3^2 = 13 \leq 16) is true, so it is inside). (g(-2, 3) = -13). 2. Find Extrema on the Boundary: Parameterize the boundary circle: (x(t) = 4\cos t), (y(t) = 4\sin t). Substitute into (g): [ h(t) = g(4\cos t, 4\sin t) = 16 + 16\cos t - 24\sin t. ] Set (h'(t) = -16\sin t - 24\cos t = 0), which gives (\tan t = -3/2). Solving for (t) yields two points. Evaluating (h(t)) at these points and at the endpoints (t=0, 2\pi) gives candidate values. The maximum on the boundary is found to be (16 + 4\sqrt{13^2 + 2^2} = 16 + 4\sqrt{173}), and the minimum is (16 - 4\sqrt{173}). 3. Compare Values: The absolute maximum on the closed disk is the larger of the interior critical value and the boundary extrema. The absolute minimum is the smaller of these values.
Example 3: Triangular Domain For (f(x, y) = x^2 - y^2 + 5) on a triangular region (S) with vertices ((-1, -2)), ((2, -2)), and ((0, 1)): 1. Find Critical Points: (fx = 2x), (fy = -2y). The only critical point is ((0, 0)), which lies inside the triangle. (f(0, 0) = 5). 2. Find Extrema on the Boundary: The boundary consists of three line segments. * Bottom edge ((y = -2), (-1 \leq x \leq 2)): (f(x, -2) = x^2 + 1). Its minimum is 1 at (x=0), maximum is 5 at (x=\pm 1). * Left edge (line from ((-1, -2)) to ((0, 1))): Parameterize and find extrema. * Right edge (line from ((2, -2)) to ((0, 1))): Parameterize and find extrema. 3. Compare Values: After evaluating all boundary segments and the interior critical point, the absolute maximum and minimum are determined by comparing all obtained function values.
Advanced Application: Optimization with Implicit Constraints In more complex problems, the domain itself may be defined by an implicit constraint. A classic example is maximizing the volume (V = LWH) of a box given a fixed surface area constraint (3LW + 2LH + 2WH = 36). Here, the objective function is (V(L, W, H)), and the constraint (g(L, W, H) = 36) must be satisfied. The method of Lagrange multipliers is ideal for this type of constrained optimization problem. One sets up the system (\nabla V = \lambda \nabla g), which yields three equations: [ \begin{align} WH &= \lambda(3W + 2H) \ LH &= \lambda(3L + 2H) \ LW &= \lambda(2L + 2W) \end{align} ] Solving this system, along with the constraint, provides the dimensions that maximize the volume. Alternatively, one can solve the constraint for one variable (e.g., (H = \frac{36 - 3LW}{2(L+W)})) and substitute it into the volume function, reducing the problem to finding critical points of (V(L, W) = \frac{36LW - 3L^2W^2}{2(L+W)}). The partial derivatives (VL) and (VW) are then set to zero to find candidate points, and the second partials test can confirm a relative maximum.
This systematic approach to finding absolute extrema—identifying interior critical points, thoroughly investigating the boundary through parameterization or Lagrange multipliers, and comparing all candidate values—provides a robust framework for solving optimization problems under constraints. It emphasizes the importance of checking all possible locations where an extremum could occur, ensuring a comprehensive solution.