A critical point can identify an important candidate, but it does not automatically produce the greatest function value. Endpoints may exceed interior values, while additional points in the domain may also matter when the function is defined there. Evaluating every relevant candidate and comparing the results prevents a merely local high point from being mistaken for the overall maximum.
An absolute maximum describes the greatest value across the entire domain or specified interval, whereas a local maximum is greatest only within a nearby portion of the domain. A function may have several local maxima, but only one value can be the largest overall, although multiple points could attain that same value. This distinction separates global behavior from nearby behavior.
The Extreme Value Theorem guarantees an absolute maximum when a function is continuous on a closed, bounded interval. The interval conditions limit the region being examined, and continuity ensures the function does not create a gap that prevents the greatest value from being attained. Under these conditions, the theorem also guarantees an absolute minimum, giving both global bounds.
The Extreme Value Theorem does not provide the same guarantee outside a closed, bounded interval with continuity. A function may still attain an absolute maximum, but its existence must be established by examining the function and its domain rather than assumed from the theorem. Consequently, interval structure, continuity, and defined domain locations become important parts of the analysis.
First identify the relevant critical points within the interval, then include both endpoints and any other domain locations where the function is defined. Evaluate the function at each candidate and compare the resulting values. The largest value is selected as the outcome of the search. This systematic process combines the calculus-based candidates with boundary and domain checks.
In an applied model, the largest function value can represent an upper outcome such as highest profit, capacity, height, temperature, or efficiency. Finding that value helps identify the strongest performance available within the modeled domain or interval. The result is meaningful only alongside the domain being studied, because changing the allowed range can change which outcome is greatest.