The Extreme Value Theorem tells us that the minimum and maximum
of a continuous function on a closed interval have to be
somewhere. But where should we look?
It turns out that extreme values can occur at the endpoints of
the intervals, or at the "top of a hill" or the "bottom of a
valley" of the graph of a function. How can we find these
hills and valleys? They can occur only at x-values where
the derivative either equals zero (where
the tangent line has slope of zero), or does not exist (where
there may be a cusp).
Fermat's
Theorem: Suppose that a<c<b. If a function
f is defined on the interval (a,b), and it has a maximum
or a minimum at c, then either f′ doesn't exist at c
or f′(c)=0.
Such x-values c are called critical
numbers.
Critical
numbers: A critical number of a function f is
a number c in the domain of f such that f′(c)=0 or
f′(c) is not defined.
What we are saying, then, is that absolute
maxima or minima can
only occur at critical numbers and endpoints of intervals.
Equivalently, if f′(c) exists and is not
zero, then f(c) is neither a maximum nor a minimum.
The process of finding such maxima and minima is not difficult in
practice: