If f has a continuous second derivative on the interval [a, b], then the error e in approximating:
b^∫_a f(x)dx using the trapezoidal rule is minimized when:
a) f′′(x)=0 for some x in (a, b)
b) f′ (x)=0 for some x in (a, b)
c) f(x) is a constant function
d) f(a)=f(b)