A quality control engineer monitors a production line where the weight of each component must strictly lie within the interval $[10.0\text{ g}, 20.0\text{ g}]$. The engineer collects a random sample of $n = 15$ components. Without knowing the individual weights, but knowing they all satisfy the interval constraint, what is the theoretical maximum possible value of the unbiased sample variance $s^2$ of this sample?