Numerical optimization has the same objective as analytical optimization: to find the values of coefficients of a function that yield best estimates of some variable. Problems can be formulated to maximize or to minimize a function. For example, where profit is the objective, the optimization will likely be one of maximization; that is, the objective is to find the values of the function that maximize the profit. The solution of a numerical optimization problem can be graphically portrayed by graphing the value of the objective function F versus each of the unknowns. Convergence is assumed when the difference in the values of the objective function from one iteration to the next becomes less than the input tolerance T. As the set of coefficients approaches the point of zero gradients, the slope of the response surface decreases and movement causes very little change in the objective function F.