This paper presents methods for the statistical analysis of plant operations optimization results with special consideration for real-time optimization (RTO) applications. The key challenge is to determine whether the results of an optimization calculation should be implemented in the plant. Since feedback data used to correct the model include noise and the effects of high-frequency disturbances, the results of the model-based optimization calculations are corrupted by a stochastic component. The methods developed in this paper apply multivariable statistical hypothesis tests based on control charts in order to distinguish between high-frequency disturbances propagated through the calculations and significant changes in the plant optimization variables with the goals of reducing the frequency of unnecessary changes in the implemented independent optimization variables and increasing plant profits. Only the statistically significant results are implemented in the plant. Case studies indicate that increased profit can be obtained by implementing fewer changes to the process because the preponderance of changes due to noise are rejected whereas most meaningful changes are implemented.