You cannot use an entire forecast distribution as the objective, but must characterize the distribution using a single summary measure for comparing and selecting one distribution over another. So, to use OptQuest, you must select a statistic of one forecast to be the objective. You must also select whether to maximize or minimize the objective, or set it to a target value.
The statistic you select depends on your goals for the objective. For maximizing or minimizing some quantity, the mean or median are often used as measures of central tendency, with the mean being the more common of the two. For highly skewed distributions, however, the mean may become the less stable (having a higher standard error) of the two, and so the median becomes a better measure of central tendency.
The X in Y Chance statistic can be used only for requirements, not objectives.
For minimizing overall risk, the standard deviation and the variance of the objective are the two best statistics to use. For maximizing or minimizing the extreme values of the objective, a low or high percentile may be the appropriate statistic. For controlling the shape or range of the objective, the skewness, kurtosis, or certainty statistics may be used. If you are working with Six Sigma or another process quality program, you may want to use process capability metrics in defining the objective. For more information on these statistics, see the Glossary, online help, and the online Oracle Crystal Ball Statistical Guide.