THE TARGET MEAN PROBLEM FOR AN ARBITRARY QUALITY CHARACTERISTIC DISTRIBUTION

We consider the general problem of determination of an optimal process mean setting and the upper screening limit in relation to a fixed low er limit for a canning problem when the amount of fill follows an arbi trary continuous distribution. We show that the optimal upper limit is given by a very simple formula regardless of the shape of the distrib ution. We obtain a general condition to solve for the optimal process mean. Some specific distributions considered in the paper include norm al, truncated normal, and truncated logistic. Comparisons are made in terms of the values of the optimal process mean, container rejection r ate, and expected profit per fill attempt. Moreover, for any continuou s fill distribution it is shown that a two-level control scheme with o ptimal upper limit is always economically superior to a single level s cheme.