This thesis is concerned with the problem of selecting the most profitable process mean for production processes where measurement errors exist. For such situations, the 'best estimator' of the true value of quality characteristic is obtained from repeated measurements of an item, and it is used to reduce measurement errors.
Two inspection schemes are proposed. One is a two-stage inspection procedure; depending on repeated measurements of the first stage, an item is accepted, rejected, or the decision is deferred to the second stage. The other is a sequential inspection procedure, where the decision to accept, reject, or take an additional inspection of an item is made at every measurement point until the number of repeated measurements reaches its upper bound. For each procedure, an expected profit model is constructed and optimal process mean, cut-off values, and number of repeated measurements(or its upper bound) are obtained when accepted(rejected) items are sold at regular(reduced) price. Sensitivity analyses are also performed to investigate the effects of errors in the estimation of process parameters.