Multiple-criteria optimal design of X¯ control charts

Abstract

Economic designs of control charts have been criticized for requiring the user to assign explicit costs to type I and type II errors, which are the errors associated with false alarms and running the process while it is out of control. In this paper an interactive multicriteria nonlinear optimization algorithm is applied to a model for the design of X¯ charts. In the proposed model, the user does not need to specify the cost of false alarms and running out of control. Only sampling costs, which usually are relatively easy to estimate, need to be specified. A case study illustrates the interactive procedure for a real set of process data.