Premium
Economic design of a two‐stage control chart based on both performance and surrogate variables
Author(s) -
Lee Jooho,
Kwon WooJoo
Publication year - 1999
Publication title -
naval research logistics (nrl)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.665
H-Index - 68
eISSN - 1520-6750
pISSN - 0894-069X
DOI - 10.1002/(sici)1520-6750(199912)46:8<958::aid-nav6>3.0.co;2-i
Subject(s) - variable (mathematics) , control variable , surrogate model , control chart , computer science , process variable , process (computing) , chart , surrogate data , control (management) , mathematical optimization , control theory (sociology) , statistics , mathematics , artificial intelligence , machine learning , mathematical analysis , physics , nonlinear system , quantum mechanics , operating system
The traditional approach to economic design of control charts is based on the assumption that a process is monitored using only a performance variable. If, however, the performance variable is costly to measure and a less expensive surrogate variable is available, the process may be more efficiently controlled by using both performance and surrogate variables. In this article we propose a model for economic design of a two‐stage control chart which uses a highly correlated surrogate variable together with a performance variable. The process is assumed to be monitored by the surrogate variable until it signals out‐of‐control behavior, then by the performance variable until it signals out‐of‐control behavior or maintains in‐control signals for a prespecified amount of time, and the two variables are used in alternating fashion. An algorithm based on the direct search method of Hooke and Jeeves [6] is used to find the optimum values of design parameters. The proposed model is applied to the end‐closure welding process for nuclear fuel to compute the amount of reduction in cost compared with the current control procedure. © 1999 John Wiley & Sons, Inc. Naval Research Logistics 46: 958–977, 1999