Degree Type

Dissertation

Date of Award

1989

Degree Name

Doctor of Philosophy

Department

Statistics

First Advisor

Stephen B. Vardeman

Abstract

A state space process control model involving deterministic drift of the process mean and adjustment error is presented. A Kalman filter estimator of the process mean is developed. Optimal adjustment strategies based on this Kalman filter estimator are developed by the technique of dynamic programming for two special cases of the model and for the general model. The optimal policy in the general case calls for an adjustment to the process when the estimate of the process mean is outside an interval bounded by lower and upper action limits. The action limits depend on the variability associated with the Kalman estimator. The optimal adjustment consists of compensation for the currently perceived misadjustment of the process mean and the anticipated drift, and possibly, depending on the cost of adjustment and the size of the drift, a small over-compensation to anticipate future drift. Computational methods for approximating the action limits and over-compensation constants are presented. The effects of parameter values on these limits and constants are described.

DOI

https://doi.org/10.31274/rtd-180813-9048

Publisher

Digital Repository @ Iowa State University, http://lib.dr.iastate.edu/

Copyright Owner

Karen Lorraine Jensen

Language

en

Proquest ID

AAI8920149

File Format

application/pdf

File Size

179 pages

Share

COinS