Research Reports from the Department of Operations
Document Type
Dissertation
Publication Date
5-1-1985
Abstract
When an experimenter performs experiments, the most important consideration is the experimental design. However, if the experimental cost is low, the experimental design is not a problem because the experimenter is at liberty to choose any design he wants. But if the cost is high and the number of experiments is restricted, then the experimental design becomes critical. Classical experimental designs are the best designs and they have unique properties. But they cannot be used when the experimenter is confronted with special problems such as a restriction on the number of experiments, restricted experimental runs, augmentation of designs, or special shapes of an experimental region. The experimenter may choose an arbitrary experimental design but a better design for his experiment is obtained by using a criterion yielding good designs. In this dissertation, the D-optimality criterion is used because designs based on this criterion are invariably good in many respects such as low variance for the parameter, low correlation among parameters and low maximum variance (x) over all candidate x. After experiments based on the D-optimal experimental design are run, the data can be easily analyzed by the computer program RSP using the backward elimination method. Finally, in order to reduce the computer time in finding the D-optimal design, we suggest a rule of locating experimental runs at j + 1 evenly spaced points of the predictor variables to estimate a polynomial model of degree j, especially when j = 1 or 2.
Keywords
Operations research, Experimental design, Response surfaces (Statistics), Mathematical optimization, Computer simulation, Regression analysis
Publication Title
Dissertation/Technical Memorandums from the Department of Operations, School of Management, Case Western Reserve University
Issue
Technical memorandum no. 561 ; Submitted in partial fulfillment of the requirements for the Degree of Doctor of Philosophy.
Rights
This work is in the public domain and may be freely downloaded for personal or academic use
Recommended Citation
Lee, Chin-Kyooh, "Improving and Expanding the Role of D-Optimality in Experimental Designs with Emphasis on Response Surfaces" (1985). Research Reports from the Department of Operations. 243.
https://commons.case.edu/wsom-ops-reports/243