Abstract
An optimization method has been developed that searches one variable at a time under the conditions that previously searched variables are at their approximate optimum and variables to be searched later are constant. This method provides the optimum and also information on the effects of the variables. The method is based on the Partan and Powell methods and on assuming linear partial derivatives of the objective function with respect to any given variable. The procedure involves searching each variable separately with other variables either constant or varied so that they remain at their estimated optimum for the given conditions. This method provides useful information on the effects of independent variables as well as locating the overall optimum and a number of partial optimums and requires a comparable number of trials. Copyright © 1974 American Institute of Chemical Engineers
Recommended Citation
M. E. Findley, "Modified One‐at‐a‐time Optimization," AIChE Journal, vol. 20, no. 6, pp. 1154 - 1160, Wiley; American Institute of Chemical Engineers (AIChE), Jan 1974.
The definitive version is available at https://doi.org/10.1002/aic.690200614
Department(s)
Chemical and Biochemical Engineering
Publication Status
Full Access
International Standard Serial Number (ISSN)
1547-5905; 0001-1541
Document Type
Article - Journal
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2023 Wiley; American Institute of Chemical Engineers (AIChE), All rights reserved.
Publication Date
01 Jan 1974