Sample size selection for improved Nelder-Mead performance

Author

Tomick, John J. ; Arnold, Steven F. ; Barton, Russell R.

Author_Institution

Dept. of Math. Sci., US Air Force Acad., Colorado Springs, CO, USA

fYear

1995

fDate

3-6 Dec 1995

Firstpage

341

Lastpage

345

Abstract

The Nelder-Mead (1965) simplex algorithm has been used for sequential optimization of simulation response functions. The rescaling operations of this algorithm can lead to inappropriate termination at non-optimal points. We have used the probabilistic characterization of this behaviour to develop special rules for determining the number of replications to take for each experimental design point. Computational experiments indicate that the quality of the solution is often improved

Keywords

design of experiments; discrete event simulation; optimisation; probability; Nelder-Mead simplex algorithm; experimental design points; improved performance; inappropriate termination; nonoptimal points; probabilistic characterization; replications; rescaling operations; sample size selection; sequential optimization; simulation response functions; solution quality; Computational modeling; Design optimization; Discrete event simulation; Manufacturing industries; Optimization methods; Reflection; Springs; Statistics; Stochastic processes; System performance;

fLanguage

English

Publisher

ieee

Conference_Titel

Simulation Conference Proceedings, 1995. Winter

Conference_Location

Arlington, VA

Print_ISBN

0-78033018-8

Type

conf

DOI

10.1109/WSC.1995.478754

Filename

478754