Design of Computer Simulation Experiments for Industrial Systems

The aim of this paper is to provide background
information on the existing literature on experimental 
design techniques which may be applicable to the design
of computer simulation experiments for industrial 
systems.  Although major emphasis is placed on analysis
of variance techniques, three other techniques 
of data analysis are considered-multiple ranking procedures,
sequential sampling and spectral analysis. 
 The paper treats four specific experimental design
problems and several techniques for solving them. 
 The four experimental design problems are: (1) the
problem of stochastic convergence, (2) the problem 
of factor selection, (3) the problem of motive
and (4) the many response problem.

CACM May, 1966

Burdick, D. S.
Naylor, T. H.

CA660502 JB March 3, 1978  10:20 AM

1435	5	1435
1435	5	1435
1435	5	1435
1517	5	1435
1435	6	1435