Correlated simulation experiments in first-order response surface design
Operations Research
Metamodels for simulation input-output relations
WSC '92 Proceedings of the 24th conference on Winter simulation
A three-class variance swapping technique for simulation experiments
Operations Research Letters
Hi-index | 0.00 |
The paper investigates the use of an effective pseudo-random number assignment strategy, which we call maximum-blocks strategy, for simulation experiments involving the estimation of quadratic response surface metamodel in 3/sup k/ factorial designs. The maximum-blocks strategy is shown to perform better than well-known existing quadratic-metamodel assignment strategies, such as the IR, CR, and modified-AR strategies.