Empirical model-building and response surface
Empirical model-building and response surface
Using central composite designs in simulation experiments
WSC '92 Proceedings of the 24th conference on Winter simulation
Design of experiments: robust design: seeking the best of all possible worlds
Proceedings of the 32nd conference on Winter simulation
Simulation with Visual Slam and Awesim
Simulation with Visual Slam and Awesim
Simulation Modeling and Analysis
Simulation Modeling and Analysis
| Hi-index | 0.01 |
This paper proposes a strategy using the common random numbers and antithetic variates in simulation experiments that utilize the central composite design for estimating the second-order model. This strategy assigns the common random numbers to the cubic points and a center, and the antithetic variates to the axial points and one center. An appropriate selection of the axial points allows this strategy to achieve variance reduction in estimation of first and second order coefficients of a given model. Under certain conditions, the proposed method yields better estimates for the model parameters than independent streams. In estimation of the intercept, independent random number assignment strategy shows better results than the proposed strategy. The simulation experiments on the (s, S) inventory system supports such results.