• Journal of Korean Institute of Industrial Engineers
• /
• v.19 no.1
• /
• pp.61-71
• /
• 1993

We focus on a way of combining the Monte Calro methods of antithetic variates and control variates to reduce the variance of the estimator of the mean response in a simulation experiment. Combined Method applies antithetic variates (partially) for driving approiate stochastic model components to reduce the vaiance of estimator and utilizes the correlations between the response and control variates. We obtain the variance of the estimator for the response analytically and compare Combined Method with control variates method. We explore the efficiency of this method in reducing the variance of the estimator through the port operations model. Combined Method shows a better performance in reducing the variance of estimator than methods of antithetic variates and control variates in the range from 6% to 8%. The marginal efficiency gain of this method is modest for the example considered. When the effective set of control variates is small, the marginal efficiency gain may increase. Though these results are from the limited experiments, Combined Method could profitably be applied to large-scale simulation models.

• Journal of the Korea Society for Simulation
• /
• v.14 no.3
• /
• pp.129-136
• /
• 2005

We develop a variance reduction technique applicable in one simulation experiment whose purpose is to estimate the parameters of a first order linear model. This method utilizes the control variates obtained during the course of simulation run under Schruben and Margolin's method (S-M method). The performance of this method is shown to be similar in estimating the main effects, and to be superior to S-M method in estimating the overall mean response in a given model. We consider that a proposed method may yield a better result than S-M method if selected control variates are highly correlated with the response at each design point.

• Journal of the Korea Society for Simulation
• /
• v.1 no.1
• /
• pp.1-16
• /
• 1992

This research develops two variance reduction methods for estimating the parameters of the experimental simulation model having multiple design points based on an approach focusing on reduction of the variances of the mean responses across multiple design points. The first method extends a combined approach of antithetic variates and control variates for a single design point to the multipopulation context with independent streams across the design points. The second method extends the same strategy in conjunction with the Schruben-Margolin method for improving the first method. We illustrate an example for implementing the second method. We expect these two approaches may improve the estimation of the parameters of interest compared with the control variates method.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1992.04b
• /
• pp.127-136
• /
• 1992

시뮬레이션 실험에서 시스템 성과에 대한 추정치의 정확도를 개선하기 위한 분산감소기법(Variance Reduction Technique)은 입력영역과 출력영역에 대한 것으로 나누어 볼 수 있다. 본 연구에서는 시스템 성과 추정량이 단일 변량인 경우에, 분산감소기법으로 많이 사용되는 Antithetic Variates방법과 Control Variates방법을 결합하여 응용가능한 시뮬레이션 실험설계기법을 제시하고 이 기법을 선택된 모형에 적용하여 시뮬레이션의 효율성을 분석하였다. 실험결과, 제안된 기법은 기존 방법들보다 추정치의 분산을 5%-8% 더 감소시켰다. 비록 제한된 실험결과이지만 이러한 효과는 대형 시뮬레이션의 경우에 적지 않으리라 기대된다. 특히 효과적인 Control Variates의 수가 적은 경우, 제안된 기법은 매우 효율적이다.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2005.05a
• /
• pp.45-49
• /
• 2005

본 연구는 시뮬레이션 반응변수가 입력 인자의 선형 1차식으로 표현된 경우에 인자의 파라미터를 효과적으로 추정하기위해 사용될 수 있는 분산감소기법을 제안하였다. 이 기법은 하나의 실험설계에 공통난수와 대조난수를 동시에 사용하는 Schruben과 Margolin의 방법과 시뮬레이션하는 도중에 얻어지는 통제변수를 활용하는 기법을 결합하는 방법으로 시뮬레이션의 효율성을 개선하고자 하였다. 시뮬레이션 결과 제안된 기법은 주어진 모형의 평균 반응치를 추정한 데는 S-M 기법보다 효과적이었으며 인자의 다른 파라미터를 추정하는 데는 S-M 기법과 비슷한 성과를 보이고 있다. 만일 시뮬레이션 과정에서 반응변수와 상관성이 높은 통제변수들을 선택할 수 있는 경우에는 제안된 기법이 S-M 기법보다 보다 파라미터 추정에 효과적일 것으로 판단된다.

• Proceedings of the Korea Society for Simulation Conference
• /
• 1992.10a
• /
• pp.8-8
• /
• 1992

Through a simulation experiment, often an experimenter is concerned with estimating the system parameters of the linear model consisting of m design points from the outputs oft the simulation model. To improve the estimation of the system parameters and reliability of these estimators, appropriate simulation techniques have been developed. For the first order linear model, Schruben and Margolin (1978) exploited the random number assignment rules which uses a combination of common random numbers and antithetic streams in a simulation experiment designed to estimate the system parameters when the design matrix of simulation model admits orthogonal blocking into two blocks. Nozari, Arnold and Pegden (1984) developed a method for appliying the method of control variates to the situation of the linear model having multiple design points. This talk deals with a different way of utilizing controls under the correlation induction strategy of Schruben and Margolin's to improve the simulation efficiency, and presents a procedure for obtaining the estimators of the system parameters analytically. Simulation results on a selected simulation model indicate a promising evidence that a proposed method may yield better results than Schruben and Margolin's method.