Selecting a Good Stochastic System for the Large Number of Alternatives

Efficient Approach for Selecting the Best Subset of Buffer Profile

“…To achieve this target Chen et al [19] proposed the optimal computing budget allocation (OCBA) approach that gives a large number of simulation samples to the designs that have a great effect on identifying the best design, whereas it gives a limited simulation sample for those designs that have little effect on identifying the best one. Most research on the same framework has focused on selecting the single best design, see Almomani and Abdul Rahman [20], Almomani and Abdul Rahman [21], Almomani et al [22], Almomani et al [23], Almomani and Alrefaei [24], Alrefaei and Almomani [25], and there has been no research involving subset selection. Chen et al [4], Chen et al [26] fill this gap by providing an efficient allocation approach for selecting the top m designs, known as (OCBA-m) approach.…”

On the optimal computing budget allocation problem for large scale simulation optimization

“…The objective of the OO procedure is to isolate a subset of good systems with high probability, then any simulation optimization procedure can be used to locate the optimal solution(s) from the isolated set. Many sequential selection procedures are proposed to select a good system when the number of alternatives is large, see Almomani and Rahman [8] , Alrefaei and Almomani [9] , Almomani and Alrefaei [10] . All these procedures are still focused on selecting a single best system or selecting a subset containing one of the best systems.…”

The Effect of Simulation Parameters on the Selection Approach

“…In fact, they are used in second stage, after the OO procedure reduces the number of alternatives, so that it will appropriate for the R&S procedures. Almomani and Abdul Rahman [1] has proposed a new sequential selection approach in order to solve the selection problems for a huge number of alternatives. In this approach, the first step is to use the OO procedure to select a subset that intersects with the set of the actual best m% system.…”

“…In this paper, we discuss the significant differences in efficiency of the selection approach by Almomani and Abdul Rahman [1] when different simulation parameters are applied. We focus on the parameters such as; the initial sample size (t 0 ), increment in simulation samples (∆), total budget (B), and the elapsed (execution) time (T ), and present the numerical illustrations for each parameter.…”

“…In Section 3, we review the Ranking and Selection procedures, whereas the Ordinal Optimization and Optimal Computing Budget Allocation are presented in Section 4. Next, we present the algorithm by Almomani and Abdul Rahman [1] in Section 5. The simulation issues and the efficiency of the selection approach are discussed together with a series of numerical examples in Section 6, a future research present in Section 7, and Section 8 concludes this study.…”