Using Ordinal Optimization Approach to Improve Efficiency of Selection Procedures

2010

Self Cite

John and Chen (IEEE Trans Reliab 55(1): [135][136][137][138][139][140][141][142][143][144][145][146][147][148] 2006) propose an exact two-stage solution based on the Least Favorable Configuration (LFC) to the ranking and selection problem of determining the best system from k lognormal populations. Lognormal density is commonly used to model certain lifetimes in reliability and survival analysis. It is known that selection procedures that are developed based on the LFC are conservative and become inefficient when the number of systems is large. We propose to take into account the differences of sample values within the parameter of interest when determining sample sizes, which can significantly increase the efficiency of the selection procedures. We also sequentialize the selection procedure and provide a procedure for estimating the constant needed to apply the solution. An experimental performance evaluation demonstrates the validity and efficiency of the enhanced lognormal selection procedure.

show abstract

Section: Discussionmentioning

confidence: 99%

“…In practice, the true values of β ik and β il are unknown; we need to use the sample valueŝ β ik andβ il to estimate d il . See Chen (2004) for details.…”

Section: Motivationmentioning

confidence: 99%

See 1 more Smart Citation

An enhanced lognormal selection procedure

2010

Self Cite

show abstract

“…Using the ordinal optimization (Ho et al 1992) approach, Chen (2004) proposes taking into account the differences of sample means and using normal approximation to compute the critical constant. That is, we are more interested in whether a given design is better than the others rather than the accuracy of the performance measures in our solution technique.…”

Section: A Sequential Indifference-zone Selection Proceduresmentioning

confidence: 99%

“…Consequently, Rinott_s procedure is conservative. Chen (2004) and Chen and Kelton (2005) approximate the difference of true means by the difference of sample means, which can significantly improve the efficiency of selection procedures. However, these procedures do not guarantee P(CS) !…”

Section: Introductionmentioning

confidence: 99%

Comparison with a Standard via All-Pairwise Comparisons

2006

Self Cite

We develop two sequential procedures to compare a finite number of designs with respect to a single standard. The goal is to identify a good design, and ensure that the standard is chosen when other alternatives are not better than the standard. We give preference to the standard since there are costs and time involved when replacing the standard. These procedures can be used when the expected performance of the standard is known or unknown, when variances across designs are unequal, and with the variance reduction technique of common random numbers. An experimental performance evaluation demonstrates the validity and efficiency of these sequential procedures.

show abstract

Some insights of using common random numbers in selection procedures

2012