Empirical model-building and response surface
Empirical model-building and response surface
A guide to simulation (2nd ed.)
A guide to simulation (2nd ed.)
First-order perturbation analysis of a simple multi-class finite source queue
Performance Evaluation
Queueing Systems: Theory and Applications
Variance property of discontinuous perturbation analysis
WSC '96 Proceedings of the 28th conference on Winter simulation
Simterpolations: estimating an entire queueing function from a single sample path
WSC '87 Proceedings of the 19th conference on Winter simulation
Discrete Event Dynamic Systems
Functional Estimation with Respect to a Threshold Parametervia Dynamic Split-and-Merge
Discrete Event Dynamic Systems
Infinitesimal perturbation analysis for queueing networks with general service time distributions
Queueing Systems: Theory and Applications
SIGMETRICS '83 Proceedings of the 1983 ACM SIGMETRICS conference on Measurement and modeling of computer systems
A max-algebra approach to modeling and simulation of tandem queueing systems
Mathematical and Computer Modelling: An International Journal
| Hi-index | 0.00 |
Weconsider queueing networks for which the performance measure J ( \theta ) depends on a parameter \theta, which can be a service time parameter or a buffer size,and we are interested in sensitivity analysis of J (\theta ) with respect to \theta . We introducea new method, called customer-oriented finite perturbation analysis(CFPA), which predicts J ( \theta + \Delta ) foran arbitrary, finite perturbation \Delta froma simulation experiment at \theta . CFPA can estimatethe entire performance function (by using a finite number ofchosen points and fitting a least-squares approximating polynomialto the observation) within one simulation experiment. We obtainCFPA by reformulating finite perturbation analysis (FPA) forcustomers. The main difference between FPA and CFPA is that theformer calculates the sensitivities of timing epochs of events,such as external arrivals or service time completions, whilethe latter yields sensitivities of departure epochs of customers.We give sufficient conditions for unbiasedness of CFPA. Numericalexamples show the efficiency of the method. In particular, weaddress sensitivity analysis with respect to buffer sizes andthereby give a solution to the problem for which perturbationanalysis was originally built.