|
Sign In to gain access to subscriptions and/or personal tools.
|
Importance Functions for RESTART Simulation of Highly-Dependable Systems
José Villén-Altamirano
Departamento de Matemática Aplicada E.U. Informática, Universidad Politécnica de Madrid calle Arboleda s/n 28031 Madrid, Spain, jvillen{at}eui.upm.es
RESTART is a widely applied accelerated simulation technique that allows the evaluation of very low probabilities. In this method a number of simulation retrials are performed when the process enters regions of the state space where the chance of occurrence of the rare event is higher. Formulas for evaluating the optimal number of regions and retrials have been provided in previous papers. Guidelines were also provided for obtaining a suitable function, the importance function, used to define the regions. This paper provides a simple importance function that can be useful for RESTART simulation of models of many highly dependable systems. Some eXamples from the literature illustrate the application of this importance function. Steady-state unavailability of balanced systems is accurately estimated within short computational times, and also the unavailability of an unbalanced system but with much more computational effort.
Key Words: Rare-event simulation • RESTART • highly-dependable systems
References
- Alexopoulos, C. and B.C. Shultes. 2001. Estimating reliability measures for highly-dependable Markov systems, using balanced likelihood ratios, IEEE Transactions on Reliability 50(3): 265- 80.[CrossRef][Web of Science]
- Nicola, V., P. Shahabuddin, and M.K. Nakayama. 2001. Techniques for fast simulation of models of highly dependable systems, IEEE Transactions on Reliability 50(3): 246-64.[CrossRef][Web of Science]
- Villén-Altamirano, M. and J. Villén-Altamirano. 2006. On the efficiency of RESTART for multidimensional state systems, ACM Transactions on Modelling and Computer Simulation 16(3): 251-79.[CrossRef]
- Kahn, H. and T.E. Harris. 1951. Estimation of particle transmission by random sampling, National Bureau of Standards Applied Mathematics Series 12: 27-30.
- Villén-Altamirano, M. and J. Villén-Altamirano. 1991. RESTART: a method for accelerating rare event simulations. In Proceedings of the 13th International Teletraffic Congress, Copenhagen. North Holland, pp. 71-6.
- Villén-Altamirano, M. and J. Villén-Altamirano. 2002. Analysis of RESTART simulation: theoretical basis and sensitivity study. European Transactions on Telecommunications 13(4): 373-386.[Web of Science]
- L'ecuyer, P., V. Demers, and B. Tuffin. 2007. Rare events, splitting, and quasi-Monte Carlo, to appear in ACM Transactions on Modelling and Computer Simulation 17(2): article 9.[CrossRef]
- Villén-Altamirano, J. 1998. RESTART method for the case where rare events can occur in retrials from any threshold. International Journal of Electronic Communications (AEU) 52(3): 183-90.
- Goyal, A., P. Shahabuddin, P. Heidelberger, et al. 1992. A unified framework for simulating Markovian models of highly dependable systems, IEEE Transactions on Computing 41(1): 36-51.[CrossRef]
- Tuffin, B. and K.S. Trivedi. 2000. Implementation of importance splitting techniques in stochastic Petri net package. In 11th International Conference; TOOLS 2000, Schaumburg, Il, USA, and Lecture Notes in Computer Science 1786: 216-29.
- Zimmermann, A., M. Knoke, A. Huck, and G. Hommel. 2006. Towards version 4.0 of TimeNET. In Proceedings of the 13th GI/ITG Conference on Measurement, Modeling, and Evaluation of Computer and Communication Systems, Nürnberg, Germany, pp. 477-80.
SIMULATION, Vol. 83, No. 12,
821-828 (2007)
DOI: 10.1177/0037549707081257
 CiteULike  Complore  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter What's this?
|
|
