Analysis of nonstationary stochastic simulations using classical time-series models

ABSTRACT

This article extends the use of classical autoregressive and moving average time-series models to the analysis of a variety of nonstationary discrete-event simulations. A thorough experimental evaluation shows that integrated and seasonal time-series models constitute very promising metamodels, especially for analyzing queueing system simulations under congested or cyclical traffic conditions. In some situations, stationarity-inducing transformations may be required before this methodology can be used. Our approach for efficient estimation of meaningful performance measures of selected responses in the target system is illustrated using a set of case studies taken from the simulation literature.

REFERENCES

• 1: Akaike, H. 1974. A new look at the statistical model identification. IEEE Trans. Autom. Contr. AC, 19, 716--723.
• 2: Bailey, N. T. J. 1957. Some further results in the non-equilibrium theory of a simple queue. J. R. Statist. Soc. B, 19, 326--333.
• 3: Bailey, N. T. J. 1964. The Elements of Stochastic Processes with Applications to the Natural Sciences. John Wiley & Sons, New York.
• 4: George Edward Pelham Box , Gwilym M. Jenkins, Time Series Analysis: Forecasting and Control, Prentice Hall PTR, Upper Saddle River, NJ, 1994
• 5: Brandão, R. M. 2006. Análise de resultados da simulações não estacionárias aperiódicas e cíclicas. Ph.D. thesis, Universidade dos Açores.
• 6: Brandão, R. M. and Porta Nova, A. M. O. 1999. An experimental evaluation of methods for simulation output analysis. In Proceedings of the European Simulation Symposium, G. Horton et al., Eds. The Society for Computer Simulation, Erlangen, 601--607.
• 7: Brandão, R. M. and Porta Nova, A. M. O. 2003a. Análise de resultados da simulação de filas de espera não stacionárias usando séries cronológicas. In Literacia e Estatística--Actas do X Congresso Anual da SPE, P. Brito et al., Eds. Sociedade Portuguesa de Estatística, 133--140.
• 8: Brandão, R. M. and Porta Nova, A. M. O. 2003b. Non-Stationary queue simulation analysis using time series. In Proceedings of the Winter Simulation Conference, S. Chick et al., Eds. IEEE, Piscataway, NJ, 408--413.
• 9: Chatfield, C. 2000. Time-Series Forecasting. Chapman & Hall/CRC, Boca Raton, FL.
• 10: Espasa, A. and Peña, D. 1995. The decomposition of forecast in seasonal ARIMA models. J. Forecasting 14, 565--583.
• 11: Fishman, G. S. 1971. Estimating sample size in computing simulation experiments. Manage. Sci. 18, 1, 21--38.
• 12: Ljung, G. M. and Box, G. E. P. 1978. On a measure of lack of fit in time series models. Biometrika 65, 67--72.
• 13: Nozari, A., Arnold, S. F., and Pegden, C. D. 1984. Control variates for multipopulation simulation experiments. IIE Trans. 16, 159--169.
• 14: Pankratz, A. 1983. Forecasting with Univariate Box-Jenkins Models: Concept and Cases. John Wiley & Sons, New York.
• 15: Peña, D., Tiao, G. C., and Tsay, R. S. 2001. A Course in Time Series Analysis. John Wiley & Sons, New York.
• 16: Schriber, T. J. and Andrews, R. W. 1984. ARMA-Based confidence intervals for simulation output analysis. Amer. J. Math. Manage. Sci. 4, 3 & 4, 345--373.
• 17: Sheth-Voss, P. A., Willemain, T. R., and Haddock, J. 2005. Estimating the steady-state mean from short transient simulations. Eur. J. Oper. Res. 162, 403--417.
• 18: Yuan, M. and Nelson, B. L. 1994. Autoregressive-output-analysis method revisited. Ann. Oper. Res. 53, 391--418.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.6 SIMULATION AND MODELING
      I.6.6 Simulation Output Analysis

Additional Classification:
  G. Mathematics of Computing
  G.3 PROBABILITY AND STATISTICS
      Subjects: Time series analysis

  I. Computing Methodologies
  I.6 SIMULATION AND MODELING
      I.6.8 Types of Simulation
          Subjects: Discrete event

General Terms:
Algorithms, Experimentation, Management, Performance

Keywords:
Discrete-event simulation, output analysis, simulation metamodels, time-series models