On the efficiency of RESTART for multidimensional state systems

RESTART (Repetitive Simulation Trials After Reaching Thresholds) is a widely applicable accelerated simulation technique that allows the evaluation of extremely low probabilities. The focus of this article is on providing guidelines for achieving a high efficiency in a simulation with RESTART. Emphasis is placed on the choice of the importance function, that is, the function of the system state for determining when retrials are made. A heuristic approach which is shown to be effective for some systems is proposed for this choice. A two-queue tandem network is used to illustrate the efficiency achieved following these guidelines. The importance function chosen in this example shows that an appropriate choice of the importance function leads to an efficient simulation of a system with multidimensional state space. Also presented are sufficient conditions for achieving asymptotic efficiency, and it is shown that they are not very restrictive in RESTART simulation.

[1]  Jan-Kees C. W. van Ommeren,et al.  On the importance function in splitting simulation , 2002, Eur. Trans. Telecommun..

[2]  T. Kuhlmann,et al.  CASE STUDIES ON MULTIDIMENSIONAL RESTART SIMULATIONS , 1998 .

[3]  J. Villen-Altamirano.  RESTART METHOD FOR THE CASE WHERE RARE EVENTS CAN OCCUR IN RETRIALS FROM ANY THRESHOLD , 1998 .

[4]  Perwez Shahabuddin.  Rare event simulation in stochastic models , 1995, WSC '95.

[5]  F. Schreiber,et al.  The restart/lre method for rare event simulation , 1996, Proceedings Winter Simulation Conference.

[6]  Michael R. Frater,et al.  Fast simulation of buffer overflows in tandem networks ofGI/GI/1 queues , 1994, Ann. Oper. Res..

[7]  Paul Glasserman,et al.  Multilevel Splitting for Estimating Rare Event Probabilities , 1999, Oper. Res..

[8]  J. Townsend,et al.  Efficient rare event simulation using DPR for multidimensional parameter spaces , 1998 .

[9]  P. Glasserman,et al.  A large deviations perspective on the efficiency of multilevel splitting , 1998, IEEE Trans. Autom. Control..

[10]  J. Keith Townsend,et al.  Efficient simulation of delay in TCP/IP networks using DPR-based splitting , 2002, 2002 IEEE International Conference on Communications. Conference Proceedings. ICC 2002 (Cat. No.02CH37333).

[11]  Kishor S. Trivedi,et al.  Implementation of Importance Splitting Techniques in Stochastic Petri Net Package , 2000, Computer Performance Evaluation / TOOLS.

[12]  P. Tsoucas.  Rare events in series of queues , 1992 .

[13]  Manuel Villén-Altamirano,et al.  Enhancement of the Accelerated Simulation Method RESTART by Considering Multiple Thresholds , 1994 .

[14]  Christian Kelling.  A framework for rare event simulation of stochastic Petri nets using “RESTART” , 1996, Winter Simulation Conference.

[15]  José Villén-Altamirano,et al.  Analysis of restart simulation: Theoretical basis and sensitivity study , 2002, Eur. Trans. Telecommun..

[16]  J. Keith Townsend,et al.  The theory of direct probability redistribution and its application to rare event simulation , 1999, TOMC.

[17]  C. Gorg,et al.  COMPARISON AND OPTIMIZATION OF RESTART RUN TIME STRATEGIES , 1998 .

[18]  Carmelita Görg,et al.  The RESTART/LRE method for rare event simulation , 1996, WSC.

[19]  Maurizio Naldi,et al.  A comparison of the GEVT and RESTART techniques for the simulation of rare events in ATM networks , 1998, Simul. Pract. Theory.

[20]  J.P.C. Kleijnen,et al.  Importance sampling in systems simulation : A practical failure? , 1979 .

[21]  Paul Glasserman,et al.  Splitting for rare event simulation: analysis of simple cases , 1996, Winter Simulation Conference.

[22]  José Villén-Altamirano,et al.  RESTART: a straightforward method for fast simulation of rare events , 1994, Proceedings of Winter Simulation Conference.

[23]  Ward Whitt,et al.  The Asymptotic Efficiency of Simulation Estimators , 1992, Oper. Res..

[24]  Dirk P. Kroese,et al.  Efficient simulation of a tandem Jackson network , 1999, TOMC.

[25]  Dirk P. Kroese,et al.  On the entrance distribution in RESTART simulation , 1999 .

[26]  Jean Walrand,et al.  A quick simulation method for excessive backlogs in networks of queues , 1989 .

[27]  C. Görg,et al.  Simulating rare event details of ATM delay time distributions with RESTART/LRE , 1999 .

[28]  José Villén-Altamirano,et al.  Rare event RESTART simulation of two-stage networks , 2007, Eur. J. Oper. Res..

[29]  Dirk P. Kroese,et al.  A comparison of RESTART implementations , 1998, 1998 Winter Simulation Conference. Proceedings (Cat. No.98CH36274).

[30]  A. J. Bayes.  Statistical Techniques for Simulation Models , 1970, Aust. Comput. J..