Gradient Estimation for Multicomponent Maintenance Systems with Age-Replacement Policy

We consider multicomponent maintenance systems with an F-failure group age-replacement policy: it keeps failed components idling until F components are failed and then replaces all failed components together with the nonfailed components whose age has passed the critical threshold age θn for components of type n. With each maintenance action, costs are associated. We derive various unbiased gradient estimators based on the measure-valued differentiation approach for the gradient of the average cost. Each estimator has its own domain of applicability. We also compare the performance of our gradient estimators when applied to stochastic optimization with other general gradient-free methods.

[1]  Bernd Heidergott,et al.  A Note on Gradient Estimation for Maintenance Systems , 2007, IEEE Transactions on Automatic Control.

[2]  Paul Glasserman,et al.  Structural Conditions for Perturbation Analysis Derivative Estimation: Finite-Time Performance Indices , 1991, Oper. Res..

[3]  Bernd Heidergott,et al.  A Perturbation Analysis Approach to Phantom Estimators for Waiting Times in the G/G/1 Queue , 2010, Discret. Event Dyn. Syst..

[4]  Leyuan Shi,et al.  An application of perturbation analysis to a replacement problem in maintenance theory , 1993, WSC '93.

[5]  F. Vázquez-Abad,et al.  Measure-Valued Differentiation for Markov Chains , 2008 .

[6]  Pierre L'Ecuyer,et al.  Two Approaches for Estimating the Gradient in Functional Form , 1993, Proceedings of 1993 Winter Simulation Conference - (WSC '93).

[7]  Paul Glasserman,et al.  Structural conditions for perturbation analysis of queueing systems , 1991, JACM.

[8]  R. Wets,et al.  Weak Convergence of Probability Measures Revisited , 1987 .

[9]  Paul Glasserman,et al.  Performance continuity and differentiability in Monte Carlo optimization , 1988, WSC '88.

[10]  J. Spall.  Multivariate stochastic approximation using a simultaneous perturbation gradient approximation , 1992 .

[11]  Jian-Qiang Hu,et al.  Conditional Monte Carlo: Gradient Estimation and Optimization Applications , 2012 .

[12]  J. Hammersley,et al.  Monte Carlo Methods , 1965 .

[13]  Michael C. Fu,et al.  Conditional Monte Carlo , 1997 .

[14]  J. Spall.  A Stochastic Approximation Technique for Generating Maximum Likelihood Parameter Estimates , 1987, 1987 American Control Conference.

[15]  P. Glasserman,et al.  Smoothed perturbation analysis for a class of discrete-event systems , 1990 .

[16]  Paul Glasserman,et al.  Gradient Estimation Via Perturbation Analysis , 1990 .

[17]  P. L’Ecuyer.  Two approaches for estimating the gradient in functional form , 1993, WSC '93.

[18]  Arie Hordijk,et al.  Measure-Valued Differentiation for Stationary Markov Chains , 2006, Math. Oper. Res..

[19]  Rajan Suri,et al.  Infinitesimal perturbation analysis for general discrete event systems , 1987, JACM.

[20]  RAJAN.  Perturbation Analysis: The State of the Art and Research Issues Explained via the GI/G/l Queue , 2004 .

[21]  F. Vázquez-Abad,et al.  Measure valued differentiation for random horizon problems , 2006 .

[22]  Bernd Heidergott.  A weak derivative approach to optimization of threshold parameters in a multicomponent maintenance system , 2001, Journal of Applied Probability.

[23]  Georg Ch. Pflug,et al.  Gradient estimation for discrete-event systems by measure-valued differentiation , 2010, TOMC.

[24]  George Ch. Pflug,et al.  Optimization of Stochastic Models , 1996 .

[25]  Rommert Dekker,et al.  Optimal maintenance of multi-component systems: a review , 2008 .

[26]  Reuven Y. Rubinstein,et al.  Modern simulation and modeling , 1998 .

[27]  Y. Ho,et al.  Smoothed (conditional) perturbation analysis of discrete event dynamical systems , 1987 .

[28]  Rommert Dekker,et al.  Maintenance of light-standards—a case-study , 1998, J. Oper. Res. Soc..

[29]  Pierre L'Ecuyer,et al.  Functional Estimation with Respect to a Threshold Parameter via Dynamic Split-and-Merge , 1997, Discret. Event Dyn. Syst..

[30]  P. L’Ecuyer,et al.  ON THE LINEAR GROWTH OF THE SPLIT-AND-MERGE SIMULATION TREE FOR A MULTICOMPONENT AGE REPLACEMENT MODEL , 2007 .

[31]  V. Nollau.  Kushner, H. J./Clark, D. S., Stochastic Approximation Methods for Constrained and Unconstrained Systems. (Applied Mathematical Sciences 26). Berlin‐Heidelberg‐New York, Springer‐Verlag 1978. X, 261 S., 4 Abb., DM 26,40. US $ 13.20 , 1980 .

[32]  P. L'Ecuyer,et al.  Approximation and bounds in discrete event dynamic programming , 1983, The 23rd IEEE Conference on Decision and Control.

[33]  Xi-Ren Cao,et al.  Perturbation analysis of discrete event dynamic systems , 1991 .

[34]  S. Strickland.  Gradient/sensitivity estimation in discrete-event simulation , 1993, WSC '93.

[35]  Arie Hordijk,et al.  Derivatives of Markov Kernels and Their Jordan Decomposition , 2003 .

[36]  F. A. van der Duyn Schouten,et al.  Replacement policies for traffic control signals , 1998 .

[37]  Xi-Ren Cao,et al.  Convergence properties of infinitesimal perturbation analysis , 1988 .

[38]  P. L’Ecuyer,et al.  Functional estimation for a multicomponent age replacement model , 1999 .

[39]  J. Spall.  Implementation of the simultaneous perturbation algorithm for stochastic optimization , 1998 .

[40]  Reuven Y. Rubinstein,et al.  Sensitivity analysis of discrete event systems by the “push out” method , 1992, Ann. Oper. Res..

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

[42]  R. Suri,et al.  Perturbation analysis: the state of the art and research issues explained via the GI/G/1 queue , 1989, Proc. IEEE.

[43]  Raghu Pasupathy,et al.  A Testbed of Simulation-Optimization Problems , 2006, Proceedings of the 2006 Winter Simulation Conference.

[44]  Pierre L'Ecuyer,et al.  Processus de décision markoviens à étapes discrètes: Application à des problèmes de remplacement d'équipement , 1983 .

[45]  Yu-Chi Ho,et al.  Functional Estimation with Respect to a Threshold Parametervia Dynamic Split-and-Merge , 1997 .

[46]  Pierre Brémaud,et al.  On the pathwise computation of derivatives with respect to the rate of a point process: The phantom RPA method , 1992, Queueing Syst. Theory Appl..