Comparison of Kriging-based algorithms for simulation optimization with heterogeneous noise

Many discrete simulation optimization techniques are unsuitable when the number of feasible solutions is large, or when the simulations are time-consuming. For problems with low dimensionality, Kriging-based algorithms may be used: in recent years, several algorithms have been proposed which extend the traditional Kriging-based methods (assuming deterministic outputs) to problems with noise. Our objective in this paper is to compare the relative performance of a number of these algorithms on a set of well-known test functions, assuming different patterns of heterogeneous noise. The conclusions and insights obtained may serve as a useful guideline for researchers aiming to apply Kriging-based methods to solve engineering and/or business problems, and may be useful in the development of future Kriging-based algorithms.

[1]  Noel A Cressie,et al.  Statistics for Spatial Data. , 1992 .

[2]  James R. Luedtke,et al.  Models and formulations for multivariate dominance-constrained stochastic programs , 2015 .

[3]  Barry L. Nelson,et al.  Discrete Optimization via Simulation , 2015 .

[4]  J. Yin,et al.  Kriging metamodel with modified nugget-effect: The heteroscedastic variance case , 2011, Comput. Ind. Eng..

[5]  Mike Preuss,et al.  High-Dimensional Model-Based Optimization Based on Noisy Evaluations of Computer Games , 2012, LION.

[6]  Barry L. Nelson,et al.  Chapter 17 Selecting the Best System , 2006, Simulation.

[7]  Lihua Sun,et al.  Balancing Exploitation and Exploration in Discrete Optimization via Simulation Through a Gaussian Process-Based Search , 2014, Oper. Res..

[8]  Jie Xu,et al.  Efficient discrete optimization via simulation using stochastic kriging , 2012, Proceedings Title: Proceedings of the 2012 Winter Simulation Conference (WSC).

[9]  Inneke Van Nieuwenhuyse,et al.  Simulation optimization in inventory replenishment: a classification , 2015 .

[10]  Nando de Freitas,et al.  A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning , 2010, ArXiv.

[11]  N. Zheng,et al.  Global Optimization of Stochastic Black-Box Systems via Sequential Kriging Meta-Models , 2006, J. Glob. Optim..

[12]  Loo Hay Lee,et al.  Simulation optimization via kriging: a sequential search using expected improvement with computing budget constraints , 2013 .

[13]  Averill M. Law,et al.  Simulation Modeling and Analysis , 1982 .

[14]  Peter Auer,et al.  Finite-time Analysis of the Multiarmed Bandit Problem , 2002, Machine Learning.

[15]  Ahmed Saif,et al.  Cold supply chain design with environmental considerations: A simulation-optimization approach , 2016, Eur. J. Oper. Res..

[16]  Alan J. Mayne,et al.  Towards Global Optimisation 2 , 1976 .

[17]  Joe Naoum-Sawaya,et al.  Simulation-optimization approaches for water pump scheduling and pipe replacement problems , 2015, Eur. J. Oper. Res..

[18]  Donald R. Jones,et al.  A Taxonomy of Global Optimization Methods Based on Response Surfaces , 2001, J. Glob. Optim..

[19]  Xi Chen,et al.  Stochastic kriging with biased sample estimates , 2014, ACM Trans. Model. Comput. Simul..

[20]  D. Ginsbourger,et al.  A benchmark of kriging-based infill criteria for noisy optimization , 2013, Structural and Multidisciplinary Optimization.

[21]  Barry L. Nelson,et al.  Using Ranking and Selection to "Clean Up" after Simulation Optimization , 2003, Oper. Res..

[22]  Warren B. Powell,et al.  The Correlated Knowledge Gradient for Simulation Optimization of Continuous Parameters using Gaussian Process Regression , 2011, SIAM J. Optim..

[23]  C. Lemieux.  Monte Carlo and Quasi-Monte Carlo Sampling , 2009 .

[24]  Graham Coates,et al.  A simulation model to enable the optimization of ambulance fleet allocation and base station location for increased patient survival , 2015, Eur. J. Oper. Res..

[25]  Barry L. Nelson,et al.  Stochastic kriging for simulation metamodeling , 2008, 2008 Winter Simulation Conference.

[26]  Jack P. C. Kleijnen,et al.  Kriging Metamodeling in Simulation: A Review , 2007, Eur. J. Oper. Res..

[27]  Victor Picheny,et al.  Noisy kriging-based optimization methods: A unified implementation within the DiceOptim package , 2014, Comput. Stat. Data Anal..

[28]  Warren B. Powell,et al.  The Knowledge-Gradient Policy for Correlated Normal Beliefs , 2009, INFORMS J. Comput..

[29]  Barry L. Nelson,et al.  Selecting the best system when systems are revealed sequentially , 2007 .

[30]  Xi Chen,et al.  The effects of common random numbers on stochastic kriging metamodels , 2012, TOMC.

[31]  Jack P. C. Kleijnen,et al.  Robustness of Kriging when interpolating in random simulation with heterogeneous variances: Some experiments , 2005, Eur. J. Oper. Res..

[32]  Jack P. C. Kleijnen,et al.  Simulation-optimization via Kriging and bootstrapping: a survey , 2014, J. Simulation.

[33]  MICHAEL C. FU,et al.  Techniques for optimization via simulation: an experimental study on an (s,S) inventory system , 1997 .

[34]  Andy J. Keane,et al.  Engineering Design via Surrogate Modelling - A Practical Guide , 2008 .

[35]  Chun-Hung Chen,et al.  Simulation Budget Allocation for Further Enhancing the Efficiency of Ordinal Optimization , 2000, Discret. Event Dyn. Syst..

[36]  Victor Picheny,et al.  Quantile-Based Optimization of Noisy Computer Experiments With Tunable Precision , 2013, Technometrics.

[37]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..