Lognormal Ordinary Kriging Metamodel in Simulation Optimization

ABSTRACT
This  paper  presents  a  lognormal  ordinary  kriging  (LOK)  metamodel  algorithm  and  its  application  to optimize a stochastic simulation problem. Kriging models have been developed as an interpolation method in geology. They have been successfully used for the deterministic simulation optimization (SO) problem. In recent  years,  kriging  metamodeling  has  attracted  a growing  interest  with  stochastic  problems.SO researchers  have  begun  using  ordinary  kriging  through  global  optimization  in  stochastic  systems.  The goals  of  this  study  are  to  present  LOK  metamodel  algorithm  and  to  analyze  the  result  of  the  application step-by-step.  The  results  show  that  LOK  is  a  powerful  alternative  metamodel  in  simulation  optimization when the data are too skewed.