Kriging metamodeling in constrained simulation optimization: an explorative study

Abstract

This paper describes two experiments exploring the potential of the Kriging methodology for constrained simulation optimization. Both experiments study an (s, S) inventory system with the objective of finding the optimal values of s and S. The goal function and constraints in these two experiments differ, as does the approach to determine the optimum combination predicted by the Kriging model. The results of these experiments indicate that Kriging offers opportunities for solving constrained optimization problems in stochastic simulation; future research will focus on further refining the methodology.

Untitled

“…[14] suggested that some neighbors of the predicted optimal point should be simulated after the kriging process to improve solution. This step is called the sensitivity analysis [36].…”

“…LHDs are particularly well suited for kriging because they can cover the design space [14]. LHD was first described for computer experiments by McKay et al [31].…”

Sensitivity Analysis in Discrete Event Simulation Using Design of Experiments

“…It is also recommended that the reader study this topic further as well [18,43,46]. In this sense, the use of Kriging metamodeling for simulation has established itself in the scientific simulation community, which can be seen in [46,47,48,49], thus demonstrating its promising research field.…”

Statistical testing of optimality conditions in multiresponse simulation-based optimization

“…Example 2 simulates the well-known ðs; SÞ inventory model, assuming that out-of-stock penalty costs are so hard to quantify that instead a prespecified service (or fill rate) constraint must be satisfied; see [18]. Example 3 integrates this ðs; SÞ inventory model with a production model; see [4]. The first two examples use parametric bootstrapping for the gradients that are estimated using a local experimental design with only the center replicated.…”