This article uses a sequentialized experimental design to select simulation input combinations for global optimization …/output data of the simulation model (computer code). This design and analysis adapt the classic "expected improvement" (EI) in …

This paper studies simulation-based optimization with multiple outputs. It assumes that the simulation model has one …

This paper presents a novel heuristic for constrained optimization of random computer simulation models, in which one … of the simulation outputs is selected as the objective to be minimized while the other outputs need to satisfy … prespecified target values. Besides the simulation outputs, the simulation inputs must meet prespecified constraints including the …

. As a novel metamodel we introduce intrinsic Kriging, for either deterministic or random simulation. For deterministic … simulation we study the famous 'e fficient global optimization' (EGO) method, substituting intrinsic Kriging for universal … Kriging. For random simulation we investigate a state-of-the-art two-stage algorithm accounting for heteroscedastic variances …

This chapter surveys two methods for the optimization of real-world systems that are modelled through simulation. These … experiment; this design fixes the input combinations of the simulation model. These regression models uses a sequence of local …-estimated through sequential designs. "Robust" optimization may use RSM or Kriging, and accounts for uncertainty in simulation inputs …

This paper uses a sequentialized experimental design to select simulation input combinations for global optimization …/output data of the simulation model (computer code). This paper adapts the classic "expected improvement" (EI) in "efficient …

simulated system. EGO treats the simulation model as a black-box, and balances local and global searches. In deterministic … simulation, EGO uses ordinary Kriging (OK), which is a special case of universal Kriging (UK). In our EGO variant we use … intrinsic Kriging (IK), which eliminates the need to estimate the parameters that quantify the trend in UK. In random simulation …

An important goal of simulation is optimization of the corresponding real system. We focus on simulation models with …, we treat the simulation model as a black box. We assume that the simulation is computationally expensive; therefore, we … use an inexpensive metamodel (approximation, emulator, surrogate) of the simulation model. A popular metamodel type is a …

models with multiple random responses.Such models arise in simulation-based optimization with multivariate outputs. This … the simulation outputs are feasible, and whether any constraints are binding. The paper applies the new procedure to both … a synthetic example and an inventory simulation; the empirical results are encouraging …