Comparative Study of Metamodeling and Sampling Design for Expensive and Semi-Expensive Simulation Models Under Uncertainty

Parnianifard, Amir; Azfanizam, A. S.; Ariffin, Mohd Khairol Anuar Mohd; Ismail, Mohd Idris Shah

doi:10.2139/ssrn.3399318

Cited by 14 publications

(31 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To overcome such computational difficulties, researchers have applied surrogate-based learning methods (e.g. polynomial regression, GP, and radial basis function) [35]- [37].…”

Section: Controlmentioning

confidence: 99%

“…queues, operations, and networks), manufacturing, medicine and biology, engineering, computer science, electronics, transportation, and logistics, see [3], [5], [36], [38]- [40]. However, several studies have systematically illustrated the applications of surrogate-based optimization algorithms [36], [37], [41]- [43].…”

Section: Controlmentioning

confidence: 99%

“…Here, we follow [37], [67], [68] and inspire Taguchi's overview of robust design [69] for dealing with uncertainty as a source of variability in the model. However, we expand…”

Section: Uncertainty Managementmentioning

confidence: 99%

See 2 more Smart Citations

A New Simulation-based Robust Optimal Design of FOPID Controller for Five Bar Linkage Robot in a Cyber-physical System

Parnianifard¹,

Wuttisittikulkij²,

Zemouche³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

This paper aims to further increase the reliability of optimal results by setting the simulation conditions to be as close as possible to the real or actual operation to create a Cyber-Physical System (CPS) view for the installation of the Fractional-Order PID (FOPID) controller. For this purpose, we consider two different sources of variability in such a CPS control model. The first source refers to changeability of a target of control model (multiple setpoints) because of environmental noise factors and the second source refers to an anomaly in sensors that is raised in a feedback loop. We develop a new approach to optimize two objective functions under uncertainty including signal energy control and response error control while obtaining the robustness among the source of variability with the lowest computational cost. A new hybrid surrogate-metaheuristic approach is developed using Particle Swarm Optimization (PSO) to update the Gaussian Process (GP) surrogate for a sequential improvement of the robust optimal result. The application of efficient global optimization is extended to estimate surrogate prediction error with less computational cost using a jackknife leave-one-out estimator. This paper examines the challenges of such a robust multi-objective optimization for FOPID control of a five-bar linkage robot manipulator. The results show the applicability and effectiveness of our proposed method in obtaining robustness and reliability in a CPS control system by tackling required computational efforts.

show abstract

“…To overcome such computational difficulties, researchers have applied surrogate-based learning methods (e.g. polynomial regression, GP, and radial basis function) [35]- [37].…”

Section: Controlmentioning

confidence: 99%

Section: Controlmentioning

confidence: 99%

See 1 more Smart Citation

A New Simulation-based Robust Optimal Design of FOPID Controller for Five Bar Linkage Robot in a Cyber-physical System

Parnianifard¹,

Wuttisittikulkij²,

Zemouche³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The term indicates a mathematical approximation that models the behavior of another model [8,12]. The objective of metamodeling is to reduce the computational cost of the simulation model during the optimization process [9,14].…”

Section: Literature Reviewmentioning

confidence: 99%

“…This makes computer simulation very tedious and impractical to run thousands of simulations for thorough design space exploration, sensitivity analysis and optimization. Therefore, This computational limitation of simulation modeling can be overcome by incorporating metamodels [6,8,9].…”

Section: Introductionmentioning

confidence: 99%

Simulation Metamodeling Approach to Complex Design of Garment Assembly Lines

Bongomin

Mwasiagi

Nganyi

et al. 2020

Preprint

View full text Add to dashboard Cite

The today competitive advantage of Ready-made garment industries depends on the ability to improve the efficiency and effectiveness of resource utilization. Ready-made garment industries have long historically adopted fewer technological and process advancement as compared to automotive, electronics and semiconductor industries. Simulation modeling of garment assembly line system has attracted a number of researchers as one way for insightful analysis of system behaviour and improving its performance. However, most of simulation studies have considered ill-defined experimental design which cannot fully explore the assembly line design alternatives and does not uncover the interaction effects of the input variables. Simulation metamodeling is an approach to assembly line design which has recently been of interest to so many researchers. However, its application in garment assembly line design has never been well explored. In this paper, simulation metamodeling of trouser assembly line with 72 operations has been demonstrated. The linear regression metamodel technique with resolution-V design was used. The effects of five factors: bundle size, job release policy, task assignment pattern, machine number and helper number on the production throughput of the trouser assembly line were studied. The increase of 28.63% of the production throughput was achieved for the best factors’ setting of the metamodel.

show abstract

Developing a two‐stage stochastic programming model in tolerance optimization using a guard band approach

Krenek

2021

Engineering Reports

View full text Add to dashboard Cite

An alternative way of looking at tolerance optimization is presented through a stochastic perspective by accounting for both manufacturing process uncertainty as well as uncertain market conditions. This can be accomplished by looking at a two‐stage stochastic programming model where the decision maker takes a course of action in the first stage, such as setting the product selling prices based on product quality, after which a random event occurs, leading to a recourse decision in the second stage. Production processes face uncertainties which impact product characteristics. The recourse decisions are the specification limits and upper and lower guard bands are necessary to monitor the process. This leads to the binning of a product based on its quality, uncertain market demand, and uncertain production process. The selling prices, the specification limits and the guard bands are then determined based on these uncertain conditions. The model presented is based on chance‐constrained programming, utilizing sample average approximation methods to solve the problem.

show abstract

Comparative Study of Metamodeling and Sampling Design for Expensive and Semi-Expensive Simulation Models Under Uncertainty

Cited by 14 publications

References 0 publications

A New Simulation-based Robust Optimal Design of FOPID Controller for Five Bar Linkage Robot in a Cyber-physical System

A New Simulation-based Robust Optimal Design of FOPID Controller for Five Bar Linkage Robot in a Cyber-physical System

Simulation Metamodeling Approach to Complex Design of Garment Assembly Lines

Developing a two‐stage stochastic programming model in tolerance optimization using a guard band approach

Contact Info

Product

Resources

About