Please use this identifier to cite or link to this item:
Files in This Item:
File SizeFormat 
Robust combination.pdf2,69 MBAdobe PDFView/Open
Title: Robust combination of the Morris and Sobol methods in complex multidimensional models
Authors: Garcia, Dorleta; Arostegui, Inmaculada; Prellezo, Raul
Abstract: Conducting global sensitivity analysis using variance decomposition methods in complex simulation models with many input factors is usually unaffordable. An alternative is to first apply a screening method to reduce the number of input factors and then apply a variance decomposition method to the reduced model. However, usually selection of input factors is not done robustly and convergence of the screening method is not ensured. We propose two new criteria, a criterion that mimics the visual selection of the input factors and a convergence criterion. In the application of the criteria to a complex model, the Morris screening method has needed 200 trajectories to converge and the visual criterion has outperformed other existing criteria. Our proposal ensures a robust combination of the Morris and the Sobol methods that provides an objective and automatic method to select the most important input factors with a feasible computing load to achieve convergence.
Keywords: Convergence criterion; Global sensitivity analysis; Morris elementary effect method; Selection criterion; Sobol variance decomposition method; GLOBAL SENSITIVITY-ANALYSIS; ENVIRONMENTAL-MODELS; SIMULATION; CONVERGENCE; OUTPUTS; DESIGN
Issue Date: 2019
Type: Article
DOI: 10.1016/j.envsoft.2019.104517
ISSN: 1364-8152
E-ISSN: 1873-6726
Funder: Department of Agriculture, Fishing and Food from the Basque Government [227390, 289257]
Department of Education, Language Policy and Culture from the Basque Government [IT1294-19]
Spanish Ministry of Economy and Competitiveness MINECO
FEDEREuropean Union (EU) [MTM2016-74931-P]
FEDER (BCAM Severo Ochoa excellence accreditation) [SEV-2017-0718]
Appears in Publication types:Artículos científicos

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.