Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates
Mathematics and Computers in Simulation - IMACS sponsored Special issue on the second IMACS seminar on Monte Carlo methods
An effective screening design for sensitivity analysis of large models
Environmental Modelling & Software
Uncertainty in the environmental modelling process - A framework and guidance
Environmental Modelling & Software
Environmental Modelling & Software
Extension and evaluation of sensitivity analysis capabilities in a photochemical model
Environmental Modelling & Software
Algebraic sensitivity analysis of environmental models
Environmental Modelling & Software
Management Option Rank Equivalence (MORE) - A new method of sensitivity analysis for decision-making
Environmental Modelling & Software
Environmental Modelling & Software
How to avoid a perfunctory sensitivity analysis
Environmental Modelling & Software
A methodology for the design and development of integrated models for policy support
Environmental Modelling & Software
Environmental Modelling & Software
Dosage planning for type 2 diabetes mellitus patients using Indexing HDMR
Expert Systems with Applications: An International Journal
Environmental Modelling & Software
Environmental Modelling & Software
An approximation method to model multivariate interpolation problems: Indexing HDMR
Mathematical and Computer Modelling: An International Journal
Sampling strategies in density-based sensitivity analysis
Environmental Modelling & Software
Hi-index | 0.00 |
The high dimensional model representation (HDMR) method is a set of tools which can be used to construct a fully functional metamodel and to calculate variance based sensitivity indices very efficiently. Extensions to the existing set of random sampling (RS)-HDMR tools have been developed in order to make the method more applicable for complex models with a large number of input parameters as often appear in environmental modelling. The HDMR software described here combines the RS-HDMR tools and its extensions in one Matlab package equipped with a graphical user interface (GUI). This makes the HDMR method easily available for all interested users. The performance of the GUI-HDMR software has been tested in this paper using two analytical test models, the Ishigami function and the Sobol' g-function. In both cases the model is highly non-linear, non-monotonic and has significant parameter interactions. The developed GUI-HDMR software copes very well with the test cases and sensitivity indices of first and second order could be calculated accurately with only low computational effort. The efficiency of the software has also been compared against other recently developed approaches and is shown to be competitive. GUI-HDMR can be applied to a wide range of applications in all fields, because in principle only one random or quasi-random set of input and output values is required to estimate all sensitivity indices up to second order. The size of the set of samples is however dependent on the problem and can be successively increased if additional accuracy is required. A brief description of its application within a range of modelling environments is given.