Sensitivity Analysis
With the help of the sensitivity analysis, the basis for a successful formulation of optimization or identification tasks is set up. Important parameters can be identified and possible optimization potential can be explored.
Method overview
Filtered linear correlation matrix
- Design variables with upper and lower bounds
- Design of Experiments (linear, quadratic, full factorial, central composite, D-optimal)
- Latin Hypercube Sampling, Monte Carlo Sampling
- Statistical postprocessing (analysis of variance, correlation analysis, principal components, coefficients of determination)
Features
Confidence intervals for estimating a coefficient of correlation of 0.5 using LHS
How many design evaluations are necessary for a sensitivity analysis?
The number of necessary design evaluations depends on the following factors:
- Sampling method used
- Number of variables
- Admissible confidence interval (error of the estimator for coefficients of correlation)
Focus
Coefficient of determination of a single response
The design space to be examined is specified by the definition of design variables with upper and lower bounds. By means of Latin Hypercube Sampling, a number of possible design realizations is produced and computed. As a result of sensitivity analysis, important parameters can be identified and possible optimization potential can be explored.
Postprocessing
Post processing window
The visualization of the results takes place in the statistic postprocessing. All important information is presented interactively within multiple windows which display visually the most important results at once.
Distributors
optiSLang is sold worldwide. A list of our distributors can be found here.
Your Contact
