Methods and systems for determining global sensitivity of a process

Patent metadata

Field	Value
Publication number	US-2016055125-A1
Application number	US-201514752268-A
Country	US
Kind code	A1
Filing date	Jun 26, 2015
Priority date	Jun 27, 2014
Publication date	Feb 25, 2016
Grant date	—

Abstract

Systems and methods for determining the sensitivity of a model to a factor are disclosed. A directional variogram corresponding to a response surface of the model is determined. The variogram is then output as an indication of the sensitivity of the model.

First claim

1 . A system for analyzing sensitivity of a model, comprising:
a computer processer; an input in communication with the computer processor; non-transitory memory in communication with the processor, the non-transitory memory storing computer executable instructions that when executed by a computer processor implement modules comprising:
an input module configured to receive data comprising a plurality of values of a model output and, for each value of the model output, corresponding values of input factors resulting in the model output;
a variogram construction module configured to construct a directional variogram based on the data received at the input module;
an output module configured to output the directional variogram as a representation of the model sensitivity.

2 . The system of claim 1 , wherein the variogram construction module is further configured to construct a directional variogram for each input factor.

3 . The system of claim 1 , wherein the output module is configured to display the variogram.

4 . The system of claim 1 , wherein the modules further comprise:
a model module configured to model a process.

5 . The system of claim 4 , wherein the modules further comprise:
a sampling module configured to generate a plurality of sample points for modeling the process.

6 . The system of claim 4 , wherein the model is configured to model an environmental system.

7 . A method for analyzing the sensitivity of a model, comprising:
receiving, by a computer, a plurality of sets of input factor values and for each set, a corresponding model output value; constructing, by a computer, a plurality of directional variograms based on the sets of input factor values and model output values; outputting the plurality of directional variograms as an indication of model sensitivity.

8 . The method of claim 7 , wherein the plurality of directional variograms comprises a variogram for each input factor in the sets of input factor values.

9 . The method of claim 8 , wherein outputting the plurality of directional variograms comprises displaying each of the variograms on a common graph.

10 . The method of claim 7 , further comprising:
generating a plurality of sample points to use as input factor values.

11 . The method of claim 7 , further comprising:
providing, by a computer, the plurality of sets of input factor values to a model and for each set of input factor values, recording an output value of the model.

12 . The method of claim 7 , wherein the model is an environmental model.

13 . The method of claim 7 , further comprising:
normalizing the plurality of variograms.

14 . The method of claim 13 , wherein normalizing the plurality of variograms comprises integrating each variogram from zero to a maximum value of an input factor.

15 . The method of claim 14 , wherein outputting the plurality of variograms comprises outputting a plurality of values for each variogram.

16 . The method of claim 15 , wherein the plurality of values comprises a value at ten percent of the factor range and fifty percent of the factor range.

17 . A method for analyzing the sensitivity of a process, comprising:
receiving, by a computer, a plurality of sets of input factor values and for each set, a corresponding process output value; constructing, by a computer, a plurality of directional variograms based on the sets of input factor values and process output values; outputting the plurality of directional variograms as an indication of process sensitivity.

18 . The method of claim 17 , wherein the plurality of directional variograms comprises a variogram for each input factor in the sets of input factor values.

19 . The method of claim 17 , further comprising:
generating a plurality of sample points to use as input factor values.

20 . The method of claim 17 , further comprising:
providing, by a computer, the plurality of sets of input factor values to a process and for each set of input factor values, recording an output value of the process.

Assignees

• Univ Arizona State

Inventors

• Razavi Seyed Saman
• Gupta Hoshin V

Classifications

• G06Q10/067 Primary

  Enterprise or organisation modelling · CPC title

• G06F30/20 Primary

  Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) · CPC title

• G06F17/18 Primary

  for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} · CPC title

• G06F17/5009

  Physics · mapped topic