Process modeling and optimization method and system

• US 8,364,610 B2
• Filed: 07/31/2007
• Issued: 01/29/2013
• Est. Priority Date: 04/08/2005
• Status: Active Grant

• Current Assignee: Caterpillar Incorporated
• Original Assignee: Caterpillar Incorporated
• Inventors: Grichnik, Anthony J., Seskin, Michael
• Primary Examiner(s): Lo, Suzanne
• Application Number:
  US11/882,189
• Publication Number:
  US 20080021681A1
• Time in Patent Office: 2,009 Days
• Field of Search: 703/2, 706/13, 706 19- 20, 706/12
• US Class Current: 706/12
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

A method is provided for model optimization. The method may include obtaining respective distribution descriptions of a plurality of input parameters to a model indicative of interrelationships between the input parameters and one or more output parameters. The method may also include specifying respective search ranges for the plurality of input parameters and simulating the model to determine a desired set of input parameters based on a zeta statistic of the model. Further, the method may include determining respective desired distributions of the input parameters based on the desired set of input parameters; determining significance levels of the input parameters in interacting with the output parameter based on the simulation and the desired distributions of the input parameters; and presenting the significance levels.

Claims

1. A computer-implemented method for model optimization, comprising:

obtaining respective distribution descriptions of a plurality of input parameters to a model, the model indicating interrelationships between the input parameters and one or more output parameters of a system design;

specifying respective search ranges for the plurality of input parameters;

simulating the model to determine a desired set of input parameters within the search ranges based on a zeta statistic of the model;

determining respective desired distributions of the input parameters based on the desired set of input parameters;

determining significance levels of the input parameters in interacting with the output parameters based on the simulation and on the desired distributions of the input parameters, including;

obtaining data records of the input parameters and the output parameters based on the simulation of the model;

building a multiple layer tree diagram based on the data records, the tree diagram including a trunk node representing the data records, a plurality of split points, and a plurality of layer nodes, wherein each split point corresponds to one of the input parameters and the layer nodes are coupled to the trunk node via the split points; and

determining the significance levels of the input parameters based on the order of the split points in relation to the trunk node; and

identifying, based on the significance levels of the input parameters, a parameter of the system design that has a threshold design tolerance;

presenting the determined significance levels, including;

displaying a graphical representation of the tree diagram to a user via a display device; and

outputting, via the graphical representation, a representation of the identified parameter of the system design that has the threshold design tolerance.

2. The computer-implemented method according to claim 1, wherein the zeta statistic ζ

is represented by;

3. The computer-implemented method according to claim 1, wherein building the tree diagram includes applying a Chi-square automatic interaction detection (CHAID) algorithm based on the data records.

4. The computer-implemented method according to claim 3, further including:

preparing a set of predictors for each input parameters;

performing a Chi-squared test to merge the set of predictors into a desired number of predictors;

calculating a p-value for the merged predictors; and

choosing an input parameter corresponding to a merged predictor with the smallest p-value as the highest level split point of the tree diagram.

5. The computer-implemented method according to claim 1, wherein simulating includes:

starting a genetic algorithm;

generating a candidate set of values of the input parameters;

providing the candidate set of values of the input parameters to the model to generate values of the one or more output parameters;

obtaining output parameter distributions based on the one or more output parameters;

calculating respective compliance probabilities of the one or more outputs; and

calculating a zeta statistic of the model.

6. The computer-implemented method according to claim 5, further including:

determining a minimum compliant probability from the respective compliant probabilities of the one or more output parameters; and

setting a goal function of the genetic algorithm to maximize a product of the zeta statistic and the minimum compliant probability, the goal function being set prior to starting the genetic algorithm.

7. The computer-implemented method according to claim 6, wherein the simulating further includes:

determining whether the genetic algorithm converges;

identifying the candidate set of input parameters as the desired set of input parameters if the genetic algorithm converges;

choosing a different candidate set of input parameters if the genetic algorithm does not converge; and

repeating the step of simulating to identify a desired set of input parameters based on the different candidate set of input parameters.

8. The computer-implemented method according to claim 7, further including:

selectively modifying the search ranges of the input parameters used in the simulation based on the determined significance levels of the input parameters.

9. A computer system, comprising:

a display device;

at least one input device; and

a processor configured to;

obtain respective distribution descriptions of a plurality of input parameters to a model, the model indicating interrelationships between the input parameters and one or more output parameters of a system design;

specify respective search ranges for the plurality of input parameters;

simulate the model to determine a desired set of input parameters within the search ranges based on a zeta statistic of the model;

determine respective desired distributions of the input parameters based on the desired set of input parameters;

determine significance levels of the input parameters in interacting with the output parameters based on the simulation and on the desired distribution of the input parameters, including;

obtaining data records of the input parameters and the output parameters based on the simulation of the model;

building a multiple layer tree diagram based on the data records, the tree diagram including a trunk node representing the data records, a plurality of split points, and a plurality of layer nodes, wherein each split point corresponds to one of the input parameters and the layer nodes are coupled to the trunk node via the split points; and

determining the significance levels of the input parameters based on the order of the split points in relation to the trunk node;

identify an input parameter having a significance level below a significance threshold corresponding to a threshold design tolerance; and

present, via a graphical user interface on the display device, an indication of the identified input parameter with the significance level below the threshold significance corresponding to the threshold design tolerance.

10. The computer system according to claim 9, wherein the zeta statistic ζ

is represented by;

11. The computer system according to claim 9, wherein the processor is configured to build the tree diagram by applying a Chi-square automatic interaction detection (CHAID) algorithm based on the data records.

12. The computer system according to claim 11, wherein the processor is further configured to:

prepare a set of predictors for each input parameters;

perform a Chi-squared test to merge the set of predictors into desired number of predictors;

calculate a p-value for all merged predictors; and

choose input parameter corresponding to a merged predictor with the smallest p-value as the highest level split point of the tree diagram.

13. The computer system according to claim 9, wherein, to simulate the model, the processor is configured to:

start a genetic algorithm;

generate a candidate set of values of the input parameters;

provide the candidate set of values of the input parameters to the model to generate values of the one or more output parameters;

obtain output distributions based on the one or more output parameters;

calculate respective compliance probabilities of the one or more output parameters; and

calculate a zeta statistic of the model.

14. The computer system according to claim 13, the processor is further configured to:

determine a minimum compliant probability from the respective compliant probabilities of the one or more output parameters; and

set a goal function of the genetic algorithm to maximize a product of the zeta statistic and the minimum compliant probability, the goal function being set prior to starting the genetic algorithm.

15. The computer system according to claim 14, wherein, to simulate the model, the processor is further configured to:

determine whether the genetic algorithm converges;

identify the candidate set of input parameters as the desired set of input parameters if the genetic algorithm converges;

choose a different candidate set of input parameters if the genetic algorithm does not converge; and

repeat the step of simulating to identify a desired set of input parameters based on the different candidate set of input parameters.

16. The computer system according to claim 15, wherein the processor is further configured to:

selectively modify the search ranges of the input parameters used in the simulation based on the determined significance levels of the input parameters.

17. A computer-implemented method for model optimization, comprising:

obtaining respective distribution descriptions of a plurality of input parameters to a model, the model indicating interrelationships between the input parameters and one or more output parameters of a system design;

specifying respective search ranges for the plurality of input parameters;

simulating the model to determine a desired set of input parameters within the search ranges based on a zeta statistic of the model;

determining respective desired distributions of the input parameters based on the desired set of input parameters;

determining significance levels of the input parameters in interacting with the output parameters based on the simulation and on the desired distributions of the input parameters, the determining including;

obtaining data records of the input parameters and the output parameters based on the simulation of the model;

building a multiple layer tree diagram based on the data records, the tree diagram including a trunk node representing the data records, a plurality of split points, and a plurality of layer nodes, wherein each split point corresponds to one of the input parameters and the layer nodes are coupled to the trunk node via the split points; and

determining the significance levels of the input parameters based on the order of the split points in relation to the trunk node;

determining input parameters that have significance levels below a significance threshold;

modifying the search ranges of the input parameters based on the determined significance levels; and

optimizing the model based on the input parameters determined to have significance levels below the significance threshold and on the modified search ranges.