Systems and methods for presenting results of experiments

• US 9,007,375 B2
• Filed: 09/29/2010
• Issued: 04/14/2015
• Est. Priority Date: 09/29/2010
• Status: Active Grant

• Current Assignee: MoreSteam.com LLC
• Original Assignee: MoreSteam.com LLC
• Inventors: Hathaway, William M., Skrivanek, Smita Kulkarni
• Primary Examiner(s): Amini, Javid A
• Application Number:
  US12/893,884
• Publication Number:
  US 20120075322A1
• Time in Patent Office: 1,658 Days
• Field of Search: 345/440, 345/440.2, 345/419
• US Class Current: 345/440
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs

Abstract

A method of automatically analyzing data from at least one data set including a plurality of process factors of interest and a process output of interest to determine the relationship between the factors of interest and the output of interest at a given significance level and preserving model hierarchy. The method includes the steps of calculating the effects of the factors of interest against the process output of interest, developing a model including the significant factors of interest and respective estimated coefficients and omitting the insignificant factors of interest, generating a representation of the model, and generating at least one graphical representation of the factors of interest.

Claims

1. A method of automatically analyzing experimental data from at least one data set including a plurality of process factors of interest and a process output of interest to model the relationship between the factors of interest and the output of interest at a given significance level and preserving model hierarchy such that the model includes the factors of interest involved in a significant interaction, the method comprising:

calculating, using a computer, effects of the factors of interest on the process output of interest, including

performing an analysis of variance test to calculate a significance of the individual effect of each factor of interest and interactions between two or more of the factors of interest,

calculating coefficients and effect estimates for each factor of interest, and

calculating model comparison statistics,

comparing the significance calculated for each factor of interest and each interaction to the given significance level and identifying a factor of interest or an interaction as having a significant effect if the significance calculated for the factor of interest or the interaction satisfies the given significance level and identifying a factor of interest or an interaction as having an insignificant effect if the significance calculated for the factor of interest or the interaction does not satisfy the given significance level;

developing, using the computer, a recommended model including each factor of interest and interaction identified as having a significant effect and each factor of interest involved in an interaction identified as having a significant effect multiplied by the coefficient for the factor of interest and omitting each factor of interest and interaction identified as having an insignificant effect, the recommended model representing the relationship between significant factors of interest and the process output of interest;

generating a representation of the model; and

generating at least one graphical representation of the factors of interest, the graphical representation providing a first representation corresponding to the factors of interest and interactions identified as having a significant effect, including factors of interest involved in a significant interaction, and a second representation corresponding to the factors of interest and interactions identified as having an insignificant effect.

2. The method of claim 1, further comprising automatically determining the relationship between the factors of interest and a second output of interest at a given significance level and preserving model hierarchy.

3. The method of claim 1, wherein the at least one graphical representation of the factors of interest includes a half normal graph.

4. The method of claim 1, wherein the at least one graphical representation of the factors of interest includes a Pareto chart.

5. The method of claim 1, further comprising displaying, at least one graphical representation of the factors of interest, the factors of interest identified as having a significant effect in a first color and displaying the factors of interest identified as having an insignificant effect in a second color, the first color different from the second color.

6. The method of claim 1, further comprising displaying, at least one graphical representation of the factors of interest, the factors of interest identified as having a significant effect in a first shape and displaying the factors of interest identified as having an insignificant effect in a second shape, the first shape different from the second shape.

7. The method of claim 1, further comprising displaying, at least one graphical representation of the factors of interest, the factors of interest identified as having a significant effect in a first color and a first shape and displaying the factors of interest identified as having an insignificant effect in a second color and a second shape, the first color different from the second color and the first shape different from the second shape.

8. The method of claim 1, further comprising grouping the factors of interest identified as having a significant effect together on the at least one graphical representation of the factors of interest.

9. The method of claim 1, wherein the method is performed by an add-in of spreadsheet program.

10. The method of claim 1, wherein the model and the at least one graphical representation of the factors of interest are displayed on a spreadsheet.

11. Non-transitory computer-readable medium encoded with a plurality of processor-executable instructions for automatically analyzing experimental data from at least one data set including a plurality of process factors of interest and a process output of interest to model the relationship between the factors of interest and the output of interest at a given significance level and preserving model hierarchy such that the model includes the factors of interest involved in a significant interaction, the instructions comprising:

calculating effects of the factors of interest on the process output of interest, including

performing an analysis of variance test to calculate a significance of the individual effect of each factor of interest and interactions between two or more of the factors of interest,

calculating coefficients and effect estimates for each factor of interest, and

calculating model comparison statistics;

comparing the significance calculated for each factor of interest and each interaction to the given significance level and identify a factor of interest or an interaction as having a significant effect if the significance calculated for the factor of interest or the interaction satisfies the given significance level and identify a factor of interest or an interaction as having an insignificant effect if the significance calculated for the factor of interest or the interaction does not satisfy the given significance level;

developing a recommended model including each factor of interest and interaction identified as having a significant effect and each factor of interest involved in an interaction identified as having a significant effect multiplied by the coefficient for the factor of interest and omitting each factor of interest and interaction identified as having an insignificant effect, the recommended model representing the relationship between significant factors of interest and the process output of interest;

generating a representation of the model; and

generating at least one graphical representation of the factors of interest, the graphical representation providing a first representation corresponding to the factors of interest and interactions identified as having a significant effect, including factors of interest involved in a significant interaction, and a second representation corresponding to the factors of interest and interactions identified as having an insignificant effect.

12. The computer-readable medium of claim 11, further comprising instructions for automatically determining the relationship between the factors of interest and a second output of interest at a given significance level and preserving model hierarchy.

13. The computer-readable medium of claim 12, further comprising instructions for developing a second model for the second output of interest including the significant factors of interest and respective estimated coefficients and omitting the insignificant factors of interest.

14. The computer-readable medium of claim 12, further comprising instructions for displaying the model in a first worksheet and displaying the second model in a second worksheet.