Measuring change distance of a factor in a decision

• US 8,429,115 B1
• Filed: 12/23/2009
• Issued: 04/23/2013
• Est. Priority Date: 12/23/2009
• Status: Expired due to Fees

• Current Assignee: Decision Lens, Inc.
• Original Assignee: Decision Lens, Inc.
• Inventors: Adams, William James Louis, Saaty, Daniel Lowell
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): Seck, Ababacar
• Application Number:
  US12/646,312
• Time in Patent Office: 1,217 Days
• Field of Search: 706/59
• US Class Current: 706/59
• CPC Class Codes: G06N 7/00 Computing arrangements based on specific mathematical models

Abstract

An analytic network process (ANP) storage memory stores an ANP weighted supermatrix representing an ANP model. A processor is in communication with the ANP storage memory. The processor fixes each one of plural nodes one at a time in the ANP model and measures a row sensitivity of each of the nodes as it is fixed, the row sensitivity being measured using one or both of (i) a predetermined increase value and (ii) a predetermined decrease value. Also, it scores how much synthesized alternatives scores found by doing ANP row sensitivity on each of the fixed nodes change from original values thereof, to generate a value indicating a change distance for each of the nodes. A same proportionality is maintained in the stored ANP weighted supermatrix as the fixed node is changed by the predetermined increase and/or decrease values.

Claims

1. An apparatus comprising:

an analytic network process (ANP) storage memory that stores an ANP weighted supermatrix representing an ANP model; and

a processor in communication with the ANP storage memory, the processor being configured to facilitate

fixing each one of plural nodes one at a time in the ANP model and measuring a row sensitivity of each one of the plural nodes as it is fixed, the row sensitivity being measured using one or both of (i) a predetermined increase value and (ii) a predetermined decrease value;

scoring how much synthesized alternatives scores found by doing ANP row sensitivity on each of the plural fixed nodes change from original values thereof, to generate a value indicating a change distance for each one of the plural nodes;

maintaining a same proportionality in the stored ANP weighted supermatrix as the fixed node is changed by the one or both of the predetermined increase and decrease values; and

utilizing a weighted average of different metrique calculations of the alternative scores on the space where a metrique is defined according to;

x=(x₁, x₂, . . . , x_n) y=(y₁, y₂, . . . , y_n)

are two vectors in Rⁿ′

, where

Rⁿ is real n-space with the ANP model,

x_i is a starting alternative score for alternative i, and y_i is an ending alternative score for the alternative i, and

d;

Rⁿ×

Rⁿ→

R a continuous function is a metrique calculation iff for all x,yε

Rⁿ

d(x,y)≧

0, and

d(x,x)=0.

2. The apparatus of claim 1, wherein the row sensitivity uses both the predetermined increase and decrease values, to measure the effect on the synthesized alternatives scores that is caused by changing an importance of each one of the fixed plural nodes by a same amount.

3. The apparatus of claim 1, further comprising comparing the synthesized alternatives scores for each one of the plural nodes in the ANP model to the original values thereof, to determine the relative influence of each one of the plural nodes with respect to the ANP model.

4. The apparatus of claim 1, wherein the alternatives scores for each of the plural fixed nodes are synthesized from an induced family of ANP models formed by row perturbations of each of the fixed nodes in the stored ANP weighted supermatrix.

5. The apparatus of claim 1, further comprising:

an output unit in communication with the processor and configured to output, for a display to a user, the value indicating a change distance for each one of the plural nodes.

6. The apparatus of claim 1, further comprising

an input unit configured to input, from an input device, pairwise comparisons, ANP ratings, or ANP client data, which are transformed into priority vectors and stored into the ANP weighted supermatrix, the pairwise comparisons representing a judgment of priority between ANP alternatives in the pair, the ANP ratings representing a rating of a choice, and the ANP client data representing real world values.

7. A method, comprising:

storing, in an analytic network process (ANP) storage memory, an ANP weighted supermatrix representing an ANP model;

in a processor in communication with the ANP storage memory, storing, in an analytic network process (ANP) storage memory, an ANP weighted supermatrix representing an ANP model;

fixing each one of plural nodes one at a time in the ANP model and measuring a row sensitivity of each one of the plural nodes as it is fixed, the row sensitivity being measured using one or both of (i) a predetermined increase value and (ii) a predetermined decrease value;

scoring how much synthesized alternatives scores found by doing ANP row sensitivity on each of the plural fixed nodes changes from original values thereof, to generate a value indicating a change distance for each one of the plural nodes;

maintaining a same proportionality in the stored ANP weighted supermatrix as the fixed node is changed by the one or both of the predetermined increase and decrease values and

utilizing a weighted average of different metrique calculations of the alternative scores on the space Rⁿ, where a metrique is defined according to;

x=(x₁, x₂, . . . , x_n) y=(y₁, y₂, . . . , y_n)

are two vectors in Rⁿ′

, where

Rⁿ is real n-space with the ANP model,

x_i is a starting alternative score for alternative i, y_i is an ending alternative score for the alternative i, and

d;

Rⁿ×

Rⁿ→

R a continuous function is a metrique calculation iff for all x, yε

Rⁿ

d(x,y)≧

0, and

d(x,x)=0.

8. The method of claim 7, wherein the row sensitivity uses both the predetermined increase and decrease values, to measure the effect on the synthesized alternatives scores that is caused by changing an importance of each one of the fixed plural nodes by a same amount.

9. The method of claim 7, further comprising comparing the synthesized alternatives scores for each one of the plural nodes in the ANP model to the original values thereof, to determine the relative influence of each one of the plural nodes with respect to the ANP model.

10. The method of claim 7, wherein the alternatives scores for each of the plural fixed nodes are synthesized from an induced family of ANP models formed by row perturbations of each of the fixed nodes in the stored ANP weighted supermatrix.

11. The method of claim 7, further comprising:

outputting, to an output unit in communication with the processor, for display to a user, the value indicating the change distance for each one of the plural nodes.

12. The method of claim 7, further comprising:

inputting, from an input device, pairwise comparisons, ANP ratings, or ANP client data, which are transformed into priority vectors and stored into the ANP weighted supermatrix, the pairwise comparisons representing a judgment of priority between ANP alternatives in the pair, the ANP ratings representing a rating of a choice, and the ANP client data representing real world values.

13. A non-transitory computer-readable storage medium encoded with a computer executable instructions, wherein execution of said computer executable instructions by one or more processors causes a computer to perform the steps of:

storing, in an analytic network process (ANP) storage memory, an ANP weighted supermatrix representing an ANP model;

fixing each one of plural nodes one at a time in the ANP model and measuring a row sensitivity of each one of the plural nodes as it is fixed, the row sensitivity being measured using one or both of (i) a predetermined increase value and (ii) a predetermined decrease value;

scoring how much synthesized alternatives scores found by doing ANP row sensitivity on each of the plural fixed nodes changes from original values thereof, to generate a value indicating a change distance for each one of the plural nodes;

maintaining a same proportionality in the stored ANP weighted supermatrix as the fixed node is changed by the one or both of the predetermined increase and decrease values; and

utilizing a weighted average of different metrique calculations of the alternative scores on the space Rⁿ, where a metrique is defined according to;

x=(x₁, x₂, . . . , x_n) y=(y₁, y₂, . . . , y_n)

are two vectors in Rⁿ′

, where

Rⁿ is real n-space with the ANP model,

x_i is a starting alternative score for alternative i, y_i is an ending alternative score for the alternative i, and

d;

Rⁿ×

Rⁿ→

R a continuous function is a metrique calculation iff for all x, yε

Rⁿ

d(x,y)≧

0, and

d(x,x)=0.

14. The computer-readable storage medium of claim 13, wherein the row sensitivity uses both the predetermined increase and decrease values, to measure the effect on the synthesized alternatives scores that is caused by changing an importance of each one of the fixed plural nodes by a same amount.

15. The computer-readable storage medium of claim 13, further comprising comparing the synthesized alternatives scores for each one of the plural nodes in the ANP model to the original values thereof, to determine the relative influence of each one of the plural nodes with respect to the ANP model.

16. The computer-readable storage medium of claim 13, wherein the alternatives scores for each of the plural fixed nodes are synthesized from an induced family of ANP models formed by row perturbations of each of the fixed nodes in the stored ANP weighted supermatrix.

17. The computer readable storage medium of claim 13, further comprising instructions for:

outputting, to an output unit in communication with the processor, for display to a user, the value indicating the change distance for each one of the plural nodes.

18. The computer readable storage medium of claim 13, further comprising instructions for:

inputting, from an input device, pairwise comparisons, ANP ratings, or ANP client data, which are transformed into priority vectors and stored into the ANP weighted supermatrix, the pairwise comparisons representing a judgment of priority between ANP alternatives in the pair, the ANP ratings representing a rating of a choice, and the ANP client data representing real world values.