METHOD AND SYSTEM FOR ANALYZING AND OPTIMIZING DISTRIBUTION OF WORK FROM A PLURALITY OF QUEUES

• US 20120004948A1
• Filed: 07/02/2010
• Published: 01/05/2012
• Est. Priority Date: 07/02/2010
• Status: Active Grant

• Current Assignee: Fis Financial Compliance Solutions LLC (Fidelity National Information Services Incorporated)
• Original Assignee: Fis Financial Compliance Solutions LLC (Fidelity National Information Services Incorporated)
• Inventors: Taintor, Robert C., Leibon, Gregory David
• Granted Patent: US 8,412,563 B2
• Time in Patent Office: Days
• US Class Current: 705/7.29
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/06316 Sequencing of tasks or work
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0206 Price or cost determination based on market factors

Abstract

A method and system for analyzing and optimizing the distribution of work from a plurality of queues includes storing historical data in a database. The historical data may include sets of different types of data items, each of the data items having an associated monetary value. The method further includes assigning a relative score to each of the data items within the sets and calculating a cumulative monetary value for each relative score, the cumulative monetary value being the summation of the monetary values for all items having relative scores up to and including the relative score. The method further includes determining the maximum cumulative monetary value calculated and its corresponding relative score. In one embodiment, the corresponding relative score is then used to determine which items to work in a set of queues that has not yet been worked.

Claims

1. A computer-implemented method for analyzing investment in time and resources in reviewing potential fraud activities, comprising acts of:

(a) storing historical data in a database, the historical data comprising a first plurality of potential fraud items having a first type and a second plurality of potential fraud items having a second type, each of the first and second plurality of potential data items having a fraud determination and an associated monetary value;

(b) assigning a relative score to each of the potential fraud items within the first and second plurality of fraud items;

(c) calculating a cumulative monetary value for each relative score, the cumulative monetary value being the summation of associated monetary values for all potential fraud items having relative scores up to and including the relative score; and

(d) determining a maximum cumulative monetary value and corresponding relative score.

2. The method of claim 1, further comprising an act of using the corresponding relative score to determine which potential fraud items in a set to analyze.

3. The method of claim 2, further comprising an act of using the corresponding relative score to determine which potential fraud items in a plurality of sets to analyze.

4. The method of claim 2, wherein a historical fraud analyzer performs acts (a)-(d) and a current fraud analyzer performs an act of using the corresponding relative score to determine which potential fraud items in the set to analyze.

5. The method of claim 1, further comprising an act of generating a graph of the cumulative monetary values as a function of relative score, prior to performing the act of determining the maximum cumulative value.

6. The method of claim 1, wherein the associated monetary value of potential fraud items having a positive fraud determination is a function of money saved in preventing a fraud event and a cost to investigate the potential fraud item.

7. The method of claim 6, wherein the associated monetary value of potential fraud items includes variable value, fixed costs and variable costs.

8. The method of claim 7, wherein the fixed costs include a system implementation cost, a system maintenance cost, and a human resource cost.

9. The method of claim 7, wherein the variable costs include a cost analyst'"'"'s time, a false action rate and a cost of a false action.

10. The method of claim 7, wherein the variable value includes a savings if a value transfer is prevented, a probability preventing a value transfer when taking action, a rate of false inaction, and a cost associated with inaction.

11. The method of claim 1, wherein the associated monetary value of potential fraud items having a negative fraud determination is a function of a cost to investigate the potential fraud item.

12. The method of claim 1, wherein the act of assigning a relative score to each of the potential fraud items further comprises acts of:

assigning first numerical scores to each potential fraud item within the first plurality of fraud items;

assigning second numerical scores to each potential fraud item within the second plurality of potential fraud items;

applying a normalizing function to the second numerical scores to generate normalized numerical scores; and

using the first numerical scores and the normalized numerical scores as the relative scores.

13. The method of claim 1, wherein the act of assigning a relative score to each of the potential fraud items further comprises an act of using a fraud detection model to detect fraud.

14. The method of claim 13, wherein the fraud detection model includes a normalizing function.

15. The method of claim 13, wherein the fraud detection model performs an act of generating a score for each of the potential fraud items.

16. The method of claim 15, further comprising the act of approximating an anticipated return on investment value associated with analyzing for each of the plurality of potential fraud items.

17. The method of claim 16, wherein the act of determining the maximum cumulative monetary value and corresponding relative score further comprises an act of using the anticipated return on investment values.

18. The method of claim 13, wherein a first fraud detection model is used to assign relative scores to each potential fraud item within the first plurality of fraud items and a second fraud detection model is used to assign relative scores to each potential fraud item within the second plurality of fraud items.

19. The method of claim 18, wherein the first fraud detection model includes a first normalizing function and the second fraud detection model includes a second normalizing function.

20. The method of claim 18, further comprising:

repeating acts (b)-(d) for a plurality of first fraud detection models to generate a plurality of maximum cumulative values and corresponding relative scores; and

comparing the plurality of first fraud detection models using the plurality of maximum cumulative values and corresponding relative scores.

21. The method of claim 12, wherein the act of assigning a relative score to each of the potential fraud items further comprises acts of:

applying a series of normalizing functions to the second numerical scores to generate a series of normalized numerical scores; and

generating a series of relative scores using the first numerical scores and the series of normalized numerical scores; and

wherein the method further comprises the acts of;

repeating acts (c) and (d) for each of the series of relative scores to generate a series maximum cumulative monetary values and corresponding relative scores; and

determining a greatest maximum cumulative monetary value and corresponding normalizing function.

22. The method of claim 21, further comprising generating a graph of the series of maximum cumulative monetary values as a function of normalizing factor, prior to the act of determining the greatest maximum cumulative value.

23. The method of claim 2, where in the act of using the corresponding relative score to determine which potential fraud items in the set to analyze further comprises selecting the potential fraud items in the set having a numerical score greater than or equal to the corresponding relative score for further analysis.

24. The method of claim 1, wherein the first type of fraud is selected from the group consisting of:

first party fraud, third party fraud and collusive fraud.

25. The method of claim 24, wherein the second type of fraud is different than the first type of fraud and is selected from the group consisting of:

first party fraud, third party fraud and collusive fraud.

26. The method of claim 13, further comprising the act of using the maximum cumulative monetary value to modify the fraud detection model.

27. The method of claim 26, wherein the fraud detection model is defined using a plurality of parameters and wherein the act of using the maximum cumulative monetary value to modify the fraud detection model further comprises deforming at least one of the plurality of parameters to change the maximum cumulative monetary value.

28. The method according to claim 1, wherein acts (a)-(d) are performed by a fraud detection system.

29. A system for analyzing investment in time and resources in reviewing potential fraud activities, comprising:

a database adapted to store historical data, the historical data comprising a first plurality of potential fraud items having a first type and a second plurality of potential fraud items having a second type, each of the first and second plurality of potential data items having a fraud determination and an associated monetary value;

a score generator adapted to assign a relative score to each of the potential fraud items within the first and second plurality of fraud items;

a calculator adapted to calculate a cumulative monetary value for each relative score, the cumulative monetary value being the summation of associated monetary values for all potential fraud items having relative scores up to and including the relative score; and

a component adapted to determine a maximum cumulative monetary value and corresponding relative score.

30. The system of claim 29, further comprising a queue generator adapted to determine, based on the corresponding relative score, an order of potential fraud items in a set to analyze.

31. The system of claim 29, further comprising a current fraud analyzer that is adapted to use the corresponding relative score to determine which potential fraud items in the set to analyze.

32. The system of claim 29, further comprising a component adapted to generate a graph of the cumulative monetary values as a function of relative score, prior to determining the maximum cumulative value.

33. The system of claim 29, wherein the associated monetary value of potential fraud items having a positive fraud determination is a function of money saved in preventing a fraud event and a cost to investigate the potential fraud item.

34. The system of claim 33, wherein the associated monetary value of potential fraud items includes variable value, fixed costs and variable costs.

35. The system of claim 34, wherein the fixed costs include a system implementation cost, a system maintenance cost, and a human resource cost.

36. The system of claim 34, wherein the variable costs include a cost analyst'"'"'s time, a false action rate and a cost of a false action.

37. The system of claim 34, wherein the variable value includes a savings if a value transfer is prevented, a probability preventing a value transfer when taking action, a rate of false inaction, and a cost associated with inaction.

38. The system of claim 29, wherein the associated monetary value of potential fraud items having a negative fraud determination is a function of a cost to investigate the potential fraud item.

39. The system of claim 29, wherein the score generator is adapted to:

assign first numerical scores to each potential fraud item within the first plurality of fraud items;

assign second numerical scores to each potential fraud item within the second plurality of potential fraud items;

apply a normalizing function to the second numerical scores to generate normalized numerical scores; and

use the first numerical scores and the normalized numerical scores as the relative scores.

40. The system of claim 29, wherein the score generator uses a fraud detection model to detect fraud.

41. The system of claim 40, wherein the fraud detection model includes a normalizing function.

42. The system of claim 40, wherein the fraud detection model performs an act of generating a score for each of the potential fraud items.

43. The system of claim 42, further comprising a component adapted to approximate an anticipated return on investment value associated with analyzing for each of the plurality of potential fraud items.

44. The system of claim 43, wherein the component adapted to determine the maximum cumulative monetary value and corresponding relative score further comprises a component that determines an anticipated return on investment values.

45. A computer-readable medium comprising computer-executable instructions that, when executed on a processor of a server, perform a method for analyzing investment in time and resources in reviewing potential fraud activities, comprising acts of:

(a) storing historical data in a database, the historical data comprising a first plurality of potential fraud items having a first type and a second plurality of potential fraud items having a second type, each of the first and second plurality of potential data items having a fraud determination and an associated monetary value;

(b) assigning a relative score to each of the potential fraud items within the first and second plurality of fraud items;

(c) calculating a cumulative monetary value for each relative score, the cumulative monetary value being the summation of associated monetary values for all potential fraud items having relative scores up to and including the relative score; and

(d) determining a maximum cumulative monetary value and corresponding relative score.

46. A computer system comprising:

a memory;

a display; and

a processor adaptively coupled to the memory and programmed to render a graphical user interface for analyzing investment in time and resources in reviewing potential fraud activities, the graphical user interface comprising;

a first display area configured to display a plurality of fraud items, wherein the first display area is adapted to display an ordered list of fraud items based on a monetary value determined for each of the respective plurality of fraud items.

47. The computer system according to claim 46, wherein the graphical user interface is adapted to display to a user an indication of at least one monetary value determined for at least one respective fraud item.

48. The computer system according to claim 46, wherein the graphical user interface is adapted to display a cumulative monetary value for at least one of the plurality of fraud items.

49. The computer system according to claim 46, wherein the graphical user interface is adapted to display cumulative monetary values for at least two respective ones of the plurality of fraud items, and wherein the graphical user interface is adapted to indicate a maximum monetary value among the displayed cumulative monetary values.