Decision management system providing qualitative account/customer assessment via point in time simulation

• US 20040107132A1
• Filed: 11/24/2003
• Published: 06/03/2004
• Est. Priority Date: 03/05/1998
• Status: Active Grant

• Current Assignee: CGI Technologies and Solutions Incorporated (CGI Group)
• Original Assignee: American Management Systems, Incorporated (CGI, Inc.)
• Inventors: Ho, Richard, Honarvar, Laurence, Burt, Len
• Granted Patent: US 7,076,475 B2
• Time in Patent Office: Days
• US Class Current: 705/11
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/0637 Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0204 Market segmentation
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Abstract

A computer-implemented decision management system which provides qualitative client assessment via point in time simulation. More specifically, the decision management system (a) simulates the effect of a strategy by applying the strategy to client data, and (b) tracks what type of client traveled through a respective decision point in the strategy during the simulation. Simulation parameters can be selected without technical intervention via, for example, a GUI and a relational data model. Such parameters can include which customers and which client data sample will be used for the simulation, and which inbound events will trigger the strategy. Moreover, detailed and aggregate results can be consolidated to determine expected results from the simulated strategies, and traditional and OLAP reporting facilities can be applied to analyze/view the consolidated results.

Claims

1. A method for applying strategy decisioning criteria to a client population of an organization, comprising:

associating each client in the client population with a category, wherein each category is comprised of a grouping of clients based upon at least a value to the organization for the grouping of clients as compared to other groupings of clients, and wherein different rules are applied in executing decisions for at least some inbound events associated with clients depending upon the category in which the respective client is associated;

creating a plurality of test groups for a plurality of categories by randomly assigning clients to a test group in each client'"'"'s respective category, wherein different strategies are utilized for applying the different rules for executing decisions;

evaluating client performance in each category across different test groups; and

refining the strategies to be utilized for applying different rules to improve client performance according to results of the client performance evaluation.

2. The method of claim 1, further comprising the steps of:

moving certain clients between categories when the outcome of an accumulation of particular decisions associated with a respective client results in changing the value to the organization attributed to the client to one that is associated with a different category; and

tracking the movement of clients between categories,

wherein strategies are refined such that clients change categories in a manner that further improves client performance as compared to an original implementation of the respective strategies.

3. The method of claim 1, wherein strategies are based upon decision tree structures, and the same tree structure is applied across each category.

4. The method of claim 3, wherein each test group experiences a different tree structure.

5. The method of claim 4, wherein the step of refining strategies is performed by implementing hybrid strategies that combine paths of decision tree structures from different test groups for different categories.

6. The method of claim 5, wherein client performance is evaluated by tagging paths through a strategy, and the hybrid strategies that are implemented utilize the best individual performance for each category across the plurality of test groups.

7. The method of claim 1, wherein the step of evaluating client performance occurs by periodically aggregating client performance data into different permutations of numerically comparable performance metrics.

8. The method of claim 1, wherein clients are customers of a credit card company, and categories correspond to different customer credit levels.

9. A method for applying strategy decisioning criteria to a client population of an organization, comprising:

associating each client in the client population with a category, wherein each category is comprised of a grouping of clients based upon at least a value to the organization for the grouping of clients as compared to other groupings of clients, and wherein different rules are applied in executing decisions for at least some inbound events associated with clients depending upon the category in which the respective client is associated;

creating a plurality of test groups for a plurality of categories by randomly assigning clients to a test group in each client'"'"'s respective category, wherein different strategies are utilized for applying the different rules for executing decisions; and

moving certain clients between categories when the outcome of an accumulation of particular decisions in different test groups associated with a respective client results in changing the value to the organization attributed to the client to one that is associated with a different category.

10. The method according to claim 9, further comprising the steps of:

tracking the movement of clients between categories; and

refining strategies such that clients change categories in a manner that further improves client performance as compared to an original implementation of the respective strategies.

11. The method of claim 10, wherein strategies are based upon decision tree structures, and the same tree structure is applied across each category.

12. The method of claim 11, wherein each test group experiences a different tree structure.

13. The method of claim 10, wherein the step of refining strategies is performed by implementing hybrid strategies that combine paths of decision tree structures from different test groups for different categories.

14. The method of claim 13, wherein client performance is evaluated by tagging paths through a strategy, and the hybrid strategies that are implemented utilize the best individual performance for each category across the plurality of test groups.

15. The method of claim 14, wherein client performance is evaluated by periodically aggregating client performance data into different permutations of numerically comparable performance metrics.

16. The method of claim 10, wherein clients are customers of a credit card company, and categories correspond to different customer credit levels.