Computer-implemented systems and methods for emerging warranty issues analysis

• US 7,912,772 B2
• Filed: 10/02/2006
• Issued: 03/22/2011
• Est. Priority Date: 09/30/2005
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Noal, Danny C., Henry, Robert Martin Jr., McLeod, Max J., Whear, Michael L.
• Primary Examiner(s): Merchant; Shahid R
• Application Number:
  US11/542,037
• Publication Number:
  US 20070088776A1
• Time in Patent Office: 1,632 Days
• Field of Search: 705/1.1, 705/7, 705/10, 705/35
• US Class Current: 705/35
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Abstract

Computer-implemented methods and systems for analyzing claims related to a warranty. Statistical analysis is performed in order to examine claim counts with respect to a time period (e.g., production period, claim period, etc.). A claim issue is detected based upon the statistical analysis with respect to the time period.

Claims

1. A computer-implemented method for analyzing warranty issues, comprising:

using statistical analysis, by one or more data processors, to detect one or more warranty issues by examining one or more claim counts within a time period;

using a robust estimator, by the one or more data processors to calculate claim intensities corresponding to the one or more claim counts;

estimating, by the one or more data processors, one or more seasonality factors using a logistic regression model, wherein the estimated seasonality factors correspond to variations in claim intensities relative to the time period;

using the estimated one or more seasonality factors to determine a type of warranty analysis to perform upon the one or more claims, by the one or more data processors, wherein warranty analysis types include a production period model analysis and a claim period model analysis;

performing, by the one or more data processors, emerging issue analysis using the warranty analysis, wherein the results of the emerging issues analysis are reconciled;

processing, by the one or more data processors, one or more alerts based upon the reconciled emerging issues analysis, wherein each alert has an associated alert index score; and

generating, by the one or more data processors, a graphical interface display including the reconciled emerging analysis results according to the associated index scores.

2. The method of claim 1, wherein the time period is a production period, sales period, time in service period, or claim period.

3. The method of claim 1, wherein the graphical interface is displayed on one screen or on multiple screens.

4. The method of claim 1, wherein the estimating one or more seasonality factors are multiplicative and account for variations in claim levels relative to a claim month.

5. The method of claim 1, further comprising:

displaying detailed information regarding why an alert received a particular alert index score, wherein the detailed information includes a graph that shows production month and months in service, and wherein the graph contains blocks that are filled in to represent where an actual claim count value met or exceeded a critical claim count value.

6. The method of claim 1, wherein the one or more warranty issues are detected within received warranty data including one or more units, and wherein an in-service date is not required to be known for each unit contained in the received warranty data.

7. The method of claim 6, wherein the in-service date identifies how long a product has been in service at a particular point in time.

8. The method of claim 6, wherein the statistical analysis estimates time in-service of units that do not have an in-service date in the received warranty data, and wherein the estimation of time in-service includes using a build date of a unit and a build-to-sale lag profile.

9. The method of claim 8, wherein the build-to-sale lag profile is a lognormal distribution of a number of days between a time when the unit was built and a time when the unit went into service, and wherein the build-to-sale lag profile is developed from units that have an in-service date in the data.

10. The method of claim 1, wherein using statistical analysis includes using call center data to detect one or more warranty issues.

11. The method of claim 1, wherein the statistical analysis includes performing an analysis on a cumulative number of claims received in a time period for products produced in a pre-specified previous number of time periods.

12. The method of claim 1, wherein the statistical analysis approximates distribution of C_t as a Poisson distribution;

wherein C_t is the cumulative number of claims received in the claim period t for products produced in the pre-specified previous number of period.

13. A computer-program product, tangibly embodied in a machine-readable storage medium, including instructions configured to cause a data processing apparatus to:

use statistical analysis to detect one or more warranty issues by examining one or more claim counts within a time period;

use a robust estimator to calculate claim intensities corresponding to the one or more claim counts;

estimate one or more seasonality factors using a logistic regression model, wherein the estimated seasonality factors correspond to variations in claim intensities relative to the time period;

use the estimated one or more seasonality factors to determine a type of warranty analysis to perform upon the one or more claims, wherein warranty analysis types include a production period model analysis and a claim period model analysis;

perform emerging issue analysis using the warranty analysis, wherein the results of the emerging issues analysis are reconciled;

process one or more alerts based upon the reconciled emerging issues analysis, wherein each alert has an associated alert index score; and

generate a graphical interface display including the reconciled emerging analysis results according to the associated index scores.

14. A system for analyzing warranty issues, comprising:

one or more processors;

a computer-readable storage medium containing instructions configured to cause the one or more processors to perform operations, including;

using statistical analysis to detect one or more warranty issues by examining one or more claim counts within a time period;

using a robust estimator to calculate claim intensities corresponding to the one or more claim counts;

estimating one or more seasonality factors using a logistic regression model, wherein the estimated seasonality factors correspond to variations in claim intensities relative to the time period;

using the estimated one or more seasonality factors to determine a type of warranty analysis to perform upon the one or more claims, wherein warranty analysis types include a production period model analysis and a claim period model analysis;

performing emerging issue analysis using the warranty analysis, wherein the results of the emerging issues analysis are reconciled;

processing one or more alerts based upon the reconciled emerging issues analysis, wherein each alert has an associated alert index score; and

generating a graphical interface display including the reconciled emerging analysis results according to the associated index scores.