Forecasting discovery costs based on interpolation of historic event patterns

• US 8,073,729 B2
• Filed: 09/30/2008
• Issued: 12/06/2011
• Est. Priority Date: 09/30/2008
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kisin, Roman, Raynaud-Richard, Pierre
• Primary Examiner(s): Dunham, Jason
• Assistant Examiner(s): SCHEUNEMANN, RICHARD N
• Application Number:
  US12/242,478
• Publication Number:
  US 20100082382A1
• Time in Patent Office: 1,162 Days
• Field of Search: 705/7
• US Class Current: 705/7.35
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0206 Price or cost determination based on market factors
• CPC Class Codes: G06Q 30/0283 Price estimation or determination

Abstract

A computer-implemented method of forecasting discovery costs captures historical business events, which are aggregated and statistically analyzed by matter type and business event types. Statistical data is used to calculate probabilities of occurrence of future business events and to extrapolate occurrence of those future business events. The method forecasts an extrapolated volume of production at the time of a forecasted export event. The method further forecasts costs for future discovery from the extrapolated volume of production.

Claims

1. A computer-implemented method of forecasting litigation discovery costs for a litigation matter that includes a number of business events, comprising:

capturing historical business events for the litigation matter;

aggregating the historical business events, with an aggregation module, and statistically analyzing the captured historical business events by matter type, time period, and business event type to produce statistical data specific to a time period;

aggregating the statistical data for matter type and event type;

extrapolating future business events using the statistical data provided by the statistical analysis to calculate probabilities of occurrence of future business events;

forecasting an extrapolated volume of production at the time of a forecasted export event, wherein an export event is an event representing when a document page is being exported; and

forecasting costs for future discovery from the extrapolated volume of production by applying a culling rate and average review cost.

2. The method of claim 1, wherein the statistical data includes Export Lag, which is the time duration from an occurrence of a given event to an export event.

3. The method of claim 2, wherein the statistical data includes a statistical distribution export lag for event types within the matter type.

4. The method of claim 1, including providing a probability of export, which is the likelihood of the export event occurring after an event of a given type has occurred in the context of a legal matter.

5. The method of claim 1, wherein aggregating the statistical data for matter type and event type is across multiple fiscal periods to provide trend and seasonal analysis.

6. The method of claim 1, wherein the step of aggregating and statistically analyzing the captured historical business events by matter type and business event types to produce statistical data includes event weighting in which recent data is weighed more heavily that older data.

7. The method of claim 1, wherein aggregating and statistically analyzing the captured historical business events by matter type and business event types to produce statistical data includes event filtering that discards events with values outside of an acceptable range and events from the matters identified as abnormal.

8. The method of claim 1, wherein extrapolating future business events using the statistical data provided by the statistical analysis to calculate probabilities of occurrence of future business events includes a weighed extrapolation algorithm in which recent events are given relatively more weight in forecasting than older events.

9. The method of claim 1, wherein forecasting an extrapolated volume of production at the time of a forecasted export event includes forecasting the number of custodians and data sources involved in collections based on the actual number of custodians and data sources in the scope of the matter.

10. The method of claim 1, wherein forecasting an extrapolated volume of production at the time of a forecasted export event includes computing the probability of the export within the forecasting period based on the historical event patterns.

11. The method of claim 1, wherein forecasting an extrapolated volume of production at the time of a forecasted export event includes applying a collection size per custodian and data source.

12. The method of claim 1, wherein forecasting an extrapolated volume of production at the time of a forecasted export event includes estimating typical make up of the collection in terms of file types.

13. The method of claim 1, including calibrating the method by making provisions and adjustments to one or more parameters of a forecasting model.

14. The method of claim 1, performing matter cost modeling for various scenarios by adjusting parameters of a forecasting model.

15. The method of claim 1, including creating reports for forecasting an extrapolated volume of production at the time of a forecasted export event.

16. The method of claim 1, including creating reports for forecasting costs for future discovery from the extrapolated volume of production.

17. The method of claim 1, including creating reports for forecasting costs for future discovery from an extrapolated volume of production.

18. A computer-implemented method of forecasting litigation discovery costs for a litigation matter that includes a number of business events, comprising:

capturing historical business events for the litigation matter;

aggregating the historical business events, with an aggregation module, and statistically analyzing the captured historical business events by matter type, time period, and business event type to produce statistical data specific to a time period;

aggregating the statistical data for matter type and event type;

extrapolating future business events using the statistical data provided by the statistical analysis to calculate probabilities of occurrence of future business events;

forecasting an extrapolated volume of production at the time of a forecasted export event, wherein an export event is an event representing when a document page is being exported; and

forecasting costs for future discovery from the extrapolated volume of production;

wherein forecasting an extrapolated volume of production at the time of a forecasted export event includes applying typical collection-to-export volume ratio to estimate the volume of export.

19. A computer-implemented method of forecasting litigation discovery costs for a litigation matter that includes a number of business events, comprising:

capturing historical business events for the litigation matter;

aggregating the historical business events, with an aggregation module and statistically analyzing the captured historical business events by matter type, time period, and business event type to produce statistical data specific to a time period;

aggregating the statistical data for matter type and event type;

extrapolating future business events using the statistical data provided by the statistical analysis to calculate probabilities of occurrence of future business events;

forecasting an extrapolated volume of production at the time of a forecasted export event, wherein an export event is an event representing when a document page is being exported; and

forecasting costs for future discovery from the extrapolated volume of production;

wherein forecasting an extrapolated volume of production at the time of a forecasted export event includes applying a gigabyte per page mapping to the collected volume according to the collection make up ratio to estimate the size of the export in pages.