Forecasting discovery costs based on complex and incomplete facts

• US 8,484,069 B2
• Filed: 09/02/2009
• Issued: 07/09/2013
• Est. Priority Date: 06/30/2008
• 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): Marcus, Leland
• Assistant Examiner(s): SCHEUNEMANN, RICHARD N
• Application Number:
  US12/553,068
• Publication Number:
  US 20090327049A1
• Time in Patent Office: 1,406 Days
• Field of Search: 705 711- 742
• US Class Current: 705/7.35
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 10/06375 Prediction of business process outcome or impact based on a proposed change
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

The discovery cost forecasting system uses incomplete information to generate forecasts of discovery expenses. Models are generated with different levels of granularity depending upon the available facts and user preferences. An expert provides real-time information to improve the accuracy of the predictions. A scenario analysis is generated that evaluates how changes to the parameters of the matter could affect the forecasted cost. An estimate is generated of the degree of advancement of the matter in its lifecycle as part of the cost forecast. The cost associated with each data source and custodian is tracked and reported. A configurable and extensible cost equation is defined. A dashboard user interface is generated to monitor the entire portfolio to extract easy to understand facts, trends, and early warning signs related to discovery cost management.

Claims

1. A system for forecasting discovery costs comprising:

a memory; and

a processor, the processor configured to implement instructions stored in the memory, the memory storing executable instructions, the processor comprising;

a trend engine that receives historical facts relating to past and current discovery and generates at least one of a plurality of models modeling discovery cost forecasts based on the historical facts; and

a forecasting engine that receives the at least one of the plurality of models and current facts, generates a discovery cost forecast for a current discovery process based on the at least one of the plurality of models and the current facts, including automatically switching between the plurality of models based on quality assessments of the plurality of models, estimates values of key lifecycle indicators that indicate a degree of advancement of the current discovery process, adjusts the values of key lifecycle indicators based on monitoring event patterns of the current discovery process for inactivity, and adjusts the discovery cost forecast based at least in part on actual status of the key lifecycle indicators received from a user, the discovery cost forecast being adjusted using a weighted average of a plurality of prediction tracks that are weighed to determine which one of the plurality of prediction tracks is most likely to reflect a current state of the current discovery process based at least in part on the lifecycle indicators, wherein each of the plurality of prediction tracks represents a model of dependencies between parameters of a matter as they exist at points in time during a lifecycle of the current discovery process.

2. The system of claim 1, wherein the trend engine generates a model that includes a default cost profile across all matter types.

3. The system of claim 1, wherein the trend engine generates a model that includes a default cost profile for each matter type.

4. The system of claim 3, wherein the model is manually configured.

5. The system of claim 1, wherein the trend engine generates a model that includes a default cost profile for each matter type based on which stage in a matter lifecycle has been reached by the discovery.

6. The system of claim 1, wherein the trend engine generates a model that distinguishes between individual data sources and custodians based on their individual cost profile or cost profile category.

7. A computer-implemented method for forecasting discovery costs on a computer comprising a processor and a memory, the processor configured to implement steps stored in the memory, comprising the steps of:

receiving, by the computer, historical facts about past and current discovery;

assessing, by the computer, a level of quantity and quality of the historical facts;

transforming, by the computer, the historical facts and the current facts into one of a plurality of models modeling discovery cost forecasts for a current discovery process based on the historical facts;

estimating values of key lifecycle indicators that indicate a degree of advancement of the current discovery process;

adjusting the values of key lifecycle indicators based on monitoring event patterns of the current discovery process for inactivity; and

adjusting the discovery cost forecast based at least in part on actual status of the key lifecycle indicators received from a user, the discovery cost forecast being adjusted using a weighted average of a plurality of prediction tracks that are weighed to determine which one of the plurality of prediction tracks is most likely to reflect a current state of the current discovery process based at least in part on the lifecycle indicators, wherein each of the plurality of prediction tracks represents a model of dependencies between parameters of a matter as they exist at points in time during a lifecycle.

8. A system for forecasting discovery costs comprising:

a memory; and

a processor, the processor configured to implement instructions stored in the memory, the memory storing executable instructions, the processor comprising;

a trend engine that receives historical facts relating to past and current discovery, and generates at least one of a plurality of models modeling discovery cost forecasts based on the historical facts,

a cost equation user interface to configure parameters to enable and adjust different components of an overall cost equation, including adjusting the overall cost equation to calibrate between different ways to account for discovery costs, and

a forecasting engine that generates a discovery cost forecast for a current discovery process using the configurable cost equation based on the model, the current facts, and the cost equation parameters, that estimates values of key lifecycle indicators that indicate a degree of advancement of the current discovery process, that adjusts the values of key lifecycle indicators based on monitoring event patterns of the current discovery process for inactivity, and that adjusts the discovery cost forecast based at least in part on actual status of the key lifecycle indicators received from a user, the discovery cost forecast being adjusted using a weighted average of a plurality of prediction tracks that are weighed to determine which one of the plurality of prediction tracks is most likely to reflect a current state of the current discovery process based at least in part on the lifecycle indicators, wherein each of the plurality of prediction tracks represents a model of dependencies between parameters of a matter as they exist at points in time during a lifecycle.

9. The system of claim 8 wherein the cost equation parameters are defined as any of:

global across all matter types, for each matter type, or for a specific matter.

10. The system of claim 9, wherein global parameters are mandatory, per matter type parameters supersede global parameters if defined, and per matter parameters supersede any of the per matter type parameters or global parameters if defined.

11. A computer-implemented method for forecasting discovery costs on a computer comprising a processor and a memory, the processor configured to implement steps stored in the memory, comprising the steps of:

receiving, by the computer, historical facts about current and past discovery;

generating, by the computer, at least one of a plurality of models modeling discovery cost forecasts based on the historical facts;

receiving, by the computer, a plurality of cost equation parameters used to configure parameters to enable and adjust different components of an overall cost equation, including adjusting the overall cost equation to calibrate between different ways to account for discovery costs;

generating, by the computer, a discovery cost forecast for a current discovery process using a configurable cost equation based on a model, the current facts, and the cost equation parameters;

estimating values of key lifecycle indicators that indicate a degree of advancement of the current discovery process;

adjusting the values of key lifecycle indicators based on monitoring event patterns of the current discovery process for inactivity; and

adjusting, by the computer, the discovery cost forecast based at least in part on actual status of the key lifecycle indicators received from a user, the discovery cost forecast being adjusted using a weighted average of a plurality of prediction tracks that are weighed to determine which one of the plurality of prediction tracks is most likely to reflect a current state of the current discovery process based at least in part on the lifecycle indicators, wherein each of the plurality of prediction tracks represents a model of dependencies between parameters of a matter as they exist at points in time during a lifecycle.

12. The method of claim 11, further comprising the step of:

receiving, by the computer, cost equation parameters defined as any of;

global across all matter types, for each matter type, or for a specific matter.

13. The method of claim 11, wherein matter parameters are used if defined, per matter type parameters are used if defined and no equivalent matter parameters are defined, and global parameters are used otherwise.

14. A system for forecasting discovery costs comprising:

a memory; and

a processor, the processor configured to implement instructions stored in the memory, the memory storing executable instructions, the processor comprising;

a trend engine that receives historical facts relating to past and current discovery, and generates at least one model modeling discovery cost forecasts based on the historical facts;

a forecasting engine that generates a discovery cost forecast for all current matters based on the at least one model, the current facts for each matter, that estimates values of key lifecycle indicators that indicate a degree of advancement of each matter, that adjusts the values of key lifecycle indicators based on monitoring event patterns of each matter for inactivity, and that adjusts the discovery cost forecast based at least in part on actual status of the key lifecycle indicators received from a user, the discovery cost forecast being adjusted using a weighted average of a plurality of prediction tracks that are weighed to determine which one of the plurality of prediction tracks is most likely to reflect a current state of each matter based at least in part on the lifecycle indicators, wherein each of the plurality of prediction tracks represents a model of dependencies between parameters of a matter as they exist at points in time during a lifecycle; and

a dashboard user interface (UI) for monitoring cost related characteristics and statistics across an entire discovery matter portfolio, including triggering a warning based on one or more of;

a number of custodians, a number of data sources, a number of collections to export for review, and a number of new matters, as compared with historical trends.

15. The system of claim 14, wherein the dashboard UI tracks particular events within a certain period in the past to detect facts that represent warning signs of a risk of increased cost in the future.

16. The system of claim 14, wherein the dashboard UI is configured to notify a user through display or alerts whenever a particular event occurs.

17. The system of claim 14, wherein the dashboard UI identifies at least one most expensive matter based on any of a cost metric, a different target period, a cost incurred, a cost estimated, a cost accrued, and a cost forecasted.

18. The system of claim 14, wherein the dashboard UI identifies any of:

a data source, a custodian, a category of data sources, and a category of custodians by discovery cost.

19. A computer-implemented method for forecasting discovery costs comprising a processor and a memory, the processor configured to implement steps stored in the memory, comprising the steps of:

receiving, by the computer, historical facts related to past and current discovery;

generating, by the computer, at least one model modeling discovery cost forecasts based on the historical facts;

generating a discovery cost forecast for all matters based on the at least one model and the current facts for each matter;

estimating values of key lifecycle indicators that indicate a degree of advancement of each matter;

adjusting the values of key lifecycle indicators based on monitoring event patterns of each matter for inactivity;

adjusting the discovery cost forecast based at least in part on actual status of the key lifecycle indicators received from a user, the discovery cost forecast being adjusted using a weighted average of a plurality of prediction tracks that are weighed to determine which one of the plurality of prediction tracks is most likely to reflect a current state of each matter based at least in part on the lifecycle indicators, wherein each of the plurality of prediction tracks represents a model of dependencies between parameters of a matter as they exist at points in time during a lifecycle; and

monitoring, by the computer, cost related characteristics and statistics across an entire discovery matter portfolio, including triggering a warning based on one or more of;

a number of custodians, a number of data sources, a number of collections to export for review, and a number of new matters, as compared with historical trends.

20. The method of claim 19, further comprising the step of:

tracking, by the computer, a particular event within a certain period in a past to detect facts that represent warning signs of a risk of increased cost in the future.

21. The method of claim 19, further comprising the step of:

identifying, by a computer, at least one most expensive matter based on any of a cost metric, a different target period, a cost incurred, a cost estimated, a cost accrued, and a cost forecasted.

22. The method of claim 19, further comprising the step of:

notifying a user, by the computer, through display or alerts whenever a particular event occurs.