Geospatial model forecasting of events and threats

• US 7,346,597 B2
• Filed: 04/04/2005
• Issued: 03/18/2008
• Est. Priority Date: 04/02/2004
• Status: Active Grant

• Current Assignee: Radiant Analytic Solutions, Inc. (Galileo Parent, Inc.)
• Original Assignee: Spadac Inc.
• Inventors: Dumas, Mark E.
• Primary Examiner(s): Hirl; Joseph P
• Application Number:
  US11/098,510
• Publication Number:
  US 20050222829A1
• Time in Patent Office: 1,079 Days
• Field of Search: 706/58, 706/45, 706/46, 700/28
• US Class Current: 706/58
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 10/02 Reservations, e.g. for tickets, services or events
• 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 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G09B 29/007 using computer methods

Abstract

A forecasting engine and method assists in rapidly and accurately forecasting occurrences of identifiable events and/or results based on signature and/or pattern matching. The present invention derives signature for event-types based on a comparison of actual event data with pre-established representational surfaces. The surfaces represent functional measurements and analysis associated with elements of the geospatial boundary being considered. The present invention provides highly refined modeling processes to assist in quickly focusing on the proper measurement type and/or variable type, and detailing analysis around the most relevant factors. In this way, the present invention allows for more rapid and more accurate assessment determinations. In one aspect, the present invention provides a decision support system for assisting in the determination of potentially successful commercial locations. In another embodiment, the present invention provides a centralized portal capable of assessing multiple problems simultaneously.

Claims

1. A geospatial event forecasting engine for forecasting the likelihood for occurrence of at least one event type, comprising:

a spatial database storing geospatial boundary information for one or more areas of interest, and further storing geospatial characteristic information pertaining to at least one variable of interest in the form of one or more variable layers associated with said one or more areas of interest, with the variable of interest being unrelated to past event type data for the at least one event type;

a boundary component for establishing a geospatial boundary pertaining to a first area of interest and a grid containing a plurality of cells within said boundary;

a function component for identifying a first functional measurement of a cell element for each cell to said at least one variable of interest for each of said one or more layers associated with said first area of interest, and for indexing said first functional measurement for each cell, thereby establishing a database of first functional measurements unrelated to the event type;

a signature derivation component for receiving geospatial information related to one or more past events of at least one event type, including location information for said one or more past events, said signature derivation component further identifying and indexing an event functional measurement from the location information for each of said one or more events to the at least one variable of interest for each of said layers, and deriving a signature pattern for said event type based upon the event functional measurement; and

an event likelihood determinant component for determining a level of signature match between said derived signature pattern and at least one cell of said plurality of cells, whereby a forecast of an event-type likelihood can be displayed on at least a portion of a display of said geospatial boundary based on said determined level of signature match.

2. The engine of claim 1 wherein said functional measurement is not a nearest neighbor value and is one from the group of measurements including:

density, concentration, average distance by actual path, Manhattan distance, visibility.

3. The engine of claim 2 wherein said functional measurement is average distance by actual path, wherein said actual path is artificial.

4. The engine of claim 2 wherein said functional measurement is average distance by actual path, wherein said actual path is a natural path.

5. The engine of claim 1 wherein said functional measurement is a manual function.

6. The engine of claim 1 wherein said at least one variable of interest is non-static.

7. The engine of claim 1 wherein said event likelihood determinant component determines whether there is a signature match by determining a score indicative of said at least one cell'"'"'s compatibility with the event type signature.

8. The engine of claim 7 wherein said event likelihood determinant component determines whether there is a signature match for each of said cells within said boundary and further plots said scores on a choropleth graph.

9. The engine of claim 1 wherein said at least one variable of interest is provided with a plurality of feature types individually indexed by a respective feature identifier.

10. The engine of claim 1 wherein said plurality of cells are of irregular shape.

11. The engine of claim 10 wherein said cell irregular shapes are determined by political boundaries.

12. The engine of claim 10 wherein said cell irregular shapes are determined by natural boundaries.

13. The engine of claim 10 wherein said cell irregular shapes are not determined by political or natural boundaries.

14. The engine of claim 1 wherein said signature derivation component represents said received geospatial information in a non-point format.

15. The engine of claim 14 wherein said non-point format is one of the formats from the group consisting of:

polygon, line, three-dimensional.

16. The engine of claim 1 wherein said event likelihood determination component determines, for each event, the individual cell element nearest said event and associates the functional measurement for said nearest individual cell element with said event.

17. The engine of claim 1 wherein said event likelihood determination component determines said level of signature match between said derived signature pattern and at least one cell of said plurality of cells through comparison with said indexed functional measurement.

18. The engine of claim 1 wherein said event likelihood determination component determines said level of signature match between said derived signature pattern and each of said plurality of cells.

19. The engine of claim 1 wherein said established boundary is a first boundary, and wherein said engine further includes a signature transfer component for applying said derived signature to indexed functional measurements of a second boundary, said second boundary having a plurality of cells therein, and determining a level of signature match between said derived signature and at least one cell of said plurality of cells in said second boundary.

20. The engine of claim 19 wherein said indexed functional measurements of said second boundary are measured from a cell element for each cell in said second boundary to at least one variable of interest, the at least one variable of interest within said first boundary being different from the at least one variable of interest within said second boundary.

21. The engine of claim 1 wherein said functional measurement is not a probability density function.

22. The engine of claim 1 further including a layer selection component for selecting at least one of said plurality of layers for imposing upon said grid.

23. The engine of claim 1 wherein said one or more variable layers are each indicative of one or more variables of interest which are different from the one or more variables of interest of the remaining layers.

24. The engine of claim 1 wherein said one or more variable layers are indicative of the same variable or variables of interest, but wherein each of said layers represents a time frame which is different from the remaining layers.

25. The engine of claim 1 wherein said cell element is a centroid.

26. The engine of claim 1 wherein at least one of said cells includes a plurality of cell elements and wherein said function component identifies and indexes a functional measurement for each cell element for each cell to said at least one variable of interest for each of said layers.

27. The engine of claim 1 further including an alert component for communicating said signature match determination to at least one remote agent operating within said geospatial boundary.

28. A method of forecasting geospatial events of at least one event type, comprising the steps of:

storing geospatial boundary information for one or more areas of interest;

storing geospatial characteristic information pertaining to at least one variable of interest in the form of one or more variable layers associated with said one or more areas of interest, with the variable of interest being unrelated to fast event type data for the at least one event type;

establishing a geospatial boundary pertaining to a first area of interest and a grid containing a plurality of cells within said boundary;

identifying a first functional measurement of a cell element for each cell to said at least one variable of interest for each of said one or more layers associated with said first area of interest, and indexing said first functional measurement for each cell thereby establishing a database of first functional measurements unrelated to the event type;

receiving geospatial information related to one or more past events of at least one event type, including location information for said one or more past events;

identifying and indexing an event functional measurement from the location information for each of said one or more events to the at least one variable of interest for each of said layers, and deriving a signature pattern for said event type based upon the event functional measurement; and

determining a level of signature match between said derived signature pattern and at least one cell of said plurality of cells, whereby a forecast of an event type likelihood can be displayed on at least a portion of a display of said geospatial boundary based on said determined level of signature match.

29. The method of claim 28 wherein said functional measurement is not a nearest neighbor value and is one from the group of measurements including:

density, concentration, average distance by actual path, Manhattan distance, visibility.

30. The method of claim 28 wherein said functional measurement is a manual function.

31. The method of claim 28 wherein said step of determining a level of signature match determines a score indicative of said at least one cell'"'"'s compatibility with the event type signature.

32. The method of claim 31 wherein said step of determining a level of signature match determines whether there is a signature match for each of said cells within said boundary and further plots said scores on a choropleth graph.

33. The method of claim 28 wherein said at least one variable of interest is provided with a plurality of feature types individually indexed by a respective feature identifier.

34. The method of claim 28 wherein said plurality of cells are of irregular shape.

35. The method of claim 34 wherein said cell irregular shapes are determined by political boundaries.

36. The method of claim 34 wherein said cell irregular shapes are determined by natural boundaries.

37. The method of claim 34 wherein said cell irregular shapes are not determined by political or natural boundaries.

38. The method of claim 28 wherein said received geospatial information is represented in a non-point format.

39. The method of claim 38 wherein said non-point format is one of the formats from the group consisting of:

polygon, line, three-dimensional.

40. The method of claim 28 wherein said step of determining a level of signature match determines, for each event, the individual cell element nearest said event and associates the functional measurement for said nearest individual cell element with said event.

41. The method of claim 28 wherein said step of determining a level of signature match determines said level of signature match between said derived signature pattern and at least one cell of said plurality of cells through comparison with said indexed functional measurement.

42. The method of claim 28 wherein said step of determining a level of signature match determines said level of signature match between said derived signature pattern and each of said plurality of cells.

43. The method of claim 28 wherein said established boundary is a first boundary, and wherein said method further includes the steps of applying said derived signature to indexed functional measurements of a second boundary, said second boundary having a plurality of cells therein, and determining a level of signature match between said derived signature and at least one cell of said plurality of cells in said second boundary.

44. The method of claim 43 wherein said indexed functional measurements of said second boundary are measured from a cell element for each cell in said second boundary to at least one variable of interest, the at least one variable of interest within said first boundary being different from the at least one variable of interest within said second boundary.

45. The method of claim 28 wherein said functional measurement is not a probability density function.

46. The method of claim 28 further including a layer selection component for selecting at least one of said plurality of layers for imposing upon said grid.

47. The method of claim 28 wherein said one or more variable layers are each indicative of one or more variables of interest which are different from the one or more variables of interest of the remaining layers.

48. The method of claim 28 wherein said one or more variable layers are indicative of the same variable or variables of interest, but wherein each of said layers represents a time frame which is different from the remaining layers.

49. The method of claim 28 wherein at least one of said cells includes a plurality of cell elements and wherein said function component identifies and indexes a functional measurement for each cell element for each cell to said at least one variable of interest for each of said layers.

50. The method of claim 28 further including the step of providing an alert component for communicating said signature match determination to at least one remote agent operating within said geospatial boundary.

51. A method for determining an event type based on an event signature, comprising the steps of:

establishing a geospatial boundary and a grid containing a plurality of cells within said boundary;

establishing at least one layer imposed upon the grid, the layer being indicative of geospatial characteristics of at least one variable of interest, with the variable of interest being unrelated to past event type data for the event signature;

identifying a first functional measurement of a cell element for each cell to the variable of interest, and for indexing said first functional measurement for each cell;

receiving geospatial information pertaining to one or more post events of a given event type, including location information for said one or more past event;

identifying an event functional measurement to the at least the variable of interest for each of said events;

determining a likelihood associating said event type'"'"'s relative proximity to the variable of interest;

establishing a signature for said event type based on said determined likelihood;

receiving an outside event signature associated with said boundary;

determining a level of signature match between said established signature and said outside event signature.

52. A decision support system, comprising:

means for establishing a group of options associated with a system decision;

means for associating each option of said group of options with an area of interest (AOI);

means for collecting training data associated with each of said options;

means for establishing a result forecast for each option; and

means for scoring said result forecasts so as to determine which option in said group of options is optimal.

53. The system of claim 52 wherein said means for establishing a result forecast for each option comprises:

a boundary component for establishing a geospatial boundary associated with said AOI and a grid containing a plurality of cells within said boundary;

a layer component for establishing at least one layer imposed upon the grid, the layer being indicative of geospatial characteristics of at least one variable of interest, with the variable of interest being unrelated to the training data;

a function component for identifying a first functional measurement of a cell element for each cell to said at least one variable of interest, and for indexing said first functional measurement for each cell;

a signature derivation component for receiving geospatial information related to one or more past events of at least one event type, including location information for said one or more past events, said signature derivation component further identifying and indexing an event functional measurement from the location information for each of said one or more events to the at least one variable of interest, and deriving a signature pattern for said event type based upon the event functional measurement;

an event likelihood determinant component for determining a level of signature match between said derived signature pattern and at least one cell of said plurality of cells.

54. The system of claim 52 wherein said group of options is a group of potential retail establishment locations.

55. The system of claim 52 wherein said group of options is a group of potential insurance candidates.

56. A decision support system for learning from at least one past meaningful result comprising:

means for establishing at least one geospatial area of interest (AOI) comprising a two-dimensional, cell-based, geospatial boundary;

means for imposing at least one layer of geospatial information on said AOI, said geospatial information including at least one variable of interest, with the variable of interest being unrelated to the past meaningful result;

means for determining the proximity of each cell in said boundary to each of said at least one variable of interest;

means for determining a location of the past meaningful result;

means for determining the proximity of the past meaningful result to said at least one variable of interest;

means for establishing a signature associated with said result; and

means for determining a closest match with said result signature among said cells.

57. A system for forecasting an event of a given event type, comprising:

an interface for receiving a problem definition;

a forecasting component having a plurality of areas of interest, a plurality of variable layers, each variable layer being associated with a respective variable of interest that is unrelated to past event type data for the given event type, and a plurality of training data elements representing at least one discrete event;

a selection interface for receiving a selection of at least one area of interest and at least one variable layer,

a training data interface for receiving a selection of training data;

an assessment component for applying said training data to said selected area of interest and said selected variable layer to determine a forecast for one or ore locations within said area of interest where a future event related to said at least one discrete event may occur.

58. The system of claim 57 wherein said training data selection is made by said user.

59. The system of claim 58 wherein said training data selection is made by said system according to said problem definition.