Aggregation and analytics for application-specific optimization based on multiple data sources

• US 10,234,598 B2
• Filed: 11/02/2017
• Issued: 03/19/2019
• Est. Priority Date: 09/13/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Eickelman, Hans-Jurgen, Liu, Ying, Osiecki, Thomas H., Villa Real, Lucas Correia
• Primary Examiner(s): Lu, Kuen S
• Application Number:
  US15/801,922
• Publication Number:
  US 20180059286A1
• Time in Patent Office: 502 Days
• Field of Search: 707602
• CPC Class Codes: G01W 1/14 Rainfall or precipitation gauges measuring volume in general G01F
• CPC Class Codes: G06F 16/244 Grouping and aggregation
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/254 Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
• CPC Class Codes: G06G 7/50 for distribution networks, e.g. for fluids G06G7/62 takes precedence
• CPC Class Codes: Y02A 10/40 Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping

Abstract

Aggregating and transforming data, and performing analytics thereupon, for application-specific optimization based on multiple data sources. The data is preferably ingressed automatically, and may originate from various public and/or private data sources. Data transformation preferably aligns the data aggregated from the various sources, to thereby allow meaningful referencing. Complex and non-aligned data can therefore be consolidated, such that it is readily digestible by simulation (or other) software. In an embodiment, risk of flooding for a supply chain is computed from the aggregated and transformed data, using data analytics based on physical computation for flood risk assessment, allowing the supply chain to be optimized with regard to threat of flooding and/or actual flooding. In another embodiment, risk of wild fire may be assessed. Other types of risk may also be assessed.

Claims

1. A method for assessing risk of flooding by aggregating and transforming data from a plurality of data sources, comprising:

determining, for a supply chain spanning a plurality of geographical locations, a location of interest within the supply chain for flooding risk analysis;

determining a plurality of data sources that describe a physical environment of the location;

automatically ingressing, from selected ones of the plurality of sources, the data that describes the physical environment;

programmatically transforming the ingressed data into data maps that are aligned to one another to allow referencing therebetween;

evaluating risk of flooding pertaining to the location by using the aligned data maps as input to an evaluator selected from the group consisting of a simulation model and an analytic process, wherein processing performed by the evaluator comprises;

watershed delineation to determine a geographical boundary of a watershed area surrounding the location;

canopy interception analysis to determine an impact of physical canopy on an amount of rainfall that reaches a ground surface in the watershed area;

soil infiltration analysis to determine how rainfall reaching the ground surface will be absorbed into soil at the ground surface in the watershed area; and

2-dimensional diffusive routing to determine a flow of rainfall that reaches the ground surface in the watershed area but is not absorbed into the soil at the ground surface in the watershed area, the 2-dimensional diffusive routing comprising, for the non-absorbed rainfall, a mass of water flow and a momentum of the water flow; and

responsive to determining that the risk of flooding pertaining to the location exceeds a predetermined threshold, determining an alternative geographical location for at least one of the plurality of geographical locations within the supply chain to thereby avoid using, at least temporarily, the location of interest, wherein the alternative geographical location is determined by repeating the processing performed by the evaluator to evaluate risk of flooding pertaining to the alternative geographical location and determining that the risk of flooding pertaining to the alternative geographical location does not exceed the predetermined threshold.

2. The method according to claim 1, wherein:

the location is provided using longitude and latitude values; and

determining the plurality of sources comprises;

reverse geocoding the longitude and latitude values to determine identifying name information associated with the location; and

locating the sources which provide suitable data for a geographical region corresponding to the identifying name information.

3. The method according to claim 1, wherein:

the location is provided using coordinates; and

determining the plurality of sources comprises;

using the coordinates to access a mapping which correlates the coordinates with identifying name information associated with the location; and

locating the sources which provide suitable data for a geographical region corresponding to the identifying name information.

4. The method according to claim 1, wherein the automatically ingressing comprises using session cookies to allow coordinated non-interactive download over a period of time, from the selected ones, of the data that describes the physical environment.

5. The method according to claim 1, wherein the transforming comprises programmatically changing a resolution of the ingressed data from at least a first of the selected ones to align with a resolution of the ingressed data from at least a second of the selected ones.

6. The method according to claim 1, wherein the transforming further comprises programmatically changing the resolution of the ingressed data from at least a first of the selected ones to align with a resolution of reference data for the location.

7. The method according to claim 1, wherein the transforming comprises programmatically removing at least one noise artifact from the ingressed data from at least a first of the selected ones.

8. The method according to claim 7, wherein:

the ingressed data from the first of the selected ones comprises topography data pertaining to the location; and

the programmatically removing comprises first identifying, in the ingressed topography data, the at least one noise artifact as being irrelevant to underlying topography of the location and therefore comprising a noise artifact to be removed from the ingressed topography data.

9. The method according to claim 1, wherein:

the ingressed data from at least a first of the selected ones comprises topography data pertaining to the location; and

the transforming comprises;

programmatically delineating, from the ingressed topography data, at least one watershed area pertaining to the location; and

programmatically removing, from the ingressed topography data, portions thereof which are separate from any of the at least one delineated watershed area.

10. The method according to claim 1, wherein:

the ingressed data from at least a first of the selected ones and a second of the selected ones comprises topography data pertaining to the location; and

the transforming comprises aligning the topography data from the first of the selected ones and the second of the selected ones to have an identical number of pixels and to match an identical geo-location of a reference topography data.

11. The method according to claim 1, wherein:

the ingressed data from at least a first of the selected ones comprises topography data pertaining to the location; and

the evaluating comprises;

determining, from the topography data, canopy interception information pertaining to the location; and

using the canopy interception information when determining the risk.

12. The method according to claim 1, wherein:

the ingressed data from the first of the selected ones comprises land use data pertaining to the location;

the transforming comprises programmatically extracting a feature from the ingressed data from at least a first of the selected ones, the feature describing use of land at or near the location; and

the programmatically extracting comprises removing, from the land use data, the feature from the ingressed data to allow focus on the land use data for the location.

13. The method according to claim 1, wherein:

the ingressed data from at least a first of the selected ones comprises soil data pertaining to the location; and

the evaluating comprises;

determining, from the soil data, soil infiltration information pertaining to the location; and

using the soil infiltration information when determining the risk.

14. The method according to claim 1, wherein:

the ingressed data from at least a first of the selected ones comprises historical data pertaining to the location; and

the transforming comprises programmatically cleansing the ingressed historical data such that a resulting one of the data maps has unsuitable information removed therefrom.

15. The method according to claim 1, wherein:

the ingressed data is organized as a plurality of grids; and

the transforming further comprises programmatically overlaying data of a first of the grids with data of a second of the grids, responsive to determining that the data of the second grid is of a better quality than the data of the first grid.

16. A method for assessing risk of wild fire by aggregating and transforming data from a plurality of data sources, comprising:

determining, for a supply chain spanning a plurality of geographical locations, a location of interest within the supply chain for wild fire risk analysis;

determining a plurality of data sources that describe a physical environment of the location;

automatically ingressing, from selected ones of the plurality of sources, the data that describes the physical environment, wherein the ingressed data comprises;

historical data pertaining to wind directions for a first area that includes the location;

vegetation type information for a second area that includes the location; and

soil moisture information for a third area that includes the location, and wherein the first area, the second area, and the third area may be overlapping or distinct;

programmatically transforming the ingressed data into data maps that are aligned to one another to allow referencing therebetween;

evaluating risk of wild fire pertaining to the location by using the aligned data maps as input to an evaluator selected from the group consisting of a simulation model and an analytic process; and

responsive to determining that the risk of wild fire pertaining to the location exceeds a predetermined threshold, determining an alternative geographical location for at least one of the plurality of geographical locations within the supply chain to thereby avoid using, at least temporarily, the location of interest, wherein the alternative geographical location is determined by repeating the evaluating risk performed by the evaluator to evaluate risk of wild fire pertaining to the alternative geographical location and determining that the risk of wild fire pertaining to the alternative geographical location does not exceed the predetermined threshold.