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

Patent metadata

Field	Value
Publication number	US-10234598-B2
Application number	US-201715801922-A
Country	US
Kind code	B2
Filing date	Nov 2, 2017
Priority date	Sep 13, 2014
Publication date	Mar 19, 2019
Grant date	Mar 19, 2019

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.

First claim

The invention claimed is:

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

Assignees

• IBM

Inventors

• Eickelman Hans-Jurgen
• Liu Ying
• Osiecki Thomas H
• Villa Real Lucas Correia

Classifications

• G06F17/30563

  Physics · mapped topic

• G06F17/30554

  Physics · mapped topic

• G06F17/30412

  Physics · mapped topic

• Y02A10/46

  Cross-Sectional Technologies · mapped topic

• G06G7/50

  for distribution networks, e.g. for fluids (G06G7/62 takes precedence) · CPC title