Method and system for predicting the financial impact of forecasted weather conditions

• US 10,520,645 B2
• Filed: 05/31/2017
• Issued: 12/31/2019
• Est. Priority Date: 05/31/2016
• Status: Active Grant

• Current Assignee: Accuweather Incorporated
• Original Assignee: Accuweather Incorporated
• Inventors: Radich, Rosemary Y., Loftus, Tim, Bowers, Jennifer, Roehsner, Paul, Raymond, Paul, Root, Michael R.
• Primary Examiner(s): Le, John H
• Application Number:
  US15/609,650
• Publication Number:
  US 20170343702A1
• Time in Patent Office: 944 Days
• Field of Search: 702 3, 345473, 701117, 705 145
• CPC Class Codes: G01W 1/10 Devices for predicting weather conditions computers per se G06; display devices G09
• 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

Abstract

A system and method of predicting the impact of forecasted weather, environmental, and geologic events (that include one or more weather/environmental/geologic conditions) by determining a recurrence interval of each past condition in each location, determining the correlation between the past condition and the observable impact of the past event, calculating a predicted observable impact of each past event, calculating a predicted impact of each past event recurring by multiplying the predicted observable impact of the past event by the recurrence interval of the past condition, grouping the past events into a plurality of groups based on the predicted impact of the past condition recurring, determining a threshold for each group, receiving forecasted conditions, and determining the predicted impact of the forecasted conditions by comparing the forecasted conditions with the thresholds.

Claims

1. A method of predicting the financial impact of forecasted weather conditions, the method comprising:

receiving information indicative of past weather events in a plurality of locations, each of the past weather events including one or more past weather conditions;

receiving information indicative of the observable financial impact of at least some of the past weather events;

for each of the plurality of locations and each of the one or more weather conditions;

determining a recurrence interval of each of the past weather conditions in the plurality of locations;

determining correlations between the past weather conditions and the observable financial impact of the past weather events;

calculating a predicted observable financial impact of each of the past weather conditions;

calculating a predicted financial impact of each of the past weather conditions recurring by multiplying the predicted observable financial impact the past weather condition by the recurrence interval of the past weather condition;

grouping the past weather events into a plurality of groups based on the predicted financial impact of the past weather condition recurring; and

determining a threshold for each of the plurality of groups;

receiving forecasted weather conditions; and

determining the predicted financial impact of the forecasted weather conditions by comparing the forecasted weather conditions with the thresholds determined for each of the groups in the locations of the forecasted weather events.

2. The method of claim 1, wherein the correlation between the past weather condition and the observable financial impact of the past weather event is determined using a regression algorithm.

3. The method of claim 1, further comprising:

weighting the observable financial impact of past weather events by population.

4. The method of claim 1, further comprising:

interpolating the thresholds for the plurality of locations to determine thresholds for additional locations.

5. The method of claim 4, wherein the thresholds are interpolated using a Kriging technique.

6. The method of claim 1, wherein determining thresholds for each of the one or more weather conditions comprises determining thresholds for each of a plurality of weather conditions and each of the plurality of locations.

7. The method of claim 6, wherein the plurality of weather conditions include at least one of snow, rain, ice, wind, or temperature.

8. A system for predicting the financial impact of forecasted weather events, comprising:

one or more databases that store;

information indicative of past weather events in a plurality of locations, each of the past weather events including one or more past weather conditions; and

information indicative of the observable financial impact of at least some of the past weather events;

an analysis unit that;

for each of the plurality of locations and each of the one or more weather conditions;

determines a recurrence interval of each of the past weather conditions in the plurality of locations;

determines correlations between the past weather conditions and the observable financial impact of the past weather events;

calculates a predicted observable financial impact of each of the past weather conditions;

calculates a predicted financial impact of each of the past weather conditions recurring by multiplying the predicted observable financial impact the past weather condition by the recurrence interval of the past weather condition;

groups the past weather events into a plurality of groups based on the predicted financial impact of the past weather condition recurring; and

determines a threshold for each of the plurality of groups;

receives forecasted weather conditions; and

determines the predicted financial impact of the forecasted weather conditions by comparing the forecasted weather conditions with the thresholds determined for each of the groups in the locations of the forecasted weather events.

9. The system of claim 8, wherein the analysis unit is further configured to determine the correlation between the past weather condition and the observable financial impact of the past weather event using a regression algorithm.

10. The system of claim 8, wherein the analysis unit is further configured to weight the observable financial impact of past weather events by population.

11. The system of claim 8, wherein the analysis unit is further configured to interpolate the thresholds for the plurality of locations to determine thresholds for additional locations.

12. The system of claim 11, wherein the analysis unit is further configured to interpolate the thresholds using a Kriging technique.

13. The system of claim 8, wherein the analysis unit is configured to determine thresholds for each of a plurality of weather conditions in each of a plurality of locations.

14. The system of claim 13, wherein the plurality of weather conditions include at least one of snow, rain, ice, wind, or temperature.

15. A non-transitory computer-readable storage medium storing instructions that, when executed by a computer process, cause a computing device to:

receive information indicative of past weather events in a plurality of locations, each of the past weather events including one or more past weather conditions;

receive information indicative of the observable financial impact of at least some of the past weather events;

for each of the plurality of locations and each of the one or more weather conditions;

determine a recurrence interval of each of the past weather conditions in the plurality of locations;

determine the correlation between the past weather condition and the observable financial impact of the past weather event;

calculate a predicted observable financial impact of each of the past weather events;

calculate a predicted financial impact of each of the past weather events recurring by multiplying the predicted observable financial impact the past weather event by the recurrence interval of the past weather condition;

group the past weather events into a plurality of groups based on the predicted financial impact of the past weather condition recurring; and

determine threshold for each of the plurality of groups;

receive forecasted weather conditions; and

determine the predicted financial impact of the forecasted weather conditions by comparing the forecasted weather conditions with the thresholds determined for each of the groups in the locations of the forecasted weather events.

16. The computer-readable storage medium of claim 15, wherein the instructions further cause the computing device to determine the correlation between the past weather condition and the observable financial impact of the past weather event using a regression algorithm.

17. The computer-readable storage medium of claim 15, wherein the instructions further cause the computing device to weight the observable financial impact of past weather events by population.

18. The computer-readable storage medium of claim 15, wherein the instructions further cause the computing device to interpolate the thresholds for the plurality of locations to determine thresholds for additional locations.

19. The computer-readable storage medium of claim 15, wherein the instructions further cause the computing device to determine thresholds for each of a plurality of weather conditions in each of a plurality of locations.

20. The computer-readable storage medium of claim 19, wherein the plurality of weather conditions include at least one of snow, rain, ice, wind, or temperature.