Systems, methods, and platform for estimating risk of catastrophic events
View Patent ↗In an illustrative embodiment, systems and methods for calculating risk scores for locations potentially affected by catastrophic events include receiving a risk score request for a location, the risk score request including a request for assessment of risk exposure related to a type of catastrophic event. Based on the type of catastrophic event, a data compression algorithm may be applied to a catastrophic risk model representing amounts of perceived risk to an area surrounding the location. In response to receiving the risk score request, a risk score for the location may be calculated that corresponds to a weighted estimation of one or more data points in a compressed catastrophic risk model. A risk score user interface screen may be generated in real-time to present the catastrophic risk score and one or more corresponding loss metrics for the location due to a potential occurrence of the type of catastrophic event.
1. A method for compressing catastrophic risk models to enable accelerated evaluation of exposure impact to properties, the method comprising:
accessing, by processing circuitry, a risk model corresponding to a catastrophic event, wherein the risk model comprises a grid of multi-dimensional spatial data for determining at least one of a likelihood of occurrence of a catastrophic event across a geography or a cost of the occurrence, the grid comprising a plurality of location data points, each location data point corresponding to a respective measure of a plurality of measures; and
compressing, by the processing circuitry, the risk model to produce a compressed risk model comprising measures corresponding to a portion of the plurality of location data points of the risk model, wherein compressing the risk model comprises, for each data point of at least a portion of the plurality of location data points,
calculating an estimated measure value for the respective data point using a set of measures corresponding to a set of data points of the plurality of location data points geographically closest to the respective data point,
comparing the estimated measure value to the measure of the respective data point, and
discarding the respective data point based on the estimated measure value being within an error tolerance of the measure of the respective data point, wherein the error tolerance is one of a set of error tolerances, each error tolerance being associated with a separate region of a plurality of geographic regions represented by the grid; and
storing, by the processing circuitry, the compressed risk model to a non-transitory computer readable medium, wherein the compressed risk model is configured to estimate, based on input comprising a given location corresponding to any one of the plurality of location data points, a likelihood of occurrence of the catastrophic event at the given location or a cost of the occurrence at the given location.
2. The method of claim 1 , wherein the plurality of measures each relate to a monetary amount or a climactic statistic.
3. The method of claim 1 , wherein a number of data points of the compressed risk model is fewer than one percent of a total number of the plurality of location data points of the risk model.
4. The method of claim 1 , wherein compressing the risk model comprises, for at least one data point of the portion of the plurality of location data points, replacing the respective measure of the plurality of measures with a fitted measure value calculated using a set of location data points geographically surrounding the at least one data point.
5. The method of claim 4 , wherein calculating the fitted measure value comprises applying spatial interpolation to the set of location data points geographically surrounding the at least one data point.
6. The method of claim 1 , wherein the set of error tolerances comprises a first error tolerance representing a high amount of tolerable variation in a first subset of the plurality of measures within a first geographic region of data points, and a second error tolerance representing a low amount of tolerable variation in a second subset of the plurality of measures within a second geographic region of data points, wherein
the first geographic region of data points is associated with high loss values due to the catastrophic event, and
the second geographic region of data points is associated with low loss values due to the catastrophic event.
7. The method of claim 1 , further comprising determining, by the processing circuitry, the set of error tolerances based at least in part on a designated compression level.
8. The method of claim 1 , further comprising receiving at least a portion of the set of error tolerances as user input.
9. The method of claim 1 , wherein each location data point of the portion of the plurality of location data points in the compressed risk model further corresponds to a respective probability of a plurality of probabilities.
10. The method of claim 9 , wherein:
compressing the risk model comprises generating, for each location data point of the portion of the plurality of location data points, a probability distribution comprising a list of measures including the measure and a corresponding list of probabilities including the probability.
11. The method of claim 1 , further comprising performing, by the processing circuitry, a diagnostic fit calculation on the compressed risk model to confirm or improve accuracy of measures at the portion of the plurality of location data points retained in the compressed risk model.
12. The method of claim 11 , wherein performing the diagnostic fit calculation comprises:
selecting a set of discarded data points of the plurality of location data points, wherein
each data point of the set of discarded data points is not included in the compressed risk model;
adding the set of discarded data points to the compressed risk model to generate an enhanced compressed risk model; and
for each data point of the set of discarded data points,
calculating a fitted measure value for the respective data point using the set of measures corresponding to the set of data points of the portion of the plurality of location data points geographically closest to the respective data point,
comparing the fitted measure value to the measure of the respective data point, and
retaining the respective data point in the enhanced compressed risk model based on the fitted measure value meeting or exceeding an improvement criterion.
13. A non-transitory computer-readable medium having instructions stored thereon, wherein the instructions, when executed on processing circuitry, cause the processing circuitry to perform operations to create a compressed catastrophic risk model, the operations comprising:
accessing a risk model corresponding to a catastrophic event, wherein the risk model comprises a grid of multi-dimensional spatial data for determining at least one of a likelihood of occurrence of a catastrophic event across a geography or a cost of the occurrence, the grid comprising a plurality of location data points, each location data point corresponding to a respective measure of a plurality of measures; and
compressing the risk model to produce a compressed risk model comprising measures corresponding to a portion of the plurality of location data points of the risk model, wherein compressing the risk model comprises, for each data point of at least a portion of the plurality of location data points,
calculating an estimated measure value for the respective data point using a set of measures corresponding to a set of data points of the plurality of location data points geographically closest to the respective data point, comparing the estimated measure value to the measure of the respective data point, and
discarding the respective data point based on the estimated measure value being within an error tolerance of the measure of the respective data point, wherein the error tolerance is one of a set of error tolerances, each error tolerance being associated with a separate region of a plurality of geographic regions represented by the grid; and
storing the compressed risk model to a non-transitory storage region, wherein the compressed risk model is configured to estimate, based on input comprising a given location corresponding to any one of the plurality of location data points, a likelihood of occurrence of the catastrophic event at the given location or a cost of the occurrence at the given location.
14. The non-transitory computer-readable medium of claim 13 , wherein a number of data points of the compressed risk model is fewer than one percent of a total number of the plurality of location data points of the risk model.
15. The non-transitory computer-readable medium of claim 13 , wherein compressing the risk model comprises, for at least one data point of the portion of the plurality of location data points, replacing the respective measure of the plurality of measures with a fitted measure value calculated using a set of location data points geographically surrounding the at least one data point.
16. The non-transitory computer-readable medium of claim 15 , wherein calculating the fitted measure value comprises applying spatial interpolation to the set of location data points geographically surrounding the at least one data point.
17. The non-transitory computer-readable medium of claim 13 , wherein the set of error tolerances comprises a first error tolerance representing a high amount of tolerable variation in a first subset of the plurality of measures within a first geographic region of data points, and a second error tolerance representing a low amount of tolerable variation in a second subset of the plurality of measures within a second geographic region of data points, wherein
the first geographic region of data points is associated with high loss values due to the catastrophic event, and
the second geographic region of data points is associated with low loss values due to the catastrophic event.
18. The non-transitory computer-readable medium of claim 13 , wherein:
each location data point of the portion of the plurality of location data points in the compressed risk model further corresponds to a respective probability of a plurality of probabilities; and
compressing the risk model comprises generating, for each location data point of the portion of the plurality of location data points, a probability distribution comprising a list of measures including the measure and a corresponding list of probabilities including the probability.
19. The non-transitory computer-readable medium of claim 13 , wherein the instructions, when executed by the processing circuitry, cause the processing circuitry to perform the operations comprising performing, by the processing circuitry, a diagnostic fit calculation on the compressed risk model to confirm or improve accuracy of measures at the portion of the plurality of location data points retained in the compressed risk model.
20. The non-transitory computer-readable medium of claim 19 , wherein performing the diagnostic fit calculation comprises:
selecting a set of discarded data points of the plurality of location data points, wherein
each data point of the set of discarded data points is not included in the compressed risk model;
adding the set of discarded data points to the compressed risk model to generate an enhanced compressed risk model; and
for each data point of the set of discarded data points,
calculating a fitted measure value for the respective data point using a set of measures corresponding to the set of data points of the portion of the plurality of location data points geographically closest to the respective data point,
comparing the fitted measure value to the measure of the respective data point, and retaining the respective data point in the enhanced compressed risk model based on the fitted measure value meeting or exceeding an improvement criterion.