Asset-level vulnerability and mitigation

1. A method comprising:

receiving a request for a damage propensity score for a parcel, the request specifying responsive to an occurrence of a real-time hazard event;

receiving hazard event data for the real-time hazard event, the hazard event data comprising a current set of hazard conditions of the real-time hazard event and including a degree of exposure of the parcel to the real-time hazard event;

receiving imaging data for the parcel, wherein the imaging data comprises street-view imaging data of the parcel;

extracting, by a machine-learned model comprising a plurality of classifiers and using object recognition, characteristics of a plurality of vulnerability features for the parcel from the imaging data, the plurality of vulnerability features comprising structures and vegetation;

determining, by the machine-learned model and from the characteristics of the plurality of vulnerability features and in response to the hazard event data for the real-time hazard event, the damage propensity score for the parcel, the damage propensity score indicating a measure of risk to the parcel including the characteristics of the plurality of vulnerability features of damage due to the real-time hazard event; and

providing a representation of the damage propensity score for the parcel responsive to the real-time hazard event for display.

2. The method of claim 1 , further comprising: generating, from the characteristics of the plurality of vulnerability features, a set of parcel characteristics.

3. The method of claim 1 , further comprising: generating, from the characteristics of the plurality of vulnerability features and imaging data for the parcel, a three-dimensional model of the parcel.

4. The method of claim 1 , wherein imaging data for the parcel comprises imaging data captured within a threshold of time from a time of the request.

5. The method of claim 1 , further comprising:

receiving, updated hazard event data for the real-time hazard event; and

determining, from the characteristics of the plurality of vulnerability features, the hazard event data, and the updated hazard event data, an updated damage propensity score for the parcel for the real-time hazard event.

6. The method of claim 1 , further comprising:

determining, by the machine-learned model and for the parcel, one or more mitigation steps;

determining, by the machine-learned model and based on the one or more mitigation steps, an updated damage propensity score; and

providing a representation of the one or more mitigation steps and the updated damage propensity score.

7. The method of claim 6 , wherein the one or more mitigation steps comprise adjustments to the characteristics of the plurality of vulnerability features extracted from the imaging data.

8. The method of claim 6 , wherein determining one or more mitigation steps further comprises:

iterating an updated damage propensity score determination based on adjusted characteristics of the plurality of vulnerability features.

9. The method of claim 8 , further comprising:

determining the updated damage propensity score meets a threshold damage propensity score.

10. The method of claim 6 , wherein determining the one or more mitigation steps comprises:

determining for a particular type of hazard event, the one or more mitigation steps, wherein one or more mitigation steps for a first type of hazard event is different than one or more mitigation steps for a second type of hazard event.

11. The method of claim 1 , further comprising:

generating training data for the machine-learned model, the generating comprising:

receiving, for a hazard event, a plurality of parcels located within a proximity of the hazard event, wherein each parcel of the plurality of parcels received at least a threshold exposure to the hazard event;

receiving, for each parcel of the plurality of parcels, imaging data for the parcel, wherein the imaging data comprises street-view imaging data; and

extracting, from the imaging data, characteristics of a plurality of vulnerability features for a first subset of parcels of the plurality of parcels that did not burn and for a second subset of parcels of the plurality of parcels that did burn during the hazard event; and

providing, to a machine-learned model, the training data.

12. The method of claim 11 , wherein extracting characteristics of the plurality of vulnerability features comprises, providing the imaging data to the plurality of classifiers.

13. The method of claim 12 , wherein extracting characteristics of the plurality of vulnerability features further comprises identifying, by the plurality of classifiers, a plurality of objects in the imaging data.

14. The method of claim 11 , further comprising:

receiving, for each parcel of the plurality of parcels, additional structural characteristics;

extracting, from the additional structural characteristics, a second plurality of vulnerability features for the first subset of parcels of the plurality of parcels that did not burn and for the second subset of parcels of the plurality of parcels that did burn during the hazard event; and

providing, to the machine-learned model, the second plurality of vulnerability features.

15. The method of claim 14 , wherein the additional structural characteristics comprises post-hazard event inspections of the plurality of parcels.

16. A non-transitory computer storage medium encoded with a computer program, the computer program comprising instructions that when executed by a data processing apparatus cause the data processing apparatus to perform operations comprising: receiving a request for a damage propensity score for a parcel, the request responsive to an occurrence of a real-time hazard event;

receiving hazard event data for the real-time hazard event, the hazard event data comprising a current set of hazard conditions of the real-time hazard event and including a degree of exposure of the parcel to the real-time hazard event;

receiving imaging data for the parcel, wherein the imaging data comprises street-view imaging data of the parcel;

extracting, by a machine-learned model comprising a plurality of classifiers and using object recognition, characteristics of a plurality of vulnerability features for the parcel from the imaging data, the plurality of vulnerability features comprising structures and vegetation;

determining, by the machine-learned model and from the characteristics of the plurality of vulnerability features and in response to the hazard event data for the real-time hazard event, the damage propensity score for the parcel, the damage propensity score indicating a measure of risk to the parcel including the characteristics of the plurality of vulnerability features of damage due to the real-time hazard event; and

providing a representation of the damage propensity score for the parcel responsive to the real-time hazard event for display.

17. The non-transitory computer storage medium of claim 16 , further comprising:

generating training data for the machine-learned model, the generating comprising:

receiving, for a hazard event, a plurality of parcels located within a proximity of the hazard event, wherein each parcel of the plurality of parcels received at least a threshold exposure to the hazard event;

receiving, for each parcel of the plurality of parcels, imaging data for the parcel, wherein the imaging data comprises street-view imaging data; and

extracting, from the imaging data, characteristics of a plurality of vulnerability features for a first subset of parcels of the plurality of parcels that did not burn and for a second subset of parcels of the plurality of parcels that did burn during the hazard event; and

providing, to a machine-learned model, the training data.

18. The non-transitory computer storage medium of claim 17 , further comprising:

receiving, for each parcel of the plurality of parcels, additional structural characteristics;

extracting, from the additional structural characteristics, a second plurality of vulnerability features for the first subset of parcels of the plurality of parcels that did not burn and for the second subset of parcels of the plurality of parcels that did burn during the hazard event; and

providing, to a machine-learned model, the training data.

19. A system comprising:

a user device; and

one or more computers operable to interact with the user device and to perform operations comprising:

receiving a request for a damage propensity score for a parcel, the request responsive to an occurrence of a real-time hazard event;

receiving hazard event data for the real-time hazard event, the hazard event data comprising a current set of hazard conditions of the real-time hazard event and including a degree of exposure of the parcel to the real-time hazard event;

receiving imaging data for the parcel, wherein the imaging data comprises street-view imaging data of the parcel;

extracting, by a machine-learned model comprising a plurality of classifiers and using object recognition, characteristics of a plurality of vulnerability features for the parcel from the imaging data, the plurality of vulnerability features comprising structures and vegetation;

determining, by the machine-learned model and from the characteristics of the plurality of vulnerability features and in response to the hazard event data for the real-time hazard event, the damage propensity score for the parcel, the damage propensity score indicating a measure of risk to the parcel including the characteristics of the plurality of vulnerability features of damage due to the real-time hazard event; and

providing a representation of the damage propensity score for the parcel responsive to the real-time hazard event for display.