Adaptable property inspection model

Patent metadata

Field	Value
Publication number	US-10572947-B1
Application number	US-201514956580-A
Country	US
Kind code	B1
Filing date	Dec 2, 2015
Priority date	Sep 5, 2014
Publication date	Feb 25, 2020
Grant date	Feb 25, 2020

Abstract

Computer executable instructions may be used to configure a special purpose computing device to perform a method for determining whether an unbound property insurance lead may be associated with a property for which a property inspection is not required. The computer executable instructions, when executed by a processor may cause the special purpose computing device to access a data repository storing a plurality of unbound property insurance leads, process each of the plurality of unbound property insurance leads using an adaptable inspection model to determine an insurability risk score for each of the unbound property insurance leads, compare each insurability risk score to a constraint, and communicate an indication of whether a property inspection is necessary for each of the plurality of unbound property insurance leads.

First claim

What is claimed is:

1. A system comprising:
a data repository including information corresponding to a plurality of property insurance leads, property inspection data for previously inspected properties, and claims data for previously submitted claims;
a property inspection computing device comprising a processor;
a network communicatively coupling the data repository to the property inspection computing device; and
a non-transitory memory device storing instructions that, when executed by the processor cause the property inspection computing device to provide a computer-implemented sell-learning adaptable inspection model, including:
analyzing, by an inspection model generator, at least a portion of the property inspection data and the claims data;
generating, by the inspection model generator, one or more adjustments to the computer-implemented self-learning adaptable inspection model based on the analysis of the at least a portion of the property inspection data and the claims data, the adaptable inspection model configured to identify one or more risks associated with providing insurance coverage for the property;
applying, by an adjustment module of the inspection model generator, the one or more adjustments to the adaptable inspection model;
for each of the property insurance leads:
determining, by one or more computing devices, an insurability risk score for a property associated with the each of the property insurance leads using the adaptable inspection model, wherein determining the insurability risk score includes:
calculate a first model risk score corresponding to an insurability risk associated with insuring the property, the first model risk score determined using a first inspection risk model;
calculate a second model risk score corresponding to an insurability risk associated with insuring the property, the second model risk score determined using a second inspection risk model; and
calculate the insurability risk score based on the first model risk score and the second model risk score; and
comparing the insurability risk score to a constraint;
determining one or more of the property insurance leads for which the insurability risk score meets the constraint;
determining whether each of the one or more of the property insurance leads is to be inspected by applying an inspection percentage target;
generating a user interface displaying a geographic map including a plurality of first visual indicators identifying a location of each of the property insurance leads, along with a second visual indicator for each of the property insurance leads indicating whether or not an inspection of each property associated with each of the plurality of property insurance leads is to be ordered, the second visual indicator having a first display color, size or pattern when inspection of a property is to be ordered, and having a second display color, size or pattern, different from the first display color, size or pattern, when inspection of a property is not to be ordered;
updating at least the property inspection data based on the indication of whether an inspection of each property is to be ordered; and
self-revise the adaptable inspection model by:
further analyzing one or more portions of the updated property inspection data and the claims data to determine an adjustment to a weighting factor of the adaptable inspection model; and
further adjusting the adaptable inspection model based on the further analysis by applying the weighting factor adjustment.

2. The system of claim 1 ,
wherein the non-transitory memory device further comprises instructions that, when executed by the processor, cause the processor to:
calculate the insurability risk score using a first weighting factor applied to the first model risk score and a second weighting factor applied to the second model risk score, the first and second weighting factors configured to minimize costs associated with providing property insurance for a plurality of properties.

3. The system of claim 1 , wherein the non-transitory memory device further comprises instructions that, when executed by the processor, cause the processor:
calculate the insurability risk score using a first weighting factor to weight the first model risk score and a second weighting factor to weight the second model risk score, the weighting factors configured to identify a maximum number of potential property issues associated with providing property insurance for a plurality of properties.

4. The system of claim 1 , wherein the non-transitory memory device further comprises instructions that, when executed by the processor, cause the processor to:
apply a first weighting factor to the first model risk score and a second weighting factor to the second model risk score, the first and second weighting factors configured to minimize costs associated with providing property insurance for a plurality of properties and to identify a maximum number of property issues associated with providing property insurance for the plurality of properties.

5. The system of claim 1 , the instructions to determine the insurability risk score further comprise instructions that, when executed by the processor, cause the processor to:
calculate a first condition hazard metric and a first insurance-to-value metric using the first inspection risk model;
calculate a second condition hazard metric and a second insurance-to-value metric using the second inspection risk model; and
calculate the insurability risk score as a combination of the first condition hazard metric, the second condition hazard metric, the first insurance-to-value metric, and the second insurance-to-value metric.

6. The system of claim 1 , wherein the non-transitory memory device further comprises instructions that, when executed by the processor, cause the processor to:
generate a user interface screen indicating whether or not to inspect each property associated with each of the plurality of insurance leads.

7. The system of claim 1 , further comprising,
a remote user device comprising a processor running an insurance application, wherein the insurance application:
accesses, via the network, information associated with the plurality of property insurance leads including an indication of whether or not an inspection of each property associated with each of the plurality of insurance leads is to be ordered;
analyzes information associated with each of the plurality of property insurance leads to determine a geographic area including a majority of properties not requiring an inspection; and
generate a user interface screen for display on a user interface of the user device, wherein the user interface screen includes a visual representation of the geographic area including a majority of properties not requiring a property inspection.

8. The system of claim 7 , wherein the insurance application is further configured to: generate a visual representation of a geographic area including a majority of properties that require a property inspection.

9. The system of claim 1 ,
wherein applying the one or more adjustments includes adjusting at least one of the first insurance risk model and the second insurance risk model.

10. The system of claim 1 , wherein the adaptable inspection model operates using a first set of parameters when a request for inspection originates from a first geographical region and a second set of parameters when the request for inspection originates from a second geographical region.

11. The system of claim 1 , wherein the non-transitory memory device further comprises instructions that, when executed, cause the processor to:
communicate, via a network, to an originator of a

Assignees

• Allstate Insurance Co

Inventors

• Berends James M
• Macinnis David
• Collins Cassie E
• Dalenta Teresa J

Classifications

• G06Q50/163

  Real estate management · CPC title

• G06Q40/08 Primary

  Insurance · CPC title