Granted Patent B2

US 12,536,593 · App. 17/391,054 · Granted Jan 27, 2026

Inventors: Jason Crabtree (Vienna, VA); Andrew Sellers (Monument, CO)

Assignee: QOMPLX LLC

G06Q40/08G06F30/20

Abstract

Claims (50)

1 . A system for risk quantification for insurance process management employing an advanced insurance management and decision platform comprising:

a high-speed data retrieval and storage module stored in a memory of and operating on a processor of a computing device and configured to:

retrieve, via network interface, a plurality of insurance contract underwriting terms pertaining to a physical asset from a plurality of data sources;

extract and identify, using a scalable expressively scriptable connection interface, a plurality of risks associated with the physical asset based on the plurality of underwriting terms;

quantify a magnitude of each of the plurality of identified risks associated with the physical asset using real-time environmental data; and

identify an upcoming risk event in the geographical location of the physical asset; and

a data analysis module stored in a memory of and operating on the processor of the computing device and configured to:

receive, via real-time data connections, a plurality of first sensor data pertaining to the physical asset and a plurality of second sensor data pertaining to one or more individuals;

extract location data pertaining to the physical asset from the plurality of first sensor data by parsing location-specific components of telemetry signals;

analyze the plurality of first sensor data to determine an activeness metric of the physical asset, wherein the activeness metric comprises at least one of a measure of density, a measure of interaction frequency, and a measure of traffic associated with the physical asset;

retrieve an individual profile associated with the one or more individuals from a graph-based information storage service;

create a cognitive map based at least on the plurality of second sensor data and the individual profile, wherein the cognitive map comprises one or more risk-related decision making processes, wherein the cognitive map represents cognitive information as data sets stored in the memory;

assign a risk propensity score using the one or more specific risk-related decision making processes by applying decision-specific heuristic analysis;

combine, using a specialized data transformation pipeline, the activeness metric and the risk-propensity score based on the location data to produce a cognitive density metric associated with the physical asset or the location of the physical asset, wherein the cognitive density metric provides a technical representation of risk that cannot be calculated through conventional risk assessment; and

send the cognitive density metric to a predictive analytics module; and

the predictive analytics module stored in a memory of and operating on the processor of the computing device and configured to:

receive the cognitive density metric;

generate, using information theory-based predictive statistics functions and machine learning algorithms, a vulnerability model that relates the insurance contract underwriting terms, the magnitude of the plurality of risks, and the cognitive density metric to determine a risk response of the physical asset to the plurality of risks; and

apply the upcoming risk event to the determined risk response to predict a loss associated with the physical asset; and

a predictive simulation module stored in a memory of and operating on the processor of the computing device and configured to:

generate a blended exposures and losses model using discrete event simulator programming configured to analyze concentration of losses in terms of time, location, and cognitive density;

generate synthetic data through a generative adversarial network based on the magnitude of each of the plurality of risks and the predicted loss associated with the physical asset, wherein the generative adversarial network comprises a generator model and a discriminator model operating in a feedback loop to create high-quality synthetic data; and

pass the synthetic data through the blended exposures and losses model to determine a concentrated risk of loss associated with a plurality of assets of the same type as the physical asset.

2 . The system of claim 1 , wherein at least one predictive simulation algorithm of the discrete event simulator programming performs historical simulations based on actual loss data.

3 . The system of claim 2 , wherein at least one of the plurality of risks is geo-political conditions at one or more sites of client business operation.

4 . The system of claim 1 , wherein the individual profile comprises contextual information, cognitive information, demographic information, socioeconomic information, insured assets, insurance policy information, and location information.

5 . The system of claim 4 , wherein the location information is determined or inferred from historical location data of one or more individuals.

6 . The system of claim 4 , wherein the location information is determined from mobile devices and smart wearables.

7 . A method for risk quantification for insurance process management employing an advanced insurance management and decision platform, the method comprising the steps of:

retrieving, via a network interface, a plurality of insurance contract underwriting terms pertaining to a physical asset from a plurality of data sources;

extracting and identifying, using a scalable expressively scriptable connection interface, a plurality of risks associated with the physical asset based on the plurality of underwriting terms;

quantifying a magnitude of each of the plurality of identified risks associated with the physical asset using real-time environmental data;

identifying an upcoming risk event in the geographical location of the physical asset;

receiving, via real-time data connections, a plurality of first senor data pertaining to the physical asset and a plurality of second sensor data pertaining to one or more individuals' data;

extracting location data pertaining to the physical asset from the plurality of first sensor data by parsing location-specific components of telemetry signals;

analyzing the plurality first sensor data to determine an activeness metric of the physical asset, wherein the activeness metric comprises at least one of a measure of density, a measure of interaction frequency, and a measure of traffic associated with the physical asset;

retrieving an individual profile associated with the one or more individuals from a graph-based information storage service;

creating a cognitive map based at least on the plurality of second sensor data and the individual profile wherein the cognitive map comprises one or more risk-related decision making processes, wherein the cognitive map represents cognitive information as data sets stored in the memory;

assigning a risk propensity score using the one or more specific risk-related decision making processes by applying decision-specific heuristic analysis;

combining, using a specialized data transformation pipeline, the activeness metric and the risk-propensity score based on the location data to produce a cognitive density metric associated with the physical asset or the location of the physical asset, wherein the cognitive density metric provides a technical representation of risk that cannot be calculated through conventional risk assessment;

generating, using information theory-based predictive statistics functions and machine learning algorithms a vulnerability model that relates the insurance contract underwriting terms, the magnitude of the plurality of risks, and the cognitive density metric to determine a risk response of the physical assets to the plurality of risks;

applying the upcoming risk event to the determined risk response to predict a loss associated with the physical asset;

generating a blended exposures and losses model using discrete event simulator programming configured to analyze concentration of losses in terms of time, location, and cognitive density;

generating synthetic data through a generative adversarial network based on the magnitude of each of the plurality of risks and the predicted loss associated with the physical asset, wherein the generative adversarial network comprises a generator model and a discriminator model operating in a feedback loop to create high-quality synthetic data; and

passing the synthetic data through the blended exposures and losses model to determine a concentrated risk of loss associated with a plurality of assets of the same type as the physical asset.

8 . The method of claim 7 , wherein at least one predictive simulation algorithm of the discrete event simulator programming performs historical simulations based on actual loss data.

9 . The method of claim 8 , wherein at least one of the plurality of risks is geo-political conditions at one or more sites of client business operation.

10 . The method of claim 7 , wherein the individual profile comprises contextual information, cognitive information, demographic information, socioeconomic information, insured assets, insurance policy information, and location information.

11 . The method of claim 10 , wherein the location information is determined or inferred from historical location data of one or more individuals.

12 . The method of claim 10 , wherein the location information is determined from mobile devices and smart wearables.

Assignments (5)

Continuity (12)

References Cited (121)