# US 11,906,960 B1 · Bayesian risk inference engine (BRIE)

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,906,960 B1
- Document ID: US11906960B1
- Status: Granted
- Kind: B1
- Application number: 17/972,023
- Filed: 2022-10-24
- Granted: 2024-02-20
- Earliest claimed: 2022-10-24
- Assignees: Uptake Technologies, Inc.
- Applicants: Uptake Technologies, Inc.
- Inventors: Ruben Gonzalez
- Web page: https://patentgrep.com/patent/US11906960B1

## CPC

- G05B 23/0283: {Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] (preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06)}
- G05B 23/0216: {Human interface functionality, e.g. monitoring system providing help to the user in the selection of tests or in its configuration}
- G05B 23/024: {Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks}
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A computing platform is configured to: (a) generate a predicted health distribution of an asset for a failure mode based on a prior health distribution and a wear rate distribution. The computing platform is further configured to (b) update, the predicted health distribution based on (i) an observed state distribution corresponding to an observed state associated with the asset and (ii) a normalized value representative of a probability of the observed state over one or more health values of the asset. The computing platform is further configured to (c) generate a survival curve of the asset based on the predicted health distribution and a set of wear rates; iteratively perform (a)-(c) for each failure mode of the set of failure modes; aggregate the survival curve for each failure mode into an aggregate survival curve; and cause a client device to display a visual representation of the aggregate survival curve.

## Claims

### Claim 1

A computing platform for generating survival curves for an asset comprising:
a communication interface;
one or more processors;
a non-transitory computer-readable medium storing a Bayesian risk inference engine (BRIE); and
program instructions stored on the non-transitory computer-readable medium that are executable by the one or more processors to cause the computing platform to:
(a) generate, by executing the BRIE, a predicted health distribution of an asset for a failure mode of a set of failure modes based on a prior health distribution of the asset for the failure mode and a wear rate distribution based on sensor data corresponding to the asset,
(b) update, by executing the BRIE, the predicted health distribution based on (i) an observed state distribution corresponding to an observed state associated with the asset that is indicative of a health state of the asset represented by the predicted health distribution and (ii) a normalized value representative of a probability of the observed state over one or more health values of the asset,
(c) generate, by executing the BRIE, a survival curve of the asset based on the predicted health distribution and a set of wear rates, the survival curve including predictions of a survival probability for the asset at a plurality of future points,
iteratively perform (a)-(c) for each failure mode of the set of failure modes,
aggregate, by executing the BRIE, the survival curve for each failure mode into an aggregate survival curve of the asset, and
cause a client device to display on a display screen a visual representation of the aggregate survival curve of the asset, the visual representation depicting on the display screen predicted points of time when the asset is predicted to fail.

### Claim 2 (depends on 1)

The computing platform of claim 1, wherein the program instructions, when executed by the one or more processors, further cause the computing platform to update the predicted health distribution by:
aggregating, by executing the BRIE, a set of observation values during a time period associated with the predicted health distribution, wherein each observation value included in the set of observation values is representative of one or more observed states of an asset component during a portion of the time period; and
generating, by executing the BRIE, the observed state distribution based on the set of observation values.

### Claim 3 (depends on 1)

The computing platform of claim 1, wherein each failure mode corresponds to a respective probability of disabling the asset, and the program instructions, when executed by the one or more processors, further cause the computing platform to aggregate the survival curve for each failure mode into the aggregate survival curve of the asset by:
applying, by executing the BRIE, a respective disabling weighting value to the survival curve for each failure mode prior to aggregating the survival curve for each failure mode into the aggregate survival curve, wherein the respective disabling weighting value is based on a probability of disabling the asset corresponding to the failure mode represented by the survival curve.

### Claim 4 (depends on 1)

The computing platform of claim 1, wherein the program instructions, when executed by the one or more processors, further cause the computing platform to update the predicted health distribution by:
comparing, by executing the BRIE, the observed state to a set of prior observed states obtained within a time period threshold relative to the observed state;
determining, by executing the BRIE, whether or not the observed state is dependent from the set of prior observed states; and
responsive to determining that the observed state is dependent from the set of prior observed states, applying, by executing the BRIE, a dependency weighting value to the observed state that reduces an impact of the observed state on the predicted health distribution.

### Claim 5 (depends on 1)

The computing platform of claim 1, wherein the program instructions, when executed by the one or more processors, further cause the computing platform to update the predicted health distribution by:
analyzing, by executing the BRIE, a set of prior observed states and a set of prior health states to evaluate a plurality of underlying causes of the observed state;
determining, by executing the BRIE, a predominant cause of the plurality of underlying causes of the observed state; and
applying, by executing the BRIE, a cause weighting value to the observed state that adjusts an impact of the observed state on the predicted health distribution, wherein the weighting cause value is based on the predominant cause.

### Claim 6 (depends on 1)

The computing platform of claim 1, wherein the survival curve for each failure mode and the aggregate survival curve of the asset include predictions of the survival probability for the asset at the plurality of future points based on a plurality of maintenance potentials that includes: (i) no maintenance, (ii) planned maintenance, (iii) unplanned maintenance, and (iv) failure prior to maintenance.

### Claim 7 (depends on 1)

The computing platform of claim 1, wherein the program instructions, when executed by the one or more processors, further cause the computing platform to:
sample, by executing the BRIE, the prior health distribution and the wear rate distribution at a same plurality of points prior to generating the predicted health distribution;
generate, by executing the BRIE, the predicted health distribution based on the same plurality of points of both the prior health distribution and the wear rate distribution;
sample, by executing the BRIE, the observed state distribution at the same plurality of points prior to updating the predicted health distribution;
update, by executing the BRIE, the predicted health distribution based on the same plurality of points of the observed state distribution;
generate, by executing the BRIE, the survival curve based on the same plurality of points of the predicted health distribution; and
interpolate, by executing the BRIE, the aggregate survival curve from the same plurality of points of each survival curve prior to causing the client device to display the visual representation of the aggregate survival curve of the asset.

### Claim 8 (depends on 1)

The computing platform of claim 1, wherein
the predicted health distribution is a first predicted health distribution generated at a first time, and
at a second time subsequent to the first time, the first predicted health distribution is the prior health distribution.

### Claim 9

A computer-implemented method for generating survival curves for an asset comprising:
(a) generating, by one or more processors executing a Bayesian risk inference engine (BRIE), a predicted health distribution of an asset for a failure mode of a set of failure modes based on a prior health distribution of the asset for the failure mode and a wear rate distribution based on sensor data corresponding to the asset;
(b) updating, by the one or more processors executing the BRIE, the predicted health distribution based on (i) an observed state distribution corresponding to an observed state associated with the asset that is indicative of a health state of the asset represented by the predicted health distribution and (ii) a normalized value representative of a probability of the observed state over one or more health values of the asset;
(c) generating, by the one or more processors executing the BRIE, a survival curve of the asset based on the predicted health distribution and a set of wear rates, the survival curve including predictions of a survival probability for the asset at a plurality of future points;
iteratively performing (a)-(c) for each failure mode of the set of failure modes;
aggregating, by the one or more processors executing the BRIE, the survival curve for each failure mode into an aggregate survival curve of the asset; and
causing, by the one or more processors, a client device to display on a display screen a visual representation of the aggregate survival curve of the asset, the visual representation depicting on the display screen predicted points of time when the asset is predicted to fail.

### Claim 10 (depends on 9)

The computer-implemented method of claim 9, wherein updating the predicted health distribution further comprises:
aggregating, by executing the BRIE, a set of observation values during a time period associated with the predicted health distribution, wherein each observation value included in the set of observation values is representative of one or more observed states of an asset component during a portion of the time period; and
generating, by executing the BRIE, the observed state distribution based on the set of observation values.

### Claim 11 (depends on 9)

The computer-implemented method of claim 9, wherein each failure mode corresponds to a respective probability of disabling the asset, and wherein aggregating the survival curve for each failure mode into the aggregate survival curve of the asset further comprises:
applying, by executing the BRIE, a respective disabling weighting value to the survival curve for each failure mode prior to aggregating the survival curve for each failure mode into the aggregate survival curve, wherein the respective disabling weighting value is based on a probability of disabling the asset corresponding to the failure mode represented by the survival curve.

### Claim 12 (depends on 9)

The computer-implemented method of claim 9, wherein updating the predicted health distribution further comprises:
comparing, by executing the BRIE, the observed state to a set of prior observed states obtained within a time period threshold relative to the observed state;
determining, by executing the BRIE, whether or not the observed state is dependent from the set of prior observed states; and
responsive to determining that the observed state is dependent from the set of prior observed states, applying, by executing the BRIE, a dependency weighting value to the observed state that reduces an impact of the observed state on the predicted health distribution.

### Claim 13 (depends on 9)

The computer-implemented method of claim 9, further comprising updating the predicted health distribution by:
analyzing, by executing the BRIE, a set of prior observed states and a set of prior health states to evaluate a plurality of underlying causes of the observed state;
determining, by executing the BRIE, a predominant cause of the plurality of underlying causes of the observed state; and
applying, by executing the BRIE, a cause weighting value to the observed state that adjusts an impact of the observed state on the predicted health distribution, wherein the weighting cause value is based on the predominant cause.

### Claim 14 (depends on 9)

The computer-implemented method of claim 9, wherein the survival curve for each failure mode and the aggregate survival curve of the asset include predictions of the survival probability for the asset at the plurality of future points based on a plurality of maintenance potentials that includes: (i) no maintenance, (ii) planned maintenance, (iii) unplanned maintenance, and (iv) failure prior to maintenance.

### Claim 15 (depends on 9)

The computer-implemented method of claim 9, further comprising:
sampling, by the one or more processors executing the BRIE, the prior health distribution and the wear rate distribution at a same plurality of points prior to generating the predicted health distribution;
generating, by the one or more processors executing the BRIE, the predicted health distribution based on the same plurality of points of both the prior health distribution and the wear rate distribution;
sampling, by the one or more processors executing the BRIE, the observed state distribution at the same plurality of points prior to updating the predicted health distribution;
updating, by the one or more processors executing the BRIE, the predicted health distribution based on the same plurality of points of the observed state distribution;
generating, by the one or more processors executing the BRIE, the survival curve based on the same plurality of points of the predicted health distribution; and
interpolating, by the one or more processors executing the BRIE, the aggregate survival curve from the same plurality of points of each survival curve prior to causing the client device to display the visual representation of the aggregate survival curve of the asset.

### Claim 16

A tangible, non-transitory computer-readable medium storing instructions for generating survival curves for an asset, that when executed by one or more processors cause the one or more processors to:
(a) generate, by executing a Bayesian risk inference engine (BRIE), a predicted health distribution of an asset for a failure mode of a set of failure modes based on a prior health distribution of the asset for the failure mode and a wear rate distribution based on sensor data corresponding to the asset;
(b) update, by executing the BRIE, the predicted health distribution based on (i) an observed state distribution corresponding to an observed state associated with the asset that is indicative of a health state of the asset represented by the predicted health distribution and (ii) a normalized value representative of a probability of the observed state over one or more health values of the asset;
(c) generate, by executing the BRIE, a survival curve of the asset based on the predicted health distribution and a set of wear rates, the survival curve including predictions of a survival probability for the asset at a plurality of future points;
iteratively perform (a)-(c) for each failure mode of the set of failure modes;
aggregate, by executing the BRIE, the survival curve for each failure mode into an aggregate survival curve of the asset; and
cause a client device to display on a display screen a visual representation of the aggregate survival curve of the asset, the visual representation depicting on the display screen predicted points of time when the asset is predicted to fail.

### Claim 17 (depends on 16)

The tangible, non-transitory computer-readable medium of claim 16, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to update the predicted health distribution by:
aggregating, by executing the BRIE, a set of observation values during a time period associated with the predicted health distribution, wherein each observation value included in the set of observation values is representative of one or more observed states of an asset component during a portion of the time period; and
generating, by executing the BRIE, the observed state distribution based on the set of observation values.

### Claim 18 (depends on 16)

The tangible, non-transitory computer-readable medium of claim 16, wherein each failure mode corresponds to a respective probability of disabling the asset, and the instructions, when executed by the one or more processors, further cause the one or more processors to aggregate the survival curve for each failure mode into the aggregate survival curve of the asset by:
applying, by executing the BRIE, a respective disabling weighting value to the survival curve for each failure mode prior to aggregating the survival curve for each failure mode into the aggregate survival curve, wherein the respective disabling weighting value is based on a probability of disabling the asset corresponding to the failure mode represented by the survival curve.

### Claim 19 (depends on 16)

The tangible, non-transitory computer-readable medium of claim 16, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to update the predicted health distribution by:
comparing, by executing the BRIE, the observed state to a set of prior observed states obtained within a time period threshold relative to the observed state;
determining, by executing the BRIE, whether or not the observed state is dependent from the set of prior observed states; and
responsive to determining that the observed state is dependent from the set of prior observed states, applying, by executing the BRIE, a dependency weighting value to the observed state that reduces an impact of the observed state on the predicted health distribution.

### Claim 20 (depends on 16)

The tangible, non-transitory computer-readable medium of claim 16, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to update the predicted health distribution by:
analyzing, by executing the BRIE, a set of prior observed states and a set of prior health states to evaluate a plurality of underlying causes of the observed state;
determining, by executing the BRIE, a predominant cause of the plurality of underlying causes of the observed state; and
applying, by executing the BRIE, a cause weighting value to the observed state that adjusts an impact of the observed state on the predicted health distribution, wherein the weighting cause value is based on the predominant cause.