Computer systems and methods for creating asset-related tasks based on predictive models

Patent metadata

Field	Value
Publication number	US-11017302-B2
Application number	US-201715469720-A
Country	US
Kind code	B2
Filing date	Mar 27, 2017
Priority date	Mar 25, 2016
Publication date	May 25, 2021
Grant date	May 25, 2021

Abstract

Computer systems, devices, and methods are provided for improving the technology related to asset condition monitoring. For instance, an asset data platform may be configured to receive data related to asset operation, ingest, process, and analyze the received data, and then provide a set of advanced tools that enable a user to monitor asset operation and take action based on that asset operation. The set of advanced tools may include (1) an interactive visualization tool, (2) a task creation tool, (3) a rule creation tool, and/or (4) a metadata tool.

First claim

The invention claimed is:

1. A computing system comprising:
a network interface that facilitates communication over a communication network with (a) a plurality of assets that are each located remote from the computing system and (b) a plurality of client stations that are each running a software application for creating a task related to the operation of an asset;
at least one processor;
a non-transitory computer-readable medium; and
program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor to cause the computing system to:
receive, from a given asset of the plurality of assets, data indicating asset events that have occurred at the given asset, wherein the given asset comprises an industrial, transportation, or utility machine that is equipped with sensors and an on-board computing system;
cause a first client station of the plurality of client stations to display a visual representation of a set of asset events that have occurred at the given asset and are selectable by a user of the first client station;
receive, from the first client station that is displaying a visual representation of the set of asset events that have occurred at the given asset, data defining a user request to create a new task for addressing at least one asset event of a given type that is selected from the displayed set of asset events that have occurred at the given asset;
in response to receiving the data defining the user request;
(i) generate a given task form that comprises a plurality of user-editable task fields for defining the new task, wherein at least one given task field of the plurality of user-editable task fields can be set to one of a plurality of predefined user-selectable options;
(ii) execute a set of predictive models for suggesting which predefined user-selectable option to select for the given task field of the given task form, wherein the set of predictive models is trained based on historical data comprising, for each of a plurality of previous tasks created for asset events of the given type, a respective identification of which one or more predefined user-selectable options were previously selected by a user for the given task field, and wherein each predictive model in the set of predictive models, when executed, functions to (a) evaluate data about the selected at least one asset event of the given type for which the new task is being created and (b) based on the evaluation, output a likelihood of a respective predefined user-selectable option being selected by the user of the first client station for the given task field in the given task form;
(iii) based on the output of the set of predictive models, generate a suggestion of one or more predefined user-selectable options to select for the given task field of the given task form; and
(iv) cause the first client station to display a visual representation of the given task form that is pre-populated with the suggested one or more predefined user-selectable options to select for the given task field of the given task form;

after causing the first client station to display the visual representation of the given task form, receive, from the first client station, task data for inclusion in the new task; and
based on the received task data, create the new task for addressing the selected at least one asset event of the given type.

2. The computing system of claim 1 , wherein the given task field comprises a field for selecting a knowledge article to include in the new task.

3. The computing system of claim 2 , wherein the suggestion of the one or more predefined user-selectable options comprises a ranked list of knowledge articles to include in the new task.

4. The computing system of claim 1 , wherein the suggestion of the one or more predefined user-selectable options to select for the given task field of the given task form includes, for each one of the suggested one or more predefined user-selectable options, a corresponding indication of the likelihood of the predefined user-selectable option being selected for the given task field in the given task form.

5. The computing system of claim 1 , wherein generating the suggestion of one or more predefined user-selectable options to select for the given task field of the given task form based on the output of the set of predictive models comprises, for each predictive model:
comparing the predictive model's output of the likelihood of the respective predefined user-selectable option being selected by the user of the first client station to a confidence threshold; and
if the predictive model's output exceeds the confidence threshold, including the respective predefined user-selectable option in the suggestion.

6. The computing system of claim 1 , further comprising program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor to cause the computing system to:
after creating the new task, cause a second client station of the plurality of client stations to display a visual representation of the new task.

7. The computing system of claim 1 , wherein the received task data for inclusion in the new task comprises task data representing a user-selected value for the given task field, and wherein the computing system further comprises program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor to cause the computing system to:
after receiving the task data for inclusion in the new task comprising the task data representing the user-selected value for the given task field, use the received task data to update the historical data to be used to train the set of predictive models for suggesting which predefined user-selectable option to select for the given task field of the given task form; and
cause the set of predictive models to be retrained based on the updated historical data.

8. The computing system of claim 1 , wherein the set of predictive models for suggesting which predefined user-selectable option to select for the given task field of the given task form comprises a first set of predictive models, the computing system further comprising program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor to cause the computing system to:
based on the data indicating asset events that have occurred at the given asset, execute a second set of predictive models for suggesting a new task to create based on the asset events that have occurred at the given asset, wherein the second set of predictive models is trained based on historical data indicating previous tasks that were created based on different combinations of asset events that had occurred at the plurality of assets, and wherein each predictive model in the second set of predictive models outputs a likelihood of a respective task being created for the asset events that have occurred at the given asset;
based on the output of the second set of predictive models, generate a suggestion of one or more tasks to create for the asset events that have occurred at the given asset; and
cause the first client station to display a visual representation of the suggested one or more tasks to create for the displayed set of asset events that have occurred at the given asset.

9. The computing system of claim 8 , wherein the data defining the user request to create the new task for addressing the at least one asset event that is selected from the displayed set of asset events comprise data defining a user request to create one of the suggested one or more tasks.

10. The computing system of claim 8 , wherein generatin

Assignees

• Uptake Tech Inc

Inventors

• Mcelhinney Adam
• Horrell Michael
• Liang Dennis

Classifications

• G06Q10/20 Primary

  Administration of product repair or maintenance · CPC title

• G06Q10/00

  Administration; Management · CPC title

• G06N5/04 Primary

  Inference or reasoning models · CPC title

• G06F9/30 Primary

  Arrangements for executing machine instructions, e.g. instruction decode (for executing microinstructions G06F9/22) · CPC title

• G06F15/16 Primary

  Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs {(coordinating program control therefor G06F9/52; in regulating and control system G05B)} · CPC title