Condition-based maintenance techniques

1. A computer-implemented method of determining a condition-based maintenance hardware solution for monitoring a physical asset, the computer-implemented method comprising:

receiving, from a plurality of different sensor types, data representing corresponding sensed physical properties of a training asset;

receiving, from a first party, training asset failure information;

receiving information that represents a hardware model of at least the plurality of different sensor types; and

in response to 1) the data representing corresponding sensed physical properties of the training asset, 2) the training asset failure information, and 3) the information that represents the hardware model of at least the plurality of different sensor types:

1) Generating and displaying an output that represents a first hardware configuration for monitoring the physical asset, or 2) selecting a second hardware configuration from a set of pre-defined hardware configurations,

wherein the first hardware configuration and the second hardware configuration include at least one of the plurality of different sensor types.

2. The computer-implemented method of claim 1 , comprising:

generating and displaying the output that represents the first hardware configuration for monitoring the physical asset.

3. The computer-implemented method of claim 1 , comprising:

selecting the second hardware configuration from a set of pre-defined hardware configurations.

4. The computer-implemented method of claim 1 , comprising:

in response to 1) the data representing corresponding sensed physical properties of the training asset, 2) the training asset failure information, and 3) the information that represents the hardware model of at least the plurality of different sensor types:

training a machine learning model to be used by the first hardware configuration or the second hardware configuration to automatically predict a failure of a component of the physical asset.

5. The computer-implemented method of claim 4 , comprising:

receiving, from either the first hardware configuration or the second hardware configuration, 4) data representing corresponding sensed physical properties of the physical asset; and

in further response to 3) the data representing corresponding sensed physical properties of the physical asset:

updating the trained machine learning model.

6. The computer-implemented method of claim 1 , comprising:

using a trained machine learning model with the first hardware configuration or the second hardware configuration to automatically predict a failure of a component of the physical asset.

7. The computer-implemented method of claim 1 , wherein a first sensor type of the plurality of different sensor types has a first precision, and wherein the at least one of the plurality of different sensor types of the first hardware configuration is of the first sensor type and has a second precision less than the first precision.

8. The computer-implemented method of claim 1 , wherein a first sensor type of the plurality of different sensor types has a first sampling rate, and wherein the at least one of the plurality of different sensor types of the first hardware configuration is of the first sensor type and has a second sampling rate different than the first sampling rate.

9. The computer-implemented method of claim 1 , wherein the plurality of different sensor types includes multiple ones of a first sensor type, and wherein the at least one of the plurality of different sensor types of the first hardware configuration includes one or more fewer of the first sensor type.

10. The computer-implemented method of claim 1 , wherein the plurality of different sensor types includes multiple ones of a first sensor type, wherein the first sensor type of the plurality of different sensor types has a first precision, wherein the first sensor type of the plurality of different sensor types has a first sampling rate, wherein the plurality of different sensor types includes multiple ones of a first sensor type, and wherein the set of pre-defined hardware configurations includes:

a first pre-defined hardware configuration including at least one sensor of the first sensor type having a second precision less than the first precision;

a second pre-defined hardware configuration including at least one sensor of the first sensor type and having a second sampling rate less than the first sampling rate; and

a third pre-defined hardware configuration including one or more fewer of the first sensor type.

11. The computer-implemented method of claim 1 , comprising:

encrypting, by the first party, the training asset failure information.

12. The computer-implemented method of claim 1 , comprising:

encrypting the data representing corresponding sensed physical properties of the training asset.

13. A system for determining a condition-based maintenance hardware solution for monitoring a physical asset, the computer-implemented method comprising:

a computing device having a processor configured to:

receive, from a plurality of different sensor types, data representing corresponding sensed physical properties of a training asset;

receive, from a first party, training asset failure information;

receive information that represents a hardware model of at least the plurality of different sensor types;

in response to 1) the data representing corresponding sensed physical properties of the training asset, 2) the training asset failure information, and 3) the information that represents the hardware model of at least the plurality of different sensor types:

1) Generate and display an output that represents a first hardware configuration for monitoring the physical asset, 2) select a second hardware configuration from a set of pre-defined hardware configurations, or 3) configure, using software, a third hardware configuration;

wherein the first hardware configuration and the second hardware configuration include at least one of the plurality of different sensor types; and

use a trained machine learning model with the first hardware configuration or the second hardware configuration to automatically predict a failure of a component of the physical asset.

14. The system of claim 13 , wherein the processor is configured to:

generate and display the output that represents the first hardware configuration for monitoring the physical asset.

15. The system of claim 13 , wherein the processor is configured to:

select the second hardware configuration from a set of pre-defined hardware configurations.

16. The system of claim 13 , wherein the processor is configured to:

configure, using software, a third hardware configuration.

17. The system of claim 13 ,

wherein in response to 1) the data representing corresponding sensed physical properties of the training asset, 2) the training asset failure information, and 3) the information that represents the hardware model of the plurality of different sensor types, the processor is configured to:

train a machine learning model to be used by the first hardware configuration or the second hardware configuration to automatically predict a failure of a component of the physical asset.

18. The system of claim 13 , wherein the processor is configured to:

receive, from either the first hardware configuration or the second hardware configuration, 4) data representing corresponding sensed physical properties of the physical asset; and

in further response to 3) the data representing corresponding sensed physical properties of the physical asset:

update the trained machine learning model.

19. A computer-implemented method of determining a condition-based maintenance hardware solution for monitoring a physical asset, the computer-implemented method comprising:

receiving, from a plurality of different sensor types, data representing corresponding sensed physical properties of a training asset;

receiving, from a first party, training asset failure information;

receiving information that represents a hardware model of at least the plurality of different sensor types;

in response to 1) the data representing corresponding sensed physical properties of the training asset, 2) the training asset failure information, and 3) the information that represents the hardware model of at least the plurality of different sensor types:

1) Selecting a first hardware configuration from a set of pre-defined hardware configurations, or 2) configure, using software, a second hardware configuration;

wherein the first hardware configuration includes at least one of the plurality of different sensor types; and

using a trained machine learning model with the first hardware configuration or the second hardware configuration to automatically predict a failure of a component of the physical asset.

20. The computer-implemented method of claim 19 , comprising:

receiving, from either the first hardware configuration or the second hardware configuration, 4) data representing corresponding sensed physical properties of the physical asset; and

in further response to 3) the data representing corresponding sensed physical properties of the physical asset:

updating the trained machine learning model.