Piping and instrumentation planning and maintenance system

1. A piping and instrumentation planning and maintenance system comprising:

an input/output (I/O) interface for receiving a target piping and instrumentation diagram (PID) from a document source system;

a processor in communication with the I/O interface; and

non-transitory computer readable media in communication with the processor that stores instruction code, which when executed by the processor, causes the processor to:

select, based on a type of the target PID, a process pipeline of a plurality of process pipelines for processing PIDs;

classify, using the selected process pipeline, entities and properties thereof within the target PID, wherein the entities include one or more assets and interconnections therebetween specified in the PID;

compare the classified entities to a knowledge base that represents relationships between a plurality of assets and interconnections between the plurality of assets;

determine, based on the comparison, whether the one or more assets in the target PID are interconnected correctly; and

when the one or more assets are not interconnected correctly, generate a report to identify the one or more assets that are not interconnected correctly.

2. The system according to claim 1 , wherein the instruction code causes the processor to:

process a plurality of PIDs through one or more models for identifying entities of each PID to thereby train the one or more models to identify the entities of each PID; and

generate the knowledge base based on the entities identified in the plurality of PIDs.

3. The system according to claim 2 , wherein the models include an asset detection model, an asset connection model, a pipe detection model, and a text tag detection model.

4. The system according to claim 2 , wherein the instruction code causes the processor to:

generate a user interface that depicts a classified representation of the target PID;

receive alternative classifications for one or more of the entities of the classified representation to thereby update the classified representation of the target PID; and

train the one or more models with the updated classified representation of the target PID.

5. The system according to claim 2 , wherein the PIDs correspond to image documents and at least one model is limited to processing images smaller than the PIDs, wherein the instruction code causes the processor to:

split each PID into smaller images capable of being processed by the at least one model;

process the smaller images through the at least one model; and

combine the results of the at least one model.

6. The system according to claim 1 , wherein the knowledge base corresponds to an ontology and the ontology includes nodes that define documentation associated with assets.

7. The system according to claim 1 , wherein the instruction code causes the processor to:

receive, from a terminal, a query related to the selection and arrangement of assets;

generate a response to the query based on information in the knowledge base; and

communicate, to the terminal, the response to the query.

8. A method for planning and maintaining piping and instrumentation comprising:

receiving a target piping and instrumentation diagram (PID) from a document source system;

selecting, based on a type of the target PID, a process pipeline of a plurality of process pipelines for processing PIDs;

classifying, using the selected process pipeline, entities and properties thereof within the target PID, wherein the entities include one or more assets and interconnections therebetween specified in the PID;

comparing the classified entities to a knowledge base that represents relationships between a plurality of assets and interconnections between the plurality of assets;

determining, based on the comparison, whether the one or more assets in the target PID are interconnected correctly; and

when the one or more assets are not interconnected correctly, generating a report to identify the one or more assets that are not interconnected correctly.

9. The method according to claim 8 , further comprising:

processing a plurality of PIDs through one or more models for identifying entities of each PID to thereby train the one or more models to identify the entities of each PID; and

generating the knowledge base based on the entities identified in the plurality of PIDs.

10. The method according to claim 9 , wherein the models include an asset detection model, an asset connection model, a pipe detection model, and a text tag detection model.

11. The method according to claim 9 , further comprising:

generating a user interface that depicts a classified representation of the target PID;

receiving alternative classifications for one or more of the entities of the classified representation to thereby update the classified representation of the target PID; and

training the one or more models with the updated classified representation of the target PID.

12. The method according to claim 9 , wherein the PIDs correspond to image documents and at least one model is limited to processing images smaller than the PIDs, wherein the method further comprises:

splitting each PID into smaller images capable of being processed by the at least one model

processing the smaller images through the at least one model; and

combining the results of the at least one model.

13. The method according to claim 8 , wherein the knowledge base corresponds to an ontology and the ontology includes nodes that define documentation associated with assets.

14. The method according to claim 8 , further comprising:

receiving, from a terminal, a query related to the selection and arrangement of assets;

generating a response to the query based on information in the knowledge base; and

communicating, to the terminal, the response to the query.

15. A non-transitory computer readable media that stores instruction code for planning and maintaining piping and instrumentation, the instruction code being executable by a machine for causing the machine to:

receive a target piping and instrumentation diagram (PID) from a document source system;

select, based on a type of the target PID, a process pipeline of a plurality of process pipelines for processing PIDs;

classify, using the selected process pipeline, entities and properties thereof within the target PID, wherein the entities include one or more assets and interconnections therebetween specified in the PID;

compare the classified entities to a knowledge base that represents relationships between a plurality of assets and interconnections between the plurality of assets;

determine, based on the comparison, whether the one or more assets in the target PID are interconnected correctly; and

when the one or more assets are not interconnected correctly, generate a report to identify the one or more assets that are not interconnected correctly, wherein the report facilitates proactive replacement or rearrangement of assets in a facility associated with the target PID.

16. The non-transitory computer readable media according to claim 15 , wherein the instruction code causes the machine to:

process a plurality of PIDs through one or more models for identifying entities of each PID to thereby train the one or more models to identify the entities of each PID; and

generate the knowledge base based on the entities identified in the plurality of PIDs.

17. The non-transitory computer readable media according to claim 16 , wherein the models include an asset detection model, an asset connection model, a pipe detection model, and a text tag detection model.

18. The non-transitory computer readable media according to claim 16 , wherein the instruction code causes the machine to:

generate a user interface that depicts a classified representation of the target PID;

receive alternative classifications for one or more of the entities of the classified representation to thereby update the classified representation of the target PID; and

train the one or more models with the updated classified representation of the target PID.

19. The non-transitory computer readable media according to claim 16 , wherein the PIDs correspond to image documents and at least one model is limited to processing images smaller than the PIDs, wherein the instruction code causes the machine to:

split the PIDs into smaller images capable of being processed by the at least one model;

process the smaller images through the at least one model; and

combine the results of the at least one model.

20. The non-transitory computer readable media according to claim 15 , wherein the knowledge base corresponds to an ontology and the ontology includes nodes that define documentation associated with assets.