Artificial intelligence-based engineering requirements analysis
Methods and systems are described that use artificial intelligence (AI)-based natural language processing (NLP) techniques to rapidly analyze, merge, and/or identify relationships between systems engineering artifacts (e.g., system requirements, system architecture descriptions, research requirements, risk assessments, etc.) to accelerate systems engineering program management, technical capability development, and acquisition initiatives associated with systems engineering projects.
1. A computer-implemented method for automated analysis of engineering requirements comprising:
displaying, within a first region of a graphical user interface on an electronic display device, a first list of one or more engineering requirement documents, wherein the one or more engineering requirement documents comprise one or more domain-specific acronyms;
displaying, within a second region of the graphical user interface on the electronic display device, a second list of one or more engineering documents uploaded by the user;
receiving, using one or more processors, a first input from the user indicating a selection of one or more engineering requirement documents from the first list;
pre-processing, using a first machine learning model trained to convert domain-specific acronyms to natural language, the one or more selected engineering requirement documents from the first list to convert the one or more domain-specific acronyms into words:
processing the one or more pre-processed engineering requirement documents using one or more artificial intelligence (AI)-based natural language processing models to identify one or more key terms;
processing the one or more engineering documents of the second list using the one or more AI-based natural language processing models to identify text corresponding to the one or more key terms, wherein words of the one or more engineering documents of the second list and a mask comprising the one or more key terms are input to a transformer encoder of the one or more AI-based natural language processing models and an output of the transformer encoder is processed by a classification layer of the one or more AI-based natural language processing models that outputs the text corresponding to the one or more key terms; and
indicating, in the display of the second list of the one or more engineering documents within the second region of the graphical user interface, those engineering documents of the second list that include the text corresponding to the one or more key terms.
2. The computer-implemented method of claim 1 , further comprising determining, using the one or more AI-based natural language processing models, a relationship between the one or more engineering requirement documents selected from the first list and those engineering documents of the second list that include the text corresponding to the one or more key terms.
3. The computer-implemented method of claim 2 , wherein the relationship comprises a priority, a dependency, a causal constraint, a resource constraint, or a discretionary constraint.
4. The computer-implemented method of claim 1 , further comprising using a graphical indicator to indicate the one or more engineering requirement documents selected from the first list in the first region of the graphical user interface.
5. The computer-implemented method of claim 4 , wherein the graphical indicator comprises highlighting the one or more selected engineering requirement documents.
6. The computer-implemented method of claim 1 , further comprising using a graphical indicator to indicate those engineering documents of the second list that include the text corresponding to the one or more key terms in the second region of the graphical user interface.
7. The computer-implemented method of claim 6 , wherein the graphical indicator comprises highlighting those engineering documents of the second list that include the text corresponding to the one or more key terms.
8. The computer-implemented method of claim 6 , wherein the graphical indicator is different depending on a relationship between the one or more engineering requirement documents of the first list and those engineering documents of the second list that include the text corresponding to the one or more key terms.
9. The computer-implemented method of claim 1 , wherein the one or more engineering documents of the second list comprise system or software requirement documents, system or software specification documents, system or software design documents, system or software architecture documents, system or software research requirements, system or software data model documents, system or software risk factor documents, system or software test case documents, system or software quality requirements, or any combination thereof.
10. The computer-implemented method of claim 1 , wherein the graphical user interface is configured to allow the user to exchange documents and data with a commercial systems engineering software package.
11. The computer-implemented method of claim 1 , further comprising outputting an updated engineering requirements document that summarizes text extracted from those engineering documents of the second list that include the text corresponding to the one or more key terms.
12. The computer-implemented method of claim 11 , wherein the updated engineering requirements document complies with a professionally-recognized standard for engineering requirements documents.
13. The computer-implemented method of claim 1 , wherein the one or more AI-based natural language processing models comprise a Support Vector Machine (SVM) algorithm, a Bayesian Network algorithm, a Maximum Entropy algorithm, a Conditional Random Field algorithm, a Neural Network or Deep Learning algorithm, a Bidirectional Encoder Representations from Transformers (BERT) algorithm, a Term Frequency-Inverse Document Frequency (TF-IDF) algorithm, or any combination thereof.
14. The computer-implemented method of claim 1 , wherein the one or more AI-based natural language processing models have been trained on a training data set comprising domain-specific vocabulary words and acronyms.
15. The computer-implemented method of claim 1 , wherein the graphical user interface comprises at least a third region configured to allow the user to enter and define domain-specific vocabulary words and acronyms, and wherein the user-entered and defined domain-specific vocabulary words and acronyms are used to update a training data set used to train the one or more AI-based natural language processing models.
16. A computer-implemented method for automated analysis of engineering requirements comprising:
receiving a first set of one or more engineering requirement documents, wherein the one or more engineering requirement documents comprise one or more domain-specific acronyms;
receiving a second set of one or more engineering documents;
receiving a first input from a user indicating a selection of one or more engineering requirement documents from the first set;
pre-processing, using a first machine learning model trained to convert domain-specific acronyms to natural language, the one or more selected engineering requirement documents from the first set to convert the one or more domain-specific acronyms into words;
processing the one or more pre-processed engineering requirement documents using one or more processors configured to run one or more artificial intelligence (AI)-based natural language processing models to identify one or more key terms;
processing the one or more engineering documents of the second set using the one or more AI-based natural language processing models to identify text corresponding to the one or more key terms, wherein words of the one or more engineering documents of the second set and a mask comprising the one or more key terms are input to a transformer encoder of the one or more AI-based natural language processing models and an output of the transformer encoder is processed by a classification layer of the one or more AI-based natural language processing models that outputs the text corresponding to the one or more key terms;
identifying those engineering documents of the second set that include the text corresponding to the one or more key terms; and
determining, using the one or more AI-based natural language processing models, a relationship between the one or more engineering requirement documents selected from the first set and those engineering documents of the second set that include the text corresponding to the one or more key terms.
17. The computer-implemented method of claim 16 , wherein the relationship comprises a priority, a dependency, a causal constraint, a resource constraint, or a discretionary constraint.
18. The computer-implemented method of claim 16 , wherein the one or more engineering documents of the second list comprise system or software requirement documents, system or software specification documents, system or software design documents, system or software architecture documents, system or software research requirements, system or software data model documents, system or software risk factor documents, system or software test case documents, system or software quality requirements, or any combination thereof.
19. The computer-implemented method of claim 16 , further comprising outputting an updated engineering requirements document that summarizes text extracted from those engineering documents of the second list that include the text corresponding to the one or more key terms.
20. The computer-implemented method of claim 19 , wherein the updated engineering requirements document complies with a professionally-recognized standard for engineering requirement documents.
21. The computer-implemented method of claim 16 , wherein the one or more AI-based natural language processing-algorithms models comprise a Support Vector Machine (SVM) algorithm, a Bayesian Network algorithm, a Maximum Entropy algorithm, a Conditional Random Field algorithm, a Neural Network or Deep Learning algorithm, a Bidirectional Encoder Representations from Transformers (BERT) algorithm, a Term Frequency-Inverse Document Frequency (TF-IDF) algorithm, or any combination thereof.
22. The computer-implemented method of claim 16 , wherein the one or more AI-based natural language processing models have been trained on a training data set comprising domain-specific vocabulary words and acronyms.
23. The computer-implemented method of claim 16 , wherein the computer-implemented method is packaged as a containerized web application.
24. A system for performing automated analysis of engineering requirements comprising:
a display;
one or more input devices;
one or more processors; and
a memory device comprising a set of stored instructions which, when executed by the one or more processors, cause the system to:
receive a first set of one or more engineering requirement documents, wherein the one or more engineering requirement documents comprise one or more domain-specific acronyms;
receive a second set of one or more engineering documents;
receive a first input from a user indicating a selection of one or more engineering requirement documents from the first set;
pre-process, using a first machine learning model trained to convert domain-specific acronyms to natural language, the one or more selected engineering requirement documents from the first set to convert the one or more domain-specific acronyms into words;
process the one or more pre-processed engineering requirement documents using one or more processors configured to run one or more artificial intelligence (AI)-based natural language processing models to identify one or more key terms;
process the one or more engineering documents of the second set using the one or more AI-based natural language processing models to identify text corresponding to the one or more key terms, wherein words of the one or more engineering documents of the second set and a mask comprising the one or more key terms are input to a transformer encoder of the one or more AI-based natural language processing models and an output of the transformer encoder is processed by a classification layer of the one or more AI-based natural language processing models that outputs the text corresponding to the one or more key terms;
identify those engineering documents of the second set that include the text corresponding to the one or more key terms; and
determine, using the one or more AI-based natural language processing models, a relationship between the one or more engineering requirement documents selected from the first set and those engineering documents of the second set that include the text corresponding to the one or more key terms.
25. A non-transitory computer-readable storage medium storing one or more programs, the one or more programs comprising instructions which, when executed by one or more processors of an electronic device or system, cause the electronic device or system to:
receive a first set of one or more engineering requirement documents, wherein the one or more engineering requirement documents comprise one or more domain-specific acronyms;
receive a second set of one or more engineering documents;
receive a first input from a user indicating a selection of one or more engineering requirement documents from the first set;
pre-process, using a first machine learning model trained to convert domain-specific acronyms to natural language, the one or more selected engineering requirement documents from the first set to convert the one or more domain-specific acronyms into words;
process the one or more pre-processed engineering requirement documents using one or more processors configured to run one or more artificial intelligence (AI)-based natural language processing models to identify one or more key terms;
process the one or more engineering documents of the second set using the one or more AI-based natural language processing models to identify text corresponding to the one or more key terms, wherein words of the one or more engineering documents of the second set and a mask comprising the one or more key terms are input to a transformer encoder of the one or more AI-based natural language processing models and an output of the transformer encoder is processed by a classification layer of the one or more AI-based natural language processing models that outputs the text corresponding to the one or more key terms;
identify those engineering documents of the second set that include the text corresponding to the one or more key terms; and
determine, using the one or more AI-based natural language processing models, a relationship between the one or more engineering requirement documents selected from the first set and those engineering documents of the second set that include the text corresponding to the one or more key terms.