Systems and methods for generating assurance cases
AssigneeGeneral Electric Company
InventorsSaswata Paul, Baoluo Meng, Michael R. Durling, Kit Siu
Life of this patent
Estimating the term…
Abstract
A system, method, and program for generating self-consistent assurance cases using one or more computer models configured to receive a query from a user, to process the query and determine the nature of the assurance case to be generated, to determine and retrieve from a normalized, curated database the evidence to support the assurance case claims, and to populate an assurance case candidate with the same. The assurance case candidate may then be evaluated by a critic against one or more constraints and then either returned to the computer model for re-synthesis or forwarded to a translator to be converted into a human-readable format.
Claims18
A system for generating self-consistent assurance cases, the system comprising:
- a normalized database stored within a computer storage device and containing, as normalized data entries, evidence, an activity, an agent, and a relationship between two or more of:
- the evidence, the activity, and the agent;
- a raw data database containing raw data entries, wherein one or more of the normalized data entries of the normalized database are based on one or more of the raw data entries;
- a computational model interface configured to facilitate entry of at least one interrogation by a user to a computational model comprising a natural language request by the user for instructions explaining how to populate at least a portion of an assurance case candidate, wherein the computational model is executed by at least one processor and configured to process the at least one interrogation to generate a recommendation responsive to the interrogation based on information contained within the raw data database;
- an assurance case interface configured to:
- facilitate entry of at least one retrieval request by the user to the normalized database for retrieval of one or more of the plurality of normalized data entries, and
- facilitate entry by the user of at least some of the one or more normalized data entries into the assurance case candidate;
- a constraint library containing at least one constraint that is stored in a critic-readable format;
- and
- at least one critic configured to process the assurance case candidate generated by the user to determine whether the assurance case candidate satisfies the at least one constraint.
A method for generating self-consistent assurance cases, the method comprising the steps of:
- storing a plurality of normalized data entries within a normalized database, wherein the plurality of normalized data entries comprises evidence, an activity, an agent, and at least one relationship between two or more of:
- the evidence, the activity, and the agent;
- storing a plurality of raw data entries within a raw database, wherein one or more of the normalized data entries of the normalized database are based on one or more of the raw data entries;
- submitting through a computational model interface at least one interrogation to a computational model, wherein the interrogation comprises a natural language request by a user for instructions explaining how to populate at least a portion of an assurance case candidate;
- processing the interrogation to generate a recommendation responsive to the interrogation based on information contained within the raw database;
- submitting, through an assurance case interface, at least one retrieval request by the user to the normalized database for retrieval of one or more of the plurality of normalized data entries;
- entering at least some of the one or more normalized data entries into the assurance case candidate;
- and
- processing the assurance case candidate with at least one critic to determine whether the assurance case candidate satisfies at least one constraint.
A non-transitory computer-readable storage medium containing instructions that, when executed by at least one processor, cause the at least one processor to operate a user interface, wherein the user interface is configured to:
- receive, from a user, an interrogation comprising a natural language request by the user for instructions explaining how to populate at least a portion of an assurance case candidate;
- transmitting the interrogation to a computational model, wherein the computational model is configured to process the at least one interrogation to generate a recommendation responsive to the interrogation based on information contained within a raw data database;
- display the recommendation to the user;
- receive, from the user, a retrieval request for one or more of a plurality of normalized data entries from a normalized database;
- receive, from the user, instructions to enter at least some of the one or more normalized data entries so received into the assurance case candidate;
- and
- transmit, to at least one critic, the assurance case candidate for evaluation as to whether the assurance case candidate satisfies at least one constraint,
- wherein the normalized database contains as normalized data entries, evidence, an activity, an agent, and a relationship between two or more of:
- the evidence, the activity, and the agent, and
- wherein one or more of the normalized data entries of the normalized database are based on one or more of the raw data entries.
Dependencies are inferred from the claim text.