# US 2019/0087529 A1 · Decisions with Big Data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0087529 A1 - Document ID: US20190087529A1 - Status: Published application - Kind: A1 - Application number: 16/182,389 - Filed: 2018-11-06 - Published: 2019-03-21 - Earliest claimed: 2014-03-24 - Assignees: Imagars LLC - Inventors: Baldur Andrew Steingrimsson; Kwan Yi; Robert L. Jones; Mikalai Kisialiou - Web page: https://patentgrep.com/patent/US20190087529A1 ## CPC - G06N 20/10: using kernel methods, e.g. support vector machines [SVM] - G06F 16/2465: {Query processing support for facilitating data mining operations in structured databases} - G06F 16/3334: {Selection or weighting of terms from queries, including natural language queries} - G06F 16/3344: {using natural language analysis} - G06F 30/00: Computer-aided design [CAD] - G06F 30/10: Geometric CAD - G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model - G06N 3/045: Combinations of networks - G06N 3/0499: Feedforward networks - G06N 3/08: Learning methods - G06N 3/09: Supervised learning - G06Q 30/0202: Market predictions or forecasting for commercial activities - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR ## Abstract This invention presents a framework for applying artificial intelligence to aid with product design, mission or retail planning. The invention outlines a novel approach for applying predictive analytics to the training of a system model for product design, assimilates the definition of meta-data for design containers to that of labels for books in a library, and represents customers, requirements, components and assemblies in the form of database objects with relational dependence. Design information can be harvested, for the purpose of improving decision fidelity for new designs, by providing such database representation of the design content. Further, a retrieval model, that operates on the archived design containers, and yields results that are likely to satisfy user queries, is presented. This model, which is based on latent semantic analysis, predicts the degree of relevance between accessible design information and a query, and presents the most relevant previous design information to the user. ## Claims ### Claim 1 An apparatus for predictive analytics, one that utilizes vector entities archived from past design or planning projects, for the purpose of providing predictions aiding new design or planning projects, an apparatus comprising of a database module for storing the vector entities, with one vector entity representing an input vector from a past design or planning project, and the another vector representing a corresponding known good output vector, with a known good output vector being defined as an output vector that fulfills original requirements imposed on the archived design or planning project, a generic system module, for relating the output vector to the input vector, whose system model is derived from the archived vector entities, and where the system module can be made capable of supporting input vectors with elements taking on continuous or categorical values, where the values are confined to a range or not, the system module can be derived using multivariate linear regression, especially in case of input data set of small to medium size, or can be implemented using neural networks, especially in case of input data set of large size, the system module can be made capable of approximating a generic continuous-valued system model with a two-layer feed-forward neural network, a prediction module for producing a guiding reference for a new design or planning project, where the guiding reference is obtained as output from the system module when applying an input vector from the new design or planning project as input, an optional external database module, for querying for pertinent reference information helpful to the context of the new design or planning project, when necessary. ### Claim 2 A method for predictive analytics, one that utilizes vector entities archived from past design or planning projects, for the purpose of providing predictions aiding new design or planning projects, a method that further utilizes a database access step, for accessing the vector entities, with one vector entity representing an input vector from a past design or planning project, and another vector representing a corresponding known good output vector, with a known good output vector being defined as an output vector that fulfills original requirements imposed on the archived design or planning project, a generic system model access step, for relating the output vector to the input vector, where the system model is derived from the archived vector entities, and where the system model can be made capable of supporting input vectors with elements taking on continuous or categorical values, where the values are confined to a continuous range or not, the system model can be derived using multivariate linear regression, especially in case of input data set of small to medium size, or can be implemented using neural networks, especially in case of input data set of large size, the system model can be made capable of approximating a generic continuous-valued system model with a two-layer feed-forward neural network, a prediction step, for producing a guiding reference for a new design or planning project, a reference obtained as output from the system model when applying a vector from the new design or planning project as input, an optional access step to an external database or databases, for querying for pertinent reference information helpful to the context of the new design or planning project, when necessary. ### Claim 3 (depends on 1) An apparatus according to claim 1 for improving the fidelity of engineering design decisions on new design projects, an apparatus that utilizes vector entities from past design projects, the composition of said vector entities includes an input vector, which captures product design specification, an optional customer object, which is a part of the input vector, a requirement object, which is also a part of the input vector, an output vector, which captures design data, an optional component object, which is part of the output vector, and an optional assembly object, which is also a part of the output vector. ### Claim 4 (depends on 2) A method according to claim 2, for improving the fidelity of engineering design decisions on new design projects, a method utilizing vector entities from past design projects, the composition of said vector entities includes an input vector, which captures product design specification, an optional customer object, which is a part of the input vector, a requirement object, which is also a part of the input vector, an output vector, which captures design data, an optional component object, which is part of the output vector, and an optional assembly object, which is also a part of the output vector. ### Claim 5 (depends on 4) A method according to claim 4, where the structure of the database objects is determined based on function, and where the definition of a customer object includes a specification of a unique identifier for the customer, a specification of a name for the customer, an optional specification of a category for the customer, an optional specification of an importance level assigned to the customer, a method capable of interfacing with Hadoop and Apache Spark software frameworks, for distributed processing of large data sets, through application program interfaces provided by these software frameworks. ### Claim 6 (depends on 4) A method according to claim 4, where the structure of the database objects is determined based on function, and where the definition of a requirement object includes a specification of a unique identifier for the requirement, a specification of a name for the requirement, an optional specification of a customer with whom the requirement is associated, an optional specification of an importance level assigned to the requirement, an optional specifications of a category for the requirement, an optional specification of a function that the requirement is addressing, an optional specification of characteristics for the requirement, an optional specification of the minima and maxima defining the range, or ranges, of values corresponding to the requirement being fulfilled, an optional specification of the unit associated with the range, or ranges, of values corresponding to the requirement being fulfilled. ### Claim 7 (depends on 4) A method according to claim 4, where the structure of the database objects is determined based on function, and where the definition of an assembly object includes a specification of a unique identifier for the assembly object, a specification of a name for the assembly object, a specification of the requirement, or requirements, that the assembly object addresses, an optional specification of an assembly, or assemblies, subordinate to the assembly which the assembly object describes, an optional specification of the input applied to the assembly which the assembly object describes, an optional specification of the output produced by the assembly which the assembly object describes, an optional specification of the process applied to the assembly which the assembly object describes, an optional specification of the hierarchical rules or constraints, that the assembly, which the object describes, is subjected to. ### Claim 8 (depends on 4) A method according to claim 4, where the structure of the database objects is determined based on function, and where the definition of a component object includes a specification of a unique identifier for the component object, a specification of a name for the component object, a specification of the requirement, or requirements, that the component object addresses, an optional specification of the input applied to the component, which the object describes, an optional specification of the output produced by the component, which the component object describes, an optional specification of the process applied to the component, which the component object describes, a specification of the dimensions that apply to the component object, an optional specification of the material characteristics for the component object. ### Claim 9 (depends on 2) A method according to claim 2, for identifying anomalies in a set of continuous-valued or categorical data, a method employing comparison of distribution of each data variable, either by using quantiles or another statistical test, to see if the variations are significantly different, or counting of occurrence of each label or category and comparing, or a distance measure, such as the Mahalanobis distance, and looking for big changes, or applying an absolute difference between new and old data, setting a threshold, and reporting everything exceeding the threshold, or a multi-dimensional technique, like correlation matrix, principal components or clustering, and looking for changes, or a statistical or machine learning model, specialized for anomaly detection, such as Support Vector Machines, t-distributed Stochastic Neighbor Embedding, Isolation Forests, Peer Group Analysis or Break Point Analysis. ### Claim 10 (depends on 4) A method according to claim 4, where a mechanical engineering design is represented in a database in the form of a hierarchical structure consisting of a master assembly object, along with the requirements that the master level object addresses, sub-assembly objects and components, that depend on the master assembly object, along with the requirements that the sub-assembly or component objects address, sub-sub-assembly objects and components, that depend on the sub-assembly objects, along with the requirements that these sub-sub-assembly or component object address, subsequent levels of hierarchical representation of assembly objects and components, together with the requirements that these objects address, as needed to represent the design with the accuracy and detail desired. ### Claim 11 (depends on 4) A method according to claim 4, where the database is structured such that component options and risks can be programmed into the database based on existing engineering knowledge. ### Claim 12 (depends on 1) An apparatus according to claim 1, for automatically acquiring and continually updating asset performance models or tactical planning knowledge, in order to improve decision support by automating mission planning in highly dynamic environments, and for enabling maintainability, an apparatus comprising of an input vector capturing requirements for a tactical mission plan, an object, which is a part of the input vector, for specifying mission goals, an optional object, which is also a part of the input vector, for specifying expected mission performance, an output vector consisting of a tactical mission plan or an asset model, an optional object, which is part of the output vector, for registering a performance capability, and an optional object, which is part of the output vector, for registering an environmental model, an apparatus capable of interfacing with a Mine Warfare Environmental Decision Aid Library, a data warehouse, or another Tactical Decision Aid used for naval warfare mission planning. ### Claim 13 (depends on 2) A method according to claim 2, for automatically acquiring and continually updating asset performance models and tactical planning knowledge, in order to improve decision support by automated mission planning systems in highly dynamic environments, and for enabling maintainability, a method utilizing an input vector capturing requirements for a mission plan, an object, which is a part of the input vector, for specifying the mission goals, an optional object, which is also a part of the input vector, for specifying the expected mission performance, an output vector consisting of a tactical mission plan or asset model, an optional object, which is part of the output vector, for registering the performance capability, and an optional object, which is part of the output vector, for registering the environmental model, a method capable of interfacing with a Mine Warfare Environmental Decision Aid Library, a data warehouse, or another Tactical Decision Aid used for naval warfare mission planning. ### Claim 14 (depends on 1) An apparatus according to claim 1, for guiding near-autonomous generation of strike mission plans, an apparatus leveraging data, models and standards in a database for mission planning, an apparatus capable of supporting multi-domain, multi-asset mission planning as well as dynamic re-planning, an apparatus comprising of an input vector, for capturing requirement definition for a mission plan, an object, which is a part of the input vector, for capturing the mission goals, an optional object, which is also a part of the input vector, for capturing waypoints on the mission route, an output vector consisting of the mission plan itself, an optional object, which is part of the output vector, for capturing weapon selected, an optional object, which is part of the output vector, for capturing waypoint selected, an optional object, which is part of the output vector, for capturing assets coordinated, an optional object, which is part of the output vector, for capturing threats assessed, an optional object, which is part of the output vector, for capturing strike composition, an optional object, which is part of the output vector, for capturing fuel usage, an optional object, which is part of the output vector, for capturing time line developed, and an optional object, which is part of the output vector, for capturing communication plan, an apparatus capable of interfacing with a Joint Mission Planning System for strike mission planning. ### Claim 15 (depends on 2) A method according to claim 2, for guiding near-autonomous generation of strike mission plans, a method leveraging data, models and standards in a database for mission planning, a method utilizing an input vector containing a requirement definition for a mission plan, an object stating the mission goals, which is a part of the input vector, an optional waypoint object, which is also a part of the input vector, an output vector consisting of the mission plan itself, an optional object, which is part of the output vector, capturing weapon selected, an optional object, which is part of the output vector, capturing waypoint selected, an optional object, which is part of the output vector, capturing assets coordinated, an optional object, which is part of the output vector, capturing threats assessed, an optional object, which is part of the output vector, capturing strike composition, an optional object, which is part of the output vector, capturing fuel usage, an optional object, which is part of the output vector, capturing time line developed, and an optional object, which is part of the output vector, capturing communication plan, a method capable of interfacing with a Joint Mission Planning System for strike mission planning. ### Claim 16 (depends on 1) An apparatus according to claim 1, for improving retail planning and supply chain management, by validating forecasted category sales and sanitizing the planning process through provision of guiding references, an apparatus comprising of an input vector containing forecasted sales by store and category, an output vector consisting of the observed sales by store and category. ### Claim 17 (depends on 2) A method according to claim 2, for improving retail planning and supply chain management, by validating forecasted category sales and sanitizing the planning process through provision of guiding references, a method utilizing an input vector containing forecasted sales by store and category, an output vector consisting of the observed sales by store and category. ### Claim 18 (depends on 2) A method according to claim 2 for integration of digital product metrology information obtained from a large database using predictive analytics through metrology feature recognition, and for generation of an inspection plan from the recognized features. ### Claim 19 A method for querying a database, for the purpose of efficiently identifying archived database items matching a new user input query, a method utilizing a representation step, involving a sequence of text processes, for purpose of identifying vocabulary of terms in the user input query or in corpus of database items, a weighting step, in which a weighting scheme is applied to the vocabulary of terms, with resultant weights becoming values in a matrix for latent semantic analysis, a comparison step, based on latent semantic analysis, in which the degree of relevance between the input query and the corpus of database items is predicted, for the purpose of efficiently matching the input query with the corpus items, a retrieval step, for retrieving the database corpus items most relevant to the user input query. ### Claim 20 (depends on 19) A method according to claim 19, in which the representation step consists of tokenization, dropping common terms and stop words, normalization of terms, and stemming. ### Claim 21 (depends on 19) A method according to claim 19, in which the semantic analysis in the comparison step involves creation of a co-occurrence matrix, where the columns represent context or documents and the rows represent words or terms, a method where an entry (i,j) in the matrix corresponds to weight of word i appearing in context j, a method in which singular value decomposition is applied to the co-occurrence matrix, for the purpose of identifying hidden semantic relationships. ### Claim 22 (depends on 2) A method according to claim 2, for guiding near-autonomous generation of strike mission plans, a method leveraging data, models and standards in a database for mission planning, a method capable of supporting multi-vehicle, multi-domain mission planning in highly dynamic environments along with appropriate security levels, a method utilizing for this purpose requirement objects featuring a domain parameter, capturing a universal index to the mission domains, an asset parameter, capturing a universal index to the mission assets, a parameter accounting for the different security levels supported, capturing a universal index to the mission assets, an optional parameter specifying mission target selected, an optional parameter specifying location of the mission target selected, an optional parameter specifying the overall mission goals.