Automated Solution Manager
The system automates the selection of software solutions for large-scale industrial assets by guiding users through a dynamic question sequence, addressing the complexity of managing multiple lifecycle stages and ensuring stakeholder needs are met, thus simplifying the selection process.
Patent Information
- Application Number
- JP2026525084
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-08-20
- Filing Date
- 2024-10-24
- Publication Date
- 2026-11-19
AI Technical Summary
Complexity in managing large-scale industrial assets across multiple lifecycle stages poses a challenge for stakeholders to identify relevant software solutions that meet the needs of end users, organizational management, and IT infrastructure, often requiring expert knowledge and manual selection processes.
A computer-implemented system with a solution database, hierarchical tree structure, and graphical user interface that guides users through a dynamic sequence of questions to generate an output report recommending appropriate software solutions for large-scale industrial facilities, utilizing artificial intelligence to automate the selection process.
Enables non-experts to efficiently identify and assemble a customized set of software solutions that meet the needs of all stakeholders, reducing the complexity and time required for selection, and ensuring interoperability and alignment with organizational goals.
Smart Images

Figure 2026539061000001_ABST
Abstract
Description
TECHNICAL FIELD
[0001] Related Applications This application claims priority to U.S. Non-Provisional Application No. 18 / 810,290, filed August 20, 2024, entitled "Style-Guide Compliant Artificial Intelligence" (Attorney Docket No. 37402-20304), naming Jason Vick, John Craighton and George L. Jackson as inventors, and U.S. Non-Provisional Application No. 18 / 808,468, filed August 19, 2024, entitled "Automated Solution Manager" (Attorney Docket No. 37402-20303), naming Jason Vick, John Craighton and George L. Jackson as inventors, and also claims priority to U.S. Provisional Application No. 63 / 545,818, filed October 26, 2023, entitled "Automated Solution Manager" (Attorney Docket No. 37402-20301), naming Jason Vick and John Craighton as inventors.
[0002] The disclosure of each of the foregoing is incorporated herein by reference in its entirety.
[0003] Exemplary embodiments of the present invention generally relate to solution management systems, and more particularly, various embodiments of the present invention relate to platforms for effectively guiding users to appropriate solutions.
[0004] BACKGROUND ART Many enterprises have a large, complex set of solutions spanning the entire lifecycle of large-scale assets, from design, construction, operation, maintenance, cybersecurity and physical security, to decommissioning. These products may be sold to nearly any of dozens or hundreds of related industry subsectors. Their extremely high complexity poses a major obstacle for stakeholders to identify relevant products that can be implemented based on end users' goals, strategies, and challenges.
[0005] Summary of the Invention According to one embodiment of the present invention, a method is performed by a computer that manages multiple software solutions, each of which is useful for building or operating large-scale industrial equipment, and this method is The steps include providing a computer with a solution database that defines multiple separate software implementation solutions, A step of providing a data model to a computer, wherein the model consists of a hierarchical tree structure from more general nodes, terminating in multiple specific solutions from multiple separate software implementation solutions, and the data model is configured to work with a solution database to facilitate changes to the hierarchical tree structure in response to user input, and A step of generating a graphical user interface on a display device that communicates with a computer, which generates a dynamic sequence of questions in response to both a data model and answers to questions received from a user, wherein the answers include user input, and the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of the large-scale industrial facility, wherein the sequence of questions is (1) An initial question designed to elicit a first answer from the user, (2) Multiple consecutive subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, each subsequent question following a preceding question and determined by the user's answer to that preceding question, and the multiple consecutive subsequent questions ending with the last question, Steps including, A step in which an automated report builder is instructed to create an output report after receiving answers to a series of questions, wherein the automated report builder is configured to create an output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from multiple distinct software implementation solutions according to the answers. Includes.
[0006] In some embodiments of this method, the step of having an automated report builder create an output report includes the step of having the automated report builder automatically create a report after a series of questions have been completed.
[0007] In some embodiments of this method, the step of causing an automated report builder to create an output report is performed in response to a computer receiving a request to create an output report.
[0008] In some embodiments of this method, the solution database is indexed according to the tools and the solutions within the tools, each tool has tool information stored in the solution database, and each tool has one or more solutions from multiple separate software implementation solutions.
[0009] In some embodiments of this method, Each question is uniquely associated with a corresponding node from multiple nodes in a hierarchical tree structure. A graphical user interface includes, for each question, a predetermined set of selectable fields having predetermined selectable displays, the selectable displays being derived from a node corresponding to the question.
[0010] In some embodiments of this method, the graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a previously displayed preceding question.
[0011] In some embodiments of this method, the step is to provide a computer-implemented artificial intelligence, wherein the artificial intelligence is Accepting the vector defined by the answers to a series of questions, From the aforementioned vectors, generate an output report containing the specified set of solutions. It is configured in this way.
[0012] In some such embodiments, a computer-implemented artificial intelligence includes a neural network trained on a training set having multiple training vectors, where each training vector is A set of training answers to a specified series of questions, A specified set of solutions selected from multiple separate software implementation solutions, It has.
[0013] In some embodiments, the computer-implemented artificial intelligence includes generative artificial intelligence configured to produce an output report in response to prompts, the prompts including a vector defined by answers to a set of questions.
[0014] Another embodiment includes a computer implementation system for managing multiple software solutions, each of which is useful for building or operating large-scale industrial equipment, and the system is Computer memory, A solution database that defines multiple separate software implementation solutions, A data model comprising a hierarchical tree structure from more general nodes, ending in multiple specific solutions from multiple separate software implementation solutions, and configured to work in conjunction with a solution database to facilitate changes to the hierarchical tree structure in response to user input, and computer memory for storing the data model. A user interface module configured to generate a graphical user interface on a computer display device that generates a dynamic sequence of questions in response to both a data model and answers to questions received from a user, wherein the answers include user input, and the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of the large-scale industrial facility, (1) An initial question designed to elicit a first answer from the user, (2) Multiple consecutive subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, each subsequent question following a preceding question and determined by the user's answer to that preceding question, and the multiple consecutive subsequent questions ending with the last question, A user interface module, including, An automated report builder configured to generate an output report after receiving answers to a series of questions, wherein the output report is configured to generate an output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from multiple separate software implementation solutions according to the answers. Includes.
[0015] In some such system embodiments, the solution database is indexed according to the tools and the solutions within the tools, each tool has tool information stored in the solution database, and each tool has one or more solutions from a plurality of separate software implementation solutions.
[0016] In some embodiments, Each question is uniquely associated with a corresponding node from multiple nodes in a hierarchical tree structure. A graphical user interface includes, for each question, a predetermined set of selectable fields having predetermined selectable displays, the selectable displays being derived from a node corresponding to the question.
[0017] In some embodiments, the graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a previously displayed preceding question.
[0018] In some embodiments, it also includes computer-implemented artificial intelligence, wherein the artificial intelligence accepts a vector defined by answers to a series of questions, generates an output report including a specified set of solutions from the vector is configured as such.
[0019] In some such embodiments, the computer-implemented artificial intelligence comprises a neural network trained on a training set having a plurality of training vectors, each training vector has a set of training answers to a specified series of questions, and an associated set of specified solutions selected from a plurality of distinct software-implemented solutions, and .
[0020] Yet another embodiment includes a non-transitory computer-readable medium having computer-executable code that, when executed by a computer system, causes the computer system to perform a method, the method comprising providing, to the computer, a solution database that defines a plurality of distinct software-implemented solutions; providing a data model to the computer, wherein the model is configured in a hierarchical tree structure from more general nodes and terminates at a plurality of specific solutions from the plurality of distinct software-implemented solutions, and the data model is configured to cooperate with the solution database to facilitate modifications to the hierarchical tree structure in response to user input; generating, on a display device in communication with the computer, a graphical user interface that generates a dynamic sequence of questions in response to both the data model and answers to questions received from a user, wherein the answers include user input, and the series of questions is configured to request information from the user regarding issues related to the construction or operation of large-scale industrial equipment, wherein the series of questions (1) An initial question designed to elicit a first answer from the user, (2) Multiple consecutive subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, each subsequent question following a preceding question and determined by the user's answer to that preceding question, and the multiple consecutive subsequent questions ending with the last question, Steps including, A step in which an automated report builder is instructed to create an output report after receiving answers to a series of questions, wherein the automated report builder is configured to create an output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from multiple distinct software implementation solutions according to the answers. Includes.
[0021] In some such embodiments, the solution database is indexed according to the tools and the solutions within the tools, each tool has tool information stored in the solution database, and each tool has one or more solutions from multiple separate software implementation solutions.
[0022] In some such embodiments, each question is uniquely associated with a corresponding node from a plurality of nodes in a hierarchical tree structure, and the graphical user interface includes for each question a predetermined set of selectable fields having predetermined selectable displays, the selectable displays being derived from the node corresponding to the question.
[0023] In some embodiments, the graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a previously displayed preceding question.
[0024] In some embodiments, the method further includes the step of providing a computer-implemented artificial intelligence, wherein the artificial intelligence is Accepting the vector defined by the answers to a series of questions, From the aforementioned vectors, generate an output report containing the specified set of solutions. It is configured in this way.
[0025] An exemplary embodiment of the present invention is implemented as a computer program product having a computer-readable medium containing computer-readable program code. The computer-readable code may be read and utilized by a computer system according to a conventional process.
[0026] According to one embodiment, a method performed by a computer is: Steps include providing multiple examples of style guide-compliant text that conform to the company's specific style guide, and A step of providing an existing trained generative artificial intelligence, wherein the existing trained generative artificial intelligence is not customized to generate text that conforms to a company's specific style guide, The process involves customizing an existing trained generative AI by training it with multiple examples of style guide-compliant text to generate a style guide-compliant generative AI configured to generate style guide-compliant content from input text that does not conform to a company's specific style guide, and Includes.
[0027] In some embodiments, an existing trained generative artificial intelligence using multiple examples of style guide-compliant text includes the step of training the existing trained generative artificial intelligence using a generative adversarial network.
[0028] In some embodiments, the step of training an existing pre-trained generative artificial intelligence using multiple examples of style guide-compliant text includes the step of training an existing pre-trained generative artificial intelligence using a variational autoencoder.
[0029] In some embodiments, the step of training an existing pre-trained generative artificial intelligence using multiple examples of style guide-compliant text includes the step of training an existing pre-trained generative artificial intelligence using a diffusion model.
[0030] In some embodiments, a company has a graphical corporate logo, a style guide specifies rules for the proper use of the graphical corporate logo with text, multiple examples of style guide compliant text include examples of the proper use of the graphical corporate logo with text defined by the company's specific style guide, and style guide compliant generating artificial intelligence is trained to create text that complies with the rules for the proper use of the graphical corporate logo.
[0031] In some embodiments, a company has a corporate trademark expressed in text, a style guide specifies rules for the proper use of the corporate trademark, multiple examples of style guide-compliant text include examples of the proper use of the corporate trademark with other text defined by the company's specific style guide, and style guide-compliant generating artificial intelligence is trained to create text that complies with the rules for the proper use of the corporate trademark.
[0032] In some embodiments, a company has a corporate trademark expressed in text, a style guide designates the trademark only as an adjective, and multiple examples of style guide-compliant text include examples of appropriately using the corporate trademark only as an adjective in conjunction with other text defined by the company's specific style guide, and a style guide-compliant generating artificial intelligence is trained to create text that appropriately uses the corporate trademark as an adjective.
[0033] Some embodiments also include the steps of: receiving a suite vector containing text that does not conform to a particular style guide of a company; providing the suite vector to a style guide compliance generating artificial intelligence; and using the style guide compliance generating artificial intelligence to convert the suite vector into a style guide compliance report.
[0034] Another embodiment includes a computer implementation system that manages multiple software solutions, each useful for building or operating large-scale industrial facilities for a company. Computer memory, A solution database that defines multiple separate software implementation solutions, A data model comprising a hierarchical tree structure from more general nodes, ending in multiple specific solutions from multiple separate software implementation solutions, and configured to work in conjunction with a solution database to facilitate changes to the hierarchical tree structure in response to user input, and computer memory for storing the data model. A user interface module configured to generate a graphical user interface on a computer display device that generates a dynamic sequence of questions in response to both a data model and answers received from a user to the questions, wherein the answers include user input, and the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of the large-scale industrial facility, and the sequence of questions is (1) An initial question designed to elicit a first answer from the user, (2) Multiple consecutive subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, each subsequent question following a preceding question and determined by the user's answer to that preceding question, and the multiple consecutive subsequent questions ending with the last question, A user interface module, including, An automated report builder configured to generate a suite vector based on answers to a series of questions, wherein the suite vector identifies a suite of software implementation solutions selected from multiple separate software implementation solutions, includes text describing each solution within the suite of software implementation solutions, and the text does not conform to the company's style guide. A style guide compliant generating artificial intelligence trained to generate style guide compliant content from text in a sweet vector, Includes.
[0035] In some embodiments, the style guide compliant generative AI is trained from an existing trained generative AI, and the existing trained generative AI is not customized to create text that conforms to a company's specific style guide. The steps include receiving multiple examples of style guide-compliant text that conform to the company's specific style guide, The process involves transforming an existing trained generative artificial intelligence (AI) by training it with multiple examples of style guide-compliant text to generate style guide-compliant generative AI configured to generate style guide-compliant content from input text that does not conform to a company's specific style guide, and This trains a style guide-compliant generative artificial intelligence.
[0036] In some embodiments, style guide compliant generative artificial intelligence is trained using a generative adversarial network.
[0037] In some embodiments, style guide compliant generative artificial intelligence is trained using a variational autoencoder.
[0038] In some embodiments, style guide compliant generative artificial intelligence is trained using a diffusion model.
[0039] Another embodiment includes a non-temporary computer-readable medium having computer-executable code, which, when executed by a computer system, causes the computer system to perform a method. This code Code to retrieve multiple examples of style guide-compliant text that conform to a company's specific style guide from computer memory, Code for providing existing trained generative artificial intelligence, where the existing trained generative artificial intelligence is not customized to create text that conforms to a company's specific style guide, and This includes code for customizing an existing trained generative AI by training it with multiple examples of style guide-compliant text to generate a style guide-compliant generative AI configured to generate style guide-compliant content from input text that does not conform to a company's specific style guide.
[0040] In some embodiments, the code for customizing an existing trained generative artificial intelligence includes code for customizing an existing trained generative artificial intelligence using a generative adversarial network.
[0041] In some embodiments, the code for customizing an existing trained generative artificial intelligence includes the code for customizing an existing trained generative artificial intelligence using a variational autoencoder.
[0042] In some embodiments, the code for customizing an existing trained generative artificial intelligence includes code for customizing an existing trained generative artificial intelligence using a diffusion model.
[0043] In some embodiments, a company has a graphical corporate logo, and a style guide specifies rules for the proper use of the graphical corporate logo with text. Multiple examples of style guide compliant text include examples of the proper use of graphical corporate logos with text defined by a company's specific style guide. Style guide compliant generating artificial intelligence is trained to create text that conforms to the rules for the proper use of graphical corporate logos.
[0044] In some embodiments, a company has a corporate trademark expressed in text, and a style guide specifies rules for the proper use of the corporate trademark. Multiple examples of style guide compliant text include examples of the proper use of corporate trademarks with other text defined by a particular company's style guide. Style guide compliant generating artificial intelligence is trained to create text that complies with the rules for the proper use of corporate trademarks.
[0045] An exemplary embodiment of the present invention is implemented as a computer program product having a computer-readable medium containing computer-readable program code. The computer-readable code may be read and utilized by a computer system according to a conventional process.
[0046] Those skilled in the art will gain a more complete understanding of the advantages of various embodiments of the present invention from the following “Modes for Carrying Out the Invention,” which are described with reference to the drawings summarized below. [Brief explanation of the drawing]
[0047] [Figure 1] This figure shows an example of a practical framework for value engineering. [Figure 2] This figure shows a schematic example of an SDR for a large-scale capital project platform. [Figure 3] This diagram shows an outline example of the pre-value discovery stage 1 (self-discovery). [Figure 4] This diagram shows an overview of the value discovery stages 1 through 6 (business valuation). [Figure 5]This diagram shows a schematic example of value delivery (interlock with services). [Figure 6] This diagram shows an overview of value realization (value realization evaluation and case studies). [Figure 7A] This is a schematic example of a master map sample image. [Figure 7B] This is a schematic example of a master map sample image. [Figure 7C] This is a schematic example of a master map sample image. [Figure 7D] This figure shows a schematic example of a master map sample image (hereinafter collectively referred to as "Figure 7"). [Figure 8A] This figure shows a schematic example of a task map. [Figure 8B] This figure shows a schematic example of a task map. [Figure 8C] This figure shows an example of a problem map (hereinafter collectively referred to as "Figure 8"). [Figure 9A] This diagram shows an example of a solution map. [Figure 9B] This diagram shows an example of a solution map. [Figure 9C] This diagram shows an example of a solution map. [Figure 9D] This figure shows an example of a solution map (hereinafter collectively referred to as "Figure 9"). [Figure 10] This figure shows an overview example of business valuation. [Figure 11] This figure shows an overview example of business valuation. [Figure 12] This figure shows an overview example of business valuation. [Figure 13] This figure shows an overview example of business valuation. [Figure 14] This figure shows an overview example of business valuation. [Figure 15] This figure shows an overview example of business valuation. [Figure 16] This diagram shows a schematic example of ROI analysis. [Figure 17] This is a diagram illustrating a large-scale capital project. [Figure 18] This diagram schematically shows a high-level block diagram of a 3D design program. [Figure 19] This diagram shows a flowchart of one embodiment of the method. [Figure 20] This diagram schematically shows one embodiment of the system. [Figure 21] This diagram shows a flowchart of one embodiment of the method.
[0048] Modes for carrying out the invention The exemplary embodiment enables non-experts to address problems in a way that previously required expertise, in the form of deep knowledge of available solutions.
[0049] Many software products include a variety of functions and technical requirements. For example, a typical spreadsheet includes the ability to perform mathematical operations on data in spreadsheet cells, the ability to spell-check text entered by the user, and the ability to create and run macros that execute a set of predefined functions available in the spreadsheet software. Another example is a typical word processor, which includes the ability to format text entered by the user, the ability to spell-check text entered by the user, and the ability to create macro executions that execute a set of predefined functions available in the word processor software. However, a user who has word processor software or spreadsheet software but requires spell-check functionality in a different software product cannot use the spell-check functionality from the word processor or spreadsheet software in that different software product.
[0050] In another example, a user with a drawing document created using Draw software (part of the OpenOffice suite of open-source software products) may be unable to open, view, or edit that drawing document using Microsoft's Visio software due to differences in file formats between Draw and Visio, for example.
[0051] Furthermore, not all available software products run under the same operating system, and some may not work on computers lacking specific computer hardware (for example, computer-aided design software may not work on computers without graphics accelerator hardware). Therefore, an assembly of software products that appears attractive to a user may actually fail to meet the user's needs if one or more of those software products are unavailable on the user's computer system, or if they cannot be used in conjunction with other software on the user's computer system.
[0052] Furthermore, multiple levels of stakeholders (or "interested parties") within the organization may have the authority to influence which solutions are added to a set of solutions.
[0053] For example, the end-user community needs to ensure that the suite of solutions includes solutions to the specific challenges that end-users face. For instance, in the context of designing, building, and operating a power plant, end-users could include, to name a few, architects, builders, and operations and maintenance engineers.
[0054] Furthermore, organizational management must ensure that a suite of solutions includes solutions that facilitate the achievement of corporate objectives, even if they are not the primary concern of the end user. In the context of power plant design, construction, and operation, for example, corporate concerns may include keeping the project on schedule, increasing power plant uptime (or reducing power plant downtime), and limiting work injuries to plant workers. The suite of solutions may include solutions selected by the end user, or solutions different from those selected by the end user.
[0055] Furthermore, the organization's information technology management team needs to assemble or guide the assembly of a suite of solutions that meet the needs of other stakeholders (e.g., end users and organizational management), while also ensuring that the solutions within that suite work in conjunction with the organization's existing information technology infrastructure.
[0056] For these purposes, various embodiments automate and guide the selection of a set of solutions from an array of available solutions. For example, an array of available solutions may include several solutions, each addressing a specific need or concern, where one of them may satisfy the end user, while some of those solutions may not satisfy the needs of the organizational management and / or organizational information technology team. Similarly, an array of available solutions may include several solutions, each addressing a specific need or concern, where one of them may satisfy organizational management, while some of those solutions may not satisfy the needs of the end user and / or organizational information technology team. Similarly, an array of available solutions may include several solutions, each addressing a specific need or concern, where one of them may satisfy the information technology team and organizational management, while some of those solutions may not satisfy the needs of the end user and / or organizational management.
[0057] Typically, the task of selecting and assembling a solution suite that satisfies all stakeholders from multiple available solutions is left to the organization's information technology team.
[0058] Assembling a suite of software products that meets a user's specific needs presents several technical challenges. This requires not only knowledge of the user's specific needs (e.g., the user requires spell-checking functionality, the ability to run macros, the ability to edit documents in a given format), but also knowledge of which available software products possess one or more of the required functions, and which available software products can work with each other or are interoperable (e.g., which software products can work with files created using different software products). In some situations, the economics of acquiring and implementing a suite of software products (e.g., return on investment) is considered in the selection of software products.
[0059] The complexity of each software product within a large software product set has traditionally made it impossible for anyone to assemble a complete set of software products (to collectively meet the specific needs of a user) without expertise in both the functionality of each such product and its technical requirements and limitations.
[0060] In contrast, the exemplary embodiment makes the ability to assemble a set of software tools available to non-technical users (for example, when such non-technical users do not have expertise in both the functionality and technical requirements and limitations of each software product from a large group of available software products), without requiring such users to learn the complex technical and functional details of each software product from a multitude of available products.
[0061] For example, an organization (e.g., a corporation, a non-profit organization, an educational institution, a government agency) may face a problem that requires a software solution on or available on the organization's computer systems. Such an organization may engage one or more people to help select one or more available software products to address the problem individually or collectively. For example, these people may include a consultant, an information technology ("IT") manager of such an organization who is familiar with the available software products, and an end user familiar with the problem the organization faces. These people may consult with each other so that the consultant understands the problem the organization faces and the end user is aware of the potential solutions available for implementation on the organization's computer systems so that the end user can choose which of the available solutions to implement. In another example, a single person may be familiar with both the problem the organization faces and the potential solutions available for implementation on the organization's computer systems and be able to choose which of the available solutions to implement. For ease of explanation, the exemplary embodiments described herein may refer to both a consultant (e.g., an IT manager) and an "end user," but in practice, such embodiments may be carried out by a single person in both roles.
[0062] An exemplary embodiment helps a consultant (e.g., an information technology manager) determine which of several available solutions can be assembled into a solution suite that meets the needs of all stakeholders. The exemplary embodiment uses one or more graphical user interfaces that provide information about the various needs and concerns faced by the organization to users (e.g., end users, representatives of organizational management, information technology managers).
[0063] For example, such a user interface may be a first graphical user interface (sometimes called a “master map”) configured to request information about solutions necessary to carry out a project. Such a graphical user interface is configured to present questions to the user and receive the user’s responses to those questions. In an exemplary embodiment, the graphical user interface may present questions with several graphically represented options (for example, each option may be a predefined potential response to the question, such as in the form of a drop-down menu or a matrix of selectable options). Its format guides the user to provide information in a restricted format (e.g., not free-form) so that the information is useful for later analysis.
[0064] The user's responses to questions in the master map's graphical user interface narrow down the list of available solutions to a set of solutions that collectively meet the project's needs. However, that set of solutions may contain two or more solutions for each such issue, and therefore, the set of solutions may not be the final set of solutions, and furthermore, other factors (e.g., end-user concerns, management concerns, IT concerns) may still need to be addressed.
[0065] In exemplary embodiments, the user interface may include, in addition to, or instead of, the graphical user interface of the master map, a second graphical user interface (which may be called a “problem map” or “problem-to-solution map”) configured to request information about the end user’s needs. The problem map collects information to identify the “right solution” from among the available solutions, according to the end user’s goals, for example, to ensure that the end user has the tools available to do what needs to be done, for example, to build and / or operate and / or maintain a plant.
[0066] In exemplary embodiments, the user interface may include, in addition to the graphical user interface for the master map and / or the issue map, a third graphical user interface (which may be called the “solution map”) configured to request information about the needs of the organization’s administrators.
[0067] The information output by the aforementioned graphical user interface can be described as a “vector” or “suite vector.” This vector can be applied (for example, as input to artificial intelligence) to generate a set of proposed solutions from an array of available solutions that meet the needs of all stakeholders. In an exemplary embodiment, the vector generated from a user’s response to one or more of the graphical user interfaces described above includes information identifying each solution (or software product containing such solutions) that meets one or more of the stakeholders’ needs, along with a text description (e.g., in prose) of each such solution or software product. The description may include a description of the functionality of the solution or software product and may include a set of benefits provided by the solution. Other elements of the templated prompt may include common challenges that prevent stakeholders from achieving the aforementioned benefits, and thoughts that prompt questions to encourage further conversation.
[0068] In some exemplary embodiments, the system is configured to quickly identify relevant solutions from a complex portfolio of products, each of which may incorporate one or more solutions. Thus, the system enables the determination of a customized set of solutions for a specific end-user of the solution. In addition to benefiting the end-user of the solution, the system also helps consultants (e.g., IT managers) research and propose solutions. Details of exemplary embodiments are described below.
[0069] Definitions: As used herein and in the appended claims, unless otherwise required by context, the following terms shall have the meanings shown.
[0070] A "set" includes at least one member.
[0071] A "solution" is a feature of a software product that provides a specific function or capability of that software product.
[0072] Specifically, various embodiments provide a structured value engineering implementation framework that enables value-oriented solution identification, value delivery, and proactive value realization through structured processes targeting key organizational or end-user touchpoints throughout the suite's lifecycle. The framework is designed to establish interlocks among stakeholders to ensure a consistent value-oriented approach to developing a suite of software solutions. See Figure 1 for an example of a value engineering implementation framework that enables these goals.
[0073] The inventors identified pockets in value-based analytical efforts being carried out within organizations. Unfortunately, these pockets are typically not repeatable, scalable, comprehensive, or sustainable. As a result of their isolated nature, the current state of the art faces one or more of the following challenges:
[0074] Challenges for the consultant • The selection cycle is too long. • Focus on communicating the benefits of a product or service at the expense of linking them to the values or business outcomes they promote. • Pricing that does not align with the organization's expectations. For example, pricing may not meet the requirements for net present value (NPV), internal rate of return (IRR), or return on investment (ROI). Differentiation tends to focus on product features that are not related to how end users define value before or after implementation. There are very few evidence points (e.g., examples) to demonstrate successful outcomes when using the consultant's products. • We cannot obtain consensus from the end-user team. There is no process to help end users measure the value they realize in terms of business outcomes. • It is not possible to demonstrate the current value of the investment.
[0075] Challenges on the end-user side • Time spent engaging with end users during the evaluation process is limited. • A budget that does not align with minimum budget requirements or organizational requirements. • Internal politics hinder consensus among various end-user personas. • The realized value cannot be measured or presented.
[0076] Exemplary embodiments provide an automated, user-friendly, and simplified process for managing some or all aspects of the “value process.” This value process may be conceivable to have three distinct stages.
[0077] Value Discovery – Both end-users and consultants need a reason to initiate a meaningful conversation. End-users need direction for potential business value, and consultants need an efficient way to do the following: o Gain insights into the business needs of end users, o Gain insights into the business outcomes expected by end users, Quantifying the business value hypothesis, o Obtain the agreement of end users who will cooperate in developing the business case.
[0078] Value Delivery – This collaborative stage includes: o Development of a transparent and justifiable business case that is consistent with the value hypothesis. o Transferring ownership of the business case to the end user, o Converting a business case into a scope of work (SoW), Implement solutions based on SoW (Solutions of Work).
[0079] Value Realization – Following successful implementation, this stage includes the following: o To measure and report the actual value achieved relative to the original value hypothesis. o To guarantee end-user satisfaction, o Incorporate case studies, evidence, empirical evidence, and new use cases.
[0080] Exemplary embodiments address each of these three stages, as described below. Note that in various embodiments, the product has one or more “solutions,” each of which may be useful in completing a task. For example, a given product may have sub-products / processes / results A, B, and C, each of which is a solution. An end-user often does not need to acquire the entire product if they only need one of the solutions. The inventors have recognized that a consultant can extract only the appropriate solutions that meet the end-user’s needs. Continuing the above example, a given end-user may only need solutions B and C. Thus, various embodiments guide the end-user on a solution-based rather than product-based basis.
[0081] Phase 1: Value Discovery Value discovery can be considered, in some embodiments, to have at least two divisions: a pre-value discovery stage 1 and value discovery stages 1 through 6. Pre-value discovery stage 1 relates to end-user self-discovery, while stages 1 through 6 relate to the time consultants spend actively engaging with certified leads. Value discovery preferably leverages a set of discovery questions that enable both consultants and end-users to discover end-user pain points and challenges in line with the so-called “smart digital reality” (SDR). Figure 2 schematically shows an example of an SDR from a company producing a large-scale capital project platform (described in more detail below). This exemplary SDR operates from the planning stage to the security stage. This is known as the entry point for value discovery and will be outlined in more detail below.
[0082] Value discovery entry point The design of econometric models supports solutions for generating hierarchies aligned with the consultant's Consultant SDR. For example, a digital asset econometric model enables consultants to position multi-product solutions to end users. The same model also supports opportunities for “operational,” “maintenance,” or “secure” solutions. Not all opportunities will be multi-product solutions, but it is preferable that the model can focus on specific products within the consultant's Consultant Portfolio of available solutions. Models are delivered over time as the VE process matures. Therefore, to ensure that product-specific models are seamlessly integrated into the SDR, the consultant's Consultant Team is responsible for adding and continuously improving econometric models.
[0083] Pre-Value Discovery Stage 1 Pre-Value Discovery Stage 1 is part of the value engineering work process that focuses on leveraging value engineering assets such as case studies, empirical points, and web value models, helping stakeholders influence end-user behavior as they interact with the website during self-discovery and potentially impact the RFP. Figure 3 schematically illustrates an example.
[0084] Value Discovery Stages 1-6 In Stage 6, two opportunity IDs are created in the reminder system (for example, some embodiments use Salesforce® (GoSell)). The consultant generates a software opportunity and a service opportunity. Ideally, a Business Value Assessment (BVA) can be assigned to the two opportunity IDs, but if not, the consultant assigns the BVA to the software opportunity. See Figure 4 for an example.
[0085] Several embodiments can utilize light-touch / heavy-touch methods for BVA. Heavy-touch BVA should typically be performed by consultant value engineers and industrial consultants, targeting a certain percentage (e.g., 15%) of net new opportunities. These are the ones with a significant impact. Heavy-touch is where SIPOC, heatmaps, and other value engineering tools come into play.
[0086] Light Touch BVA must be executable by the stakeholders or account manager. Light Touch has similarities to the value hypothesis model. Consultants may target a portion of the net new opportunities (e.g., 85%) covered by the Light Touch methodology. The completion criteria for the net new software opportunities in Stage 2 should include a business valuation (light or heavy).
[0087] Phase 2: Value Delivery Value delivery is part of the work process that defines how both light and heavy business valuations are linked to delivery. Process design also allows for triggering revisions to the original business case as scope changes during implementation. See Figure 5 for an example of value delivery.
[0088] Phase 3: Value Realization After delivery is complete, the end-user's success manager (CSM), account manager, or end-user should be able to perform a value realization assessment. In an exemplary embodiment, the CSM is the professional responsible for ensuring that the end-users of the company or organization achieve the desired results and derive the maximum value from the product or service they have implemented. Some embodiments also use light-touch and heavy-touch techniques during this part of the process. If the BVA was a heavy-touch assessment, the value realization assessment (VRA) may also be a heavy-touch assessment. If the BVA was a light-touch assessment, it may be either a heavy-touch or light-touch assessment for the VRA. See, for example, Figure 6, which is an example of value realization.
[0089] The interaction of the components is important. For that purpose, the exemplary embodiment is a set of “mind maps” developed to help end users understand how their goals, concepts, strategies, and challenges can be addressed by utilizing (i.e., portfolio hierarchy) and various related solution areas, by enabling end users to derive value from significant value leakage areas within their existing work processes. For illustrative purposes, the exemplary embodiment may consider the use of the Hexagon Smart Digital Reality® suite. Therefore, the general description of smart digital reality, such as the Smart Digital Reality® suite, is for illustrative purposes only and is not intended to limit all embodiments.
[0090] An exemplary embodiment maintains three main mind maps. • Master map: Hierarchical structure of all available solutions, • Problem Map: A hierarchical structure that provides solutions to various questions / problems. • Solution Map: A hierarchical structure that allows end users to arrive at a set of appropriate solutions.
[0091] The following is a description of all three mind maps, their various embodiments, and how they relate to each other.
[0092] Master Map A master map refers to a physical mapping of a product from its execution strategy to its industrial work processes. Some or all execution strategies involve one or more business strategies related to multiple end-user types in the management of a plant or facility across various types of industries. Under these key strategies, organizations develop written work standards that define supporting work processes for properly performing duties based on organizational goals, industry best practices, and specific requirements for various job roles. Identifying specific products that provide the greatest value to end-users falls within the scope of these work processes.
[0093] Consultants (e.g., internal consultants) use this map to reference products in the appropriate context of how and where our end users apply them within the business. This map can be easily accessed via a publicly available URL and can be navigable with filters and tags that allow consultants to focus on the products or product groups of interest most to them and their end users. In addition to simply referencing the appropriate work processes with available products, it also serves as a repository of documented work processes in the form of "swimlane diagrams" that internal consultants can click to see and understand 1) how work processes are performed by different job roles and 2) how other products interact within those work processes.
[0094] As an administrator, this can be used as a “universe map” to record the work processes and workflows of a consultant as they develop models with different portfolio segments of the business. Specifically, in an exemplary embodiment, the universe map is a visual representation designed to illuminate multiple dimensions of the consultant’s solution related to the end user’s business objectives, strategies, challenges, operational workflows, and the tangible benefits that can be achieved by implementing the consultant’s techniques within these workflows.
[0095] As part of the model development workflow, consultants collaborate with internal SMEs, end-users, and analysts to identify industry-standard work processes related to smart digital reality for operation, maintenance, safety, design, planning, and execution. After these are recorded, consultants with expertise in the product and industry in question identify how their respective products fit into each recorded work process. Exemplary embodiments also include appropriate swimlane diagrams related to the standard work processes and embedding them in the map. Figure 7 shows an example of a master map.
[0096] Map updates can be completed in any number of ways, including using a product known as MindManager™, which allows you to create, edit, and publish master maps. Updates are typically performed during the model building process. Ideas for modifying or enhancing the mapping can also be submitted by any employee using our idea platform called Aha!. These ideas are tracked and reviewed by value engineering stakeholders and implemented if approved.
[0097] Issue map The challenge map concerns the physical mapping of available products from a business strategy to the challenges and benefits derived from the product's econometric model. Every business strategy has one or more business strategies related to multiple end-user types in the management of plants or facilities in various types of industries. Under these key strategies, organizations develop written work standards that define supporting work processes. At the work process standard level, several embodiments can identify specific products derived from the master map, along with common challenges encountered by end-users and the benefits of the products in addressing those challenges.
[0098] Among other methods, consultants can use this issue map to see how modeled benefits and consultant products connect to the smart digital reality at a finer level. The map is easily accessible via a publicly available URL and may be navigable with filters and tags that allow consultants to focus on products or groups of products as well as their relevant benefits and issues of greatest interest to the consultant and their end users. It also provides an internal consultant's perspective on how various solutions can address different aspects of common challenges within various work process standards. Thus, exemplary embodiments provide true solutions to end-user challenges. As an administrator, this can be used as a modeling map to record product / benefit relationships during the model building process. As part of the model development workflow, consultants can collaborate with internal SMEs, end users, and analysts to develop talk tracks that use these issues and benefits to assist end users and consultants during the business valuation process.
[0099] Map updates can be performed using the MindManager product mentioned earlier for creating, editing, and publishing maps. Updates are typically performed during the model building process. Ideas for modifying or enhancing the mapping can also be submitted by any employee using our idea platform, Aha!. These ideas are tracked and reviewed by value engineering stakeholders and implemented if approved. Figure 8 schematically shows an example of a problem map.
[0100] Solution Map This map physically maps the products within the consultant's platform from the company's objectives to the benefits of the solutions selected by the end user during the completion of the business valuation. The map presents the end user with three key questions that focus on their specific company objectives, business strategies, and challenges. The relationships between challenges, benefits, and products are derived from the challenge map, culminating in the selection of the appropriate solution.
[0101] Consultants use this map to lay out and reference the appropriate BVA sequencing on the consultant platform during the model building process. This map is easily accessible via a publicly available URL and is navigable using filters and tags, allowing listed internal consultants to focus on products or groups of products as well as related benefits and challenges. It also provides filtering capabilities to narrow down to specific end-user types and challenges. All of this helps in building the final template / model on the platform. Figure 9 shows an example of a solution map.
[0102] Solution map updates are performed by HxGN's strategic intelligence solution architects. They use a product called MindManager to create, edit, and publish the maps. Updates are typically performed during the model building process. Ideas for modifying or enhancing the mapping can also be submitted by any employee using our idea platform called Aha!. These ideas are tracked and reviewed by end-user value engineering stakeholders and implemented upon approval.
[0103] Therefore, unlike prior art solutions known to the inventors, the exemplary embodiment can simplify a large and complex portfolio of solutions across many different industry sub-sectors to disseminate product evaluation to a point where consultants and end-users can have a consistent, value-oriented dialogue, while building a robust and justifiable business case within hours rather than weeks.
[0104] Figures 10, 11, 12, 13, 14, and 15 (collectively, Figures 10-15) illustrate various examples of business value assessments, some of which involve questionnaires. These assessments are based on various mind maps, and in practice, later questions can be omitted or added depending on the answers to earlier questions. See Figure 16 for an example of ROI analysis.
[0105] As illustrated, some or all of these mind maps can be considered to form a hierarchical model extending from the highest level on the left to the lowest level on the right in the illustrated example. The model has selectable nodes to zoom in or out on the data shown (it can form a graphical user interface). Other embodiments use these mind maps to form a customized mind map for a given end user (i.e., a customized model) that reflects the end user's solution needs.
[0106] The exemplary embodiments are applicable to any of the various industries or business types. In fact, some of these business types have specific needs, such as businesses involving large-scale capital projects. As an example, Figure 17 provides an overall illustration of an example of a large-scale capital project 10 (or more generally, “Capital Project 10”) in which exemplary embodiments of the present invention can be implemented. More specifically, as is known to those skilled in the art, a capital project 10 is generally a long-term investment undertaken to build, enhance, add to, or improve a highly capital-intensive project, requiring the undertaking of both substantial amounts of financial and labor capital, and often taking several years to complete. A capital project 10 is often defined by its large cost compared to other investments that require less planning and resources (e.g., building a house or a truck). Both the private and public sectors can be involved in a capital project 10. Some examples of capital projects include the development and maintenance of roads, railways, power plants, ships, offshore oil platforms, dams, and factories.
[0107] The capital project 10 shown in Figure 17 is a power plant, and as is known to those skilled in the art, it has a vast number of different components that work together to achieve its power generation function. For example, among other things, the power plant in this figure has multiple large and small buildings, chimneys, pipes, valves, fuel tanks, ladders, and electrical systems. In fact, the design, construction, maintenance, and decommissioning of such a project requires a vast amount of planning and coordination. Without careful planning and coordination, the power plant may never be built or put into operation.
[0108] To this end, those skilled in the art have developed 3D design programs / products ("3D design program 12," also known as the "plant design program," Figure 18) to assist in the planning / design, development, maintenance, and decommissioning of capital projects 10 as shown in Figure 1. As mentioned above, one example of such a widely used 3D design program 12 is known as the SmartPlant® Enterprise product (hereinafter, the "SmartPlant® product"), distributed by Intergraph, Inc. of Huntsville, Alabama. Like other such products, the SmartPlant® product has at least the following interrelated functions and components: • 3D modeling and visualization, • Engineering and schematic diagrams, · Information management; Procurement, manufacturing, and construction • Open integration with other proprietary and open systems.
[0109] Therefore, among other people, designers, engineers, developers, and managers use these and other functions of 3D design programs 12, such as SmartPlant® products, to design, update, manage, and decommission capital projects 10, such as the power plant shown in Figure 17. In use, the 3D design program 12 generates many detailed three-dimensional representations of the components within the capital project 10. For example, the 3D design program 12 can generate a detailed 3D rendering of the piping system in the power plant in Figure 17. To achieve some function, the user can manipulate or modify the 3D rendering of specific components of the piping system, such as valves and brackets. For example, the user can render the 3D representation of the aforementioned pipe bracket on a display device, rotate it to determine or verify certain characteristics, and modify certain of its attributes.
[0110] Figure 18 schematically shows a high-level block diagram of a database 14 (e.g., a model database) containing attributes of various components of a 3D design program 12, such as a SmartPlant® product, and the capital project 10 it manages. Specifically, data on various systems and components of the capital project 10 is stored in the database 14. For example, the pipe bracket of the power plant shown in Figure 17 may have several different attributes, such as its shape, topology, geometric shape, dimensions, and material. Each of these attributes is stored in the database 14 and can be considered to have a particular "state" at any given time. In other words, the state of a component such as a pipe bracket is defined by the attributes in this database 14 at a given time. Thus, users and / or logic can access these attributes for any of several purposes, such as generating a visual representation of the bracket or modifying the bracket. Thus, users can modify the state of a component by modifying its attributes. As part of the 3D design program 12, the database 14 has a complete 360-degree dataset of information related to many of these components and systems, which can, for example, be rotated on a display device and managed in multiple orientations.
[0111] According to one embodiment of the present invention, an automated method and / or system provides a hierarchical data model having multiple solutions (e.g., master maps), a solution database having information on solutions and the benefits of solutions for a particular end user, and generates a graphical user interface having a set of questions. The method and / or system then generates a report based on the answers and / or questions based on the answers and / or end user profile / type. The report is generated automatically or in response to a user request. Based on the end user's profile and answers to business value assessments, the method and / or system has the ability to provide less than all the benefits that a given solution offers (in the report or elsewhere).
[0112] Some embodiments are similar to those described above. The focus is on solutions related to large-scale industrial projects (e.g., the construction of a ship or power plant from planning to construction, operation, and dismantling—not necessarily including the entire process, but relevant). These embodiments can also generate a second data model (e.g., an end-user specific model) depending on both the main / master data model and the answers from questions derived from business valuation.
[0113] The above automated methods and / or systems for large-scale capital can use a hierarchical data model having nodes and a graphical user interface having selectable fields. Each selectable field may have a predetermined selectable item derived from a node in the data model. Preferably, the graphical user interface automatically displays previously undisplayed questions after receiving an answer to the currently displayed question.
[0114] In some embodiments, a method and / or system can generate value representing improvements to a large-scale project when a user uses one of the recommended solutions. This value is generated by receiving input from the user into a process improvement model, and a graphical user interface also displays fields for receiving user input to generate this value.
[0115] According to other embodiments, a method and / or system focused on configuring a large-scale project platform uses a hierarchical structure having a first node related to the functionality of the project. This method / system generates requirement nodes downstream of the first node, determines solutions (at several lower levels) from multiple distinct products, and those solutions return to the top-level node, forming a hierarchical project data model with other levels. Furthermore, this method / system displays the top-level node with markers for collapsing or expanding lower-level branches and generates reports with one or more values representing improvements to the large-scale management platform from the solutions.
[0116] Exemplary embodiments may be implemented at least partially using a network. For that purpose, at runtime, such embodiments may request access to a remote computer system over a wide-area packet-based network operating at least partially on a network layer protocol. This access request may be performed by sending a synchronization message and receiving an acknowledgment message to initiate the formation of a bidirectional virtual connection to the remote computer system. Multiple packets can then be encrypted and forwarded to the remote computer device to generate a secure encrypted connection.
[0117] Some embodiments relate to a system for recommending solutions for large-scale projects. This system comprises a solution database for large-scale projects and a rule engine that traverses a hierarchical tree of questions and uses the answers and nodes of the tree to traverse towards one or more solutions in the solution database.
[0118] The features and / or aspects of the various embodiments described above can be combined.
[0119] Figure 19 is a flowchart of one embodiment of the method. Exemplary embodiments are implemented on a computer system and are sometimes referred to as “computer implementation” methods.
[0120] Step 1910 includes providing a solution database to the computer system. Providing a solution database may include adding the solution database to the computer system's memory or making the solution database available to the computer system, for example, via a communication link from a remote database.
[0121] A solution database defines or describes multiple separate software implementation solutions. For example, each separate software implementation solution may have associated functions or capabilities that can be described in the solution database for that software implementation solution.
[0122] In some embodiments, the solution database is indexed according to the tools and the solutions within the tools, each tool has tool information stored in the solution database, and each tool has one or more solutions from multiple separate software implementation solutions.
[0123] Step 1920 involves providing a data model to a computer system, the model consisting of a hierarchical tree structure from more general nodes and ending with multiple specific solutions from multiple separate software implementation solutions. In an exemplary embodiment, the data model is configured to work with a solution database to facilitate changes to the hierarchical tree structure in response to user input. Providing the data model may include adding the data model to the computer system's memory or making the data model available to the computer system via a communication link from, for example, a remote database.
[0124] A data model can include data specific to the corresponding industry, or to a particular type of project within a specific industry. For example, the power generation industry may have concerns and challenges not relevant to the pharmaceutical manufacturing industry, and this may have concerns and challenges not relevant to the shipping industry.
[0125] Step 1930 includes generating a graphical user interface on a display device that communicates with a computer system, which generates a dynamic set of questions in response to both a data model and answers to questions received from a user of the computer system. Such answers may be referred to as "user input".
[0126] In some embodiments, such a graphical user interface can be generated by inputting information from a data model into a tree. The generated graphical user interface is then customized for a corresponding industry or a corresponding project within a specific industry. The model may be stored in memory or a remote database, accessed by a computer system, and used by the computer system to generate the customized graphical user interface.
[0127] In an exemplary embodiment, each question is uniquely associated with a corresponding node from a plurality of nodes in a hierarchical tree structure, and the graphical user interface includes for each question a predetermined set of selectable fields having predetermined selectable marks, the selectable marks being derived from the node corresponding to the question.
[0128] A series of questions generated by a graphical user interface are structured to elicit information from the user regarding the challenges of the project. For example, the project might involve constructing or operating a large-scale industrial facility.
[0129] For example, a series of questions may include (1) an initial question structured to elicit a first answer from the user, and (2) a series of subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, where each subsequent question follows a preceding question and is determined by the user's answer to that preceding question, and the series of subsequent questions ends with a final question.
[0130] In exemplary embodiments, the series of questions posed by the graphical user interface are not static. Rather, the series of questions presented by the graphical user interface are dynamic in that one or more subsequent questions are determined based on the user's responses to one or more previous questions.
[0131] In some embodiments, the graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a previously displayed preceding question.
[0132] In some embodiments, a collected set of answers to questions presented by a graphical user interface forms a vector defined by the answers to a set of questions.
[0133] Step 1940 includes generating an output report after receiving answers to a series of questions, or having an automated report builder generate one. The automated report builder is configured to create an output report based on the answers, and the output report includes a specified set of solutions selected from a solution database based on the answers. In other words, each such specified solution is selected from multiple distinct software implementation solutions (e.g., from the solution database) based on the answers. In some embodiments, having the automated report builder create an output report includes automatically having the automated report builder generate a report after completing a series of questions. In some embodiments, having the automated report builder create an output report is done in response to the computer receiving a request to create an output report.
[0134] Automatically generating an output report may include as part of the output report a list of solutions identified using a graphical user interface, and / or a list of software products that provide those solutions. In an exemplary embodiment, the solution database may also include text descriptions (e.g., in prose) of the identified solutions, or text descriptions of software products containing the identified solutions, and / or text descriptions of the benefits associated with each solution, and automatically generating an output report may include listing or fitting such descriptions together with their associated solutions.
[0135] Some embodiments include providing a computer-implemented artificial intelligence (which may be called “reporting artificial intelligence”), the AI configured to accept a vector defined by answers to a set of questions (including, for example, text descriptions as described above), and to produce an output report from the vector containing a specified set of solutions. In some embodiments, the computer-implemented artificial intelligence comprises a neural network trained on a training set having multiple training vectors, each training vector containing a set of trained answers to a specified set of questions, and a corresponding set of specified solutions selected from a set of separate software implementation solutions. In some embodiments, the “reporting artificial intelligence” includes generative artificial intelligence configured to produce an output report in response to a prompt, the prompt containing a vector defined by answers to a set of questions.
[0136] Figure 20 schematically shows an embodiment of System 2000 configured to carry out embodiments of the methods and processes described herein. In exemplary embodiments, System 2000 includes a set of modules. Each module may be implemented by hardware (e.g., electrical circuits), software running on a microprocessor, or a combination of hardware and software. In some embodiments, two or more such modules can communicate data and / or control over a communication backbone 2001 (e.g., a “bus”).
[0137] System 2000 includes a communications module 2010 configured to communicate with one or more remote computers 2002 or databases 2005. Such communication may take place over a network (e.g., a local area network, a wide area network, or the Internet).
[0138] System 2000 also includes a solution database as described herein (for example, defining or describing multiple separate software implementation solutions) and computer memory 2060 for storing the data model as described herein. The data model consists of a hierarchical tree structure from more general nodes, terminating in multiple specific solutions from multiple separate software implementation solutions, and is configured to work with the solution database to facilitate changes to the hierarchical tree structure in response to user input. In other embodiments, the solution database and / or data model may be stored remotely from System 2000, for example, on a remote computer 2002 and / or database 2005.
[0139] System 2000 also includes a user interface module 2020 configured to generate a graphical user interface on a computer display device that generates a dynamic series of questions (e.g., as described above) in response to both a data model and answers to questions received from a user (which may be called "user inputs"), the series of questions configured to solicit information from the user regarding challenges relating to the construction or operation of the large-scale industrial facility. In an exemplary embodiment, the series of questions includes (1) an initial question configured to solicit a first answer from the user, and (2) a series of subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, where each subsequent question follows a preceding question, determined by the user's answer to the preceding question, and the series of subsequent questions ends with a final question.
[0140] System 2000 also includes an automated report builder module 2030 configured to generate an output report after receiving answers to a set of questions. The automated report builder 2030 is configured to generate an output report based on the answers, and includes a specified set of solutions, each of which is selected from several separate software implementation solutions based on the answers.
[0141] Some embodiments of System 2000 also include a report artificial intelligence 2040 configured (i.e., trained) to produce an output report (containing a specified set of solutions) based on a suite vector defined by (e.g., consisting of) user answers to a set of questions.
[0142] A step of training a neural network to produce an output report from vectors, wherein the output report includes a set of suggested solutions from a solution database, the step may include (a) accessing a set of training datasets, each training dataset containing test vectors and having a set of associated solutions for each such test vector.
[0143] Training a neural network includes (b) creating a first training set containing a set of training datasets, and (c) training the neural network in a first stage using the first training set to produce an output report. In an exemplary embodiment, the first training set contains at least 5,000 such training datasets.
[0144] In some embodiments, the step of training a neural network to produce output reports from vectors further includes the steps of (d) providing a second training set comprising a second training dataset, each having an associated incorrect output report, and (e) training the neural network in a second stage using the second training set, wherein the second stage improves the neural network's ability to correctly produce output reports.
[0145] In some embodiments, the system 2000 also includes a generative artificial intelligence 2050 module configured to generate text and / or figures based on answers to a set of questions. Such a generative artificial intelligence 2050 may be considered part of the report builder 2030 or may be a separate module 2050.
[0146] In exemplary embodiments, a vector (which may be called a “suite vector”) generated from a user’s response to one or more graphical user interfaces described above includes information identifying each solution (or software product containing such solutions) that meets one or more of the stakeholders’ needs, along with a textual description (e.g., in prose) of each such solution or software product. The description may include a description of the functionality of the solution or software product and may include a set of benefits provided by the solution. Other elements of the templated prompt may include common challenges that prevent the stakeholders from achieving the benefits, and thoughts that prompt questions to encourage further conversation.
[0147] Stakeholders may desire written descriptions of each solution, or written descriptions of a set of solutions generated by one or more graphical user interfaces. Furthermore, if stakeholders are employed by or associated with an organization (e.g., a corporation), stakeholders may desire that such written descriptions conform to the organization's style guide, which is specific to that organization.
[0148] For example, a company's CEO may need to give a presentation to the company's board of directors or the engineering community, and may want the presentation content (e.g., text, images) to conform to the company's style guide.
[0149] However, a simple compilation of a set of descriptions from a suite vector may, unfortunately, not fit into an organization's specific style guide. A particular style guide for getting started can specify certain elements of style that are unique to that organization, individually or collectively. For example, an organization's style guide may set standards that define the organization's brand, including, to name a few, grammar usage, logo usage, color, visual representation, fonts, word usage, and tone of writing.
[0150] Some organizations have graphical corporate logos, and their style guides specify rules for the proper use of graphical corporate logos accompanied by text. In such embodiments, multiple examples of style guide-compliant text include examples of the proper use of graphical corporate logos accompanied by text defined by the organization's specific style guide, and style guide-compliant generating artificial intelligence is trained to generate text that conforms to the rules for the proper use of graphical corporate logos.
[0151] In some organizations, corporate trademarks are expressed in text, and style guides specify rules for the proper use of corporate trademarks. In such embodiments, multiple examples of style guide-compliant text include examples of the proper use of corporate trademarks together with other text defined by the company's specific style guide, and style guide-compliant generating artificial intelligence is trained to generate text that conforms to the rules for the proper use of corporate trademarks.
[0152] In some organizations, corporate trademarks are expressed in text, and style guides specify rules for the proper use of corporate trademarks. In such embodiments, multiple examples of style guide-compliant text include examples of the proper use of corporate trademarks only as adjectives in other text defined by the company's specific style guide, and style guide-compliant generating artificial intelligence is trained to generate text that includes the proper use of corporate trademarks as adjectives.
[0153] Accordingly, some embodiments include generative artificial intelligence configured to generate a set of solution descriptions from a suite vector, and more specifically, to generate a set of solution descriptions in which the descriptions conform to the organization's style guide. Such generative artificial intelligence is primarily configured to generate content, the content entities are provided by the suite vector, and the format of the content conforms to the organization's style guide. In other words, such generative artificial intelligence is primarily configured to recast input content (text and / or images) into generated content having a format conforming to the organization's style guide, and in some embodiments, is exclusively configured to recast input content (text and / or images) into generated content having a format conforming to the organization's style guide without generating new substantial content.
[0154] Such generative artificial intelligence can be developed from scratch by developing and training a large language model, or, in a preferred embodiment, by customizing an existing generative artificial intelligence (e.g., ChatGPT) or a large language model on which an existing generative artificial intelligence is built (e.g., a large language model based on ChatGPT).
[0155] The inventors note that existing generative artificial intelligence and / or existing large-scale language models inherently possess styles. For example, existing generative artificial intelligence may inherently develop relevant styles through the input material used in its training, or it may possess relevant styles that have been intentionally imprinted upon it through that training. For example, existing generative artificial intelligence may possess styles defined by Strunk & White's The Elements of Style. However, such styles are not specific styles defined by an organization's style guide.
[0156] Building a large-scale language model, or an entire generative artificial intelligence, from scratch requires an enormous amount of time and effort, but this is within the capabilities of someone proficient in the field of generative artificial intelligence who can provide this explanation. Customizing an existing large-scale language model or an existing generative artificial intelligence may require less time and resources, and this is also within the capabilities of someone proficient in the field of generative artificial intelligence who can provide this explanation.
[0157] In exemplary embodiments, training a new (or custom) large-scale language model or generative artificial intelligence, or customizing an existing large-scale language model or generative artificial intelligence, involves collecting examples of content (text and / or figures) that conform to an organization's style guide, creating a training set from those examples, and then using that training set to train the large-scale language model or generative artificial intelligence.
[0158] To create a new (or custom) large-scale language model or generative artificial intelligence, the training set may contain millions or billions of unique style guide-compliant examples. In contrast, to customize an existing large-scale language model or generative artificial intelligence, the training set may contain only a few hundred or a few thousand unique style guide-compliant examples.
[0159] Large-scale language models and / or generative artificial intelligence can be trained by proving a training set to a training model and having the training model train or customize the large-scale language model and / or generative artificial intelligence. Such training is within the capabilities of those experienced in the field of generative artificial intelligence with this description. Such training models can include, to name a few, generative adversarial networks ("GANs"), variational autoencoders ("VAEs"), and diffusion models.
[0160] As a result, a method performed by a computer to train artificial intelligence to generate content that conforms to an organization's specific style guide includes: (1) collecting a set of content examples in which each such content example conforms to an organization's specific style guide; (2) creating a first training set containing the collected content examples; and (3) training the artificial intelligence using the first training set.
[0161] In some embodiments, training artificial intelligence using a first training set includes providing a first training set to a generative adversarial network having a generator neural network and a discriminator neural network, wherein the first training set serves as a reference for the discriminator neural network, and operating the generative adversarial network to train the generator neural network to transform the generator neural network into generative artificial intelligence configured to transform input content into output content having a format that conforms to an organization-specific style guide.
[0162] In some embodiments, training artificial intelligence using a first training set includes providing the first training set to a variational autoencoder.
[0163] In some embodiments, training artificial intelligence using a first training set includes providing the first training set to a diffusion model.
[0164] Once trained or customized, artificial intelligence can be configured to accept input content in a format that does not conform to the organization's style guide (e.g., text and / or diagrams) and convert that input content into output content that conforms to the organization's style guide, and can be used in that manner.
[0165] The application scenarios include providing such a customized artificial intelligence with input content in a format that does not conform to the organization's style guide (e.g., content generated by one or more graphical user interfaces, e.g., vectors as described above), and causing the customized artificial intelligence to generate output content that conforms to the organization's style guide by converting the input content into output content that conforms to the organization's style guide.
[0166] Figure 21 is a flowchart of one embodiment of a method relating to such generative artificial intelligence.
[0167] Step 2110 includes providing or obtaining (for example, by accessing from remote computer memory) several examples of style guide compliant text that conform to the company's specific style guide. Such compliant text may include text-based company trademarks and / or graphical company logos.
[0168] Step 2120 includes providing or obtaining existing trained generative artificial intelligence. In an exemplary embodiment, the existing trained generative artificial intelligence is not customized to generate text that conforms to a company's specific style guide.
[0169] Step 2130 includes customizing an existing trained generative artificial intelligence by training it with multiple examples of style guide compliant text to generate a style guide compliant generative artificial intelligence configured to generate style guide compliant content from input text that does not conform to a company's specific style guide.
[0170] Some embodiments include step 2140, which involves converting non-compliant content (text and / or graphics) into content that conforms to the company's style guide, and this content may be referred to as style guide compliance content.
[0171] Unfortunately, many people lack the ability to write clearly and concisely. Even those who possess this ability may not be able to create content that conforms to a company's style guide.
[0172] Therefore, the generative artificial intelligence and its use described herein can provide any one or more of several benefits, for example, in situations where written content must strictly adhere to a company's style guide, ensuring consistency in tone, language, and format. This includes using specific font types and font sizes, which may limit the flexibility of content presentation. The text must be easy to understand and avoid jargon and overly complex sentences. In some embodiments, the content must fit precisely within a given text box on an automatically generated PowerPoint presentation slide. This constraint requires the content to be concise and comprehensive, covering all key points without exceeding the available space. In some embodiments, the text must be tailored to the company's specific industry, highlighting relevant benefits and use cases. This requires a deep understanding of industry-specific challenges and opportunities.
[0173] To address some or all of these concerns, generative artificial intelligence standardizes formatting to conform to the company's style guide, automatically adjusting font type, size, and other formatting elements. This ensures that all text aligns with the company's standards and appearance experts, and makes the text clear and easy to understand by rewriting (e.g., "transforming") the content to conform to the company's voice and tone. It eliminates jargon and complex sentences, ensuring that financial benefits are effectively communicated to the end user.
[0174] In some embodiments, generative artificial intelligence translates technical product features into clearly quantifiable benefits.
[0175] In some embodiments, generative artificial intelligence creates text tailored to the end user's specific industry, highlighting relevant benefits and use cases. It leverages industry-specific knowledge to address unique challenges and opportunities, making the talk track more focused and engaging.
[0176] For example, by standardizing formatting, refining style and tone, and ensuring the inclusion of quantitative and industry-specific elements, generative artificial intelligence (e.g., ChatGPT)-driven systems streamline the creation of profit-sharing texts. This results in professional and persuasive writing that aligns with corporate standards and effectively communicates the benefits of software products to end users.
[0177] The benefits and advantages of various embodiments may include one or more of the following:
[0178] Generative artificial intelligence ensures that all text consistently adheres to the company's style guide, maintaining a uniform format, style, and tone across all documents. This is in contrast to manual drafts, which can have inconsistencies due to individual variations in style and may not conform to company standards.
[0179] Generative artificial intelligence ensures that text is clear and easy to understand, avoiding jargon and complex language, and thus improving end-user comprehension and engagement. In contrast, individual human writing styles can vary in clarity, and some texts may contain jargon or complex sentences that are difficult for readers to understand.
[0180] Generative artificial intelligence accurately incorporates quantitative data and evidence into text, ensuring accurate and compelling descriptions of financial benefits. In contrast, manual processes may introduce errors or inconsistencies into data representation, potentially undermining the credibility of the text.
[0181] Generative artificial intelligence can tailor text to specific industries by leveraging industry-specific knowledge to address unique challenges and opportunities, making content more relevant and persuasive. In contrast, manually customizing text for different industries requires extensive research and expertise, which can be resource-intensive and inconsistent.
[0182] Generative artificial intelligence can handle large volumes of text and is scalable for large-scale projects and multiple end users. In contrast, scaling manual drafting work requires more resources and can lead to bottlenecks in the production process.
[0183] Various embodiments may be characterized by potential claims enumerated in subsequent paragraphs (and before the actual claims provided at the end of this application). These potential claims form part of the description of this application. The subject matter of the following potential claims may be presented as actual claims in subsequent proceedings, including this application or any application claiming priority based on this application. The inclusion of such potential claims should not be interpreted as meaning that the actual claims do not cover the subject matter of the potential claims. Accordingly, a decision not to present these potential claims in subsequent proceedings should not be interpreted as an offering of the subject matter to the public.
[0184] Without limitation, potential subject matter that could be claimed (beginning with the letter "P" to avoid confusion with the actual claims presented below) includes:
[0185] P1. A method performed by a computer for managing multiple software solutions, each software solution being useful for building or operating large-scale industrial equipment, and the method is: The steps include providing a computer with a solution database that defines multiple separate software implementation solutions, A step of providing a data model to a computer, wherein the model consists of a hierarchical tree structure from more general nodes, terminating in multiple specific solutions from multiple separate software implementation solutions, and the data model is configured to work with a solution database to facilitate changes to the hierarchical tree structure in response to user input, and A step of generating a graphical user interface on a display device that communicates with a computer, which generates a dynamic sequence of questions in response to both a data model and answers to questions received from a user, wherein the answers include user input, and the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of the large-scale industrial facility, wherein the sequence of questions is (1) An initial question designed to elicit a first answer from the user, (2) A step comprising a series of consecutive subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, each subsequent question following a preceding question, determined by the user's answer to the preceding question, and the series of consecutive subsequent questions ending with a final question. A step in which an automated report builder is instructed to create an output report after receiving answers to a series of questions, wherein the automated report builder is configured to create an output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from multiple distinct software implementation solutions according to the answers. A method performed by a computer, including [this].
[0186] P2. The step of having the Automated Report Builder create an output report is performed by the computer described in P1, including the step of having the Automated Report Builder automatically create the report after a series of questions have been completed.
[0187] P3. The step of causing the automated report builder to create an output report is performed by the computer described in P1, in response to the computer receiving a request to create an output report.
[0188] P4. A method executed by a computer as described in any of P1 to P3, wherein the solution database is indexed according to the tools and solutions within the tool, each tool has tool information stored in the solution database, and each tool has one or more solutions from multiple separate software implementation solutions.
[0189] P5. Each question is uniquely associated with a corresponding node from multiple nodes in a hierarchical tree structure. A graphical user interface includes, for each question, a predetermined set of selectable fields having a predetermined selectable display, the selectable display being derived from a node corresponding to the question. A method performed by a computer as described in P1 to P4.
[0190] P6. A method performed by a computer as described in any of P1-P5, wherein a graphical user interface automatically displays subsequent sequential questions that were not previously displayed after receiving an answer to a previously displayed question.
[0191] P7. A step of providing computer-implemented artificial intelligence, wherein the artificial intelligence is Accepting the vector defined by the answers to a series of questions, The steps further include being configured to generate an output report containing a specified set of solutions from the aforementioned vectors, The method described on page P1 to P6 is performed by a computer.
[0192] P8. A computer-implemented artificial intelligence includes a neural network trained on a training set having multiple training vectors, where each training vector is... A set of training answers to a specified series of questions, A specified set of solutions selected from multiple separate software implementation solutions, A method performed by a computer as described in P7, having the following characteristics.
[0193] P9. Computer-implemented artificial intelligence includes generative artificial intelligence configured to produce an output report in response to prompts, the method of which is performed by the computer as described on P7, and the prompts include a vector defined by the answers to a set of questions.
[0194] P10. A computer implementation system for managing multiple software solutions, each of which is useful for building or operating large-scale industrial equipment, and the system is Computer memory, A solution database that defines multiple separate software implementation solutions, A data model comprising a hierarchical tree structure from more general nodes, ending in multiple specific solutions from multiple separate software implementation solutions, and configured to work in conjunction with a solution database to facilitate changes to the hierarchical tree structure in response to user input, and computer memory for storing the data model. A user interface module configured to generate a graphical user interface on a computer display device that generates a dynamic sequence of questions in response to both a data model and answers to questions received from a user, wherein the answers include user input, and the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of the large-scale industrial facility, (1) An initial question designed to elicit a first answer from the user, (2) Multiple consecutive subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, each subsequent question following a preceding question and determined by the user's answer to that preceding question, and the multiple consecutive subsequent questions ending with the last question, A user interface module, including, An automated report builder configured to generate an output report after receiving answers to a series of questions, wherein the output report is configured to generate an output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from multiple separate software implementation solutions according to the answers. A computer-implemented system including this.
[0195] P11. The computer implementation system described in P10, wherein the solution database is indexed according to the tools and the solutions within those tools, each tool has tool information stored in the solution database, and each tool has one or more solutions from a plurality of separate software implementation solutions.
[0196] P12. Each question is uniquely associated with a corresponding node from multiple nodes in a hierarchical tree structure. A graphical user interface includes, for each question, a predetermined set of selectable fields having a predetermined selectable display, the selectable display being derived from a node corresponding to the question. Computer implementation systems as described on any of pages P10 to P11.
[0197] P13. A computer implementation system as described in any of P10-P12, wherein the graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a preceding question that has been displayed.
[0198] P14. Computer-implemented artificial intelligence, wherein the artificial intelligence is Accepting the vector defined by the answers to a series of questions, From the aforementioned vectors, generate an output report containing the specified set of solutions. A computer-implemented artificial intelligence configured in such a way Furthermore, the computer implementation systems described in any of pages P10 to P13.
[0199] P15. A computer-implemented artificial intelligence includes a neural network trained on a training set having multiple training vectors, where each training vector is... A set of training answers to a specified series of questions, A specified set of solutions selected from multiple separate software implementation solutions, Having, Computer implementation system as described on page 14.
[0200] P16. A non-temporary computer-readable medium having computer-executable code, wherein the computer-executable code, when executed by a computer system, causes the computer system to execute a method, and this method The steps include providing a computer with a solution database that defines multiple separate software implementation solutions, A step of providing a data model to a computer, wherein the model consists of a hierarchical tree structure from more general nodes, terminating in multiple specific solutions from multiple separate software implementation solutions, and the data model is configured to work with a solution database to facilitate changes to the hierarchical tree structure in response to user input, and A step of generating a graphical user interface on a display device that communicates with a computer, which generates a dynamic sequence of questions in response to both a data model and answers to questions received from a user, wherein the answers include user input, and the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of the large-scale industrial facility, wherein the sequence of questions is (1) An initial question designed to elicit a first answer from the user, (2) Multiple consecutive subsequent questions, each subsequent question associated with a node in a hierarchical tree structure, each subsequent question following a preceding question and determined by the user's answer to that preceding question, and the multiple consecutive subsequent questions ending with the last question, Steps including, A step in which an automated report builder is instructed to create an output report after receiving answers to a series of questions, wherein the automated report builder is configured to create an output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from multiple distinct software implementation solutions according to the answers. Non-temporary computer-readable media, including [specific examples of such media].
[0201] P17. A non-temporary computer-readable medium as described on P16, in which the solution database is indexed according to the tools and solutions within the tool, each tool has tool information stored in the solution database, and each tool has one or more solutions from multiple separate software implementation solutions.
[0202] P18. Each question is uniquely associated with a corresponding node from multiple nodes in a hierarchical tree structure. A graphical user interface includes, for each question, a predetermined set of selectable fields having a predetermined selectable display, the selectable display being derived from a node corresponding to the question. A non-temporary computer-readable medium as described on any of pages P16 to P17.
[0203] P19. A non-temporary computer-readable medium as described on any of P16-P18, in which a graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a preceding question that has been displayed.
[0204] P20. The method further includes the step of providing computer-implemented artificial intelligence, wherein the artificial intelligence is Accepting the vector defined by the answers to a series of questions, From the aforementioned vectors, generate an output report containing the specified set of solutions. A non-temporary computer-readable medium configured as described in any of pages 16-19.
[0205] P101: A method performed by a computer to provide artificial intelligence trained to generate output reports, including software solutions, based on information from a user, The step includes generating the artificial intelligence by training a neural network, and the step is: A step of accessing a set of training datasets, where each training dataset includes a set of training vectors and corresponding software solutions, The steps include creating a first training set containing multiple training datasets from a set of datasets, In order to create an output report containing a list of software solutions selected from multiple available software solutions, the first step involves training a neural network using a first training set, and A method performed by a computer.
[0206] P102: The steps for training a neural network are, A step of providing a second training set, wherein the second training set includes a plurality of second training datasets, A step of training a neural network in a second stage using a second training set, wherein the second stage improves the neural network's ability to correctly generate output reports. The computer-based methods described on page 101, further including the following:
[0207] P103: Each training vector contains answers to multiple questions, and the answers are provided by the user. (a) User goals within a project being executed by the user, (b) Challenges faced by the user within the project being executed by the user, A method performed by a computer as described in any of pages P101-P102, describing one or more of the above.
[0208] P201. A method performed by a computer, Steps include providing multiple examples of style guide-compliant text that conform to the company's specific style guide, and A step of providing an existing trained generative artificial intelligence, wherein the existing trained generative artificial intelligence is not customized to generate text that conforms to a company's specific style guide, The process involves customizing an existing trained generative AI by training it with multiple examples of style guide-compliant text to generate a style guide-compliant generative AI configured to generate style guide-compliant content from input text that does not conform to a company's specific style guide, and A method performed by a computer, including [this].
[0209] P202. The step of training an existing trained generative artificial intelligence using multiple examples of style guide-compliant text is: Steps to train an existing pre-trained generative artificial intelligence using a generative adversarial network. The method performed by a computer as described on page 201, including the method described on page 201.
[0210] P203. The step of training an existing trained generative artificial intelligence using multiple examples of style guide-compliant text is: Steps to train an existing pre-trained generative artificial intelligence using a variational autoencoder. The method performed by a computer as described on page 201, including the method described on page 201.
[0211] P204. The step of training an existing trained generative artificial intelligence using multiple examples of style guide-compliant text is: Steps to train an existing pre-trained generative artificial intelligence using a diffusion model The method performed by a computer as described on page 201, including the method described on page 201.
[0212] P205. A company has a graphical corporate logo, and the style guide specifies rules for the proper use of graphical corporate logos accompanied by text. Multiple examples of style guide compliant text include examples of the proper use of graphical corporate logos with text defined by a company's specific style guide. Style guide compliant generating artificial intelligence is trained to create text that complies with the rules for the proper use of graphical corporate logos. The method described on page 201 to page 204 is performed by a computer.
[0213] P206. A company has a corporate trademark expressed in text, and the style guide specifies rules for the proper use of the corporate trademark. Multiple examples of style guide compliant text include examples of the proper use of corporate trademarks with other text defined by a particular company's style guide. The style guide-compliant generative artificial intelligence is trained to produce text that complies with the rules for proper use of corporate trademarks, The computer-implemented method according to any one of P201 to P204.
[0214] P207. The enterprise has a corporate trademark represented in text, and the style guide specifies the use of the trademark only as an adjective, The plurality of examples of style guide-compliant text include examples of properly using the corporate trademark only as an adjective with other text defined by the enterprise's specific style guide, The style guide-compliant generative artificial intelligence is trained to produce text that includes proper use of the corporate trademark as an adjective, The computer-implemented method according to any one of P201 to P204.
[0215] P208. receiving a suite vector containing text that does not comply with the enterprise's specific style guide, and providing the suite vector to the style guide-compliant generative artificial intelligence, and using the style guide-compliant generative artificial intelligence to convert the suite vector into a style guide compliance report, and The computer-implemented method according to any one of P201 to P207, further comprising:
[0216] P209. A computer-implemented system for managing a plurality of software solutions, each software solution being useful for constructing or operating large-scale industrial facilities of an enterprise, the system comprising: a computer memory, a solution database defining a plurality of distinct software-implemented solutions, and a computer memory storing a data model, the data model being constituted by a hierarchical tree structure from more general nodes and terminating at a plurality of specific solutions from a plurality of distinct software implementation solutions, the data model being configured to cooperate with a solution database to facilitate changes to the hierarchical tree structure in response to user input; a user interface module configured to generate, on a computer display device, a graphical user interface that generates a dynamic sequence of questions in response to both the data model and answers to questions received from a user, wherein the answers include user input, the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of large-scale industrial facilities, and the sequence of questions comprises (1) an initial question configured to solicit a first answer from the user; and (2) a plurality of consecutive subsequent questions, each subsequent question being associated with a node in the hierarchical tree structure, each subsequent question following a preceding question and being determined by the user's answer to the preceding question, the plurality of consecutive subsequent questions terminating at a final question; the user interface module comprising the above; an automated report builder configured to generate a suite vector based on the answers to the sequence of questions, wherein the suite vector identifies a suite of software implementation solutions selected from the plurality of distinct software implementation solutions, and comprises text describing each solution in the suite of software implementation solutions, wherein the text does not comply with a corporate style guide; and a style-guide-compliant generative artificial intelligence trained to generate style-guide-compliant content from the text of the suite vector; a computer-implemented system comprising the above.
[0217] P210. The style guide compliant generative AI is trained from an existing trained generative AI, and the existing trained generative AI is not customized to create text that conforms to the company's specific style guide. The steps include receiving multiple examples of style guide-compliant text that conform to the company's specific style guide, The process involves transforming an existing trained generative artificial intelligence (AI) by training it with multiple examples of style guide-compliant text to generate style guide-compliant generative AI configured to generate style guide-compliant content from input text that does not conform to a company's specific style guide, and The computer implementation system described on page 209 is used to train a style guide-compliant generating artificial intelligence.
[0218] P211. The computer implementation system described in P210, wherein the generative artificial intelligence compliant with the style guide is trained using a generative adversarial network.
[0219] P212. The computer implementation system described in P210, wherein the style guide-compliant generative artificial intelligence is trained using a variational autoencoder.
[0220] P213. The computer implementation system described in P210, wherein the generative artificial intelligence conforming to the style guide is trained using a diffusion model.
[0221] P215. A non-temporary computer-readable medium having computer-executable code, wherein the computer-executable code, when executed by a computer system, causes the computer system to execute a method, and the code Code to retrieve multiple examples of style guide-compliant text that conform to a company's specific style guide from computer memory, Code for providing existing trained generative artificial intelligence, where the existing trained generative artificial intelligence is not customized to create text that conforms to a company's specific style guide, and Code for customizing an existing trained generative AI by training it with multiple examples of style guide-compliant text to generate style guide-compliant generative AI configured to generate style guide-compliant content from input text that does not conform to a company's specific style guide, and Non-temporary computer-readable media, including [specific examples of such media].
[0222] P216. Non-temporary computer-readable media as described on P215, including code for customizing existing trained generative artificial intelligence using generative adversarial networks.
[0223] P217. A non-temporal computer-readable medium as described on P215, which includes code for customizing existing trained generative artificial intelligence using a variational autoencoder.
[0224] P218. Non-temporal computer-readable media as described on P215, including code for customizing existing trained generative artificial intelligence using a diffusion model.
[0225] P219. A company has a graphical corporate logo, and its style guide specifies rules for the proper use of graphical corporate logos accompanied by text. Multiple examples of style guide compliant text include examples of the proper use of graphical corporate logos with text defined by a company's specific style guide. Style guide compliant generating artificial intelligence is trained to create text that complies with the rules for the proper use of graphical corporate logos. Non-temporary computer-readable media as described on page 215.
[0226] P220. A company has a corporate trademark expressed in text, and the style guide specifies rules for the proper use of the corporate trademark. Multiple examples of style guide compliant text include examples of the proper use of corporate trademarks with other text defined by a particular company's style guide. Style guide compliant generating artificial intelligence is trained to create text that complies with the rules for the proper use of corporate trademarks. Non-temporary computer-readable media as described on page 215.
[0227] Various embodiments of this disclosure may be implemented at least partially in any conventional computer programming language. For example, some embodiments may be implemented in a procedural programming language (e.g., "C"), or an object-oriented programming language (e.g., "C++"), or in Python, R, Java, LISP, or Prolog. Other embodiments of this disclosure may be implemented as pre-programmed hardware elements (e.g., application-specific integrated circuits, FPGAs, and digital signal processors) or other related components.
[0228] In alternative embodiments, the disclosed apparatus and methods may be implemented as computer program products for use in a computer system. Such implementations may include a set of computer instructions fixed to any tangible medium, such as a non-temporary computer-readable medium (e.g., a diskette, CD-ROM, ROM, flash memory, or fixed disk). The set of computer instructions may embody all or part of the functions described herein with respect to the system.
[0229] Those skilled in the art will appreciate that such computer instructions can be written in a number of programming languages for use with many computer architectures or operating systems. Furthermore, such instructions may be stored in any memory device, such as semiconductor, magnetic, optical or other memory devices, and may be transmitted using any communication technology, such as optical, infrared, microwave, or other transmission technologies.
[0230] Among other things, such computer program products may be distributed as removable media with accompanying printed or electronic documentation (e.g., shrink-wrapped software), preloaded on a computer system (e.g., on system ROM or a fixed disk), or distributed from a server or electronic bulletin board over a network (e.g., the Internet or World Wide Web). Of course, some embodiments of the present disclosure may be implemented as a combination of both software (e.g., a computer program product) and hardware. Still other embodiments of the present disclosure are implemented entirely as hardware, or entirely as software.
[0231] The computer program logic implementing all or part of the functions previously described herein may be executed at different times on a single processor (e.g., concurrently), or at the same or different times on multiple processors, and may be executed under a single operating system process / thread, or under different operating system processes / threads. Accordingly, the term "computer process" generally refers to the execution of a set of computer program instructions, regardless of whether different computer processes are executed on the same or different processors, and regardless of whether different computer processes are executed under the same operating system process / thread or different operating system processes / threads.
[0232] The embodiments of the present invention described above are intended to be merely illustrative, and numerous variations and modifications will be apparent to those skilled in the art. Such variations and modifications are intended to fall within the scope of the present invention as defined by any of the appended claims.
Claims
1. A method performed by a computer that manages multiple software solutions, each of which is useful for building or operating large-scale industrial equipment, and the method is The steps include providing a computer with a solution database that defines multiple separate software implementation solutions, A step of providing a data model to the computer, wherein the model consists of a hierarchical tree structure from more general nodes, ends in a plurality of specific solutions from a plurality of separate software implementation solutions, and the data model is configured to work with the solution database to facilitate changes to the hierarchical tree structure in response to user input, A step of generating a graphical user interface on a display device that communicates with the computer, which generates a dynamic sequence of questions in response to both the data model and the answers to questions received from the user, wherein the answers include user input, and the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of the large-scale industrial facility, wherein the sequence of questions is (1) An initial question designed to elicit a first answer from the user, (2) A set of consecutive subsequent questions, each subsequent question associated with a node in the hierarchical tree structure, each subsequent question following a preceding question, determined by the user's answer to the preceding question, and the set of consecutive subsequent questions ending with the last question, Steps including, A step of causing an automated report builder to create an output report after receiving answers to the set of questions, wherein the automated report builder is configured to create the output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from a plurality of separate software implementation solutions according to the answers, A method performed by a computer, including [this].
2. The method performed by a computer according to claim 1, wherein the step of causing an automated report builder to create an output report includes the step of causing the automated report builder to automatically create the report after completing the series of questions.
3. The method performed by a computer according to claim 1, wherein the step of causing an automated report builder to create an output report is performed in response to the computer receiving a request to create the output report.
4. The method executed by a computer according to claim 1, wherein the solution database is indexed according to the functions of the tools and the solutions within the tools, each tool has tool information stored in the solution database, and each tool has one or more solutions from the plurality of separate software implementation solutions.
5. Each question is uniquely associated with a corresponding node from the plurality of nodes in the hierarchical tree structure, The graphical user interface includes, for each question, a predetermined set of selectable fields having a predetermined selectable display, wherein the selectable display is derived from a node corresponding to the question. The method performed by a computer as described in claim 1.
6. The method performed by a computer according to claim 1, wherein the graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a previously displayed question.
7. A step of providing computer-implemented artificial intelligence, wherein the artificial intelligence is Accepting the vector defined by the answers to the aforementioned series of questions, From the aforementioned vectors, generate the output report containing the specified set of solutions. A computer-operated method according to claim 1, further comprising the steps configured as follows:
8. The computer-implemented artificial intelligence includes a neural network trained on a training set having multiple training vectors, where each training vector is A set of training answers to a specified series of questions, A set of specified solutions selected from the aforementioned multiple separate software implementation solutions, A computer-based method according to claim 7, comprising:
9. The method performed by a computer according to claim 7, wherein the computer-implemented artificial intelligence includes generative artificial intelligence configured to produce the output report in response to prompts, the prompts including the vector defined by the answers to the set of questions.
10. A computer implementation system for managing multiple software solutions, each software solution being useful for building or operating large-scale industrial equipment, wherein the system Computer memory, A solution database that defines multiple separate software implementation solutions, A data model comprising a hierarchical tree structure from more general nodes, terminating in multiple specific solutions from multiple separate software implementation solutions, wherein the data model is configured to cooperate with the solution database to facilitate changes to the hierarchical tree structure in response to user input, and computer memory for storing the data model. A user interface module configured to generate a graphical user interface on a computer display device that generates a dynamic sequence of questions in response to both the data model and the answers to questions received from the user, wherein the answers include user input, and the sequence of questions is configured to request from the user information relating to the construction or operation of the large-scale industrial facility, and the sequence of questions is (1) An initial question designed to elicit a first answer from the user, (2) A set of consecutive subsequent questions, each subsequent question associated with a node in the hierarchical tree structure, each subsequent question following a preceding question, determined by the user's answer to the preceding question, and the set of consecutive subsequent questions ending with the last question, A user interface module, including, An automated report builder configured to generate an output report after receiving answers to the aforementioned series of questions, wherein the output report is configured to generate the output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from a plurality of separate software implementation solutions according to the answers; A computer-implemented system including this.
11. The computer implementation system according to claim 10, wherein the solution database is indexed according to the tools and solutions within the tool, each tool has tool information stored in the solution database, and each tool has one or more solutions from the plurality of separate software implementation solutions.
12. Each question is uniquely associated with a corresponding node from the plurality of nodes in the hierarchical tree structure, The graphical user interface includes, for each question, a predetermined set of selectable fields having a predetermined selectable display, wherein the selectable display is derived from a node corresponding to the question. The computer implementation system according to claim 10.
13. The computer implementation system according to claim 10, wherein the graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a previously displayed question.
14. Computer-implemented artificial intelligence, wherein the artificial intelligence is Accepting the vector defined by the answers to the aforementioned series of questions, Further including a computer-implemented artificial intelligence configured to produce the output report containing the specified set of solutions from the aforementioned vectors, The computer implementation system according to claim 10.
15. The computer-implemented artificial intelligence includes a neural network trained on a training set having multiple training vectors, where each training vector is A set of training answers to a specified series of questions, A set of specified solutions selected from the aforementioned multiple separate software implementation solutions, A computer implementation system according to claim 14, having the following features.
16. A non-temporary computer-readable medium having computer-executable code, wherein the computer-executable code, when executed by a computer system, causes the computer system to execute a method, and the method is The steps include providing the computer with a solution database that defines multiple separate software implementation solutions, A step of providing a data model to the computer, wherein the model consists of a hierarchical tree structure from more general nodes, ends in a plurality of specific solutions from a plurality of separate software implementation solutions, and the data model is configured to work with the solution database to facilitate changes to the hierarchical tree structure in response to user input, A step of generating a graphical user interface on a display device that communicates with the computer, which generates a dynamic sequence of questions in response to both the data model and the answers to questions received from the user, wherein the answers include user input, and the sequence of questions is configured to request information from the user regarding issues related to the construction or operation of the large-scale industrial facility, wherein the sequence of questions is (1) An initial question designed to elicit a first answer from the user, (2) A set of consecutive subsequent questions, each subsequent question associated with a node in the hierarchical tree structure, each subsequent question following a preceding question, determined by the user's answer to the preceding question, and the set of consecutive subsequent questions ending with the last question, Steps including, A step of causing an automated report builder to create an output report after receiving answers to the set of questions, wherein the automated report builder is configured to create the output report according to the answers, and the output report includes a specified set of solutions, each of which is selected from a plurality of separate software implementation solutions according to the answers, Non-temporary computer-readable media, including [specific examples of such media].
17. The non-temporary computer-readable medium according to claim 16, wherein the solution database is indexed according to tools and solutions within the tools, each tool has tool information stored in the solution database, and each tool has one or more solutions from the plurality of separate software implementation solutions.
18. Each question is uniquely associated with a corresponding node from the plurality of nodes in the hierarchical tree structure, The graphical user interface includes, for each question, a predetermined set of selectable fields having a predetermined selectable display, wherein the selectable display is derived from a node corresponding to the question. The non-temporary computer-readable medium according to claim 16.
19. The non-temporary computer-readable medium according to claim 16, wherein the graphical user interface automatically displays a series of subsequent questions that were not previously displayed after receiving an answer to a previously displayed question.
20. The method further includes the step of providing a computer-implemented artificial intelligence, wherein the artificial intelligence is Accepting the vector defined by the answers to the aforementioned series of questions, From the aforementioned vectors, generate the output report containing the specified set of solutions. A non-temporary computer-readable medium according to claim 16, configured as described above.
21. A method performed by a computer, Steps include providing multiple examples of style guide-compliant text that conform to the company's specific style guide, and A step of providing an existing trained generative artificial intelligence, wherein the existing trained generative artificial intelligence is not customized to generate text that conforms to the company's specific style guide, The steps include customizing the existing trained generating artificial intelligence by training it with multiple examples of the style guide compliant text in order to generate style guide compliant generating artificial intelligence configured to generate style guide compliant content from input text that does not conform to the specific style guide of the said company, A method performed by a computer, including [this].
22. The step of training the existing trained generative artificial intelligence using multiple examples of the style guide-compliant text is: Steps to train the existing pre-trained generative artificial intelligence using a generative adversarial network. A computer-based method according to claim 21, including the following:
23. The step of training the existing trained generative artificial intelligence using multiple examples of the style guide-compliant text is: Steps to train the existing pre-trained generative artificial intelligence using a variational autoencoder. A computer-based method according to claim 21, including the following:
24. The step of training the existing trained generative artificial intelligence using multiple examples of the style guide-compliant text is: Steps to train the existing pre-trained generative artificial intelligence using a diffusion model. A computer-based method according to claim 21, including the following:
25. The said company has a graphical corporate logo, and the said style guide specifies rules for the proper use of the graphical corporate logo with text, The aforementioned examples of style guide compliant text include examples of appropriate use of the graphical corporate logo with text defined by the company's specific style guide, The artificial intelligence that generates style guide-compliant text is trained to create text that conforms to the rules for the proper use of the graphical corporate logo. The method performed by a computer according to claim 21.
26. The said company has a corporate trademark expressed in text, and the said style guide specifies rules for the proper use of the said corporate trademark, The aforementioned examples of style guide-compliant text include examples of appropriate use of the company trademark with other text defined by the company’s specific style guide, The aforementioned style guide-compliant generating artificial intelligence is trained to create text that complies with the aforementioned rules for the proper use of the company trademark. The method performed by a computer according to claim 21.
27. The aforementioned company has a corporate trademark expressed in text, and the style guide specifies the use of the trademark only as an adjective. The aforementioned examples of style guide-compliant text include examples in which the company trademark is used appropriately only as an adjective accompanied by other text defined by the company’s specific style guide, The aforementioned style guide-compliant generating artificial intelligence is trained to create text that includes the appropriate use of the aforementioned company trademark as an adjective. The method performed by a computer according to claim 21.
28. The steps include receiving a suite vector containing text that does not conform to the specific style guide of the aforementioned company, The steps include providing the suite vector to the style guide-compliant generation artificial intelligence, The steps include using the style guide compliance generating artificial intelligence to convert the aforementioned suite vector into a style guide compliance report, A computer-based method according to claim 21, further comprising:
29. A computer implementation system for managing multiple software solutions, each software solution being useful for building or operating large-scale industrial facilities of an enterprise, wherein the system Computer memory, A solution database that defines multiple separate software implementation solutions, A data model comprising a hierarchical tree structure from more general nodes, terminating in multiple specific solutions from multiple separate software implementation solutions, wherein the data model is configured to cooperate with the solution database to facilitate changes to the hierarchical tree structure in response to user input, and computer memory for storing the data model. A user interface module configured to generate a graphical user interface on a computer display device that generates a dynamic sequence of questions in response to both the data model and the answers to questions received from the user, wherein the answers include user input, and the sequence of questions is configured to request from the user information relating to the construction or operation of the large-scale industrial facility, and the sequence of questions is (1) An initial question designed to elicit a first answer from the user, (2) A set of consecutive subsequent questions, each subsequent question associated with a node in the hierarchical tree structure, each subsequent question following a preceding question, determined by the user's answer to the preceding question, and the set of consecutive subsequent questions ending with the last question, A user interface module, including, An automated report builder configured to generate a suite vector based on answers to the aforementioned series of questions, wherein the suite vector identifies a suite of software implementation solutions selected from the plurality of separate software implementation solutions, includes text describing each solution within the suite of software implementation solutions, and the text does not conform to the company's style guide; A style guide compliant generating artificial intelligence trained to generate style guide compliant content from the text of the aforementioned suite vector, A computer-implemented system including this.
30. The aforementioned style guide-compliant generative artificial intelligence is trained from an existing trained generative artificial intelligence, and the existing trained generative artificial intelligence is not customized to produce text that conforms to the company's specific style guide. The steps include receiving multiple examples of style guide-compliant text that conform to the company's specific style guide, The computer implementation system according to claim 29, wherein the style guide compliant generating artificial intelligence is trained by the step of transforming the existing trained generating artificial intelligence by training the existing trained generating artificial intelligence using the plurality of examples of the style guide compliant text, in order to generate style guide compliant generating artificial intelligence configured to generate style guide compliant content from input text that does not conform to a specific style guide of the said company.
31. The computer implementation system according to claim 30, wherein the style guide-compliant generative artificial intelligence is trained using a generative adversarial network.
32. The computer implementation system according to claim 30, wherein the style guide-compliant generating artificial intelligence is trained using a variational autoencoder.
33. The computer implementation system according to claim 30, wherein the style guide-compliant generative artificial intelligence is trained using a diffusion model.
34. The said company has a graphical corporate logo, and the said style guide specifies rules for the proper use of the graphical corporate logo with text, The aforementioned examples of style guide compliant text include examples of appropriate use of the graphical corporate logo with text defined by the company's specific style guide, The computer implementation system according to claim 30.
35. A non-temporary computer-readable medium having computer-executable code, wherein the computer-executable code, when executed by a computer system, causes the computer system to execute a method, and the code Code to retrieve multiple examples of style guide-compliant text that conform to a company's specific style guide from computer memory, Code for providing existing trained generative artificial intelligence, wherein the existing trained generative artificial intelligence is not customized to produce text that conforms to the company's specific style guide, Code for customizing an existing trained generating artificial intelligence by training it with multiple examples of the style guide compliant text in order to generate a style guide compliant generating artificial intelligence configured to generate style guide compliant content from input text that does not conform to the specific style guide of the said company, Non-temporary computer-readable media, including [specific examples of such media].
36. The non-temporary computer-readable medium according to claim 35, wherein the code for customizing the existing trained generative artificial intelligence includes code for customizing the existing trained generative artificial intelligence using a generative adversarial network.
37. The non-temporal computer-readable medium according to claim 35, wherein the code for customizing the existing trained generative artificial intelligence includes code for customizing the existing trained generative artificial intelligence using a variational autoencoder.
38. The non-temporary computer-readable medium according to claim 35, wherein the code for customizing the existing trained generative artificial intelligence includes code for customizing the existing trained generative artificial intelligence using a diffusion model.
39. The said company has a graphical corporate logo, and the said style guide specifies rules for the proper use of the graphical corporate logo with text, The aforementioned examples of style guide compliant text include examples of appropriate use of the graphical corporate logo with text defined by the company's specific style guide, The artificial intelligence that generates style guide-compliant text is trained to create text that conforms to the rules for the proper use of the graphical corporate logo. The non-temporary computer-readable medium according to claim 35.
40. The said company has a corporate trademark expressed in text, and the said style guide specifies rules for the proper use of the said corporate trademark. The aforementioned examples of style guide-compliant text include examples of appropriate use of the company trademark with other text defined by the company’s specific style guide, The aforementioned style guide-compliant generating artificial intelligence is trained to create text that complies with the aforementioned rules for the proper use of the company trademark. The non-temporary computer-readable medium according to claim 35.
