Information generation apparatus, information generation system, and information generation method

Abstract

To provide an information generation apparatus, an information generation system, and an information generation method for simply generating structured information on KPI in a plant.SOLUTION: An information generation apparatus includes a control unit (104) configured to: acquire past case data related to operation of a plant and accumulates the data in a KPI information DB (103a); register tag data, which is information serving as elements of KPI, based on the past case data, on the KPI information DB (103a); register a KPI list (103ab) including calculation formulae corresponding to the tag data and relational expressions between the tag data, on the KPI information DB (103a); request, when receiving an input of desired information for the KPI of the plant from a user, generative AI using a large language model to generate generation information according to the input using the KPI information DB (103a); and present, to the user, the generation information generated by the generative AI in accordance with the request.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to an information generating device, an information generating system, and an information generating method. [Background technology]

[0002] Conventionally, Key Performance Indicators (KPIs) are known, which are a group of metrics that help define the degree of goal achievement of an organization, etc. Regarding KPIs, Patent Document 1 discloses a technology that can create a new KPI when a user provides information necessary for creating the KPI. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2006-260557 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the above-mentioned conventional techniques have room for further improvement in terms of more easily structuring and generating information related to KPIs in a plant.

[0005] Setting appropriate KPIs is essential for the operation of various plants, such as factories, power plants, substations, and chemical plants. However, there are a wide variety of plant types, and each plant typically controls a wide variety of processes using a wide variety of equipment and devices. As a result, there are a large number of categories and processes for which KPIs must be set in a plant. In addition, there are often many cases where KPIs are related to one another.

[0006] Therefore, it takes a lot of time and effort to understand and structure the relationships between KPIs in a plant and set appropriate KPIs.

[0007] An object of the present invention is to provide an information generating device, an information generating system, and an information generating method that can more easily structure and generate information related to KPIs in a plant. [Means for solving the problem]

[0008] An information generation device according to one aspect includes a control unit that acquires past case data related to plant operations and stores it in a database, registers tag data, which is information that constitutes a KPI based on the past case data, in the database, registers definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data, and, when receiving input of desired information related to the KPI of the plant from a user, requests a generation AI using a large-scale language model to generate generated information corresponding to the input using the database, and presents the generated information generated by the generation AI in response to the request to the user.

[0009] According to one aspect, an information generation system includes a system for managing plant operations and an information generation device. The system transmits past case data related to the plant operations to the information generation device. The information generation device acquires the past case data and stores it in a database. The information generation device registers tag data, which is information that constitutes a KPI, in the database based on the past case data. The information generation device registers definition information in the database, including a calculation formula corresponding to the tag data and a relational formula between the tag data. Upon receiving desired information related to the plant KPI from a user, the information generation device requests a generation AI using a large-scale language model to generate generated information corresponding to the input using the database. The information generated by the generation AI in response to the request is presented to the user.

[0010] In one aspect of the information generation method, a computer acquires past case data related to plant operations and stores it in a database, registers tag data, which is information that becomes an element of a KPI based on the past case data, in the database, registers definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data, and, when receiving input of desired information related to the KPI of the plant from a user, requests a generation AI using a large-scale language model to generate generated information corresponding to the input using the database, and presents the generated information generated by the generation AI in response to the request to the user. [Effects of the Invention]

[0011] According to one embodiment, it is possible to provide an information generating device, an information generating system, and an information generating method that can more easily structure and generate information related to KPIs in a plant. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating an outline of an information generating method according to an embodiment. [Figure 2] FIG. 1 is a diagram illustrating an example of classification of KPIs in a plant. [Figure 3] FIG. 10 is a diagram illustrating an example of a past case list. [Figure 4] FIG. 10 is a diagram illustrating an example of a KPI list. [Figure 5] FIG. 10 is a diagram illustrating an example of generation information. [Figure 6] 1 is a block diagram showing an example of the configuration of a KPI information generation device according to an embodiment. [Figure 7] FIG. 10 is a diagram illustrating an example of KPI calculation formula data. [Figure 8] FIG. 1 shows an example of creating KPIs for maximizing production volume in an ethylene plant (part 1). [Figure 9] FIG. 2 shows an example of creating KPIs for maximizing production volume in an ethylene plant. [Figure 10] FIG. 10 is a diagram showing an example of creating a KPI regarding material loss relative to the feed amount in an ethylene plant. [Figure 11] 10A and 10B are diagrams illustrating an example of a template of an output request prompt and an input example. [Figure 12] FIG. 2 is a diagram showing a processing sequence executed by the KPI information generation system according to the embodiment. [Figure 13] FIG. 2 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the KPI information generation device according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the information generating device, information generating system, and information generating method disclosed in the present application will be described in detail with reference to the accompanying drawings. Note that the present invention is not limited to these embodiments. Furthermore, the same elements are given the same reference numerals, redundant descriptions are omitted as appropriate, and the embodiments can be combined as appropriate within a consistent range.

[0014] In the following, when it is necessary to distinguish between multiple identical elements, the symbol indicating the element may be followed by a number in the form "-n" (n is a natural number). When no particular distinction is necessary, this numbering will not be used.

[0015] In the following, the information generation system according to the embodiment is assumed to be a KPI information generation system 1 (see FIG. 1) that generates information related to KPIs in a plant P1 (see FIG. 1). The information generation device according to the embodiment is assumed to be a KPI information generation device 100 (see FIG. 1). The KPI information generation device 100 is a device that generates and outputs information desired by a user U related to KPIs of the plant P1 in the KPI information generation system 1.

[0016] In the KPI information generation system 1, the KPI information generation device 100 acquires past case data related to the operation of plant P1 and stores it in the KPI information DB 103a. The KPI information generation device 100 also registers tag data, which is information that serves as elements of KPIs, in the KPI information DB 103a based on the past case data, and registers a KPI list 103ab, which includes calculation formulas corresponding to the tag data and relational formulas between the tag data, in the KPI information DB 103a. When the KPI information generation device 100 receives input of desired information related to KPIs of plant P1 from user U, it requests a generation AI using a large-scale language model to generate generated information in accordance with the input using the KPI information DB 103a. The KPI information generation device 100 also presents the generated information generated by the generation AI in response to the request to user U.

[0017] [Outline of information generation method according to this embodiment] An overview of the information generation method according to this embodiment will be described with reference to Figs. 1 to 5. Fig. 1 is an explanatory diagram of the overview of the information generation method according to this embodiment. Fig. 2 is a diagram showing an example of KPI classification in plant P1. Fig. 3 is a diagram showing an example of a past case list 103aa. Fig. 4 is a diagram showing an example of a KPI list 103ab. Fig. 5 is a diagram showing an example of generated information.

[0018] As shown in Fig. 1, a KPI information generation system 1 according to an embodiment is a system that uses past case data from a Plant Information Management System (PIMS) 10 that manages the operations of various plants P1. The PIMS 10 is a system that extracts and analyzes information (e.g., process data) related to the operation of each plant P1, i.e., production and operation, in real time, and is set up to enable visualization and sharing of the information.

[0019] The KPI information generation system 1 includes a PIMS 10, a KPI information generation device 100, and a generation AI server 200. The KPI information generation device 100 has a KPI information DB (Database) 103a. The KPI information DB 103a includes a past case list 103aa and a KPI list 103ab.

[0020] In the information generation method according to this embodiment, the KPI information generation device 100 acquires past case data from the PIMS 10 and registers it in a past case list 103aa (step S1). Based on this past case list 103aa, the KPI information generation device 100 structures and visualizes KPIs to generate desired information related to KPIs in the plant P1.

[0021] In plant P1, there are a large number of categories and processes for which KPIs are set. In addition, there are also many cases where KPIs are related to one another. For example, as shown in Figure 2, one of the targets for setting KPIs in plant P1 is category 1, "Business Objective." "Business Objective" is categorized into, for example, "Production," "Profit," "Energy," "RAM (Reliability, Availability, Maintainability)," and "Safety."

[0022] On the other hand, there is category 2, "Target Process," for which KPIs are set in plant P1. As shown in Figure 2, there are an extremely wide variety of "Target Processes." In plant P1, just as shown in Figure 2, there can be as many different KPI groups as there are by multiplying these "Business Objectives" and "Target Processes." In addition, there are often relationships between KPIs.

[0023] Therefore, in the information generation method according to the embodiment, information that will be the elements of the KPI is extracted based on past case data and input as tag data. In addition, KPI calculation formula data including a formula for calculating the KPI is associated with the tag data.

[0024] 1, in the information generation method according to the embodiment, the KPI information generation device 100 receives input of tag data for past case data from, for example, user U-1 (step S2). User U-1 extracts various pieces of information that were used as KPIs or that became elements of KPIs in any past case data in the past case list 103aa, and inputs the information as tag data.

[0025] The past case list 103aa is data that lists past case data for various types of plant P1, as shown in Fig. 3. When any past case data is selected from the list, the past case list 103aa can expand detailed information about that past case data. As detailed information, various types of information that were used as KPIs or that became elements of KPIs in that past case data can be selected, and user U-1 inputs this information as tag data by selecting it.

[0026] Returning to the explanation of Figure 1, the user U-1 also inputs KPI calculation formula data (step S3). The KPI calculation formula data is data consisting of a calculation formula for calculating each KPI. The KPI calculation formula data can be input as, for example, various general calculation formulas in chemical engineering, etc., corresponding to each tag data.

[0027] Each tag data associated with a calculation formula can be treated as a KPI or an element that constitutes a KPI. KPI calculation formula data can also be input as a relational formula between tag data associated with this calculation formula (i.e., between KPIs).

[0028] The KPI information generation device 100 registers the input in step S3 in a KPI list 103ab. The KPI list 103ab is a list of KPIs that define the calculation formulas for tag data that can be treated as KPIs and the relationships between KPIs. An example of the KPI list 103ab is shown in FIG. 4. Note that in FIG. 4 and FIGS. 7 to 10 shown later, oval rectangles (so-called rounded rectangles) represent each piece of tag data or KPI.

[0029] 4, for example, it can be seen that the tag data "Naphtha Feed (Plan %)" is linked to the tag data "Naphtha Feed Rate (Plan)" and "Naphtha Feed Rate (Actual)." In other words, when calculating the tag data "Naphtha Feed (Plan %)" as a KPI, it can be seen that it is sufficient to combine the calculation formulas associated with the tag data "Naphtha Feed Rate (Plan)" and "Naphtha Feed Rate (Actual)."

[0030] Returning to the explanation of Figure 1, the KPI information generation device 100 has a KPI information DB 103a that includes the past case list 103aa and the KPI list 103ab. In the information generation method according to the embodiment, for example, a user U-2 inputs desired information regarding the KPIs of the plant P1 to the KPI information generation device 100 (step S4). Note that the user U-2 may be the same user as the user U-1.

[0031] User U-2 inputs desired information regarding the KPIs of plant P1, for example, using a preset template. Based on this input, the KPI information generation device 100 generates an output request prompt that requests the generation AI (generation AI server 200) to generate the information desired by user U-2 while using the KPI information DB 103a.

[0032] The generation AI server 200 is a server device that functions as a generation AI. The generation AI server 200 is realized, for example, as a private cloud. The generation AI server 200 has a generation AI model (not shown). The generation AI server 200 loads and operates this generation AI model as part of a program, thereby functioning as a generation AI that generates information corresponding to an output request prompt from the KPI information generation device 100 and transmits the information to the KPI information generation device 100.

[0033] A generative AI model is, for example, a unimodal large-scale language model that can accept only natural language text as a modality. Examples of generative AI models include transfer-based models and recurrent neural network (RNN)-based models.

[0034] Examples of transfer-based models include, but are not limited to, Generative Pre-trained Transformer (GPT) and Bidirectional Auto Regressive Dialogues (BARD). Examples of RNN-based models include, but are not limited to, Receptance Weighted Key Value (RWKV).

[0035] The generation AI model of the generation AI server 200 may be individualized (which may also be read as "fine tuning") according to the generation of information related to KPIs in the plant P1.

[0036] The KPI information generation device 100 generates an output request prompt in accordance with the interface to this generation AI server 200. At this time, the KPI information generation device 100 generates an output request prompt for the generation AI server 200 so as to generate information according to the request using each piece of data registered in the KPI information DB 103a.

[0037] Then, the KPI information generation device 100 transmits an output request prompt to the generation AI server 200 (step S5). In response, the generation AI server 200 generates information according to the request of the output request prompt and transmits it as generated information to the KPI information generation device 100 (step S6).

[0038] Then, the KPI information generation device 100 receives the generation information from the generation AI server 200 and generates output information to be visualized on an arbitrary digital dashboard (hereinafter referred to as "dashboard DD1" as appropriate) based on the generation information (step S7).

[0039] An example of the generation information generated by the generation AI server 200 is shown in Figure 5. The KPI information generation device 100 causes the generation AI server 200 to generate a script that takes the generation information shown in Figure 5 as input and can output it as output information to any application software (here, EXCEL (registered trademark)). Hereinafter, the application software will be referred to as an "app" where appropriate.

[0040] This allows the user U-2 to check information about the KPIs of the plant P1 that he or she desires as visualized output information on the dashboard DD1 corresponding to any application, and the user U-2 can make decisions about the KPIs of the plant P1 based on the checked output information.

[0041] As described above, in the information generation method according to the embodiment, the KPI information generation device 100 acquires past case data related to the operation of the plant P1 and stores it in the KPI information DB 103a. The KPI information generation device 100 also registers tag data, which is information that constitutes KPI elements, in the KPI information DB 103a based on the past case data, and registers a KPI list 103ab, which includes calculation formulas corresponding to the tag data and relational expressions between the tag data, in the KPI information DB 103a. When the KPI information generation device 100 receives desired information input from a user U regarding KPIs for the plant P1, it requests a generation AI using a large-scale language model to generate generated information corresponding to the input using the KPI information DB 103a. The KPI information generation device 100 then presents the generated information generated by the generation AI in response to the request to the user U. This allows for easier structuring and generation of information related to KPIs in the plant P1.

[0042] A configuration example of the KPI information generation system 1 to which the information generation method according to the embodiment is applied will be described in more detail below.

[0043] [Configuration example of KPI information generation device 100] Fig. 6 is a block diagram showing an example of the configuration of a KPI information generation device 100 according to an embodiment. Note that Fig. 6 shows, in functional blocks, only the components necessary for explaining this embodiment, and omits the description of general components.

[0044] In addition, in the description using FIG. 6, the description of components that have already been described will be appropriately simplified or omitted.

[0045] As shown in FIG. 6, the KPI information generation device 100 includes a communication unit 101, an HMI (Human Machine Interface) unit 102, a storage unit 103, and a control unit 104.

[0046] The communication unit 101 is realized by a network adapter, etc. The communication unit 101 communicably connects the KPI information generation device 100 with the PIMS 10 and the generation AI server 200 via wireless communication and / or wired communication.

[0047] The HMI unit 102 is a component that mediates the exchange of information between the KPI information generation device 100 and a user U who uses the KPI information generation device 100. The HMI unit 102 includes an input interface that accepts information input from the user U. The input interface is realized by, for example, a keyboard, a mouse, or the like.

[0048] The HMI unit 102 also includes an output interface that outputs information to the user U. The output interface is realized by, for example, a display, a microphone, a speaker, etc. The input interface and the output interface may be integrated into a touch panel display, etc. The input interface or the output interface may also include software components such as a GUI or a display screen that are presented to the user U.

[0049] The storage unit 103 is realized by a storage device such as a RAM (Random Access Memory), a flash memory, or an HDD (Hard Disk Drive). The storage unit 103 stores a program according to the embodiment executed by the control unit 104. The storage unit 103 also stores various types of information used in the information processing executed by the control unit 104.

[0050] 6, the storage unit 103 stores a KPI information DB 103a, template information 103b, and generated AI interface information 103c. The KPI information DB 103a includes a past case list 103aa and a KPI list 103ab.

[0051] The KPI information DB 103a has already been explained, so its explanation will be omitted here. The template information 103b is information including a template of an output request prompt to be sent to the generation AI server 200. A specific example of the template will be described later using FIG. 11.

[0052] The generation AI interface information 103c is information about the interface through which the KPI information generation device 100 exchanges output request prompts and the generation information generated in response to these prompts with the generation AI server 200. The generation AI interface information 103c includes, for example, an API (Application Programming Interface) for the paired generation AI.

[0053] The control unit 104 corresponds to a so-called processor and is realized by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a GPU (Graphical Processing Unit), or the like.

[0054] The control unit 104 reads the program according to the embodiment stored in the storage unit 103 and executes it using the RAM as a work area. The control unit 104 can also be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0055] The control unit 104 has an acquisition unit 104a, a registration unit 104b, a reception unit 104c, a pair generation AI processing unit 104d, an output information generation unit 104e, and an application execution unit 104f, and realizes or executes the information processing functions and actions described below.

[0056] The internal configuration of the control unit 104 is not limited to the configuration shown in Fig. 6, and may be any other configuration as long as it is capable of executing the information processing described below. Furthermore, the connection relationship between the processing units included in the control unit 104 is not limited to the connection relationship shown in Fig. 6, and may be any other connection relationship.

[0057] The acquisition unit 104a acquires past case data from the PIMS 10 via the communication unit 101. The acquisition unit 104a also acquires input data input by the user U via the HMI unit 102.

[0058] The registration unit 104b registers the past case data acquired by the acquisition unit 104a in the past case list 103aa of the KPI information DB 103a. The registration unit 104b also registers tag data in the past case list 103aa based on input data from the user U acquired by the acquisition unit 104a. The registration unit 104b also registers KPI calculation formula data in the KPI list 103ab based on input data from the user U acquired by the acquisition unit 104a.

[0059] Here, a specific example of KPI calculation formula data will be described. Fig. 7 is a diagram showing an example of KPI calculation formula data. KPI calculation formula data is data consisting of calculation formulas for calculating each KPI. KPI calculation formula data can be input as various general calculation formulas such as chemical engineering corresponding to each tag data, for example.

[0060] 7 shows various calculation formulas used to calculate the duties around the distillation tower in plant P1. For example, when calculating the duties around the distillation tower, general calculation formulas for calculating each of the tag data "Condenser Duty (Gcal / h)," "Steam Reboiler Duty (Gcal / h)," and "Hot Oil Reboiler Duty (Gcal / h)" are input and associated with each other. Registration unit 104b appropriately presents a GUI for inputting these calculation formulas to user U via HMI unit 102, and user U can easily input the calculation formulas using this GUI.

[0061] Each tag data associated with a calculation formula can be treated as a KPI or an element that constitutes a KPI. KPI calculation formula data can be input as a relational expression between tag data associated with this calculation formula (i.e., between KPIs).

[0062] Fig. 8 is a diagram (part 1) showing an example of creating KPIs for maximizing production volume in an ethylene plant, and Fig. 9 is a diagram (part 2) showing an example of creating KPIs for maximizing production volume in an ethylene plant.

[0063] To maximize production volume in an ethylene plant, the user U first inputs a calculation formula corresponding to each tag data item related to the ethylene plant, as shown in Fig. 8. In this case, the user U inputs a calculation formula for calculating each tag data item, for example, "Naphtha Feed (Plan %)," "Fresh Ethane Feed (Plan %)," "Ethylene Production Rate (Plan %)," and "Propylene Production Rate (Plan %)."

[0064] At this time, the user U can input the relational expressions between the tag data as shown in Fig. 8. The registration unit 104b appropriately presents a GUI for inputting these relational expressions, for example, a list of tag data associated with calculation expressions, to the user U via the HMI unit 102, and the user U can easily input the relational expressions while selecting tag data using this GUI.

[0065] Then, based on the input relational equation, the registration unit 104b links the tag data "Naphtha Feed (Plan %)", "Fresh Ethane Feed (Plan %)", "Ethylene Production Rate (Plan %)", and "Propylene Production Rate (Plan %)" to the tag data that constitute each element, as shown in Figure 9.

[0066] That is, the registration unit 104b defines the relationships between tag data (or KPIs). If the user U creates a KPI "Total Production" for maximizing production volume in an ethylene plant as a function with "Naphtha Feed," "Fresh Ethane Feed," "Ethylene Production Rate," and "Propylene Production Rate" as inputs, the KPI "Total Production" can be calculated based on the defined relationships.

[0067] Another example of creating a KPI is shown in Fig. 10. Fig. 10 is a diagram showing an example of creating a KPI regarding material loss relative to the feed amount in an ethylene plant.

[0068] As shown in FIG. 10, it is assumed that the user U has input the relational expression (1) for the tag data "Material Loss (% Feed)" via the HMI unit 102 in the same manner as shown in FIGS. 8 and 9. Similarly, it is assumed that the user U has input the relational expression (2) for the tag data "Material Total Inlet," "Material Total Outlet," and "Purged Gas Rate to Flare." The registration unit 104b then defines a relationship based on these relational expressions. This makes it possible to calculate the KPI "Material Loss (% Feed)," which relates to the material loss relative to the feed amount in the ethylene plant, based on the defined relationship.

[0069] Returning to the description of Fig. 6, the reception unit 104c presents the template included in the template information 103b to the user U via the HMI unit 102. The reception unit 104c also receives input of information desired by the user U regarding the KPIs of the plant P1, which information has been input using the template and acquired by the acquisition unit 104a. The reception unit 104c also causes the pair generation AI processing unit 104d to execute pair generation AI processing according to the received input.

[0070] Here, a specific example of a template for an output request prompt will be described. Fig. 11 is a diagram showing an example of a template for an output request prompt and an input example. The accepting unit 104c presents the template for an output request prompt such as that shown in Fig. 11 to the user U.

[0071] The user U inputs information into the input fields A1 to A6 of this template as appropriate. In the input field A1, the user U specifies, for example, the perspective from which the user wants to obtain information. Here, an example of input is shown in which information from the perspective of a "professional process consultant" at the plant P1 is desired.

[0072] In the input area A2, the user U specifies an arbitrary dashboard to which visualized output information is to be output. In the input area A2, the user U specifies the name of an application such as EXCEL (registered trademark) or TABLEAU (registered trademark) that corresponds to the dashboard DD1.

[0073] In the input area A3, the user U specifies the name of a database that accumulates past cases. In this embodiment, the user U specifies the name of the KPI information DB 103a. In the input areas A4 and A5, the user U inputs a set of highly relevant elements for the information the user U desires regarding the KPIs of the plant P1. In this embodiment, in the input area A4, the user U inputs, for example, a set of elements corresponding to each classification of the "Business Objective" described above. Also, in this embodiment, in the input area A5, the user U inputs, for example, a set of elements corresponding to each classification of the "Target Process" described above.

[0074] The input area A6 is an optional input field. In this embodiment, the user U appropriately inputs information to the input area A6, for example, that the user U desires information about an ethylene plant.

[0075] Returning to the explanation of Figure 6, the pair generation AI processing unit 104d executes pair generation AI processing based on the input accepted by the accepting unit 104c. Specifically, the pair generation AI processing unit 104d generates an output request prompt to be sent to the generation AI server 200 based on the accepted input data and the generation AI interface information 103c.

[0076] The pair generation AI processing unit 104d also transmits the generated output request prompt to the generation AI server 200 via the communication unit 101. The pair generation AI processing unit 104d also acquires, via the communication unit 101, generation information generated in response to the output request prompt from the generation AI server 200.

[0077] The output information generation unit 104e generates output information that visualizes the generation information from the generation AI server 200 acquired by the pair generation AI processing unit 104d on the dashboard DD1 as input, and outputs the output information to the HMI unit 102. The output information generation unit 104e generates output information so that visualization is performed according to the dashboard DD1 corresponding to an application arbitrarily executed by the application execution unit 104f. The application execution unit 104f executes the application corresponding to the dashboard DD1.

[0078] [Processing procedure executed by KPI information generation system 1] Next, the processing procedure executed by the KPI information generation system 1 will be described with reference to Fig. 12. Fig. 12 is a diagram showing a processing sequence executed by the KPI information generation system 1 according to the embodiment.

[0079] 12, first, the KPI information generation device 100 acquires past case data from the PIMS 10 (step S101). The timing for acquiring past case data may be constant, at predetermined intervals, or at any arbitrary timing.

[0080] The KPI information generation device 100 then registers the past case data acquired from the PIMS 10 (step S102). By repeating steps S101 to S102, the past case data is accumulated in the KPI information DB 103a as a past case list 103aa.

[0081] Meanwhile, the user U acquires any past case data from the past case list 103aa via the HMI unit 102 (step S103). Then, the user U extracts information that will become each element of the KPI from the acquired past case data via the HMI unit 102 and inputs it as tag data (step S104). The KPI information generation device 100 registers the input tag data in the past case list 103aa of the KPI information DB 103a (step S105).

[0082] Furthermore, the user U inputs KPI calculation formula data for the plant P1 via the HMI unit 102 (step S106). As described above, the KPI calculation formula data is data consisting of calculation formulas for calculating each KPI.

[0083] As described above, the KPI calculation formula data can be easily input using a GUI as various general calculation formulas corresponding to each tag data and as relational formulas between tag data (i.e., between KPIs) with which the calculation formulas are associated. The KPI information generation device 100 registers the input KPI calculation formula data in the KPI list 103ab of the KPI information DB 103a (step S107).

[0084] By repeating steps S103 to S107, a list of KPIs in which the calculation formulas for each KPI and the relationships between KPIs are defined is accumulated in the KPI information DB 103a as a KPI list 103ab.

[0085] Then, when the user U inputs desired information regarding the KPIs of the plant P1 via the HMI unit 102 at any timing (step S108), the input data is passed to the KPI information generation device 100 (step S109).

[0086] The KPI information generation device 100 generates an output request prompt for the generation AI based on the input data passed to it (step S110).The KPI information generation device 100 then transmits the generated output request prompt to the generation AI, i.e., the generation AI server 200 (step S111).

[0087] Upon receiving this output request prompt, the generation AI server 200 generates information according to the request using the generation AI model (step S112), and transmits the generated information to the KPI information generation device 100 (step S113).

[0088] The KPI information generation device 100 generates output information that visualizes the generation information received from the generation AI server 200 as input (step S114), and outputs the output information to the HMI unit 102 (step S115).

[0089] [Variations] In the above-described embodiment, an example was given in which the KPI information generation device 100 and the generation AI server 200 were separate devices, but they may also be integrated into a single KPI information generation device 100. In this case, the KPI information generation device 100 has a generation AI model, and by reading this generation AI model, it also functions as a generation AI in place of the generation AI server 200.

[0090] In addition, in the above-described embodiment, an example was given in which the generative AI model was a unimodal large-scale language model, but the generative AI model may also be a multimodal large-scale language model that can accept inputs such as voice and images in addition to text as modalities.

[0091] [effect] As described above, the KPI information generation device 100 (corresponding to an example of an "information generation device") according to the embodiment includes a control unit 104. The control unit 104 acquires past case data related to the operation of the plant P1 and stores it in a KPI information DB 103a (corresponding to an example of a "database"). The control unit 104 registers tag data, which is information that serves as elements of KPIs, in the KPI information DB 103a based on the past case data. The control unit 104 also registers a KPI list 103ab (corresponding to an example of "definition information"), which includes calculation formulas corresponding to the tag data and relational expressions between the tag data, in the KPI information DB 103a. When desired information related to KPIs of the plant P1 is input from a user U, the control unit 104 requests a generation AI using a large-scale language model to generate information corresponding to the input using the KPI information DB 103a. The control unit 104 then presents the information generated by the generation AI in response to the request to the user U. This allows for more easily structuring and generating information related to KPIs in the plant P1.

[0092] Furthermore, the control unit 104 visualizes the generated information on the dashboard DD1 corresponding to any application, thereby allowing the user U to receive information about the desired KPI visualized in accordance with any application executed by the user U.

[0093] The control unit 104 also requests the generation AI to generate the generation information as a script that can be visualized on the dashboard DD1 using the generation information as input, thereby enabling the generation AI to provide the generation information in the form of a visualizeable script using a unimodal large-scale language model.

[0094] Furthermore, the control unit 104 accepts input of desired information regarding the KPIs of the plant P1 from the user U via a pre-defined template. This allows the user U to easily input desired information regarding the KPIs of the plant P1 via the template.

[0095] Furthermore, the above formula is a general formula at least in chemical engineering, so that even if the user U is not an experienced engineer in, for example, a chemical plant, he or she can easily set the KPI list 103ab by inputting a general formula in chemical engineering.

[0096] Furthermore, the plant P1 is a chemical plant. This allows the user U to more easily structure and generate information relating to KPIs in, for example, an ethylene plant.

[0097] Furthermore, the control unit 104 presents a GUI for inputting the above-mentioned calculation formulas and relational expressions to the user U based on the KPI list 103ab, which allows the user U to easily input the calculation formulas and relational expressions using this GUI.

[0098] Furthermore, the control unit 104 registers in the KPI information DB 103a a KPI list 103ab for each combination of the classification based on the business objectives of the plant P1 and the classification based on the target processes of the plant P1. This makes it possible to easily set up at least each combination of the classification based on the business objectives of the plant P1 and the classification based on the target processes for a wide range of KPI setting targets of the plant P1.

[0099] [Other embodiments] Although the embodiments of the present invention have been described above, the present invention may be embodied in various different forms other than the above-described embodiments.

[0100] [Past case data] In the above-described embodiment, an example was given in which the acquisition unit 104a of the KPI information generation device 100 acquires past case data that forms the basis of the past case list 103aa from the PIMS10 via the communication unit 101, but the acquisition unit 104a may also acquire past case data in other ways.

[0101] For example, the acquisition unit 104a may acquire past case data via a recording medium on which past case data output from the PIMS 10 is recorded. Alternatively, for example, the acquisition unit 104a may acquire past case data manually input by the user U based on data output in some format from the PIMS 10 via the HMI unit 102. Alternatively, for example, the acquisition unit 104a may acquire past case data manually input by the user U based on analog data recorded on paper or the like, regardless of whether it was output from the PIMS 10.

[0102] [system] The information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified.

[0103] Furthermore, the components of each device shown in the figure are functional concepts and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown. In other words, all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0104] Furthermore, all or any part of the processing functions performed by each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware using wired logic.

[0105] [Hardware] The KPI information generation device 100 according to the embodiment described above is realized by, for example, a computer 1000 configured as shown in Fig. 13. Fig. 13 is a hardware configuration diagram showing an example of the computer 1000 that realizes the functions of the KPI information generation device 100 according to the embodiment.

[0106] 13, the computer 1000 includes a communication device 1000a, a secondary storage device 1000b, a memory 1000c, and a processor 1000d. The components shown in FIG. 13 are interconnected by a bus or the like.

[0107] The communication device 1000a is a network interface card (NIC) or the like, and communicates with other devices. The secondary storage device 1000b is realized by a flash memory, a hard disk drive, or the like, and stores programs and databases that operate the functions shown in FIG.

[0108] The processor 1000d reads a program that executes the same processes as the respective processing units shown in FIG. 6 from the secondary storage device 1000b or the like and loads it into the memory 1000c, thereby running a thread that executes each function described in FIG. 6 or the like. For example, this thread executes the same functions as the respective processing units of the KPI information generation device 100. Specifically, the processor 1000d reads a program having the same functions as the acquisition unit 104a, the registration unit 104b, the reception unit 104c, the pair generation AI processing unit 104d, the output information generation unit 104e, and the application execution unit 104f from the secondary storage device 1000b or the like. Then, the processor 1000d executes a thread that executes the same processes as the acquisition unit 104a, the registration unit 104b, the reception unit 104c, the pair generation AI processing unit 104d, the output information generation unit 104e, the application execution unit 104f, etc.

[0109] In this way, the computer 1000 operates as an information processing device that executes various processing methods by reading and executing the program. The computer 1000 can also realize functions similar to those of the above-described embodiments by reading the program from a recording medium using a medium reading device and executing the read program. Note that the program referred to here is not limited to being executed solely by the computer 1000. For example, the present invention can be similarly applied to cases where a computer or server having a different hardware configuration executes the program, or where these execute the program in cooperation with each other.

[0110] This program can be distributed via a network such as the Internet. This program can also be recorded on a computer-readable recording medium such as a hard disk drive (HDD), a flexible disk (FD), a CD-ROM, a magneto-optical disk (MO), or a digital versatile disk (DVD), and can be executed by being read from the recording medium by a computer. A recording medium on which such a program is recorded is also an aspect of the present disclosure.

[0111] [others] Some examples of combinations of the disclosed technical features are listed below.

[0112] (1) Past case data on plant operations is collected and stored in a database. registering tag data, which is information that will be an element of the KPI, based on the past case data in the database; registering definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data in the database; When receiving an input of desired information regarding the KPI of the plant from a user, a request is made to a generation AI using a large-scale language model to generate generation information according to the input using the database; a control unit that presents the generated information generated by the generation AI in response to the request to the user; An information generating device comprising: (2) The control unit Visualizing the generated information into a digital dashboard compatible with any application software; The information generating device according to (1). (3) The control unit Requesting the generation AI to generate the generated information as a script that can be visualized on the digital dashboard using the generated information as input; The information generating device according to (2). (4) The control unit receiving input of desired information regarding the KPIs of the plant from the user via a pre-defined template; An information generating device according to (1), (2) or (3). (5) The formula is at least a general formula in chemical engineering. The information generating device according to any one of (1) to (4). (6) The plant is a chemical plant. The information generating device according to (5). (7) The control unit presenting a GUI for inputting the calculation formula and the relational formula to the user based on the definition information; The information generating device according to any one of (1) to (6). (8) The control unit registering the definition information for each combination of a classification based on business goals in the plant and a classification based on the target process in the database; The information generating device according to any one of (1) to (7). (9) A system for managing plant operations and an information generating device, The system comprises: transmitting past case data relating to the operation of the plant to the information generating device; The information generating device The past case data is acquired and stored in a database, registering tag data, which is information that will be an element of the KPI, based on the past case data in the database; registering definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data in the database; When receiving an input of desired information regarding the KPI of the plant from a user, a request is made to a generation AI using a large-scale language model to generate generation information according to the input using the database; presenting the generated information generated by the generation AI in response to the request to the user; Information generation system. (10) The computer Past case data on plant operations is collected and stored in a database. registering tag data, which is information that will be an element of the KPI, based on the past case data in the database; registering definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data in the database; When receiving an input of desired information regarding the KPI of the plant from a user, a request is made to a generation AI using a large-scale language model to generate generation information according to the input using the database; presenting the generated information generated by the generation AI in response to the request to the user; Information generation methods that perform processing. (11) On the computer, Past case data on plant operations is collected and stored in a database. registering tag data, which is information that will be an element of the KPI, based on the past case data in the database; registering definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data in the database; When receiving an input of desired information regarding the KPI of the plant from a user, a request is made to a generation AI using a large-scale language model to generate generation information according to the input using the database; presenting the generated information generated by the generation AI in response to the request to the user; A program that executes a process. (12) A computer-readable recording medium on which a program is recorded, The program On the computer, Past case data on plant operations is collected and stored in a database. registering tag data, which is information that will be an element of the KPI, based on the past case data in the database; registering definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data in the database; When receiving an input of desired information regarding the KPI of the plant from a user, a request is made to a generation AI using a large-scale language model to generate generation information according to the input using the database; presenting the generated information generated by the generation AI in response to the request to the user; A computer-readable recording medium that causes processing to be performed. [Explanation of symbols]

[0113] 1 KPI information generation system 100 KPI information generation device 101 Communications Department 102 HMI section 103 Storage section 103a KPI information DB 103aa Past Case List 103ab KPI List 103b Template Information 103c Generated AI interface information 104 Control Unit 104a Acquisition Department 104b Registration Department 104c Reception 104d Pair generation AI processing unit 104e Output information generation section 104f Application execution section 200 Generation AI Server P1 Plant U User

Claims

1. Past case data on plant operations is collected and stored in a database. registering tag data, which is information that will be an element of KPI, based on the past case data in the database; registering definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data in the database; When receiving an input of desired information regarding the KPI of the plant from a user, a request is made to a generation AI using a large-scale language model to generate generation information according to the input using the database; a control unit that presents the generated information generated by the generation AI in response to the request to the user; An information generating device comprising:

2. The control unit Visualizing the generated information on a digital dashboard compatible with any application software; The information generating device according to claim 1 .

3. The control unit Requesting the generation AI to generate the generated information as a script that can be visualized on the digital dashboard using the generated information as input; The information generating device according to claim 2 .

4. The control unit receiving input of desired information regarding the KPI of the plant from the user via a pre-defined template; The information generating device according to claim 1 .

5. The formula is at least a general formula in chemical engineering. The information generating device according to claim 1 .

6. The plant is a chemical plant. The information generating device according to claim 5 .

7. The control unit presenting a GUI for inputting the calculation formula and the relational formula to the user based on the definition information; The information generating device according to claim 1 .

8. The control unit registering the definition information for each combination of a classification based on business goals in the plant and a classification based on the target process in the database; The information generating device according to claim 1 .

9. A system for managing plant operations and an information generating device, The system comprises: transmitting past case data relating to the operation of the plant to the information generating device; The information generating device The past case data is acquired and stored in a database, registering tag data, which is information that will be an element of KPI, based on the past case data in the database; registering definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data in the database; When receiving an input of desired information regarding the KPI of the plant from a user, a request is made to a generation AI using a large-scale language model to generate generation information according to the input using the database; Presenting the generated information generated by the generation AI in response to the request to the user; Information generation system.

10. The computer Past case data on plant operations is collected and stored in a database. registering tag data, which is information that will be an element of KPI, based on the past case data in the database; registering definition information including a calculation formula corresponding to the tag data and a relational formula between the tag data in the database; When receiving an input of desired information regarding the KPI of the plant from a user, a request is made to a generation AI using a large-scale language model to generate generation information according to the input using the database; Presenting the generated information generated by the generation AI in response to the request to the user; Information generation methods that perform processing.

Images