Information processing system, information processing method, and program

Abstract

[Problem] To provide an information processing system and the like capable of optimizing an entire manufacturing line. [Solution] According to one aspect of the present invention, an information processing system is provided, the information processing system comprising a processor configured to execute following steps by reading a program, wherein the processor executes: a model acquisition step to acquire a pre-process model in which a pre-process for generating an intermediate product from a raw material is simulated, and a post-process model in which a post-process for generating a final product from the intermediate product is simulated; a condition acquisition step to acquire a process condition for each of the pre-process model and the post-process model; and an inference step to acquire, as an inference value, at least one of a characteristic value of the intermediate product, a characteristic value of the final product, a state value of a device used in the pre-process, and a state value of a device used in the post-process, by inputting the process conditions into a linked model in which at least the pre-process model and the post-process model are linked.

Description

Information processing system, information processing method and program

[0001] The present invention relates to an information processing system, an information processing method, and a program.

[0002] Patent Document 1 discloses a system for optimizing production plans and logistics plans using a plurality of manufacturing processes as a unit of optimization.

[0003] JP 2024-58363 A

[0004] The above system can be optimized using production cost as an objective function, but the cost of each manufacturing process is fixed and unrelated to other processes, so there is room for improvement in optimizing the entire manufacturing line.

[0005] In view of the above circumstances, the present invention provides an information processing system and the like that enables the overall optimization of a manufacturing line.

[0006] According to one aspect of the present invention, there is provided an information processing system including a processor configured to execute the following steps by reading a program: in a model acquisition step, an upstream process model that simulates an upstream process of producing an intermediate product from raw materials and a downstream process model that simulates a downstream process of producing a final product from the intermediate product are acquired; in a condition acquisition step, process conditions for each of the upstream process model and the downstream process model are acquired; and in an inference step, by inputting the process conditions into a linked model in which at least the upstream process model and the downstream process model are linked, at least one of a characteristic value of the intermediate product, a characteristic value of the final product, a status value of an apparatus used in the upstream process, and a status value of an apparatus used in the downstream process is acquired as an inferred value.

[0007] According to this aspect, the entire manufacturing line can be optimized by adjusting the process conditions in the upstream and / or downstream processes with reference to the inferred values ​​obtained from the linked model.

[0008] 1 is a configuration diagram showing an information processing system 1. FIG. 2 is a block diagram showing the hardware configuration of an information processing device 10. FIG. 3 is a block diagram showing the hardware configuration of a user terminal 20. FIG. 4 is a block diagram showing functions realized by the information processing device 10 (control unit 11) and the user terminal 20 (control unit 21). FIG. 5 is a diagram schematically showing the input and output of information in a linked model in which an upstream process model and a downstream process model are linked. FIG. 6 is an activity diagram showing an example of the flow of information processing (process condition adjustment processing) executed by the information processing system 1.

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the accompanying drawings. Various features shown in the following embodiments can be combined with each other.

[0010] Incidentally, a program for realizing the software appearing in one embodiment may be provided as a non-transitory computer-readable recording medium, or may be provided so as to be downloadable from an external server, or may be provided so that the program is started on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0011] Furthermore, various information processing according to an embodiment may realize input and output corresponding to the input. Here, the form of information referenced in such information processing (hereinafter referred to as reference information) is not limited as long as an output is obtained as a result of the input. The reference information may be, for example, rule-based information such as a database, a lookup table, or a predetermined function (including a decision formula such as a regression formula constructed using a statistical method), a trained model that has previously trained the correlation between input and output, or a large-scale language model that can output a desired result by inputting a prompt.

[0012] In one embodiment, the term "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In one embodiment, various information is handled, and this information is represented, for example, by physical values ​​of signal values ​​representing voltage or current, high or low signal values ​​as a binary bit set consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculations can be performed on the circuit in the broad sense.

[0013] Furthermore, a circuit in a broad sense is a circuit realized by at least an appropriate combination of a circuit, circuitry, a processor, a memory, etc. The processor may be a general-purpose processor or a dedicated circuit. That is, it includes application specific integrated circuits (ASICs), programmable logic devices (e.g., simple programmable logic devices (SPLDs), complex programmable logic devices (CPLDs), and field programmable gate arrays (FPGAs)), etc.

[0014] 1. Hardware Configuration This section describes the hardware configuration.

[0015] <Information Processing System 1> Fig. 1 is a configuration diagram showing an information processing system 1. The information processing system 1 includes a communication line 2, an information processing device 10, and a plurality of user terminals 20. The information processing device 10 and the user terminals 20 are configured to be able to communicate with each other via the communication line 2. The information processing device 10 and the user terminals 20 may be connected by wire or wirelessly.

[0016] In one embodiment, the information processing system 1 comprises one or more devices or components, which will now be described.

[0017] <Information Processing Device 10> Fig. 2 is a block diagram showing the hardware configuration of the information processing device 10. As shown in Fig. 2, the information processing device 10 includes a control unit 11, a storage unit 12, a communication unit 13, and a communication bus 14. The control unit 11, the storage unit 12, and the communication unit 13 are electrically connected via the communication bus 14 inside the information processing device 10.

[0018] <Control Unit 11> The control unit 11 processes and controls the overall operations related to the information processing device 10. The control unit 11 is, for example, a central processing unit (CPU). The control unit 11 realizes various functions related to the information processing device 10 by reading out predetermined programs stored in the storage unit 12. In other words, information processing by software stored in the storage unit 12 is specifically realized by the control unit 11, which is an example of hardware, and can be executed as each functional unit included in the control unit 11. These will be described in more detail in the next section. Note that the control unit 11 is not limited to being a single unit, and the information processing device 10 may have multiple control units 11 for each function. Furthermore, the information processing device 10 may have a configuration that combines these.

[0019] <Storage Unit 12> The storage unit 12 stores various pieces of information defined above. This may be implemented, for example, as a storage device such as a solid state drive (SSD) that stores various programs and the like related to the information processing device 10 executed by the control unit 11, or as a memory such as a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to program calculations. The storage unit 12 stores various programs, variables, etc. related to the information processing device 10 executed by the control unit 11.

[0020] <Communication Unit 13> The communication unit 13 is preferably a wired communication means such as USB, IEEE 1394, Thunderbolt (registered trademark), or wired LAN network communication, but may also include wireless LAN network communication, mobile communication such as LTE / 5G, BLUETOOTH (registered trademark) communication, etc. as needed. In other words, it is more preferable to implement the communication unit 13 as a collection of multiple communication means. In other words, the information processing device 10 may communicate various information from the outside via the communication unit 13 and the network.

[0021] The information processing device 10 may be an on-premise type or a cloud type. As the information processing device 10 in the cloud type, the above-described functions and processes may be provided in the form of, for example, SaaS (Software as a Service) or cloud computing.

[0022] <User Terminal 20> The user terminal 20 is an information processing terminal used by a user who utilizes the production line optimization service provided by the information processing device 10. Fig. 3 is a block diagram showing the hardware configuration of the user terminal 20. As shown in Fig. 3, the user terminal 20 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, an output unit 25, and a communication bus 26. The control unit 21, the storage unit 22, the communication unit 23, the input unit 24, and the output unit 25 are electrically connected via the communication bus 26 inside the user terminal 20. The description of the control unit 21, the storage unit 22, and the communication unit 23 is omitted because they are the same as the description of each unit in the information processing device 10.

[0023] <Input Unit 24> The input unit 24 accepts an action input made by the user. The action input is transferred as a command signal to the control unit 21 via the communication bus 26. The control unit 21 can execute predetermined control or calculation based on the transferred command signal as necessary. The input unit 24 may be included in the housing of the user terminal 20 or may be externally attached. For example, the input unit 24 may be implemented as a touch panel integrated with the output unit 25. When the input unit 24 is implemented as a touch panel, the user can input a tap action, a swipe action, or the like to the input unit 24. Instead of a touch panel, a switch button, a mouse, a trackpad, a QWERTY keyboard, or the like can be used as the input unit 24.

[0024] <Output Unit 25> The output unit 25 displays a graphical user interface (GUI) screen on which the user can perform actions. The output unit 25 may be included in the housing of the user terminal 20 or may be attached externally. Specifically, the output unit 25 may be implemented as a display device such as a CRT display, a liquid crystal display, an organic EL display, or a plasma display. It is preferable that these display devices are used appropriately depending on the type of user terminal 20.

[0025] 2. Functional Configuration This section describes the functional configuration of this embodiment. Information processing by software stored in the storage unit 12 is specifically realized by the control unit 11, which is an example of hardware, and can be executed as each functional unit included in the control unit 11 (a processor provided in the information processing system 1).

[0026] FIG. 4 is a block diagram showing functions realized by the information processing device 10 (controller 11) and the user terminal 20 (controller 21).

[0027] 4A, the information processing device 10 (control unit 11) includes a model registration unit 111, a model acquisition unit 112, a condition acquisition unit 113, an inference unit 114, a goal acquisition unit 115, a condition adjustment unit 116, and an artificial intelligence unit 120. As shown in FIG. 4B, the user terminal 20 (control unit 21) includes a display unit 211 and an operation acquisition unit 212.

[0028] <Model Registration Unit 111> The model registration unit 111 is configured to accept registration of a production process model that simulates a process for producing a product. Specifically, the model registration unit 111 accepts uploading of the production process model from the user terminal 20, input of the storage location on the network of the production process model from the user terminal 20, etc., and registers the model in a model database stored in the storage unit 12, for example.

[0029] A "process for producing a product" refers to a production process in which raw materials are received and a desired product is obtained by chemically reacting the raw materials using equipment, mechanically processing the raw materials, or the like. Specifically, the "process for producing a product" includes a process for synthesizing multiple raw materials, a process for dispersing raw materials, a process for mixing (kneading) multiple raw materials, a process for separating specific components from raw materials, a process for modifying raw materials, a process for molding (casting) raw materials, a process for cutting raw materials, a process for painting raw materials, a process for dyeing raw materials, a process for surface processing raw materials (polishing, plating, thin film formation), a process for attaching other parts to raw materials, a process for attaching raw materials to other parts, a process for packaging raw materials, and the like. Note that multiple types of raw materials may be input into one production process. Furthermore, multiple types of products may be obtained from one production process.

[0030] A "production process model that simulates a process" is a numerical model that defines the correlation between input parameters that represent the characteristics of raw materials and process conditions (equipment setting conditions) and output parameters that represent the characteristics of the product and the state of the equipment used in the process.

[0031] The raw material characteristics, which are input parameters, include, for example, the material properties (composition ratio, purity, quality), temperature distribution, stress distribution, volume, density, shape, etc. The process conditions, which are input parameters, include, for example, the temperature, pressure, magnetic field, voltage (current), processing time, processing sequence, etc. when performing processes such as chemical reactions and mechanical processing.

[0032] More specific process conditions include, for example, the temperature sequence, partial pressure sequence, timing of switching gas types, output value sequence of the heating power supply, rotation sequence of each axis of the processing device, feed rate sequence of each axis, processing pressure sequence, device shape, amount of material input, type of consumable parts and their physical properties before the process is performed, cumulative number of times the consumable parts are used before the process is performed, cumulative number of times the device is used before the process is performed, cumulative number of times the device is used after the most recent maintenance, the installation environment of the device (temperature, humidity, etc.), processing position relative to the raw materials, etc.

[0033] The characteristics of the product, which are output parameters, include, for example, temperature distribution, stress distribution, defect density distribution, shape (surface shape), material composition ratio, etc. The state of the equipment, which is an output parameter, includes temperature distribution (heat generation distribution) inside the equipment, pressure distribution (gas partial pressure distribution, gas flow rate distribution), magnetic field distribution, type of consumable parts and their physical properties after execution of a process, cumulative number of times consumable parts have been used after execution of a process, cumulative number of times the equipment has been used after execution of a process, cumulative number of times the equipment has been used after the latest maintenance, the installation environment of the equipment (temperature, humidity, etc.), degree of deterioration of the equipment, degree of deterioration of consumable parts, degree of contamination of the equipment, time-series data (sensor values) such as temperature, gas flow rate, and current inside the equipment, electricity consumption, production speed, CO ２ This includes emissions, production costs, production variations, estimated equipment maintenance times, and replacement times for consumable parts.

[0034] A production process model is an estimator that is constructed to receive input parameters and output output parameters. The production process model may be, for example, a table, a function, a simple algorithm, or the like that indicates the correlation between the input parameters and the output parameters. The correlation included in the production process model can be constructed, for example, by statistically analyzing actual measurement data from the production process that the production process model simulates.

[0035] The production process model may be a learning model that is trained to receive input parameters and output output parameters. In this case, the production process model is trained using training data that is a combination of the learning input parameters and the output parameters corresponding to the input parameters. The parameters of the learning model that are calculated, tuned, etc. through such learning constitute the correlations of the production process model.

[0036] The production process model may be a physical simulator (physics calculation engine). In this case, the production process model uses input parameters as initial conditions, performs physical calculations on raw materials for processing included in the production process, and outputs the results as output parameters.

[0037] The model registration unit 111 may register model information related to the production process model along with the production process model. The model information is input from the user terminal 20. The model information includes, for example, the dimensions of each of the input parameters and output parameters (the number of elements of the parameter), element information about each element of the input parameters and output parameters, meta information about the production process model, and usage range information.

[0038] Element information includes, for example, the name of the element, the numerical range (maximum and minimum values), numerical variation, units, significant digits, a flag indicating whether it can be adjusted (whether it is a fixed value or not), and private information indicating whether it can be made public to other users.

[0039] An element whose private information includes information (flag) indicating "not to be disclosed" becomes a private element, and the specific numerical value of the private element is not disclosed to other users who use the production process model (i.e., users other than the user who registered the production process model). Furthermore, if the private element is an input parameter, the private information includes information specifying the adjustment method for adjusting the numerical value of the private element, for example, either (1) or (2) below. (1) The numerical value of the private element is fixed and cannot be adjusted (the numerical value is not displayed to other users). (2) The numerical value of the private element can be selected from pre-prepared candidates (numerals listed). However, the candidates are displayed with labels such as "A," "B," and "C," and the numerical value is not displayed to other users.

[0040] The meta-information includes information related to the construction of the production process model (learning curve, training data, hyperparameters, etc.), equations used by the production process model, boundary conditions, approximation conditions, assumptions, definitions of physical properties, etc.

[0041] The usage range information includes information about other users who are permitted to use the production process model as part of a linked model of the inference unit 114, which will be described later. A production process model registered by a user is permitted to be selected as an element of a production line (linked model) to be optimized only by other users who are set as available users in the usage range information.

[0042] The model registration unit 111 may accept registration of a first production process model that simulates a process for producing a product from a first user, and may also accept registration of a second production process model that simulates a process for producing a product from a second user. The second production process model is a model that simulates a production process different from the first production process model, for example, a model that simulates a process subsequent to the production process simulated by the first production process model. For example, in a production line for a certain product, the second user is a company that is responsible for a process downstream from the first user.

[0043] The model registration unit 111 may accept an update from the user for an already registered production process model. Specifically, the model acquisition unit 112 accepts an input of an updated production process model from the user and replaces the production process model registered in the model database with the updated production process model.

[0044] The model registration unit 111 also registers configuration information of the linked model in the model database as information indicating a production line in which multiple production process models are linked. The linked model is a model that simulates a production line that includes multiple production processes. The configuration information of the linked model is registered, for example, by the user selecting, on the user terminal 20, multiple production process models that make up the production line to be optimized in accordance with the order of the processes.

[0045] The model registration unit 111 receives information identifying the selected multiple production process models (model name, ID, etc.) and information indicating the order of the processes from the user terminal 20, and registers them in the model database as configuration information of the linked model. The production process models that a user can select as elements of the linked model are not only the production process models that the user has registered, but also production process models that have been registered by other users and are set as selectable (i.e., set as available users).

[0046] When a production process model is updated, the model registration unit 111 may notify a user who has registered a linked model that incorporates the production process model that the production process model has been updated. For example, the model registration unit 111 may send information about the updated production process model (model name, update date and time, update content, etc.) to the user terminal 20 of the user who has registered a linked model that incorporates the updated production process model.

[0047] <Model Acquisition Unit 112> The model acquisition unit 112 is configured to acquire a model (i.e., a model included in the linked model) to be used for inferring a specified manufacturing line from among the production process models registered by the model registration unit 111 (i.e., registered in the model database).

[0048] Specifically, the model acquisition unit 112 acquires an upstream process model that simulates an upstream process in which an intermediate product is produced from raw materials, and a downstream process model that simulates a downstream process in which a final product is produced from the intermediate product. The raw materials input into the upstream process model may have already undergone some production process (i.e., a product produced in another production process). The final product produced by the downstream process model may be input into yet another production process (i.e., raw materials for another production process). In other words, the "final product" simply means the product obtained from the downstream process included in the "linked model," and does not mean the final product of the entire manufacturing line including the downstream process.

[0049] The intermediate product generated in the preceding process corresponds to the raw material for the succeeding process. Therefore, in the inference unit 114 described below, some of the output parameters of the preceding process model (typically, some of the characteristics of the product) are used as some of the input parameters of the succeeding process model (typically, the characteristics of the raw material).

[0050] The model acquisition unit 112 may acquire the production process model registered by the model registration unit 111 as a front-end process model or a back-end process model. This allows the production line to be optimized using a production process model prepared by the user. Note that the model acquisition unit 112 may also acquire a production process model other than the production process model registered by the user (for example, a production process model registered by a provider of a service provided by the information processing system 1) as a front-end process model or a back-end process model.

[0051] The model acquisition unit 112 may also acquire a first production process model registered by a first user as a front-end process model, and a second production process model registered by a second user as a back-end process model, thereby enabling optimization of a manufacturing line that includes production processes performed by different entities (i.e., production processes performed by multiple users).

[0052] The manufacturing line handled by the information processing system 1 may include three or more production processes. For example, the downstream process may include an intermediate process for producing a secondary intermediate product from an intermediate product and a final process for producing a final product from the secondary intermediate product. In this case, the model acquisition unit 112 may acquire a upstream process model, an intermediate process model that simulates the intermediate process, and a final process model that simulates the final process. The intermediate process model and the final process model may be production process models registered by the same user or may be production process models registered by different users. The model acquisition unit 112 may acquire a production process model other than the production process model registered by the user (a production process model registered by a service provider) as the intermediate process model or the final process model.

[0053] <Condition Acquisition Unit 113> The condition acquisition unit 113 is configured to acquire process conditions for each of the upstream process model and the downstream process model, which are conditions for the inference unit 114 to perform inference. Specifically, the condition acquisition unit 113 accepts input or selection of each element included in the process conditions of each production process model at the user terminal 20. The condition acquisition unit 113 may also accept input of raw material characteristics in addition to the process conditions for the upstream process model.

[0054] Furthermore, when the subsequent process includes an intermediate process model and a final process model, the condition acquisition unit 113 acquires the process conditions for each of the previous process model, the intermediate process model, and the final process model.

[0055] The condition acquisition unit 113 may acquire process conditions adjusted by a condition adjustment unit 116 (to be described later) instead of the process conditions input or selected on the user terminal 20 .

[0056] <Inference Unit 114> The inference unit 114 is configured to infer values ​​of output parameters in a linked model in which a plurality of process models are linked. Specifically, the inference unit 114 inputs process conditions into a linked model in which at least an upstream process model and a downstream process model are linked, and thereby acquires, as an inferred value, at least one of a characteristic value of an intermediate product, a characteristic value of a final product, a state value of a device used in the upstream process, and a state value of a device used in the downstream process.

[0057] Furthermore, when the subsequent process includes an intermediate process model and a final process model, the inference unit 114 may input process conditions into a linked model in which the previous process model, the intermediate process model, and the final process model are linked, and thereby acquire at least one of the characteristic values ​​of the intermediate product, the characteristic values ​​of the secondary intermediate product, the characteristic values ​​of the final product, the status values ​​of the equipment used in the previous process, the status values ​​of the equipment used in the intermediate process, and the status values ​​of the equipment used in the final process as an inferred value. This makes it possible to acquire inferred values ​​for each process even for a manufacturing line that includes three or more production processes.

[0058] More specifically, the inference unit 114 inputs the characteristic values ​​of the raw materials into the first production process model of the linked model (i.e., the upstream process model), and inputs the respective process conditions acquired by the condition acquisition unit 113 into each production process model (the upstream process model and the downstream process model) that make up the linked model, thereby causing each production process to output an inferred value (the characteristic value of the product and / or the status value of the equipment).

[0059] The number and type of process condition elements differ for each production process, and the number and type of element for the characteristic values ​​of the products and the state values ​​of the equipment also differ for each production process.

[0060] The inference unit 114 may input process conditions into the linked model to acquire characteristic values ​​of the final product, state values ​​of equipment used in the preceding process, and state values ​​of equipment used in the following process as inferred values. This makes it possible to infer the quality of the product obtained from the production line simulated by the linked model, and to infer optimization indicators such as the energy efficiency and consumption costs of the production line based on the state values ​​of the equipment in each process.

[0061] The inference unit 114 may acquire characteristic values ​​of the intermediate product and state values ​​of the equipment used in the upstream process by inputting characteristic values ​​and process conditions of the raw materials into the upstream process model, and acquire characteristic values ​​of the final product and state values ​​of the equipment used in the downstream process by inputting characteristic values ​​and process conditions of the intermediate product into the downstream process model. This makes it possible to infer the state of the equipment in each process while reliably transferring the execution results of the upstream process model (information on the intermediate product) to the downstream process model.

[0062] The inference unit 114 may input only some of the elements included in the characteristic values ​​of the intermediate product to the subsequent process model. In other words, the inference unit 114 does not necessarily need to use all of the elements of the characteristic values ​​of the intermediate product as input parameters of the subsequent process model. Elements defined as input parameters in the model information of the subsequent process model are input to the subsequent process model.

[0063] The inference unit 114 does not necessarily need to acquire all of the characteristic values ​​of the intermediate product, the characteristic values ​​of the final product, the status values ​​of the devices used in the previous process, and the status values ​​of the devices used in the subsequent process. For example, the inference unit 114 may acquire only the characteristic values ​​of the final product, or may acquire only the status values ​​of the devices used in the previous process and the status values ​​of the devices used in the subsequent process.

[0064] The inference unit 114 records the inferred values ​​(such as characteristic values ​​of the final product) acquired from the linked model in the model database, for example, by linking them to the linked model used for the inference. The inference unit 114 also displays the inferred values ​​on the user terminal 20. This allows the user to check each inferred value on the user terminal 20.

[0065] 5 is a diagram showing the input and output of information in a linked model that links an upstream process model and a downstream process model. As shown in FIG. 5, a characteristic value CV1 indicating the characteristics of raw materials and a process condition PC1 for the upstream process are input as input parameters to the upstream process model PM1. In response to the input of these input parameters, the upstream process model PM1 outputs a characteristic value CV2 indicating the characteristics of the intermediate product and a status value SV1 indicating the status of the equipment used in the upstream process during or after the upstream process.

[0066] A characteristic value CV2 indicating the characteristics of the intermediate product and a process condition PC2 in the downstream process are input as input parameters to the downstream process model PM2 connected to the upstream process model PM1. In response to the input of these input parameters, the downstream process model PM2 outputs a characteristic value CV3 indicating the characteristics of the final product and a status value SV2 indicating the status of the equipment used in the downstream process during or after the downstream process.

[0067] <Target Acquisition Unit 115> The target acquisition unit 115 is configured to acquire a target value related to the inference value acquired by the inference unit 114. For example, the target acquisition unit 115 accepts input, selection, etc. of a target value at the user terminal 20, associates the accepted target value with a linked model that optimizes process conditions, and registers the linked value in the model database.

[0068] The target value is an objective function when optimizing the process conditions input to the linked model, and is a numerical value indicating the upper or lower limit (target range) of the inferred value itself or the evaluation index derived from the inferred value. Examples of the evaluation index include productivity indexes such as manufacturing time, CO ２ Examples include decarbonization indicators such as emissions, energy saving indicators such as power consumption and fuel consumption, cost indicators such as manufacturing costs and raw material or fuel consumption efficiency, defect rate indicators such as product yield, and product quality indicators such as stability of product performance.

[0069] The productivity index, decarbonization index, energy saving index, and cost index are calculated using, for example, the status values ​​of the equipment in each production process and first reference information. The first reference information is information including the correlation between the status values ​​of the equipment in each production process and each index. The first reference information includes, for example, a formula, a function, a table, a learning model of the artificial intelligence unit 120, a simulator, etc.

[0070] The defect rate index and the product quality index are calculated using, for example, characteristic values ​​of the products of each production process and second reference information. The second reference information is information including correlations between the characteristic values ​​of the products of each production process and each index. The second reference information includes, for example, a formula, a function, a table, a learning model of the artificial intelligence unit 120, a simulator, etc.

[0071] Each evaluation index may be calculated using both the state value of the equipment in each production process and the characteristic value of the product, as well as third reference information. The third reference information is information including correlations between the state value of the equipment in each production process and the characteristic value of the product, and each index. The third reference information includes, for example, a formula, a function, a table, a learning model of the artificial intelligence unit 120, a simulator, etc.

[0072] For example, the user creates and inputs an index to be used as a target for optimizing the production line, or selects an index from pre-prepared options, on the user terminal 20. The user may set multiple indexes for one production line. The user also inputs a target value for each index on the user terminal 20.

[0073] The target acquisition unit 115 may acquire search conditions for optimal process conditions in addition to the target values. The search conditions are input by the user via the user terminal 20. The search conditions include, for example, elements to be optimized (for which optimal values ​​are searched) among elements included in the process conditions of each step, adjustment ranges of the elements to be optimized (for example, upper and lower search limits), conditions to be set in the fourth reference information (described later) (for example, the type of algorithm), search time, etc.

[0074] When a linked model for which a user optimizes process parameters includes a production process model registered by another user, the user can set search conditions for non-private elements among the process parameters of the production process model in the same way as for the process parameters of the production process model registered by the user. On the other hand, private elements among the process parameters of a production process model registered by another user are treated as follows, depending on the other user's specification: (1) If the value of the private element is fixed, the value of the private element will not be searched (optimized). (2) If the value of the private element can be selected from pre-prepared candidates (values ​​listed), the candidate will be searched (optimized).

[0075] <Condition Adjustment Unit 116> The condition adjustment unit 116 is configured to adjust the process conditions in the link model based on the inference value acquired by the inference unit 114, the target value acquired by the target acquisition unit 115, and the fourth reference information. With this configuration, it is possible to optimize the process conditions of the production line using the link model based on the target value set by the user.

[0076] The fourth reference information includes correlations between the inferred values ​​and target values ​​and the process conditions. The fourth reference information is stored, for example, in the storage unit 12. The fourth reference information is an estimator configured to receive the inferred values ​​and target values ​​and output process conditions adjusted to achieve the target values. The fourth reference information may include, for example, a formula, a function, a table, a simple algorithm, or the like, which indicates the correlation between the combined data of the inferred values ​​and the target values ​​and the process conditions. The correlations included in the fourth reference information can be constructed, for example, by statistically analyzing data recording the process conditions used in an actual production process and measurements such as product characteristic values ​​and equipment status values ​​observed in the actual production process.

[0077] The fourth reference information may also be a search algorithm that searches for an optimal value for each element included in the process condition. When using such a search algorithm, the condition adjustment unit 116 adjusts the current process condition in accordance with the search algorithm so that the inferred value approaches the target value, and then causes the inference unit 114 to update the inferred value using the adjusted process condition (specifically, by having the condition acquisition unit 113 acquire the adjusted process condition). Furthermore, the condition adjustment unit 116 repeatedly adjusts the process condition and updates the inferred value using the updated inferred value. The adjustment of the process condition is repeated until a search termination condition is satisfied. The search termination condition may be, for example, when the difference between the inferred value and the target value becomes less than a threshold value or when a search time set as a search condition has elapsed. The parameters (branch condition, judgment condition, etc.) that constitute the search algorithm constitute the correlation of the fourth reference information.

[0078] The fourth reference information may also include a condition adjustment model that has been trained in advance by machine learning so as to input an inferred value and a target value and output process conditions for achieving the target values. In this case, the condition adjustment unit 116 inputs the inferred value and the target value into the condition adjustment model included in the artificial intelligence unit 120 and causes the condition adjustment model to output process conditions. The condition adjustment model is constructed by machine learning using, for example, training data that corresponds process conditions used in the production process in the past with measured values ​​such as characteristic values ​​of products and status values ​​of equipment observed in the production process. In the condition adjustment model, parameters calculated, tuned, etc. through learning constitute the correlations of the fourth reference information.

[0079] The condition adjusting unit 116 does not necessarily need to adjust all elements of the process conditions. For example, the condition adjusting unit 116 may fix the process conditions for some production processes and adjust only the process conditions for other production processes. Alternatively, in each production process, some elements of the process conditions may be fixed and only the other elements may be adjusted. The range of process conditions or elements to be adjusted is set, for example, as search conditions.

[0080] The condition adjustment unit 116 links the adjusted process conditions (optimized process conditions) to, for example, the optimized linked model and records them in the model database. The condition adjustment unit 116 also displays the optimized process conditions and the inferred values ​​and / or target values ​​when the optimized process conditions are used on the user terminal 20. This allows the user to check the optimized process conditions and the inferred values ​​and / or target values ​​after optimization on the user terminal 20.

[0081] The condition adjuster 116 may also calculate the contribution rate of each element included in the process conditions to the inferred value and / or the target value and display the calculated contribution rate on the user terminal 20. For example, the contribution rate of the temperature sequence of the preceding process to the reduction of the decarbonization index may be given as an example. The contribution rate is calculated or estimated based on data acquired during the search for optimized process conditions (correlations between each element of the process conditions and the inferred value and / or the target value).

[0082] <Artificial Intelligence Unit 120> The artificial intelligence unit 120 is configured to receive input from each functional unit and return the instructed output. The artificial intelligence used by each functional unit of the information processing device 10 may be a common one, or may be prepared individually for each functional unit.

[0083] The artificial intelligence unit 120 may include a learning model such as a language model, including a generative AI, such as a Transformer or a Recurrent Neural Network (RNN).

[0084] The language model is an example of a learning model based on a machine learning algorithm. Specific examples of machine learning algorithms include nearest neighbor methods, naive Bayes methods, decision trees, support vector machines, and deep learning using neural networks. The artificial intelligence unit 120 can apply the above algorithms as appropriate.

[0085] The artificial intelligence unit 120 may have a trained model constructed by a learning method such as supervised learning, unsupervised learning, or self-supervised learning. In supervised learning, machine learning is performed using training data (training data). The training data consists of pairs of input data for learning and output data (correct answer data). Furthermore, the language model may not only be trained for a specific task, but may also be a general-purpose model that can be used for a wide range of tasks. The learning model included in the artificial intelligence unit 120 can undergo additional training as transfer learning or fine tuning.

[0086] <Display Unit 211 > The display unit 211 of the user terminal 20 is configured to display on the output unit 25 a screen indicated by the screen data transmitted from the information processing device 10 .

[0087] <Operation Acquisition Unit 212 > The operation acquisition unit 212 of the user terminal 20 is configured to accept operations by the user of the user terminal 20 .

[0088] 3. Information Processing Method This section describes an information processing method of the information processing device 10. This information processing method is executed by a computer, with each unit of the information processing device 10 acting as each step.

[0089] Specifically, this information processing method includes a model registration step, a model acquisition step, a condition acquisition step, an inference step, a target acquisition step, and a condition adjustment step. The model registration step accepts registration of a model simulating a process for producing a product. The model acquisition step acquires an upstream process model simulating an upstream process for producing an intermediate product from raw materials and a downstream process model simulating a downstream process for producing a final product from the intermediate product. The condition acquisition step acquires process conditions for each of the upstream process model and the downstream process model. The inference step inputs process conditions into a linked model in which at least the upstream process model and the downstream process model are linked, thereby acquiring at least one of a characteristic value of the intermediate product, a characteristic value of the final product, a status value of a device used in the upstream process, and a status value of a device used in the downstream process as an inferred value. The target acquisition step acquires a target value related to the inferred value. The condition adjustment step adjusts the process conditions based on the inferred value, the target value, and reference information.

[0090] 6 is an activity diagram showing an example of the flow of information processing (process condition adjustment processing) executed by the information processing system 1. The information processing will be described below along with each activity in this activity diagram.

[0091] The process condition adjustment process starts with a user providing a production process model. The user provides the production process model to the information processing device 10 by uploading it from the user terminal 20 (activity A110). The information processing device 10 registers the production process model provided by the user terminal 20 (activity A120).

[0092] After registering the production process model, the user creates a linked model by specifying at least the upstream process model and the downstream process model on the user terminal 20 (activity A130). The information processing device 10 acquires the upstream process model and the downstream process model specified on the user terminal 20 from the registered production process models (activity A140).

[0093] After creating the linkage model, the user inputs initial process conditions and target values ​​into the user terminal 20 (activity A150). The information processing device 10 acquires the process conditions and target values ​​input into the user terminal 20 (activity A160). Next, the information processing device 10 inputs the acquired process conditions into the linkage model and acquires inferred values ​​(such as the characteristics of the final product) (activity A170).

[0094] After acquiring the inferred value, the information processing device 10 determines whether the termination condition for the process condition adjustment is satisfied (activity A180). If the termination condition is not satisfied (for example, if the inferred value has not reached the target value), the information processing device 10 adjusts the process condition using a predetermined algorithm or the like (activity A190). After adjusting the process condition, the information processing device 10 re-executes the acquisition of the inferred value in activity A170. On the other hand, if the termination condition is satisfied, the information processing device 10 outputs the final process condition and inferred value to the user terminal 20 (activity A200). As a result, the final process condition and inferred value are displayed on the user terminal 20 (activity A210).

[0095] 4. Function The function of this embodiment can be summarized as follows: By referring to the inferred value obtained from the linked model and adjusting the process conditions in the front-end process and / or the back-end process, the entire manufacturing line can be optimized.

[0096] Although the embodiment of the present invention has been described above, the present invention is not limited to this and can be modified as appropriate within the scope of the technical idea of ​​the invention.

[0097] 5. Others In the above embodiment, the information processing device 10 performs various storage and control operations, but multiple external devices may be used instead of the information processing device 10. In other words, various information and programs may be distributed and stored in multiple external devices using blockchain technology or the like.

[0098] The aspect of the present embodiment is not limited to the information processing system 1, and may be an information processing method or a program. The information processing method includes steps executed by the information processing system 1. The program causes a computer to execute the steps of the information processing system 1.

[0099] At least one of the devices included in the information processing system 1 may be installed outside the country in which the functions of the information processing system 1 are performed.

[0100] The information processing system 1 may be configured only by the information processing device 10. In other words, the information processing system 1 does not necessarily have to include the user terminal 20.

[0101] The information processing system 1 does not necessarily have to include the condition adjustment unit 116. For example, the information processing system 1 may be one in which a user manually searches for optimal process conditions for obtaining a predetermined inferred value (target value).

[0102] It may be provided in the following manner.

[0103] (1) An information processing system comprising a processor configured to execute each of the following steps by reading a program, wherein in a model acquisition step, an upstream process model that simulates an upstream process of producing an intermediate product from raw materials and a downstream process model that simulates a downstream process of producing a final product from the intermediate product are acquired, in a condition acquisition step, process conditions for each of the upstream process model and the downstream process model are acquired, and in an inference step, at least one of a characteristic value of the intermediate product, a characteristic value of the final product, a status value of a device used in the upstream process, and a status value of a device used in the downstream process is acquired as an inferred value by inputting the process conditions into a linked model that links at least the upstream process model and the downstream process model.

[0104] (2) In the information processing system described in (1) above, in the inference step, the characteristic values ​​of the final product, the state values ​​of the equipment used in the upstream process, and the state values ​​of the equipment used in the downstream process are obtained as the inference values ​​by inputting the process conditions into the linked model.

[0105] (3) In the information processing system described in (2) above, in the inference step, the characteristic values ​​of the intermediate product and the state values ​​of the equipment used in the upstream process are obtained by inputting the characteristic values ​​of the raw materials and the process conditions into the upstream process model, and the characteristic values ​​of the final product and the state values ​​of the equipment used in the downstream process are obtained by inputting the characteristic values ​​of the intermediate product and the process conditions into the downstream process model.

[0106] (4) In the information processing system described in any one of (1) to (3) above, the subsequent process includes an intermediate process of generating a secondary intermediate product from the intermediate product and a final process of generating the final product from the secondary intermediate product; the model acquisition step acquires the upstream process model, an intermediate process model that simulates the intermediate process, and a final process model that simulates the final process; the condition acquisition step acquires the process conditions for the upstream process model, the intermediate process model, and the final process model; and the inference step inputs the process conditions into a linked model that connects the upstream process model, the intermediate process model, and the final process model, thereby acquiring at least one of a characteristic value of the intermediate product, a characteristic value of the secondary intermediate product, a characteristic value of the final product, a state value of an apparatus used in the upstream process, a state value of an apparatus used in the intermediate process, and a state value of an apparatus used in the final process as the inferred value.

[0107] (5) In the information processing system described in any one of (1) to (4) above, the processor is configured to further perform the following steps: in the target acquisition step, a target value related to the inferred value is acquired; in the condition adjustment step, the process conditions are adjusted based on the inferred value, the target value, and reference information; and the reference information includes a correlation between the inferred value, the target value, and the process conditions.

[0108] (6) In the information processing system described in any one of (1) to (5) above, the processor is configured to further execute the following steps: in the model registration step, accepting registration of a model that simulates a process for generating a product; and in the model acquisition step, acquiring the model as the upstream process model or the downstream process model.

[0109] (7) In the information processing system described in (6) above, the model registration step accepts registration of a first model simulating a process of generating a product from a first user and accepts registration of a second model simulating a process of generating a product from a second user, and the model acquisition step acquires the first model as the previous process model and acquires the second model as the next process model.

[0110] (8) An information processing method comprising steps executed by the information processing system according to any one of (1) to (7) above.

[0111] (9) A program for causing a computer to execute each step of the information processing system described in any one of (1) to (7) above. Of course, this is not a limitation.

[0112] 1: Information processing system, 2: Communication line, 10: Information processing device, 11: Control unit, 12: Memory unit, 13: Communication unit, 14: Communication bus, 20: User terminal, 21: Control unit, 22: Memory unit, 23: Communication unit, 24: Input unit, 25: Output unit, 26: Communication bus, 111: Model registration unit, 112: Model acquisition unit, 113: Condition acquisition unit, 114: Inference unit, 115: Target acquisition unit, 116: Condition adjustment unit, 120: Artificial intelligence unit, 211: Display unit, 212: Operation acquisition unit, CV1: Characteristic value, CV2: Characteristic value, CV3: Characteristic value, PC1: Process condition, PC2: Process condition, PM1: Front-end process model, PM2: Back-end process model, SV1: State value, SV2: State value

Claims

1. An information processing system comprising a processor configured to execute the following steps by reading a program: in a model acquisition step, an upstream process model that simulates an upstream process of producing an intermediate product from raw materials and a downstream process model that simulates a downstream process of producing a final product from the intermediate product are acquired; in a condition acquisition step, process conditions for the upstream process model and the downstream process model are acquired; and in an inference step, the process conditions are input into a linked model that links at least the upstream process model and the downstream process model, thereby acquiring at least one of a characteristic value of the intermediate product, a characteristic value of the final product, a status value of a device used in the upstream process, and a status value of a device used in the downstream process as an inferred value.

2. An information processing system according to claim 1, wherein in the inference step, the characteristic values ​​of the final product, the status values ​​of the equipment used in the preceding process, and the status values ​​of the equipment used in the following process are obtained as the inferred values ​​by inputting the process conditions into the linked model.

3. An information processing system according to claim 2, wherein in the inference step, the characteristic values ​​of the intermediate product and the state values ​​of the equipment used in the upstream process are obtained by inputting the characteristic values ​​of the raw materials and the process conditions into the upstream process model, and the characteristic values ​​of the final product and the state values ​​of the equipment used in the downstream process are obtained by inputting the characteristic values ​​of the intermediate product and the process conditions into the downstream process model.

4. An information processing system according to any one of claims 1 to 3, wherein the subsequent process includes an intermediate process of generating a secondary intermediate product from the intermediate product, and a final process of generating the final product from the secondary intermediate product; the model acquisition step acquires the previous process model, an intermediate process model that simulates the intermediate process, and a final process model that simulates the final process; the condition acquisition step acquires the process conditions for each of the previous process model, the intermediate process model, and the final process model; and the inference step inputs the process conditions into the linked model that connects the previous process model, the intermediate process model, and the final process model, thereby acquiring at least one of the characteristic values ​​of the intermediate product, the characteristic values ​​of the secondary intermediate product, the characteristic values ​​of the final product, the state values ​​of equipment used in the previous process, the state values ​​of equipment used in the intermediate process, and the state values ​​of equipment used in the final process as the inferred value.

5. An information processing system according to any one of claims 1 to 4, wherein the processor is configured to further perform the following steps: in a target acquisition step, a target value related to the inferred value is acquired; and in a condition adjustment step, the process conditions are adjusted based on the inferred value, the target value, and reference information, wherein the reference information includes a correlation between the inferred value, the target value, and the process conditions.

6. An information processing system according to any one of claims 1 to 5, wherein the processor is configured to further execute the following steps: in the model registration step, accepting registration of a model that simulates a process for generating a product; and in the model acquisition step, acquiring the model as the pre-process model or the post-process model.

7. An information processing system according to claim 6, wherein the model registration step accepts registration of a first model simulating a process of generating a product from a first user and accepts registration of a second model simulating a process of generating a product from a second user, and the model acquisition step acquires the first model as the previous process model and acquires the second model as the next process model.

8. An information processing method comprising the steps executed by the information processing system according to any one of claims 1 to 7.

9. A program for causing a computer to execute each step of the information processing system according to any one of claims 1 to 7.

Images