Methods for using data for digital transformation in manufacturing

Abstract

A method for utilizing data for digital transformation in manufacturing based on the artificial intelligence framework in manufacturing according to an embodiment of the present invention, comprising: utilizing AI data obtained by preprocessing and processing plant data and factory data in an artificial intelligence framework in manufacturing implemented via the computer device, when the plant data is acquired from a first data acquisition system installed in each plant in the factory via a computing device acting as a plant server, and when the factory data is acquired from a second data acquisition system installed in an edge device in the factory via a computing device acting as a factory server, wherein the artificial intelligence framework in manufacturing is a decentralized data space framework that enables the acquisition,The management and use of plant and factory data is enabled, and in the step of utilizing the AI ​​data, at least one manufacturing AI service related to product manufacturing, a digital twin service for simulating product manufacturing, and an LCA (Life Cycle Assessment) service that quantifies and evaluates the environmental impacts of the product and the product manufacturing process line are provided.

Description

[0001] The present invention relates to a method for processing data from a device for digital transformation in manufacturing and, in particular, a method for using data for the digital transformation of manufacturing, which is a decentralized framework for artificial intelligence in manufacturing, in which various services are provided using plant data acquired from the plant, wherein the plant has become the subject of data acquisition by installing servers based on the Industrial Asset Data Model in each plant.

[0002] The data generated in the manufacturing industry is vast and varied, including factory management data, plant data, and inventory data. Furthermore, within a factory, multiple plants are interconnected to form a production line, and multiple production lines are combined to create a single factory, resulting in enormous volumes and diverse types of data.

[0003] As a core element of the fourth industrial revolution (Industry 4.0), the digital transformation in the manufacturing industry has become a hot topic, and there are key elements such as the Industrial Asset Data Model, the assetization and trading of data, artificial intelligence in manufacturing, and digital twins.

[0004] For digital transformation in manufacturing, data generated in the manufacturing industry must be captured and managed based on the industrial asset data model. Furthermore, technology is needed to utilize the captured manufacturing data, artificial intelligence, and digital twins, and to enable data exchange transactions with other companies or factories.

[0005] As a state-of-the-art solution for implementing this digital transformation in manufacturing, Basyx, an existing system developed by the Eclipse Global Working Group, consists of servers and systems that manage factory and plant data based on the Asset Administration Shell (ASS) and are characterized by its compatibility with multiple communication protocols such as OPC-UA, HTTP / REST and MQTT, enabling it to exchange data with other companies or factories when using the EDC (Eclipse Dataspace connector).

[0006] However, Basyx is problematic because if a problem occurs on the server, the service related to the data collection system for all facilities fails and the service cannot be made available to the user, as there is only one central server for data storage.

[0007] Furthermore, Basyx is not compatible with systems that use an industrial asset data model other than ASS, and does not include AI services and data acquisition systems for manufacturing, so the problem is that a separate third-party system must be connected. [Literature on the state of the art][Patent document] (Patent document 1) Korean Public Patent Gazette No. 10-2023-0099904 (published on July 5, 2023) (Patent document 2) Korean Public Patent Gazette No. 10-2024-0014757 (published on February 2, 2024) [Purpose of the invention]

[0008] Therefore, the present invention was created to solve the aforementioned problems, and the object of the present invention is to provide a method for utilizing data for digital transformation in manufacturing, which is a decentralized framework for artificial intelligence in manufacturing that enables each plant to capture and manage structured and unstructured data generated during product manufacturing according to the manufacturing process. For this purpose, a server based on a standard industrial asset data model is installed in each plant and can provide various services using plant data collected from each plant.

[0009] Furthermore, the object of the present invention is to provide a method for utilizing data for digital transformation in manufacturing, which is a decentralized framework for artificial intelligence in manufacturing, wherein structured data with a variety of industrial asset data models such as AAS, SM profile or the like can be converted via an asset core for use in an artificial intelligence framework in manufacturing, so that AI services in manufacturing can be provided in connection with the manufacture of products, a digital twin service for simulating product production and at least one of the LCA (Life Cycle Assessment) services, which quantifies and evaluates the environmental impacts of process lines for product production.

[0010] However, the technical problem to be solved by the present invention is not limited to the technical problem mentioned above, and other unmentioned technical problems are clearly understandable to the person skilled in the art from the following description. [Solution to the task]

[0011] The method for data utilization for digital transformation according to an embodiment of the present invention as a technical method for solving the above-mentioned problem is a method for utilizing data for digital transformation in manufacturing based on an artificial intelligence framework for manufacturing, comprising the following: utilizing AI data that has been pre-processed and processed from plant and factory data by the artificial intelligence framework for manufacturing implemented in the computing device, when plant data is acquired from a first data acquisition system installed in each plant in the factory via a computing device that acts as a plant server, and when factory data is acquired from a second data acquisition system installed in an edge device in the factory via a computing device that acts as a factory server.where the framework for artificial intelligence in manufacturing is a decentralized data space framework that enables the collection, management, and use of plant and factory data, with the step of using AI data providing at least one manufacturing AI service related to product manufacturing, a digital twin service for simulating product manufacturing, and a life cycle assessment (LCA) service that quantifies and evaluates the environmental impacts of the product and the product manufacturing process line.

[0012] Furthermore, the plant and factory data can be transformed into a format corresponding to the industrial asset data model using an asset core tool and then fed into the manufacturing AI framework. This framework can then generate AI data by preprocessing and processing the plant and factory data to provide the manufacturing AI service, the digital twin, and the life cycle assessment (LCA) service. Based on the AI ​​data within the manufacturing AI framework, the manufacturing AI service can provide the following services to the plant.These services include: the intelligent visual product inspection service for detecting and classifying defects in products, the data-based quality prediction service for predicting and preventing defects in products in advance, the data-driven process optimization service for modifying or maintaining the operation of the plant, and the video-based industrial security service for monitoring the working environment, including the product manufacturing process line.

[0013] Furthermore, the plant server can store the output results of the manufacturing AI service, provided by the plant's manufacturing artificial intelligence framework, as an Industrial Asset Data Model file.

[0014] The digital twin service can be a service that creates a digital twin model that performs a simulation of product manufacturing based on AI data in an artificial intelligence framework in the factory, with the artificial intelligence framework in manufacturing being installed on the computing device that acts as a factory server in the factory, which is integrated into the factory process along with the equipment and the edge device.

[0015] Furthermore, the artificial intelligence framework for factory manufacturing can perform an update of the digital twin model by preprocessing and processing the plant and factory data received via the industrial asset data model and the asset management tool with each update of the plant and factory data.

[0016] The factory server can store the output results of the digital twin service, provided by the factory's artificial intelligence manufacturing framework, as a file of an industrial asset data model.

[0017] The LCA service can be a service provided by an artificial intelligence framework in the factory, with the artificial intelligence framework in the factory being installed on the computing device that acts as a factory server in the factory, which is involved in the factory process along with the equipment and edge device.

[0018] In addition, the factory server can store the output results of the LCA service, provided by the factory's artificial intelligence manufacturing framework, as an industrial asset data model file.

[0019] Furthermore, the plant can implement decentralization by setting up the plant server via the computing device and making it the entity that collects and manages the plant data. Effect of the invention

[0020] The present invention can serve as a basis for the free exchange of data in the safety-sensitive manufacturing industry by providing a decentralized framework that transfers the acquisition and management of data to each plant.

[0021] Furthermore, the present invention can implement digital transformation in manufacturing by providing an end-to-end framework that considers data acquisition and management as well as the use of the acquired data in manufacturing AI services, digital twin services and LCA services.

[0022] However, the effects achievable with the invention are not limited to those mentioned above, and other effects not mentioned are clearly understandable to the person skilled in the art based on the following description. The following are shown: Fig. 1 a block diagram of an application structure of a framework for artificial intelligence in manufacturing according to an embodiment of the invention; Fig. 2 a conceptual diagram of a framework for artificial intelligence in manufacturing according to an embodiment of the invention; Fig. 3 a block diagram showing the structure of an internal data space connector according to an embodiment of the invention; Fig. 4. A presentation of an example of a framework for artificial intelligence in manufacturing based on the Asset Administration Shell (AAS). Fig. 5. A flowchart of the process of the implementation procedure for digital transformation; Fig. 6 a flowchart of the detailed, in Fig. 5 steps shown for processing plant and factory data; Fig. 7 a flowchart of the detailed, in Fig. 5 steps shown for capturing plant and factory data; Fig. 8 is a flowchart of the steps for exchanging data within the system. Fig. 5 frameworks for artificial intelligence in manufacturing shown. [Specific details regarding the implementation of the invention]

[0023] In the following, embodiments of the present invention are described in detail with reference to the accompanying drawings to facilitate their application by those skilled in the art. However, the description of the invention is only an example of a structural or functional description, so the scope of the invention should not be interpreted as being limited by the embodiments described herein. That is to say, since the embodiments are subject to various modifications and can take different forms, the scope of the invention should be understood as including equivalents that can realize the technical idea. Furthermore, the purposes or effects set forth in the invention do not imply that a particular embodiment must include all or only such effects, and therefore the scope of the present invention should not be understood as being limited by this.

[0024] The meaning of the terms described in the present invention should be understood as follows.

[0025] The terms "first," "second," and the like are used to distinguish one component from another, and the scope of rights should not be limited by these terms. For example, a first component may be referred to as a second component, and similarly, a second component may be referred to as a first component. When a component is said to be "connected" to another component, this means that it may be directly connected to the other component, but there may also be other components in between. Conversely, when it is said that a component is "directly connected to another component," it should be assumed that there are no other components in between.However, other expressions describing the relationship between the components, such as "between" and "directly between" or "adjacent to" and "directly adjacent to", should be interpreted in the same way.

[0026] Singular expressions are to be understood as including the plural unless the context clearly indicates otherwise, and terms such as "includes" or "has" are intended to denote the presence of the listed features, numbers, steps, processes, components, parts or combinations thereof, and not to exclude the possibility of the presence or addition of one or more other features, numbers, steps, processes, components, parts or combinations thereof.

[0027] All terms used herein have the same meaning as they are commonly understood by a person skilled in the art in the field to which the present invention relates, unless otherwise defined. Terms defined in commonly used dictionaries should be interpreted as corresponding to the meaning they have in the context of the associated technology and may not be interpreted as having an ideal or overly formal meaning unless expressly defined in this invention. Framework for AI in manufacturing

[0028] The configuration of a preferred embodiment is described in detail below with reference to the accompanying drawings.

[0029] Fig. Figure 1 is a block diagram illustrating an application structure of an artificial intelligence framework for manufacturing according to an embodiment of the invention. Fig. Figure 2 is a conceptual diagram of a framework for artificial intelligence in manufacturing according to an embodiment of the invention, and Fig. Figure 3 is a block diagram showing the structure of an internal data space connector according to an embodiment of the invention. 1) Data acquisition and management within the framework of AI in manufacturing

[0030] In the factory process 100 in the manufacturing industry, several plants 110 for the production of products are linked together to form a product manufacturing process line, and the several product manufacturing process lines form a factory 130.

[0031] The invention includes a plurality of systems 110 in Fig. 1 is shown as the first to third system 110-1 to 110-3, but the number is not limited.

[0032] Furthermore, factory processes 100 must implement a digital transformation in manufacturing to capture and manage the large quantities and types of data that are generated (or created) before and during the production of products in the factory 130.

[0033] To implement digital transformation in manufacturing, an industrial asset data model 20 is used for data collection and management, and it is known as a conventional industrial asset data model 20, such as Asset Administration Shell in Germany and SM Profile in the USA.

[0034] Furthermore, Basyx represents a state-of-the-art solution for implementing digital transformation in manufacturing. Developed by the Eclipse Global Working Group, it is an existing system of servers and systems that structure and manage factory and plant data based on the Asset Administration Shell (ASS). Its key feature is its compatibility with multiple communication protocols such as OPC UA, HTTP / REST, and MQTT, enabling data exchange with other companies or factories when using the EDC (Eclipse Dataspace connector).

[0035] However, Basyx has the problem that there is only one central server for data storage, so if the server malfunctions, the services connected to the data collection system fail in all systems and no services can be provided to users.

[0036] Furthermore, Basyx is not compatible with systems that use an industrial asset data model other than the Asset Administration Shell (ASS), and does not include AI services in manufacturing or data acquisition systems, so the problem is that a separate third-party system must be connected.

[0037] The framework for artificial intelligence in manufacturing 10 according to an embodiment of the invention can be a decentralized data space framework that enables the acquisition, management, exchange and use of plant and factory data generated in a factory 130 to improve existing problems in implementing digital transformation in manufacturing.

[0038] Furthermore, the Artificial Intelligence in Manufacturing Framework 10 of the invention can transform data structured with various types of industrial asset data models, such as AAS and SM profiles, so that it can be used with the Asset Management Tool (Asset Core) in the Artificial Intelligence in Manufacturing Framework 10, thus enabling it to be a decentralized data space framework that provides at least one of the following services: a manufacturing AI service related to the production of products, a digital twin service that simulates the production of products, and a life cycle assessment (LCA) service that quantifies and evaluates the environmental impacts of products and product manufacturing process lines.

[0039] With reference to Fig. 1. A computing device can be installed in each plant 110, which acts as a plant server to implement the framework for artificial intelligence in manufacturing 10 for manufacturing according to an embodiment of the invention.

[0040] The object that installs the computing device on the Edge device 120 can be a user of the Artificial Intelligence in Manufacturing Framework 10.

[0041] In addition, a file of an Industrial Asset Data Model 20 generated on the basis of factory data and a second management system 11b for managing the file can be installed on the computing device of the Edge device 120.

[0042] Then a computing device of each plant 110 can be connected to the first management system 11a, so that a first data acquisition system 12a, which acquires plant data from each plant 110, can be installed together with the first management system 11a.

[0043] In the present invention, the plant data are understood to be the conditions and parameters required for the manufacture of a product, wherein the setpoints of the plant 110 may include speed, temperature, pressure and time of the device (machine) for the manufacture of the product.

[0044] The first data acquisition system 12a of the invention can, in addition to the plant data, also acquire plant status data from each plant 110.

[0045] Within the scope of the present invention, the term "plant condition data" refers to the condition values ​​of the plants 110, i.e., data relating to the condition of each plant 110, wherein the condition values ​​include operating condition information, performance information, fault information, maintenance information, and the like of the plant 110.

[0046] As described above, the framework for artificial intelligence in manufacturing 10 of the present invention is the subject of which each plant 110 collects and manages the data (plant data) by setting up plant servers via a computing device in each plant 110, so that the data acquisition and management does not take place on a single server and thus decentralization can be achieved.

[0047] The Edge device 120, installed in factory 130, can be a device (e.g., a sensor, a camera, etc.) for collecting and processing factory data generated (or created) in factory 130 in an edge computing environment.

[0048] The edge device 120 of the present invention is integrated into the factory process 100, and a computing device serving as a factory server can be installed to implement the framework for artificial intelligence in manufacturing 10 according to an embodiment of the present invention.

[0049] The person installing the computing device on the Edge Device 120 can be a user of the Artificial Intelligence in Manufacturing Framework 10.

[0050] In addition, a file of an Industrial Asset Data Model 20 based on factory data and a second management system 11b for its management can be installed on the computing device of the Edge device 120.

[0051] The computing unit of the Edge device 120 is connected to the second management system 11b, so that a second data acquisition system 12b, which collects factory data, can be installed together with the second management system 11b.

[0052] The factory data to be recorded by the second data acquisition system 12b refers to data that can be recorded inside and outside the factory 130, with the exception of the plant condition data, which is recorded by the data acquisition system 12a in each plant 110.

[0053] As a concrete example, factory data can include product quality data, product information data, environmental data, factory operating data, carbon dioxide emission data, company data and the like, which are collected by the second data acquisition system 12b, as well as all structured and unstructured data that will be generated during the product manufacturing process.

[0054] In the present invention, product quality data is information relating to the quality of products manufactured by the product manufacturing process line of factory 130 and may include the results of visual product inspection, the defect rate, the quality level determined on the basis of the defect rate, the product quality control plan, inspection records, quality indicators, quality assurance data and the like.

[0055] Furthermore, the product information data consists of information about products manufactured in the product manufacturing process line of Factory 130, and may include size, weight, material, power requirements, 2D or 3D model, product drawing, product yield, production time, production costs according to the phases of the product manufacturing process and the product manufacturing process itself, data about the parts that make up the product (name, manufacturer, product number, quantity of the part consumed, etc.), as well as information about installation, maintenance and repair history, the product's repair warranty, and the like.

[0056] The environmental data refers to information for monitoring the environmental impacts generated during product manufacturing by the production process lines in Factory 130. This may include information on air pollution regarding emissions and concentrations of air pollutants generated by the production process lines; information on water pollution regarding the properties and pollutant emissions of wastewater generated by the production process lines; information on soil pollution regarding soil contamination by waste and pollutants generated by the production process lines; and information on noise and vibration regarding noise and vibration generated by the production process line.This includes information on energy consumption regarding the quantity and type of energy consumed in the product manufacturing process line, information on waste management regarding the quantity and type of waste generated in the product manufacturing process line, and the like.

[0057] The factory's operational data refers to information relating to product manufacturing and the operational process of the product manufacturing process line. This may include information on production volume and efficiency, including the quantity and production trend of manufactured products, the operating rate and operating time of the product manufacturing process line, indicators of production efficiency (e.g., Overall Equipment Effectiveness, OEE) for each product manufacturing process, information on the operating status of plant 110 (in operation, out of operation, or stopped), the malfunction and repair history of plant 110, and condition monitoring of plant 110 (e.g., temperature, pressure, vibration, etc.).), quality information about product quality, information about the product manufacturing plan and planning as well as the manufacturing plan and planning in relation to the actual manufacturing results compared to the manufacturing plan, information about energy consumption in relation to the energy consumed in the factory 130 (e.g. electricity, gas, steam etc.) as well as energy efficiency indicators, information about the product manufacturing process in relation to the progress of each product manufacturing process, parameters of the product manufacturing process as well as events and alarms of the product manufacturing process and the like.

[0058] Carbon dioxide emission data can include correlation information for analyzing the correlation between the daily or monthly emissions of carbon dioxide (CO2), a greenhouse gas generated in Factory 130 during the production of the product in the product manufacturing process line, the emissions after the product manufacturing process, the emissions per unit of product production, and the energy consumption of Factory 130 and the CO2 emissions.

[0059] Furthermore, the company data includes information about the company operating Factory 130, and may include production information about the company's product manufacturing volume, inventory information about the stock levels and inventory turnover of raw materials and parts for the products and the manufacturing of these products, sales information (e.g., sales volume, revenue, price of products, etc.), customer information about the customers who have purchased the products, financial information, employee information (e.g., personnel, wages, hours worked, etc.), cost information about the manufacturing costs of a product (e.g., material costs, labor costs, transportation costs, etc.), information about the management of facilities and equipment in Plant 110 and Factory 130 (e.g., maintenance history, equipment utilization, energy consumption, etc.), and the like.

[0060] Factory 130 is included in factory process 100, and to implement the framework for artificial intelligence in manufacturing 10 according to an embodiment of the present invention, a computing device acting as a factory server can be installed in a device (e.g. a computer or computing equipment) to monitor a product manufacturing process line and provide the digital twin service and the LCA service of the present invention.

[0061] Then the object that installs the computing device in a fixture of Factory 130 is a user of the Artificial Intelligence Framework in Manufacturing 10. The following describes the installation of a computing device in Factory 130.

[0062] According to Fig. 2. The data acquisition systems 12a and 12b are connected to the factory process 100 via a communication protocol, so that plant and factory data can be acquired from the factory process 100.

[0063] In this context, data acquisition systems 12a and 12b do not restrict the communication protocol to the connection with factory process 100, but the communication protocol for acquiring plant and factory data can be at least one of OPC UA, HTTP / REST, MQTT, Modbus RTU, or Modbus TCP. The communication protocols mentioned above are common technologies, so for the sake of simplicity, a detailed description is omitted.

[0064] Furthermore, the plant and factory data collected by the data acquisition systems 12a and 12b can be stored on the plant server or factory server after they have undergone the sorting and structuring process by the Industrial Asset Data Model 20, as described in Fig. 2 shown.

[0065] The Industrial Asset Data Model 20 of the present invention is not limited, but in one embodiment it may include an AAS (Asset Administration Shell), an SM profile and the like, as shown in Fig. 2 shown.

[0066] The framework for artificial intelligence in manufacturing 10 of the present invention can implement decentralization by separately creating servers, such as device servers and factory servers, that manage the files of the Industrial Asset Data Model 20. 2) Use of data within an artificial intelligence framework in manufacturing

[0067] In the case of data acquired by the data acquisition system 12a, 12b of the present invention (plant and factory data), the form of the plant data and the factory data may differ depending on the type of industrial asset data model 20, and the factory process 100 and the type of factory process 100 and the industrial asset data model 20 adopted by the company may differ.

[0068] The Asset Core Tool 30 can transform the structure of any Industrial Asset Data Model 20 registered in the Artificial Intelligence in Manufacturing Framework 10 of the present invention into a structure that can be used in the Artificial Intelligence in Manufacturing Framework 10.

[0069] Furthermore, the Asset Core Tool 30 can transform the form of the plant and factory data captured by the data acquisition systems 12a, 12b so that it corresponds to the structurally transformed industrial asset data model 20, and then input the transformed plant and factory data into the framework for artificial intelligence in manufacturing 10, as described in Fig. 2 shown.

[0070] In this case, the Asset Core Tool 30 transforms the form of the plant and factory data acquired by the data acquisition systems 12a, 12b so that the Manufacturing Artificial Intelligence Framework 10 can provide the invention's manufacturing AI services based on the plant and factory data, thereby ensuring compatibility.

[0071] When converted plant and factory data is entered via the asset management tool 30, the artificial intelligence framework for manufacturing 10 of the present invention can preprocess and process the plant and factory data to provide the manufacturing AI service of the present invention; in particular, the artificial intelligence framework for manufacturing 10 can provide a manufacturing AI service to plant 110.

[0072] The artificial intelligence framework for manufacturing 10 of the present invention can preprocess and process the plant and factory data converted by the asset core tool 30 in order to provide the artificial intelligence services for manufacturing of the present invention. More specifically, the manufacturing AI service can be provided in the manufacturing AI framework 10a of the devices 110.

[0073] Then the manufacturing AI service of the present invention is not limited to the type if it can provide the user with information relating to the manufacture of the product, but in one embodiment the manufacturing AI service of the present invention, which is made available to the user by the framework for artificial intelligence in manufacturing 10a, may include an intelligent visual product inspection service, a data-based quality prediction service, a data-based process optimization service and a video-based industrial security service.

[0074] In this invention, an intelligent visual product inspection service can be a service that checks the shape, color, size and defects of a product by identifying and analyzing the visual features of a product being manufactured in a product manufacturing process line using cameras and sensors installed in a plant 110, and which can also detect and classify defects in a product based on artificial intelligence technologies such as machine learning and deep learning.

[0075] Furthermore, the data-based quality prediction service can be a service for predicting the quality of a product to be manufactured in a product manufacturing process line, based on plant and factory data collected by the data acquisition systems 12a, 12b, and it can be used to predict and prevent potential problems or product defects that may occur in the product manufacturing process line.

[0076] Furthermore, video-based industrial security services can monitor the working environment, including the product manufacturing line, using cameras and video surveillance systems installed in facilities, etc., to prevent safety accidents in advance.

[0077] Such manufacturing services can be provided in the same way in factory process 100, but the manufacturing services provided for each plant 110 of the present invention can be different. This is because each plant 110 plays a different role in product manufacturing.

[0078] The output results of the manufacturing AI service provided by the Framework for Artificial Intelligence in Manufacturing 10 can be defined as an industrial asset data model 20 and stored on the plant's server.

[0079] In this case, the fact that the output results of the manufacturing AI service are defined as an Industrial Asset Data Model 20 means that they are stored as an Industrial Asset Data Model 20 file.

[0080] The artificial intelligence framework for manufacturing 10 of the present invention can preprocess and process plant and factory data to provide not only manufacturing AI services but also digital twin services, and in particular, as described in Fig. 2 shown, providing digital twin services in the framework for artificial intelligence in manufacturing 10c of factory 130.

[0081] A digital twin service is a service provided for real-time monitoring and analysis of a product or factory process 100, wherein the digital twin service of the present invention can generate a digital twin model in which the framework for artificial intelligence in manufacturing 10 performs a simulation of product manufacturing based on pre-processed and processed plant and factory data.

[0082] In addition, the framework for artificial intelligence in manufacturing can perform 10 updates to the digital twin model by preprocessing and processing the plant and factory data received via the Industrial Asset Data Model 20 and the Asset Core Tool 30 with each update of the plant and factory data.

[0083] This digital twin model can predict the product production volume, product production speed, etc., as well as the lifespan and energy consumption of plant 110 by simulating the product production process in various scenarios, and each plant 110 can be optimized by controlling its operation to improve the product production volume based on the simulation of the product production process of the digital twin model.

[0084] Thus, the output results (or simulation results) of the digital twin service provided by the framework for artificial intelligence in manufacturing 10 can be defined as an industrial asset data model 20 and stored on a factory server.

[0085] Then the fact that the output results of the digital twin service are defined as an Industrial Asset Data Model 20 means that they are stored as a file of the Industrial Asset Data Model 20.

[0086] To provide the user with at least one of the manufacturing AI service, digital twin service and LCA service, the AI ​​server can be stored on the plant server or the factory server, with the plant and factory data pre-processed by the Manufacturing Artificial Intelligence Framework 10 representing the AI ​​data.

[0087] The AI ​​data, together with the output results of at least one of the manufacturing AI service, the digital twin service and the LCA service provided by the framework for artificial intelligence in manufacturing 10, can be defined as an industrial asset data model 20 and stored on the plant server or the factory server, respectively. 3) Exchange of data within an artificial intelligence framework in manufacturing

[0088] The data (plant data, factory data) acquired by the data acquisition systems 12a, 12b of the present invention, as well as the AI ​​data, can be in the form of plant servers that own and manage the respective plant server 110.

[0089] With reference to Fig. 1. In the internal data space, a data exchange process, including the sharing of data and apps, can take place between the individual plant servers and between the plant server and the factory server.

[0090] The term “internal data room” means a platform or system for sharing and exchanging data within an organization or company and is preferably to be understood as a system for exchanging plant data, factory data and / or AI data in one embodiment.

[0091] Furthermore, the data exchange process can be carried out using the Asset Registry Server 40 and the internal data room connector 50, which are configured in the Artificial Intelligence in Manufacturing Framework 10.

[0092] The internal data room connector 50 of the present invention can comprise an authentication unit and a data provision unit.

[0093] Then the authentication unit of the internal data room connector 50 can authenticate another internal data room connector 50 that has requested at least one of the plant data, factory data and AI data, using the authentication information stored in the identity hub 41, which will be described later.

[0094] Furthermore, the data provision unit of the internal data room connector 50 can transfer at least one of the plant data, factory data and AI data to the internal data room connector 50, which has been certified by the authentication unit.

[0095] The asset registry server 40 of the present invention comprises an identity hub 41, an edge device 120 and a factory 130, which form the factory process 100, as well as an identity hub 41 that stores each data exchange process and the authentication information, and can provide a catalog service 42 that provides brief information about asset data, factory data and AI data.

[0096] The Identity Hub 41 is a system that allows users to use the same ID and password for multiple services.

[0097] According to Fig. 1. The plant server and the factory server of the present invention are equipped with an internal data room connector 50, and as described in Fig. As shown in Figure 3, authentication and exchange of plant and factory data can be carried out through the interaction between the Asset Registry Server 40 and the internal data room connector 50.

[0098] According to Fig. 3. The method for authentication and exchange of plant and factory data of the present invention can be carried out between a data consumer A, who needs data, and a data provider B, who provides data to the data consumer A.

[0099] It is desirable that the data consumer A is understood to be at least one of the device 110, edge device 120 or factory 130 that requires plant data or factory data.

[0100] Furthermore, it is desirable that the data provider B includes at least one of the factory process 100, the other plant 110, the edge device 120 and the factory 130 of the data consumer A, in order to provide the plant data or the factory data to the data consumer A.

[0101] The internal data room connector 50 provided in data consumer A can determine the storage location of the data required by data consumer A by searching the catalog service 42.

[0102] Furthermore, the internal data room connector 50 provided in data consumer A can request the transfer of data to the internal data room connector 50 of data provider B, which is in possession of the data required by data consumer A.

[0103] The internal data room connector 50 provided in data provider B can authenticate the internal data room connector 50 of data consumer A via the identity hub 41 41 and then authorize the transfer of the data.

[0104] In this way, the data requested by data consumer A (or the data exchange) can be transmitted and received (or exchanged) via the data provider of the internal data room connector 50, provided the data transmission is approved. Examples of the manufacturing AI framework

[0105] The following describes in detail the AAS-based framework for artificial intelligence in manufacturing 10 of the present invention.

[0106] Fig. Figure 4 shows an example of a framework for artificial intelligence in manufacturing based on the Asset Administration Shell (AAS).

[0107] With reference to Fig. 4. The AASX Explorer 200 defines and modifies the structure of the AAS, and the AAS Server 300 can structure and manage the plant and factory data based on the AAS defined or modified by the AASX Explorer 200.

[0108] The AASX Explorer 200 can be part of the AASX (Autonomous Asset and Service Exchange) platform, which offers solutions for asset management.

[0109] The data acquisition systems 12a, 12b of the present invention acquire plant data from the plant 110, and the framework for artificial intelligence in manufacturing 10 of the present invention modifies the plant data acquired by the data acquisition systems 12a, 12b, adapts it to the AAS structure and transfers it to the AAS server 300, so that the plant data of the AAS structure is stored in the AAS server 300.

[0110] Furthermore, each plant 110 of the present invention can become the subject of plant data management and use through the interaction of the internal data space connector 50 with the framework for artificial intelligence in manufacturing 10 and the plant register server 40, which can enable the transmission and reception of plant data.

[0111] The fact that each Plant 110 becomes the subject of the management and use of plant data means that the subject of the assetization and transaction of plant data is also the owner of Plant 110. Methods for data acquisition and management for digital transformation in manufacturing

[0112] The following section describes in detail, with reference to the drawings, the process of implementing a method for digital transformation in manufacturing S100 based on the framework for artificial intelligence in manufacturing 10 of the present invention.

[0113] Fig. Figure 5 is a flowchart illustrating a process of the implementation procedure for digital transformation. Fig. 6 is a flowchart that shows a detailed process of the in Fig. The 5 steps shown illustrate how to process plant and factory data. Fig. 7 is a flowchart that shows a detailed process of the in Fig. The 5 steps shown for processing plant and factory data are illustrated, and Fig. 8 is a flowchart that shows a detailed process of the steps for exchanging data within the system. Fig. The 5 manufacturing frameworks shown illustrate this.

[0114] In the Fig. 5. The method S100 for implementing digital transformation in manufacturing according to an embodiment of the present invention can be carried out in the sequence of a step for acquiring data S110, a step for using data S120 and a step for exchanging data S130.

[0115] The data utilization step S120 is a process of the method for utilizing data for digital transformation in manufacturing of the present invention. It can be performed after the completion of the data acquisition step S110 and is a step for converting, preprocessing, and processing the plant and factory data acquired in the data acquisition step S110 so that it can be used in the artificial intelligence in manufacturing framework 10 of the present invention. The detailed process for converting, preprocessing, and processing plant and factory data is described in Fig. 6 shown.

[0116] As in Fig. As shown in Figure 6, the Asset Core Tool 30 can transform the structure of all types of industrial asset data models 20 registered in the framework for artificial intelligence in manufacturing 10 so that they are usable for the framework for artificial intelligence in manufacturing 10, and transform the form of the plant and factory data captured by the data acquisition systems 12a, 12b so that it corresponds to the industrial asset data model 20 (S121).

[0117] After that, the Asset Core Tool 30 can input the converted plant and factory data into the Artificial Intelligence in Manufacturing Framework 10 (S122).

[0118] Subsequently, the artificial intelligence framework for manufacturing 10 of the present invention can preprocess and process the converted plant and factory data via the asset core tool 30 (S123) to provide at least one of the manufacturing AI service, the digital twin service and the LCA service.

[0119] Subsequently, the framework for artificial intelligence in manufacturing 10 can provide at least one of the manufacturing AI service, the digital twin service and the LCA service (S124), and the output results of at least one of the manufacturing AI service, the digital twin service and the LCA service provided to the user by the framework for artificial intelligence in manufacturing 10 can be defined as the Industrial Asset Data Model 20 and stored on the plant server (S125).

[0120] After the aforementioned storage step S125, a data exchange step S130 can be performed.

[0121] The data acquisition step S110 precedes the data utilization step S120. Simultaneously, plant data from multiple plants 110, comprising the factory process 100, and factory data from the edge device 120 are acquired. The detailed sequence for plant and factory data acquisition is described in Fig. 7 shown.

[0122] In Fig. 7 Each plant 110 can be equipped with a computing device acting as a plant server to implement the artificial intelligence manufacturing framework 10 of the present invention, and a first management system 11a and a first data acquisition system 12a can be installed via this computing device (S111).

[0123] Before or after the above installation step (S111), Edge device 120 can be equipped with a computing device acting as a factory server to implement the artificial intelligence in manufacturing framework 10 of the present invention, and a second management system 11b and a second data acquisition system 12b can be installed via this computing device (S112).

[0124] Subsequently, the data acquisition systems 12a, 12b can acquire plant and factory data (S113).

[0125] Afterwards, the plant and factory data recorded by the data acquisition systems 12a and 12b can be organized (or sorted) and structured using the Industrial Asset Data Model 20 and then stored on the plant server or the factory server (S114).

[0126] During the operation of a product manufacturing process line for the production of a product, equipment 110 may be added or additional sensors may be attached to the equipment 110, which may result in the addition of equipment data items to be captured by the first data acquisition system 12a (S115).

[0127] For example, if plant 110 is added to factory 130, or additional sensors are attached to plant 110, thereby adding plant data to be captured by the first data acquisition system 12a (S115-Yes), the Industrial Asset Data Model file 20 can be automatically modified by the user or the Manufacturing Artificial Intelligence Framework 10, and the first data acquisition system 12a can capture plant data that is added to an item in connection with the modification of the Industrial Asset Data Model file 20 (S116).

[0128] If the plant data is modified in the modification and acquisition step S116 above, the storage step S114 above must be executed again to save the additionally acquired plant data on the plant server.

[0129] If, on the other hand, plant 110 is retained in factory 130 or no additional sensors are attached to plant 110 (S115-No), the step to using data S120 can be continued.

[0130] The data exchange step S130 is a step in which data is exchanged (sent and received) through the interaction between the asset registry server 40 and the internal data room connector 50, the details of which are described in Fig. 8 are shown.

[0131] Furthermore, the detailed procedure of the above-mentioned step for exchanging data S130 is described below using the authentication data exchange procedure of the first Annex 110-1 and the second Annex 110-2, which constitute the majority of Annex 110.

[0132] As in Fig. As shown in Figure 8, the internal data room connector 50, which is present in the first plant 110-1 and in the second plant 110-2, can register the plant data in the identity hub 41 of the asset registry server 40 (S131).

[0133] Subsequently, the internal data room connector 50 of the first asset 110-1 can search the catalog service 42 provided by the asset registry server 40 to determine the location of the asset data required by the first asset 110-1 (S132).

[0134] Afterwards, the internal data room connector 50 of the first system 110-1 can request the transfer of the system data required by the internal data room connector 50 of the second system 110-2 (S133).

[0135] Subsequently, the internal data room connector 50 of the second system 110-2 can authenticate and authorize the internal data room connector 50 of the first system 110-1 via the identity hub 41 (S134).

[0136] Afterwards, the first system 110-1 and the second system 110-2 can each send and receive (or exchange) the system data required by the first system 110-1 via the provided internal data space connector 50 (S135). Effects of the invention

[0137] The artificial intelligence framework for manufacturing 10 of the present invention can serve as a basis for the free exchange of data in the safety-sensitive manufacturing industry by providing a decentralized framework that designates each plant as the subject of data acquisition and management.

[0138] Furthermore, the artificial intelligence in manufacturing framework 10 of the present invention can implement the digital transformation in manufacturing by providing an end-to-end framework that considers data acquisition and management as well as the use of the acquired data in manufacturing AI services, digital twin services and LCA services.

[0139] As described above, a detailed description of preferred embodiments of the present invention is provided to facilitate its implementation and execution. Although the foregoing has been described with reference to preferred embodiments of the present invention, a person skilled in the art will understand that the present invention can be modified and altered in various ways within its scope without departing from the scope of the present invention. For example, a person skilled in the art can use any configuration described in the embodiments described above by combining them. Accordingly, the present invention is not intended to be limited to the embodiments shown herein, but is intended to be granted the broadest possible scope consistent with the principles and novel features disclosed herein.

[0140] The present invention may be implemented in other specific forms, provided they do not deviate from the technical ideas and essential features of the present invention. Therefore, the detailed description above is not to be understood as limiting in every respect, but serves only for illustration. The scope of the invention is to be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalents of the invention are included in the scope of the invention. The invention is not intended to be limited to the embodiments shown herein, but rather to have the broadest possible scope consistent with the principles and novel features disclosed herein. Furthermore, claims not expressly included in the claims may be combined into an embodiment or included as new claims by amendment after filing. Reference symbol list 10 Frameworks for Artificial Intelligence in Manufacturing 11 administrative systems, 12 data acquisition systems, 20 Industrial Asset Data Model 30 Asset Core Tools 40 Asset Registry Servers 41 Identity Hub 42 Catalogue service 50 data room connectors 51 first connector 52 Second connector 100 factory process 110 Plant 120 Edge device 130 factory 200 AASX Explore, 300 ASS servers, A data consumer B Data providers QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] KR 10-2023-0099904

[0007] KR 10-2024-0014757

[0007]

Claims

[1] Method for using data for digital transformation in manufacturing based on the framework for artificial intelligence in manufacturing, wherein the method comprises the following steps: - Utilizing AI data obtained through preprocessing and processing of plant and factory data in an artificial intelligence framework for manufacturing, implemented via the computer device, when the plant data is collected from a first data acquisition system installed in each plant in the factory via a computing device acting as a plant server, and when the factory data is collected from a second data acquisition system installed in an edge device in the factory via a computing device acting as a factory server, characterized by , that The framework for artificial intelligence in manufacturing is a decentralized data space framework that enables the collection, management, and use of plant and factory data, and In the step of using the AI ​​data, at least one manufacturing AI service related to product manufacturing, a digital twin service for simulating product manufacturing, and an LCA (Life Cycle Assessment) service that quantifies and evaluates the environmental impacts of the product and the product manufacturing process line using the AI ​​data is provided. [2] Method according to claim 1, characterized by , that The plant data and the factory data are converted into their form according to the industrial asset data model by an asset core tool and then entered into the framework for artificial intelligence in manufacturing. where the framework for artificial intelligence in manufacturing generates the AI ​​data by preprocessing and processing the plant data and the factory data in order to provide it to the manufacturing AI service, the digital twin and the LCA service. [3] Method according to claim 2, characterized by , that the manufacturing AI service provides the following services based on AI data within the framework of artificial intelligence in the manufacturing of the plant, including: the intelligent visual product inspection service for detecting and classifying defects in products, the data-driven quality prediction service for predicting and preventing defects in products in advance, the data-driven process optimization service for modifying or maintaining the operation of the plant and the video-based industrial security service for monitoring the work environment, including the product manufacturing process line. [4] Method according to claim 3, characterized by that the plant server stores the output results of the manufacturing AI service, provided by the plant's manufacturing artificial intelligence framework, as an Industrial Asset Data Model file. [5] Method according to claim 2, characterized by , that the digital twin service is a service that creates a digital twin model that performs a simulation of product manufacturing based on AI data in an artificial intelligence framework in the factory manufacturing environment, with the artificial intelligence framework in the factory being installed on a computing device that acts as a factory server in the factory, which is integrated into the factory process along with the equipment and edge device. [6] Method according to claim 5, characterized by, that the framework for artificial intelligence in manufacturing of the factory performs an update of the digital twin model by preprocessing and processing the plant and factory data received via the industrial asset data model and the asset management tool on every update of the plant and factory data. [7] Method according to claim 5, characterized by that the factory server stores the output results of the digital twin service provided by the factory's artificial intelligence manufacturing framework as an industrial asset data model file. [8] Method according to claim 2, characterized bythat the LCA service is a service provided by the Artificial Intelligence in Manufacturing framework of the factory, the Artificial Intelligence in Manufacturing framework being installed on a computing device that acts as a factory server in the factory, which is integrated into the factory process along with the facilities and the edge device. [9] Method according to claim 8, characterized by that the factory server stores the output results of the LCA service provided by the factory's Artificial Intelligence Manufacturing Framework as an industrial asset data model file. [10] Method according to claim 1, characterized by that the systems implement decentralization by setting up the system server via the computing device and becoming the object that captures and manages the system data.

Images