Manufacturing operation indicator device

Abstract

This technology provides customized work instructions that are tailored to the work situation and the skills of the workers. [Solution] The manufacturing work instruction device of the present invention comprises a storage unit and a processing unit, the storage unit stores worker skill information, which is information relating to the work skills of workers, manufacturing knowledge information, which is information structuring countermeasures for manufacturing defects based on the similarity of products and countermeasures, and manufacturing performance information, which is information relating to the manufacturing performance of products, the processing unit comprises a first generation AI, which generates work instruction information by searching the manufacturing knowledge information based on information relating to workers, information relating to work, worker skill information, and manufacturing performance information. The processing unit may also comprise a second generation AI that generates manufacturing knowledge information.

Description

[Technical Field]

[0001] The present invention relates to a manufacturing work instruction device. [Background Art]

[0002] Regardless of the type of product, ensuring and improving manufacturing quality is an important issue at manufacturing sites. For example, Patent Document 1 discloses a technique for providing a computing system for fault diagnosis in an industrial environment having a plurality of components. [Prior Art Documents] [Patent Documents]

[0003] [Patent Document 1] Japanese National Publication of International Patent Application No. 2024-519533 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] However, in the technique disclosed in Patent Document 1, although component failures and malfunctions can be detected by processing a plurality of sensor data values, determining patterns, and acquiring an industrial environment digital twin for fault diagnosis in an industrial environment, for example, work instructions that prevent failures and malfunctions in advance cannot be appropriately presented according to the work situation and the skill of the worker.

[0005] Therefore, an object of the present invention is to provide a technique for generating work instructions customized according to the work situation and the skill of the worker. [Means for Solving the Problem]

[0006] To solve the above problems, one representative manufacturing work instruction device of the present invention comprises a storage unit and a processing unit, the storage unit stores worker skill information, which is information relating to the work skills of workers, manufacturing knowledge information, which is information structuring countermeasures for manufacturing defects based on the similarity of products and countermeasures, and manufacturing performance information, which is information relating to the manufacturing performance of products, and the processing unit comprises a first generation AI, the first generation AI generates work instruction information by searching the manufacturing knowledge information based on worker information, work information, worker skill information and manufacturing performance information. [Effects of the Invention]

[0007] According to the present invention, work quality can be improved by generating work instructions that are tailored to the work situation and the skills of the workers. Other issues, configurations, and effects not mentioned above will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 shows an example of the configuration of a welding work instruction device according to Embodiment 1. [Figure 2] Figure 2 shows an example of product knowledge information related to Example 1. [Figure 3] Figure 3 shows an example of manufacturing performance information related to Example 1. [Figure 4] Figure 4 shows an example of defect information related to Example 1. [Figure 5] Figure 5 shows an example of manufacturing knowledge information related to Example 1. [Figure 6] Figure 6 shows an example of worker skill information related to Example 1. [Figure 7] Figure 7 shows an example of work instruction information related to Example 1. [Figure 8] Figure 8 shows an example of the data flow of the welding work instruction device according to Example 1. [Figure 9]Figure 9 shows an example of a flowchart of the manufacturing knowledge structuring processing unit according to Example 1. [Figure 10] Figure 10 shows an example of a flowchart of the work instruction generation processing unit according to Example 1. [Figure 11] Figure 11 shows an example of the display of work instruction information according to Example 1. [Figure 12] Figure 12 shows an example of the hardware configuration of the welding work instruction device according to Embodiment 1. [Figure 13] Figure 13 shows an example of the data flow of the manufacturing knowledge structuring processing unit according to Example 2. [Figure 14] Figure 14 shows an example of the data flow of the work instruction generation processing unit according to Example 3. [Figure 15] Figure 15 shows an example of the data flow of the work instruction generation processing unit according to Embodiment 4. [Modes for carrying out the invention]

[0009] <Explanation of terms, etc.> First, we will explain the terminology used in this disclosure.

[0010] In the following explanation, field data included in product manufacturing information mainly refers to 4M data (Man, Machine, Material, Method), but it is not limited to this. For example, field data may also be 5M data (4M data + Measurement) or 5M+E data (5M data + Environment).

[0011] Furthermore, in the following description, the "input unit", "output unit", "display device", and "interface device" may be one or more interface devices. Said one or more interface devices include one or more I / O (Input / Output) interface devices and one or more communication interface devices. An I / O interface device is an interface device for at least one of an I / O device and a remote display computer. Further, the communication interface device may be one or more communication interface devices of the same type (for example, one or more NICs (Network Interface Cards)), or two or more communication interface devices of different types (for example, a NIC and an HBA (Host Bus Adapter)).

[0012] Furthermore, in the following description, the "storage unit" may be either a memory or both a memory and a persistent storage device. The memory may typically be a main storage device. At least one memory device in the memory may be a volatile memory device or a non-volatile memory device. Further, the persistent storage device may typically be a non-volatile storage device (for example, an auxiliary storage device), and specifically may be, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), an NVME (Non-Volatile Memory Express) drive, or an SCM (Storage Class Memory).

[0013] Further, in the following description, the "processing unit" may be one or more processor devices. At least one processor device may typically be a microprocessor device such as a CPU (Central Processing Unit), and may also be another type of processor device such as a GPU (Graphics Processing Unit). At least one processor device may be single-core or multi-core. At least one processor device may be a processor core. At least one processor device may be a processor device in a broad sense, such as a circuit that is an assembly of gate arrays (e.g., FPGA (Field-Programmable Gate Array), CPLD (Complex Programmable Logic Device) or ASIC (Application Specific Integrated Circuit)) implemented by a hardware description language that performs part or all of the processing.

[0014] Further, in the following description, functions are sometimes described using the expression " ~ unit". Such functions may be implemented by one or more computer programs executed by a processor, or may be implemented by one or more hardware circuits (e.g., FPGA or ASIC), or may be implemented by a combination thereof.

[0015] Next, embodiments of the present invention will be described by taking welding work for railway vehicles as an example of manufacturing work. However, it is also clear from the following description that the manufacturing work targeted by the present invention is not limited to welding work for railway vehicles.

Examples

[0016] Example 1 of the present invention will be described with reference to FIGS. 1 to 12. Example 1 customizes information related to welding work instructions by combining two generative AIs with different roles.

[0017] <Configuration of Welding Work Instruction Device> Figure 1 shows an example of the configuration of a welding work instruction device according to Embodiment 1. As shown in Figure 1, the welding work instruction device 100 includes a storage unit 110, a processing unit 120, a communication unit 130, an input unit 140, and an output unit 150 as its components. Each component of the welding work instruction device 100 may be installed at the manufacturing site or outside the manufacturing site.

[0018] The welding work instruction device 100 may include a group of devices depending on the usage environment, such as a display computer connected via a network (not shown) that enables communication. The network may be, for example, a LAN (Local Area Network), WAN (Wide Area Network), VPN (Virtual Private Network), a communication network that uses public lines such as the Internet in part or in whole, a mobile phone communication network, or a combination thereof. The network may also be a wireless communication network such as Wi-Fi (registered trademark) or 5G (Generation).

[0019] Next, we will describe the individual components of the welding work instruction device 100 of Embodiment 1. <Storage section> The memory unit 110 stores product knowledge information 111, manufacturing performance information 112, defect case information 113, manufacturing knowledge information 114, worker skill information 115, and work instruction information 116. Examples of each type of information stored in the memory unit 110 are shown below.

[0020] Figure 2 shows an example of product knowledge information related to Example 1. Product knowledge information 111 is structured information that describes the component relationships of the product being manufactured and the similarities between products. For structuring the information, for example, graph structures or table-format data based on E-BOM (Electric Bill of Materials) or M-BOM (Manufacturing Bill of Materials) can be used.

[0021] Figure 2 shows product knowledge information 111 related to railway vehicles, using the graph structure (tree diagram) of a vehicle system as an example. In this example, vehicle systems are broadly divided into commuter trains and express trains, with multiple types of vehicles positioned within each system. According to this tree diagram, for example, commuter train A has a higher product similarity to commuter train B than express trains X and Y.

[0022] Furthermore, product knowledge information 111 may be organized not by railway vehicle units such as vehicle systems, but by structural material units such as underframes and ceilings that make up the vehicle.

[0023] Figure 3 shows an example of manufacturing performance information related to Example 1. Manufacturing performance information 112 is information that holds the details of the product manufacturing process and management performance in a table format. In this example, each record of manufacturing performance information 112 consists of manufacturing date and time 112a, railway vehicle type 112b, product to be manufactured 112c, process 112d, management item 112e, performance 112f, and worker ID 112g (multiple IDs are possible).

[0024] Specifically, for example, records #2001 and #2002 in Figure 3 manage the measured dimensions before and after welding for the same product in the same process. In addition, records #2001 and #2011 manage products that are different between car 2 of train set 4 and car 3 of train set 4, even though the process and control items are the same.

[0025] Manufacturing performance information 112 is updated as needed, reflecting the progress of work at the manufacturing site.

[0026] Furthermore, the manufacturing performance information 112 may be managed in separate tables by defining a unique management number for each product. In addition, pass / fail thresholds and their judgment results may also be stored for the management items.

[0027] Figure 4 shows an example of defect case information related to Example 1. Defect case information 113 is information that summarizes the causes and countermeasures for defect cases that occurred in the manufacturing process. In addition to the post-incident countermeasures that were implemented, defect case information 113 may also include permanent countermeasures considered to prevent future defects. In this example, each record of defect case information 113 consists of the type of railway vehicle 113a, the product to be manufactured 113b, the process 113c, the defect case 113d, the cause 113e, the post-incident countermeasure 113f, and the permanent countermeasure 113g.

[0028] Specifically, for example, record #3001 in Figure 4 stores information about a welding distortion defect that occurred during the frame welding process. It indicates that the cause of the welding distortion was that the width of the end frame attached to the frame was larger than the control value, and that repairs were carried out in the vertical assembly process, which is a subsequent process after frame welding, as a corrective measure. Furthermore, as a permanent solution, a policy is indicated to increase the number of clamps used as fixing jigs during welding.

[0029] Figure 5 shows an example of manufacturing knowledge information related to Example 1. Manufacturing knowledge information 114 is information that structures countermeasures for defects in manufacturing based on the similarity of products and countermeasures. In this example, each record of manufacturing knowledge information 114 consists of the type of railway vehicle 114a, process 114b, defect example 114c, cause 114d, and countermeasure 114e.

[0030] Specifically, for example, record #4001 in Figure 5 stores the causes and countermeasures for a defect case involving welding distortion of the underframe, a structural material common to commuter vehicles. The causes and countermeasures for #4001 are generated and stored based on the information in records #3001 and #3002 of defect case information 113. In addition, record #4002 stores the causes and countermeasures for a defect case involving the side, a structural material that differs depending on the vehicle type, but in this example, this is generated and stored as information specific to commuter vehicle B. The generation of manufacturing knowledge information 114 will be described later.

[0031] Figure 6 shows an example of worker skill information related to Embodiment 1. The worker skill information 115 is management information regarding an individual worker's welding skills. In this example, each record of the worker skill information 115 consists of the work name 115a, years of welding experience 115b, and welding skill rank 115c.

[0032] Furthermore, worker skill information 115 may not be a broad category such as "welding work," but rather detailed management of welding skills for specific processes such as "frame" or "side." Also, welding skills may be managed using quantitative evaluation values ​​rather than ranks such as A, B, or C.

[0033] Figure 7 shows an example of work instruction information related to Embodiment 1. Work instruction information 116 is text information about welding instructions generated according to the work situation and worker skills. In this example, four patterns of work instruction information are shown, depending on the combination of whether or not there are abnormalities in the production situation from the 4M perspective and the level of the worker's welding skills.

[0034] Specifically, for example, for skilled welders with an A-rank welding skill level, if there are no abnormalities in the production situation from a 4M perspective, minimal information is provided to prevent disruption to the work and improve production efficiency. On the other hand, if there are abnormalities in the production situation from a 4M perspective, the cause and countermeasures are communicated concisely to ensure work quality. Furthermore, for beginners with a C-rank welding skill level, if there are abnormalities in the production situation from a 4M perspective, detailed countermeasures are communicated to ensure work quality.

[0035] <Processing> The processing unit 120 includes a manufacturing knowledge structuring processing unit 121 and a work instruction generation processing unit 122. The processing unit will be described with reference to Figures 8 to 11.

[0036] Figure 8 shows an example of the data flow of the welding work instruction device according to Example 1.

[0037] <Manufacturing Knowledge Structuring Process> Figure 8(a) shows the data flow related to the manufacturing knowledge structuring process. The manufacturing knowledge structuring processing unit 121 includes a generation AI (second generation AI) and generates manufacturing knowledge information 114 using product knowledge information 111, manufacturing performance information 112, and defect case information 113, and stores it in the storage unit 110.

[0038] The manufacturing knowledge structuring processing unit 121 is executed upon user instruction to start. It may also be executed periodically, such as daily or weekly, based on user settings.

[0039] Figure 9 shows an example of a flowchart of the manufacturing knowledge structuring processing unit according to Example 1. The processing of the manufacturing knowledge structuring processing unit 121 includes steps S101 to S109. Each step will be described below.

[0040] The processing flow of the manufacturing knowledge structuring processing unit 121 starts when it receives a start command from the user via an interface device or the like. The processing flow may also be automatically started periodically based on an execution plan set by the user.

[0041] In steps S101 to S103, the manufacturing knowledge structuring processing unit 121 first acquires product knowledge information 111, manufacturing performance information 112, and defect case information 113. The order in which each piece of information is acquired is arbitrary, and they may be acquired simultaneously.

[0042] In step 104, the manufacturing knowledge structuring processing unit 121 generates structured information (manufacturing knowledge information) using a generation AI based on the similarities within the acquired information.

[0043] In step 105, the manufacturing knowledge structuring processing unit 121 evaluates whether there are any differences between the manufacturing knowledge information generated in step S104 and the manufacturing knowledge information 114 before the update. If there are differences, the process proceeds to S106; if there are no differences, the process ends without updating the manufacturing knowledge information 114.

[0044] In step 106, the manufacturing knowledge structuring processing unit 121 outputs the manufacturing knowledge information generated in this step to the user evaluation screen (not shown) via the output unit 150.

[0045] In steps 107 to 109, first in step S107, the manufacturing knowledge structuring processing unit 121 evaluates whether or not the user has entered any correction information for the manufacturing knowledge information generated this time. If no correction information has been entered, the process proceeds to step S108 and ends, with the manufacturing knowledge information generated this time being used as the updated manufacturing knowledge information 114. If correction information has been entered, the process proceeds to step 109 and ends, with the corrected manufacturing knowledge information being used as the updated manufacturing knowledge information 114.

[0046] In step S107, the user's evaluation and correction process for the generated manufacturing knowledge information 114 is part of the countermeasures against hallucination (a phenomenon in which information that is different from events or facts, including misidentification and logical inconsistencies, is created), which is a concern when using generation AI, but it is not necessarily an essential process.

[0047] <Work instruction generation process> Figure 8(b) shows the data flow related to the work instruction generation process. The work instruction generation processing unit 122 includes a generation AI (first generation AI) and generates work instruction information 116 using the worker skill information 115, manufacturing knowledge information 114, and manufacturing performance information 112 stored in the storage unit 110, as well as the worker information and work information entered by the user, and stores it in the storage unit 110 and outputs it.

[0048] Here, the worker information entered by the user is information for identifying the individual worker, and the work information is information about the process, including welding, that is the subject of the work instruction. The work information includes work status, similar to the manufacturing performance information 112. The work status may include 4M data, 5M data, or 5M+E data as on-site data. For example, measurement data in 5M data may include measurement data such as product dimensions, and environmental data in 5M+E data may include the heatstroke risk index (WGPT) at the work site.

[0049] The work instruction generation processing unit 122 performs its operations at the start of each manufacturing operation, based on input from the user.

[0050] Figure 10 shows an example of a flowchart of the work instruction generation processing unit according to Embodiment 1. The processing of the work instruction generation processing unit 122 includes steps S201 to S207. Each step will be described below.

[0051] The processing flow of the work instruction generation processing unit 122 starts when it receives a start instruction from the user via an interface device or the like.

[0052] In steps S201 and 202, the work instruction generation processing unit 122 first receives worker information and work information as input. The order in which each piece of information is entered is arbitrary, and they may be entered simultaneously.

[0053] In step S203, the work instruction generation processing unit 122 retrieves the corresponding worker skill information 115 from the worker information entered in step S201.

[0054] In step S204, the work instruction generation processing unit 122 retrieves the corresponding manufacturing performance information 112 from the work information entered in step S202. Here, measurement data such as product dimensions may be obtained from the work information and the work status of the manufacturing performance information 112 from the perspective of 5M. Alternatively, values ​​such as the heatstroke risk index (WGPT) and other environmental values ​​may be obtained from the perspective of 5M+E.

[0055] In step S205, the work instruction generation processing unit 122 generates a search prompt for the generating AI to search for manufacturing knowledge information.

[0056] In step S206, the work instruction generation processing unit 122 performs an augmented search of the manufacturing knowledge information 114 based on the search prompt generated in step S205, using, for example, a generation AI using RAG (Retrieval Augmented Generation).

[0057] In step 207, the work instruction generation processing unit 122 generates work instruction information 116 using the generation AI based on the manufacturing knowledge information 114 related to the target process that was searched in step S206 and the worker skill information 115 that was retrieved in step S203.

[0058] The generated work instruction information 116 is displayed on a display device, such as an external terminal, via the output unit 150 described later. Figure 11 shows an example of the display of work instruction information according to Embodiment 1. In this example, the work instruction information 116 is displayed as a pop-up on a digital work instruction diagram using a 3D model, showing the alert, its cause, and specific countermeasures.

[0059] <Communication section, input section, output section> Returning to Figure 1, the communication unit 130, input unit 140, and output unit 150 will be described.

[0060] The communication unit 130 has the function of sending and receiving various types of information to and from external devices via a network.

[0061] The input unit 140 has an information input function. The input unit 140 consists of, for example, a terminal that can be displayed and operated on a screen, or a keyboard or mouse.

[0062] The output unit 150 has the function of creating screen information that includes output information obtained by predetermined processing. The output unit 150 outputs the screen information to, for example, the display device (not shown) of the welding work instruction device 100 or an external display computer via the communication unit 130.

[0063] <Hardware Configuration> Next, with reference to Figure 12, the hardware configuration of the welding work instruction device 100 of Embodiment 1 will be described. Figure 12 shows an example of the hardware configuration of the welding work instruction device according to Embodiment 1.

[0064] The welding work instruction device 100 can be implemented as a general-purpose computer 300, or a network system comprising multiple such computers 300, which includes a processor 301, memory 302, storage 303 such as a hard disk drive (HDD), a storage medium read / write device 305 for reading or writing information to a portable storage medium 304 such as a CD (Compact Disk) or DVD (Digital Versatile Disk), an input device 306 such as a keyboard, mouse, or barcode reader, an output device 307 such as a display, and a communication device 308 for communicating with other computers via a communication network such as the Internet.

[0065] For example, the processing unit 120 can be implemented by loading a predetermined program stored in the storage 303 into the memory 302 and executing it with the processor 301; the input unit 140 and output unit 150 can be implemented by the processor 301 utilizing the input device 306 and the output device 307; and the storage unit 110 can be implemented by the processor 301 utilizing the memory 302 or the storage 303. The predetermined program described above may be downloaded from the storage medium 304 via the storage medium read / write device 305, or from the network via the communication device 308, to the storage 303, and then loaded onto the memory 302 and executed by the processor 301.

[0066] Alternatively, a predetermined program may be loaded directly onto the memory 302 from the storage medium 304 via the storage medium read / write device 305, or from the network via the communication device 308, and executed by the processor 301.

[0067] The welding work instruction device 100 is not limited to this configuration, and may also be a wearable computer that can be worn by the worker, such as a headset, goggles, glasses, or intercom.

[0068] According to Example 1, individual work instructions can be provided according to the work situation and the worker's skills, thereby improving work quality. Furthermore, if Example 1 is applied to a training device for workers, it is expected to shorten the training period.

[0069] Furthermore, according to Example 1, based on product knowledge information 111, manufacturing performance information 112, and defect case information 113, the generating AI can be used to structure work knowledge regarding defect cases and countermeasures for similar vehicle models, thereby improving the accuracy of work instructions. In addition, since the structuring of manufacturing knowledge information can be processed in advance, the responsiveness to the creation of work instruction information is improved, and the number of calls to the generating AI can be reduced, thereby suppressing computation costs. [Examples]

[0070] Next, Example 2 of the present invention will be described. Example 2 is an extension of the manufacturing knowledge structuring process of Example 1.

[0071] Information related to manufacturing operations includes a great deal of tacit knowledge, such as on-site know-how, that is not explicitly documented as specific measures or instructions. If this know-how can be incorporated into manufacturing knowledge, the quality of work instruction information can be further improved. Example 2 involves linking with external devices such as a welding traceability system that can analyze the characteristics of work that affect quality by analyzing camera footage of welding operations, in order to incorporate work know-how such as work procedures as manufacturing knowledge information 114.

[0072] Figure 13 shows an example of the data flow of the manufacturing knowledge structuring process according to Embodiment 2. Embodiment 2 includes a welding traceability linkage unit 202 that works in conjunction with a welding traceability device to acquire work know-how information such as work procedures. The acquired work know-how information is stored in the storage unit 110 as manufacturing knowledge information 114 and used as input information to the manufacturing knowledge structuring processing unit 121.

[0073] The term "cooperation unit" as used herein refers to a function or means that enables the exchange of information with the target device for coordination, and it is not necessary to specify whether the target device is located inside or outside the welding work instruction device 100 (the same applies hereinafter).

[0074] According to Example 2, by acquiring work know-how such as work procedures in cooperation with external devices such as a welding traceability system, it is possible to improve the accuracy of work instruction information. [Examples]

[0075] Next, we will describe Embodiment 3 of the present invention. Embodiment 3 is an extension of the work instruction generation process of Embodiment 1.

[0076] Generative AI generally struggles to handle numerical data precisely, and the process of obtaining analysis results is a black box, making it difficult to verify the basis of the output. Example 3 addresses this by collaborating with conventional AI equipped with statistical models and machine learning algorithms to make the basis for the results of numerical data analysis and processing a white box.

[0077] Figure 14 shows an example of the data flow of the work instruction generation process according to Embodiment 3. As shown in Figure 14, Embodiment 3 includes an AI collaboration unit 203 for collaboration with a conventional AI capable of analyzing and processing large amounts of numerical data. Information regarding the numerical data analyzed and processed by the conventional AI is input to the generation AI of the work instruction generation processing unit 122 in an appropriate format.

[0078] According to Example 3, by entrusting the analysis and processing of numerical data to conventional AI, it becomes possible to white-box a portion of the work instruction generation process. [Examples]

[0079] Next, we will describe Embodiment 4 of the present invention. Embodiment 4 is also an extension of the work instruction generation process of Embodiment 1.

[0080] In manufacturing sites for large products, it is necessary to provide work instructions for tasks involving multiple people. Example 4 integrates with a work assignment optimization device to enable work instructions that include not only individual work instructions but also work assignments that determine the division of tasks among multiple workers.

[0081] Figure 15 shows an example of the data flow of the work instruction generation process according to Embodiment 4. As shown in Figure 15, Embodiment 4 includes a work assignment coordination unit 204 for coordinating with a work assignment optimization device that determines the division of work among multiple workers. The work assignment information determined by the work assignment optimization device is input to the generation AI of the work instruction generation processing unit 122.

[0082] According to Example 4, it is possible to provide work instructions that include work assignments, which determine the division of labor among multiple workers performing tasks simultaneously, thereby improving productivity and work quality.

[0083] <Example of changes> Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above, and various modifications are possible without departing from the spirit of the present invention.

[0084] For example, as mentioned at the beginning, the above embodiment described the work instruction device using the welding work of railway vehicles as an example of manufacturing work, but it is not limited to this. The present invention can be applied to any manufacturing work of any manufactured product (for example, processing work or assembly work).

[0085] Furthermore, in the above-described embodiment, the defect case information 113 showed countermeasures for defects related to product quality that occurred in the product manufacturing process. However, this could also be countermeasures for safety defects that occurred in the product manufacturing process. For example, examples of unsafe behavior and countermeasures, examples of wearing protective clothing and points to note, and examples of heatstroke and countermeasures could be considered.

[0086] The present invention also includes the following embodiments.

[0087] (Aspect 1) A manufacturing work instruction device comprising a memory unit and a processing unit, The memory unit stores worker skill information, which is information relating to the worker's work skills; manufacturing knowledge information, which is information structuring countermeasures for manufacturing defects based on the similarity of products and countermeasures; and manufacturing performance information, which is information relating to the manufacturing performance of products. The processing unit comprises a first generation AI, The first generation AI generates work instruction information by searching the manufacturing knowledge information based on information about the worker, information about the work, the worker's skill information, and the manufacturing performance information. A manufacturing work instruction device characterized by the following features.

[0088] (Aspect 2) A manufacturing work instruction device according to embodiment 1, The memory unit further stores product knowledge information, which is information regarding the structural relationships of product components or the similarities between products, and defect case information, which is information regarding countermeasures for defects that occurred in the product manufacturing process. The processing unit further comprises a second generation AI, The second generation AI generates the manufacturing knowledge information based on the product knowledge information, the manufacturing performance information, and the defect case information. A manufacturing work instruction device characterized by the following features.

[0089] (Aspect 3) In the manufacturing work instruction device of embodiment 1 or 2, The system further includes an output unit that generates screen information for the aforementioned work instruction information. A manufacturing work instruction device characterized by the following features.

[0090] (Aspect 4) In the manufacturing work instruction device of embodiment 2 or 3, The aforementioned defect case information includes measures taken to address product quality defects that occurred during the product manufacturing process, or measures taken to address safety defects that occurred during the product manufacturing process. A manufacturing work instruction device characterized by the following features.

[0091] (Aspect 5) In any of the manufacturing work instruction devices in embodiments 2 to 4, The system further includes an interface that allows the user to evaluate and modify the manufacturing knowledge information generated by the second generation AI. A manufacturing work instruction device characterized by the following features.

[0092] (Aspect 6) In any of the manufacturing work instruction devices in embodiments 1 to 5, It is further equipped with a traceability integration unit that works in conjunction with traceability equipment to track manufacturing operations. A manufacturing work instruction device characterized by the following features.

[0093] (Aspect 7) In any of the manufacturing work instruction devices in embodiments 1 to 6, It also includes an AI integration unit that works in conjunction with AI for analyzing and processing numerical data. A manufacturing work instruction device characterized by the following features.

[0094] (Pattern 8) In any of the manufacturing work instruction devices in embodiments 1 to 7, It further includes a work assignment optimization linkage unit that works in conjunction with a work assignment optimization device that creates work assignments for multiple workers based on personnel allocation plans and work plans. A manufacturing work instruction device characterized by the following features. [Explanation of symbols]

[0095] 100: Welding work instruction device 110: Storage section 111: Product Knowledge Information 112: Manufacturing Performance Information 113: Malfunction Case Information 114: Manufacturing Knowledge Information 115: Worker Skill Information 116: Work Instruction Information 120: Processing Unit 121: Manufacturing Knowledge Structuring Processing Unit 122: Work Instruction Generation Processing Unit 130: Communications Department 140: Input section 150: Output section

Claims

1. A manufacturing work instruction device comprising a memory unit and a processing unit, The memory unit stores worker skill information, which is information relating to the worker's work skills; manufacturing knowledge information, which is information structuring countermeasures for manufacturing defects based on the similarity of products and countermeasures; and manufacturing performance information, which is information relating to the manufacturing performance of products. The processing unit comprises a first generation AI, The first generation AI generates work instruction information by searching the manufacturing knowledge information based on information about the worker, information about the work, the worker's skill information, and the manufacturing performance information. A manufacturing work instruction device characterized by the following features.

2. A manufacturing work instruction device according to claim 1, The memory unit further stores product knowledge information, which is information regarding the structural relationships of product components or the similarities between products, and defect case information, which is information regarding countermeasures for defects that occurred in the product manufacturing process. The processing unit further comprises a second generation AI, The second generation AI generates the manufacturing knowledge information based on the product knowledge information, the manufacturing performance information, and the defect case information. A manufacturing work instruction device characterized by the following features.

3. In the manufacturing work instruction device according to claim 1, The system further includes an output unit that generates screen information for the aforementioned work instruction information. A manufacturing work instruction device characterized by the following features.

4. In the manufacturing work instruction device according to claim 2, The aforementioned defect case information includes measures taken to address product quality defects that occurred during the product manufacturing process, or measures taken to address safety defects that occurred during the product manufacturing process. A manufacturing work instruction device characterized by the following features.

5. In the manufacturing work instruction device according to claim 2, The system further includes an interface that allows the user to evaluate and modify the manufacturing knowledge information generated by the second generation AI. A manufacturing work instruction device characterized by the following features.

6. In the manufacturing work instruction device according to claim 2, It is further equipped with a traceability integration unit that works in conjunction with traceability equipment to track manufacturing operations. A manufacturing work instruction device characterized by the following features.

7. In the manufacturing work instruction device according to claim 1, It also includes an AI integration unit that works in conjunction with AI for analyzing and processing numerical data. A manufacturing work instruction device characterized by the following features.

8. In the manufacturing work instruction device according to claim 1, It further includes a work assignment optimization linkage unit that works in conjunction with a work assignment optimization device that creates work assignments for multiple workers based on personnel allocation plans and work plans. A manufacturing work instruction device characterized by the following features.

Images