system

Abstract

We provide a system that significantly improves the efficiency of asset management and reduces the workload and costs involved. [Solution] A system comprising: means for acquiring asset data from an existing database; means for training a generative AI model to predict the state of assets based on the imported asset data; means for generating an asset inventory schedule based on the prediction results of the AI ​​model; means for distributing the generated inventory schedule to a terminal; means for receiving inventory results entered from the terminal and updating the database; means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model; and means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including: receiving a user utterance; adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot; encoding the prompt; and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In existing asset management systems, the man-hours and costs involved in monthly physical inventories and the removal of unnecessary assets are high. Therefore, it is necessary to streamline this process and reduce labor and costs. In addition, there is a need for a method to more accurately predict and manage physical inventories and the deterioration status of assets.

Means for Solving the Problems

[0005] To solve this problem, the present invention provides the following means. First, a server is provided with means for acquiring asset data from an existing database. Next, a means is provided for training a generative AI model for predicting the state of assets based on the imported asset data. Furthermore, a means is provided for generating an asset inventory schedule based on the prediction results of the AI ​​model, and for distributing the generated schedule to a terminal. The invention also provides means for receiving inventory results entered from the terminal and updating the database. In addition, a means is provided for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model. Finally, the invention provides a system that includes means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database.

[0006] A "database" is a system that organizes and stores data, enabling efficient searching and management.

[0007] "Asset data" refers to data that describes information about assets such as equipment and fixtures, and typically includes attributes such as type, quantity, location, and condition.

[0008] A "server" is a computer system that acquires, processes, and stores data, and provides services to other devices and users.

[0009] A "generative AI model" is an artificial intelligence model that uses machine learning algorithms to learn patterns from large amounts of data and then uses those patterns to make predictions about new data.

[0010] "Training" is the process of optimizing the parameters of a machine learning model using a dataset to improve the accuracy of predictions.

[0011] An "inventory schedule" is a schedule that outlines the plan for verifying assets and checking their quantity and condition.

[0012] A "terminal" is a device that a user operates to input information or receive information from a server, and is usually a computer or mobile device.

[0013] "Inventory results" refer to data that records the quantity and condition of assets after they have been physically inspected.

[0014] "Removal timing" refers to the optimal time to remove an asset when its deterioration or discontinuation of use is anticipated.

[0015] A "notification" is a message intended to inform a user of specific information or instructions. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Modes for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of a plurality of arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of a plurality of types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0020] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0021] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] This invention provides a method for streamlining asset management in SB / WCP systems using generative AI, thereby reducing the time and costs associated with monthly inventory and asset disposal operations. The system's program processing is explained below in natural language, followed by a specific operational example.

[0038] First, the server retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. This data is periodically imported using SQL queries or other data retrieval methods. The imported data is converted to an internal data format (e.g., CSV or JSON) and processed efficiently.

[0039] Next, the server trains a generative AI model based on the imported data. This AI model is designed to predict asset usage frequency and degradation, and uses machine learning algorithms (e.g., random forests or neural networks). Training involves splitting the data into training and validation datasets and tuning hyperparameters to evaluate and improve the model's accuracy.

[0040] Once the AI ​​model is trained, the server generates an asset inventory schedule based on the model's predictions. The inventory aims to prioritize the identification of assets that are at risk of deterioration or are of high importance. The generated schedule is delivered to the terminal and serves as a guideline for the inventory work performed by the user.

[0041] Users use a terminal to check the inventory schedule and perform the verification of designated assets. During the inventory, users input the actual quantity and condition of assets into the terminal and send the inventory results to the server. The server receives these results, updates its database, and uses it to generate forecasts and schedules for the next inventory.

[0042] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions. By notifying users of the optimal disposal timing before critical assets deteriorate and become unusable, proactive management becomes possible. Users receive notifications, perform the actual disposal work, and input the results into their terminals. The server then receives the disposal results and updates its database.

[0043] Specific example:

[0044] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training predicts that older air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and distributes it to the terminal. The user checks the inventory schedule and performs the inventory work. The inventory results are entered into the terminal and sent to the server. The server reflects the results and notifies the user of the removal timing. Based on the notification, the user removes the old air conditioning equipment and enters the removal results. Finally, the server makes predictions and plans for the following month based on the updated data.

[0045] This entire process significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, repeated training and schedule generation improve the accuracy of the AI ​​model, enabling increasingly efficient asset management.

[0046] The following describes the processing flow.

[0047] Step 1:

[0048] The server retrieves asset data from an existing database. Specifically, the server periodically executes SQL queries to extract the latest asset data from the database and convert it into an internal data format (e.g., CSV or JSON).

[0049] Step 2:

[0050] The server trains a generative AI model based on the imported data. Specifically, the server splits the data into a training dataset and a validation dataset, and trains the model using a machine learning algorithm (e.g., random forest, neural network). If there are any deficiencies in the training results, it tunes the hyperparameters and retrains the model.

[0051] Step 3:

[0052] The server generates an asset inventory schedule based on a trained AI model. Specifically, it creates a schedule that prioritizes the inventory of assets with a high risk of deterioration or those of high importance, based on the model's predictions.

[0053] Step 4:

[0054] The server distributes the generated inventory schedule to the terminals. Specifically, it sends the schedule file, converted to a format usable by the terminals, over the network, making it accessible to the users.

[0055] Step 5:

[0056] Users perform asset verification tasks based on the inventory schedule. Specifically, they check the condition and quantity of assets on-site according to the inventory schedule and input the results into their handheld terminals.

[0057] Step 6:

[0058] The terminal sends the inventory results entered by the user to the server. Specifically, it sends the input data to the server via the network, and the server stores the received data in its internal database.

[0059] Step 7:

[0060] The server updates the database based on the inventory results. Specifically, it evaluates the received inventory results and updates the existing asset database with the latest information.

[0061] Step 8:

[0062] The server calculates the optimal timing for asset decommissioning based on the updated database and predictions from the AI ​​model. Specifically, it uses a trained AI model to predict deterioration and identify assets that need to be decommissioned and when.

[0063] Step 9:

[0064] The server notifies the user of the timing for disposal. Specifically, it generates a notification message that lists the assets that need to be disposed of and the timing for disposal, and delivers this message to the terminal.

[0065] Step 10:

[0066] The user receives a notification and carries out the removal work. Specifically, they remove the unwanted assets on-site according to the notification and enter the results (removal date, method, etc.) into the terminal.

[0067] Step 11:

[0068] The terminal sends the user-entered removal results to the server. Specifically, it sends the input data to the server via the network, and the server stores the received data in its internal database.

[0069] Step 12:

[0070] The server updates the database based on the disposal results. Specifically, it evaluates the received disposal results and updates the existing asset database with the latest information.

[0071] (Example 1)

[0072] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0073] Traditional asset management systems required manual inventory and disposal processes, consuming considerable effort and time. Furthermore, accurately predicting asset deterioration and usage frequency was difficult, leading to decreased asset management efficiency. This could potentially result in increased costs due to asset deterioration and improper management.

[0074] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0075] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for dividing the training into a training dataset and a validation dataset and adjusting hyperparameters, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, and means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database. This significantly improves the efficiency of asset management and reduces the man-hours and costs of the work.

[0076] An "existing database" is a collection of information that has already been built and is in operation, and is a system in which data related to assets is stored.

[0077] "Asset data" refers to information about physical or digital assets owned by a company or organization, including attributes such as type, quantity, location, and status.

[0078] A "generative AI model" is a program that uses machine learning algorithms to learn patterns and rules from given data, and is used to predict the deterioration status and usage frequency of assets.

[0079] A "training dataset" is a collection of data used by generative AI models for learning, and it serves as the foundational data for giving the model predictive capabilities.

[0080] A "validation dataset" is a collection of data used to evaluate the prediction accuracy of generative AI models, and is managed separately from the training dataset.

[0081] "Hyperparameters" are settings or adjustment items that affect the performance and results of a machine learning model, and are adjusted to optimize the model.

[0082] An "inventory schedule" is a plan or schedule for verifying assets, and is created by prioritizing tasks based on the condition of the assets and their risk of deterioration.

[0083] A "terminal" refers to a computer or mobile device used by a user, and is hardware used for sending and receiving data with a system.

[0084] "Decommissioning timing" refers to the optimal time to dispose of or transfer a particular asset before it becomes unusable, and is calculated for preventative management purposes.

[0085] This invention provides an efficient asset management system using a generative AI model. The overall system processing flow and specific operational examples are described below.

[0086] First, the server retrieves asset data from an existing database. This data includes attribute information such as asset type, quantity, location, and status. The data is periodically imported into the server using SQL queries or other data retrieval methods.

[0087] Next, the server converts the acquired data into an internal data format (e.g., CSV or JSON). This conversion is performed using libraries such as Python's Pandas library, and the converted data is stored in storage for efficient processing.

[0088] The server trains a generative AI model based on the imported data. Machine learning algorithms (e.g., random forests or neural networks) are used for training. The server splits the data into training and validation datasets and tunes the hyperparameters. This process improves the prediction accuracy of the AI ​​model.

[0089] Once the AI ​​model is trained, the server generates an asset inventory schedule based on the model's predictions. The generated schedule prioritizes the verification tasks based on the asset's deterioration risk and importance. The generated schedule is delivered to the terminal and serves as a guideline for the inventory work performed by the user.

[0090] Users use a terminal to check the inventory schedule and perform verification tasks for designated assets. During the process, users input the actual quantity and condition of the assets into the terminal and send the results to the server. The server updates its database based on the received inventory results and uses this information to generate forecasts and schedules for the next inventory.

[0091] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions. Before important assets deteriorate and become unusable, the server notifies the user of this information. Based on the notification, the user performs the disposal work and inputs the results into their terminal, sending them to the server. After receiving the disposal results, the server updates its database.

[0092] Specific example

[0093] At the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training predicts that older air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and distributes it to the terminal. The user checks the inventory schedule and performs the inventory work. The inventory results are entered into the terminal and sent to the server. The server reflects the results and notifies the user of the removal timing. Based on the notification, the user removes the old air conditioning equipment and enters the removal results. Finally, the server makes predictions and plans for the following month based on the updated data.

[0094] This process significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, repeated training and schedule generation improve the accuracy of the AI ​​model, enabling increasingly efficient asset management.

[0095] Example of a prompt

[0096] An example of a prompt statement to illustrate this series of processes is as follows:

[0097] "In the asset management system, prompts are used to generate specific inventory schedules and disposal timings:

[0098] The server retrieves data from the asset database, converts it to CSV format, and inputs it as training data into a generative AI model. This model uses a random forest algorithm to predict the deterioration status of assets. Based on the predictions, the server generates the inventory schedule and disposal timing for the current month and distributes it to the terminal. After the inventory is completed, the results are sent from the terminal to the server, and the database is updated.

[0099] This clarifies the specific embodiments of the invention, enabling implementers to use the invention appropriately and efficiently.

[0100] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0101] Step 1: Obtain asset data

[0102] The server periodically retrieves asset data from an existing database. The input includes attribute information such as asset type, quantity, location, and status. This data is retrieved using SQL queries or other means. Specifically, the server executes an SQL query like "SELECT FROM assets WHERE month='current_month'". The output of this query is imported into the server.

[0103] Step 2: Data conversion and saving

[0104] The server converts the acquired asset data into an internal standard format (e.g., CSV or JSON). The data acquired in step 1 is used as input. Specifically, the server uses the Python Pandas library to convert the data into a DataFrame and saves it to a file such as "assets_data.csv". As output, the converted data is saved to storage in a format that can be processed efficiently.

[0105] Step 3: Training the AI ​​model

[0106] The server trains a generative AI model based on the imported data. The data saved in step 2 is used as input. Specifically, the server uses the Scikit-learn library to split the data into a training dataset and a validation dataset, and then trains a random forest model. Hyperparameter tuning is also performed. The output is a trained AI model.

[0107] Step 4: Generate and distribute the inventory schedule.

[0108] The server generates an inventory schedule based on the predictions of the trained AI model. The input includes the AI ​​model and asset data obtained in step 3. Specifically, the server uses the model to predict asset deterioration and creates an inventory schedule based on the results. The output is the generated inventory schedule, which is then delivered to the terminal.

[0109] Step 5: Conduct inventory work

[0110] The user uses a terminal to check the inventory schedule and perform the verification of the specified assets. The input includes the inventory schedule distributed in step 4. Specifically, the user checks the quantity and condition of assets such as air conditioning equipment according to the schedule and inputs the results into the terminal. The output is the verification results. These results are sent from the terminal to the server.

[0111] Step 6: Update the database of inventory results

[0112] The server receives the inventory results sent from the terminal and updates the database. The input used is the confirmation result obtained by the user in step 5. Specifically, the server integrates the new data into the existing database and uses it to generate future predictions and schedules. The output is the updated database.

[0113] Step 7: Planning and Implementing Asset Disposal

[0114] The server calculates and notifies the optimal timing for asset decommissioning based on the prediction results of the AI ​​model. Inputs include the trained AI model obtained in step 3 and the latest data. Specifically, the server predicts the deterioration status of old air conditioning equipment and calculates the decommissioning timing. This information is notified to the terminal. An output is generated: a decommissioning notification.

[0115] Step 8: Implement asset disposal and enter the results.

[0116] The user performs the appropriate asset removal work based on notifications from the server. The input includes the removal timing notified in step 7. Specifically, the user removes the old air conditioning equipment and inputs the result into the terminal. The output is the removal result, which is then sent from the terminal to the server.

[0117] Step 9: Update the database of removal results

[0118] The server receives the decommissioning results sent from the terminal and updates the database. The input used is the decommissioning results obtained in step 8. Specifically, the server integrates the decommissioning data into the existing database and uses it for future forecasts and planning. The output is the updated database. This series of steps enables efficient asset management and significantly reduces labor and costs.

[0119] (Application Example 1)

[0120] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0121] In modern logistics centers, asset management and inventory management are extremely complex and time-consuming tasks. They require a great deal of manual work, are prone to errors, and lack preventative management leads to increased losses due to asset deterioration. Traditional systems struggle to accurately predict asset deterioration and usage frequency, making it difficult to efficiently generate inventory and disposal schedules. To address these challenges, a more efficient and accurate asset management system is needed.

[0122] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0123] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, and means including an application on the terminal that notifies the user of the predicted inventory schedule and disposal timing, and allows the user to confirm and input the work. This makes it possible to accurately predict the state and frequency of use of assets and efficiently generate and notify inventory and disposal schedules, significantly improving the efficiency and accuracy of asset management.

[0124] A "server" is a device that retrieves asset data from an existing database, trains an AI model based on that data, and generates and distributes inventory schedules and disposal timings based on the prediction results.

[0125] "Asset data" refers to data that includes information about the type, quantity, location, condition, and frequency of use of goods and equipment.

[0126] A "generative AI model" is a machine learning model that uses training data to predict the deterioration status and usage frequency of assets, and to calculate appropriate inventory schedules and disposal timings.

[0127] "Training" is the process of using existing data to train a generative AI model so that it can accurately predict the state of an asset.

[0128] An "inventory schedule" is a plan for conducting inventory work at specific times, based on the deterioration status and importance of the assets.

[0129] A "terminal" is a device used by users to check inventory schedules and disposal timings, and to input actual work results.

[0130] "Removal timing" refers to the optimal period for properly removing an asset before it deteriorates and becomes unusable.

[0131] "Inventory results" refer to information about the quantity and condition of assets obtained by the user through actual inventory work.

[0132] A "database" is a database management system that stores asset data and allows servers to retrieve the information they need.

[0133] "User notifications" is a system that communicates information such as inventory schedules and disposal timings to users via their devices.

[0134] This invention is a system for improving the efficiency of asset management in logistics centers. Specifically, it acquires asset data, predicts asset status using a generative AI model, generates and distributes inventory schedules, and notifies users of asset disposal timing. The main components of this system involve servers, terminals, and users.

[0135] The server first retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. The retrieved data is periodically imported and converted into an internal CSV or JSON format. The server then uses this data to train a generative AI model. This AI model is implemented using machine learning libraries such as Scikit-learn, and its hyperparameters are tuned by splitting it into training and validation datasets.

[0136] A trained AI model predicts the deterioration status and usage frequency of assets, and generates an inventory schedule based on these predictions. The generated schedule is delivered to the user's device, which may include a smartphone or tablet. The user uses this device to check the notified inventory schedule and perform the inventory work for the specified assets. The work results are entered into the device and sent to the server, updating the database.

[0137] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the user. The user receives the notification, enters the result of asset disposal into their terminal, and this data is sent to the server, updating the database again.

[0138] As a concrete example, at the beginning of the month, the server retrieves all asset data, and an AI model predicts the deterioration of older forklifts. Based on this prediction, an inventory schedule indicating the need to remove forklifts is generated and delivered to the terminal. The user performs the inventory work, enters the results into the terminal, and sends them to the server, updating the database. Removal is carried out in a similar manner.

[0139] Examples of prompts for a generative AI model:

[0140] Train an AI model to predict asset degradation and usage frequency using the following dataset. The data is provided in CSV format, with each row representing a single asset.

[0141] The column names are as follows: Asset ID, Asset Type, Quantity, Location, Status, Usage Frequency.

[0142] The model should use this data to identify assets that are expected to deteriorate in the following month and predict when they will need to be removed.

[0143] To implement this invention, the server side performs data import, AI model training, schedule generation, notification, and database updates, while the terminal side inputs inventory results and receives notifications. This series of operations significantly improves the efficiency and accuracy of asset management in logistics centers.

[0144] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0145] Step 1:

[0146] The server retrieves asset data from an existing database. It sends SQL queries to the database to extract asset data, which includes asset type, quantity, location, status, etc. It receives asset data from the database as input and obtains raw data from the database as output.

[0147] Step 2:

[0148] The server converts the acquired asset data into an internal format (CSV or JSON). This allows for efficient data processing. It receives raw data from the database as input and obtains converted data as output.

[0149] Step 3:

[0150] The server trains a generative AI model based on the transformed asset data. In this process, the data is split into training and validation datasets, and the model is trained using Scikit-learn's RandomForestRegressor. It accepts asset data in an internal format as input and produces a trained AI model as output.

[0151] Step 4:

[0152] The server uses a trained AI model to predict the deterioration status and usage frequency of assets. This results in the asset status being obtained as a prediction. It receives a trained AI model and asset data in an internal format as input and obtains the prediction result as output.

[0153] Step 5:

[0154] The server generates an asset inventory schedule based on the predictions of an AI model. This is intended to prioritize the identification of assets at risk of deterioration or those of high importance. It takes prediction results as input and obtains an inventory schedule as output.

[0155] Step 6:

[0156] The server distributes the generated inventory schedule to the terminal. This allows users to perform inventory tasks according to the schedule. It receives the inventory schedule as input and receives notifications to the terminal as output.

[0157] Step 7:

[0158] The user uses a terminal to check the inventory schedule and perform the verification of the specified assets. The terminal receives the inventory schedule as input and inputs the inventory results as output.

[0159] Step 8:

[0160] The server receives inventory results entered from the terminal and updates the database. This ensures that the latest asset data is reflected in the database. The system receives inventory results as input and obtains an updated database as output.

[0161] Step 9:

[0162] The server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the user. It receives prediction results as input and outputs a notification of the disposal timing.

[0163] Step 10:

[0164] The user receives a notification, enters the result of asset disposal into the terminal, and sends that data to the server. The user receives a notification of the disposal timing as input and inputs the disposal result into the terminal as output.

[0165] Step 11:

[0166] The server receives the disposal results entered from the terminal and updates the database. This provides the latest asset data to be used for future predictions and planning. It receives disposal results as input and obtains an updated database as output.

[0167] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0168] This invention provides a system that supports users' efficient and effective work by combining an emotion engine with asset management in SB / WCP systems that utilize generative AI. The following describes the program processing of this system in natural language, along with specific examples.

[0169] First, the server retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. This data is periodically imported using SQL queries, etc., and converted to CSV or JSON format for efficient internal processing.

[0170] Next, the server trains a generative AI model based on the imported data. This training involves splitting the data into a training dataset and a validation dataset, evaluating the model's accuracy, and tuning hyperparameters as needed. Possible machine learning algorithms used include random forests and neural networks.

[0171] Once the model training is complete, the server generates an asset inventory schedule based on the AI ​​model's predictions. This schedule is designed to focus on assets with high depreciation risk and those of high importance. The generated schedule is delivered to the terminal for the user to use.

[0172] The server also features an emotion engine to recognize user emotions. When a user performs inventory work, the emotion engine uses the camera and microphone to identify emotions from the user's facial expressions and voice. This emotion data is transmitted to the server in real time, and the server adjusts the way inventory schedules are presented and notifications are sent as needed.

[0173] Users perform asset verification tasks based on an inventory schedule. Once users have completed asset verification, they enter the results into their terminal and send them to the server. The server receives the inventory results, updates the database, and uses them as feedback to improve the prediction accuracy of the model.

[0174] Furthermore, the server calculates and notifies the user of the optimal timing for asset disposal based on the AI ​​model's predictions. When the user receives the notification, performs the asset disposal, and enters the results into their terminal, the server receives the disposal results and updates the database. This entire process ensures that the user can always manage their assets with the latest information.

[0175] Specific example:

[0176] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predicting that old air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes the user's facial expressions and voice to analyze their emotions. For example, if the user is tired, the workload can be adjusted to reduce it.

[0177] The inventory results are entered into a terminal, and the data is sent to the server. The server reflects the results and notifies the user when the time for removal approaches. Based on the notification, the user removes old air conditioning equipment, enters the results into the terminal, and sends them to the server. The server updates the database and uses it for future predictions and planning.

[0178] This system significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, the use of an emotional engine reduces the user's workload, providing a more comfortable work environment.

[0179] The following describes the processing flow.

[0180] Step 1:

[0181] The server retrieves asset data from an existing database. This involves periodically collecting the latest asset information using SQL queries and other data retrieval methods. This data includes asset type, quantity, location, and status.

[0182] Step 2:

[0183] The server trains a generative AI model based on imported asset data. First, it splits the data into a training dataset and a validation dataset. Then, it trains the model using machine learning algorithms such as random forests and neural networks. It evaluates the model's accuracy and tunes hyperparameters as needed.

[0184] Step 3:

[0185] The server generates an asset inventory schedule based on a trained AI model. This includes a process of creating a schedule that prioritizes checking assets at high risk of deterioration or those of high importance.

[0186] Step 4:

[0187] The server distributes the generated inventory schedule to the terminals. It sends the schedule file, converted to a format usable by the terminals, over the network, making it accessible to the users.

[0188] Step 5:

[0189] The device uses an emotion engine to recognize the user's emotions while they are performing inventory tasks. It analyzes the user's facial expressions and voice through the camera and microphone to acquire emotional data.

[0190] Step 6:

[0191] The server receives emotional data sent from the emotion engine and adjusts inventory schedules and notification methods. If the user is fatigued, measures such as reducing the workload are taken.

[0192] Step 7:

[0193] Users perform asset verification according to the inventory schedule. They inspect assets on-site and input their condition and quantity into a terminal.

[0194] Step 8:

[0195] The terminal sends the inventory results entered by the user to the server. The collected data is sent to the server via the network, and the server receives it.

[0196] Step 9:

[0197] The server updates the database based on the received inventory results. The latest asset information is reflected in the database, which is then used for future predictions and planning.

[0198] Step 10:

[0199] The server calculates the optimal timing for asset decommissioning based on the predictions of the AI ​​model. This includes a process that uses a deterioration prediction model to identify which assets need to be decommissioned and when.

[0200] Step 11:

[0201] The server notifies the user of the timing for removal. It generates a notification message and delivers it to the terminal. This helps the user perform the removal process at the appropriate time.

[0202] Step 12:

[0203] The user receives a notification and carries out the removal process. Following the notification, they physically remove the unwanted assets and enter the results (removal date, method, etc.) into the terminal.

[0204] Step 13:

[0205] The terminal sends the removal results entered by the user to the server. The collected data is sent to the server via the network, and the server receives it.

[0206] Step 14:

[0207] The server updates the database based on the received disposal results. The database is updated with the latest information, which is then reflected in future forecasts and plans. This cycle continuously improves asset management, leading to greater accuracy and efficiency.

[0208] (Example 2)

[0209] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0210] Traditional asset management systems often involve manual asset data management and inventory processes, resulting in low efficiency and a high risk of human error. Furthermore, they fail to consider user emotional states, leading to uneven workload distribution and an unoptimized user work environment.

[0211] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0212] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, and means for analyzing user emotion data acquired from the terminal's camera and microphone and adjusting the inventory schedule and notification method based on the results. This improves the efficiency of asset management, reduces human error, and provides an optimal work environment that takes into account the user's emotional state.

[0213] An "existing database" is a data storage system that has already been built and is in operation by a company or organization.

[0214] "Asset data" refers to information about physical or digital assets owned by a company or organization, including attribute information such as type, quantity, location, and status.

[0215] A "generative AI model" is a type of artificial intelligence that learns from imported data and makes predictions and classifications about new data.

[0216] "Training methods" refer to the process of using training data to improve the performance of a generative AI model.

[0217] An "inventory schedule" is a plan for checking the status and location of assets and managing them appropriately.

[0218] "Terminal" refers to input and output devices such as personal computers, tablets, and smartphones that are operated by the user.

[0219] "Emotional data" refers to information that indicates a user's psychological state, obtained from their facial expressions, voice, and other similar data.

[0220] "Means of analysis" refers to methods for analyzing specific data and interpreting its meaning and trends.

[0221] "Hardware" refers to the physical components that make up computer systems and electronic devices.

[0222] "Software" refers to a set of programs and procedures that run on computers and electronic devices.

[0223] "Methods for updating a database" refer to methods for adding new information to an existing database or modifying existing information.

[0224] "Means of notification" refers to a communication method used to transmit specified information to the user.

[0225] "Optimal disposal timing" is a criterion for measuring the time when an asset is most cost-effective or before it loses its functionality.

[0226] Modes for carrying out the invention

[0227] This invention relates to a system that combines generative AI models and emotion recognition technology to streamline asset management. The following describes the program processing of this system in natural language, along with specific examples.

[0228] The server first retrieves asset data from an existing database. This data includes asset type, quantity, location, status, etc. The data is retrieved using SQL queries and converted to CSV or JSON format using the Python pandas library. For example, asset data is extracted using the following SQL query: "SELECT FROM assets WHERE status = 'active';".

[0229] Next, the server trains a generative AI model based on the imported data. This training uses Scikit-learn's Random Forest or TENSORFLOW® neural networks. The accuracy is evaluated by dividing the dataset into training and validation datasets, and hyperparameters are tuned as needed. For example, the following command is used for training: `RandomForestClassifier.fit(training_data, training_labels)`.

[0230] Once training is complete, the server generates an asset inventory schedule based on the predictions of the generative AI model. This schedule is designed to focus on assets with high depreciation risk and high importance. The generated schedule is saved in JSON format and delivered to the terminal. For example, a schedule saved as "schedule.json" is delivered to the terminal.

[0231] Furthermore, the device collects user emotion data through its camera and microphone while the user is performing inventory tasks. It uses the OpenCV library for facial recognition and Google's Speech-to-Text API for speech emotion analysis. The acquired emotion data is sent to the server in real time and analyzed by an emotion engine.

[0232] Users verify assets based on an inventory schedule. The verification results are entered from a terminal and sent to the server. For example, a user might fill out a form on a web application and press the "submit" button. This data is received by the server, and the database is updated.

[0233] Furthermore, the server calculates and notifies the user of the optimal timing for asset disposal based on the AI ​​model's predictions. Upon receiving the notification, the user performs the asset disposal task, inputs the results into their terminal, and sends them to the server. The server receives the disposal results and updates its database. This information is then used for future predictions and planning.

[0234] Specific example

[0235] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predicting that old air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes the user's facial expressions and voice to analyze their emotions. For example, if the user is tired, the workload can be adjusted to reduce it.

[0236] Inventory results are entered into a terminal and the data is sent to the server. The server reflects the results and notifies the user when the time for decommissioning is approaching. Based on the notification, the user decommissions old air conditioning equipment, enters the results into the terminal, and sends them to the server. The server updates the database and uses it for future predictions and planning. This system significantly improves the efficiency of asset management and reduces labor and costs. In addition, the use of an emotional engine reduces the workload on the user and provides a more comfortable working environment.

[0237] Examples of prompts to input into a generative AI model

[0238] 1. "Predict the deterioration state of the air conditioning equipment."

[0239] 2. "Please generate the schedule for the next inventory count."

[0240] 3. "Adjust the workload based on the user's fatigue level."

[0241] By leveraging such generative AI models and emotion engines, asset management processes are streamlined, and the user's work environment is optimized. This allows companies and organizations to achieve digitized asset management while minimizing asset degradation and operational burden.

[0242] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0243] Step 1:

[0244] First, the server retrieves asset data from an existing database. The input here is an SQL query and database connection information, and the output is asset data (type, quantity, location, status, etc.). This allows the server to obtain the necessary data for use in the next step. Specifically, it executes the SQL query "SELECT FROM assets WHERE status = 'active';".

[0245] Step 2:

[0246] Next, the server imports the acquired asset data and converts it to CSV or JSON format. The input here is the asset data acquired in step 1, and the output is a data file converted to CSV or JSON format. Specifically, this is done by executing a command like "df.to_csv('assets.csv')" using the Python pandas library.

[0247] Step 3:

[0248] The server trains a generative AI model based on the imported data. The input here is asset data in CSV or JSON format, and the output is the trained generative AI model. Specifically, it executes the command "RandomForestClassifier.fit(training_data, training_labels)" using Scikit-learn's random forest algorithm. The training data and validation data are also split in this step.

[0249] Step 4:

[0250] Once the model training is complete, the server generates an asset inventory schedule based on the AI ​​model's predictions. The input here is the trained generative AI model and imported asset data, and the output is the generated inventory schedule. Specifically, it executes "prediction = model.predict(asset_data)" and saves the result to a JSON file.

[0251] Step 5:

[0252] The generated inventory schedule is delivered from the server to the terminal. The input here is the inventory schedule in JSON format, and the output is the completion of schedule delivery to the terminal. Specifically, the data is sent to the terminal using an HTTP request.

[0253] Step 6:

[0254] The device collects emotional data through the camera and microphone while the user performs inventory tasks. The input here is the user's facial expressions and voice, and the output is analyzed emotional data. Specifically, it uses the OpenCV library and Google's Speech-to-Text API. Camera video is acquired using "cv2.VideoCapture(0)", and audio data is analyzed in real time.

[0255] Step 7:

[0256] The device sends the acquired emotion data to the server in real time. The input here is the analyzed emotion data, and the output is the completion of data transmission to the server. Specifically, it sends the emotion data to the server using an HTTP POST request.

[0257] Step 8:

[0258] The user verifies assets based on the inventory schedule. Inputs here are the inventory schedule and the actual asset status, while output is the verification results. The user checks the status and quantity of each asset and enters this information into a dedicated form on the terminal.

[0259] Step 9:

[0260] The terminal sends the inventory results to the server. The input here is the inventory results entered by the user, and the output is the completion of the data transmission to the server. Specifically, the inventory results are sent using an HTTP POST request.

[0261] Step 10:

[0262] The server reflects the received inventory results in the database. The input here is the inventory results, and the output is the updated database. Specifically, it executes an SQL query such as "UPDATE assets SET status = 'checked' WHERE asset_id = ?".

[0263] Step 11:

[0264] The server calculates and notifies the user of the optimal timing for asset removal based on the prediction results. The input here is the prediction result of a generative AI model, and the output is a notification to the user. Specifically, the function "notify(user_id, 'Asset ID: 123 needs to be removed')" is used.

[0265] Step 12:

[0266] The user performs asset disposal work based on the notification, inputs the results into the terminal, and sends them to the server. The input here is the user's disposal result, and the output is the completion of data transmission to the server.

[0267] Step 13:

[0268] The server updates the database with the received removal results. The input here is the removal results, and the output is the database with the removals reflected. Specifically, it executes the SQL query "DELETE FROM assets WHERE asset_id = ?".

[0269] This series of steps will improve the efficiency of asset management and reduce the workload on users.

[0270] (Application Example 2)

[0271] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0272] Traditional asset management systems have problems such as difficulty in efficiently predicting asset deterioration and disposal timing, resulting in a high burden on workers. Furthermore, there is a lack of means to grasp asset status in real time, hindering improvements in the efficiency and accuracy of asset management. Additionally, scheduling and notification adjustments that do not take into account the emotional state of workers are not taken into account, further hindering efforts to reduce the workload.

[0273] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0274] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, means for collecting information on equipment and parts acquired by industrial sensors in real time and transmitting it to the server, and means for recognizing the emotions of workers using an emotion engine and adjusting the inventory schedule and notification method based on that data. This improves the efficiency and accuracy of asset management and reduces the burden on workers.

[0275] An "existing database" refers to a database already in operation within the system that holds information about assets.

[0276] "Asset data" refers to a collection of data that includes detailed information about assets, such as type, quantity, location, and condition.

[0277] A "generative AI model" is an artificial intelligence model used to predict the state of an asset based on imported data.

[0278] "Training methods" refer to the techniques and methods used to perform the processes necessary to improve the accuracy of generative AI models using data.

[0279] An "inventory schedule" is a plan that outlines the timing and sequence for verifying and inspecting assets.

[0280] A "terminal" refers to a device or equipment used by workers that can receive schedules and input results.

[0281] "Imported data" refers to asset data retrieved from an existing database and incorporated into the system.

[0282] "CSV or JSON" is a type of file format used for data exchange and storage.

[0283] "Data splitting means" refers to methods and techniques for splitting into a training dataset and a validation dataset.

[0284] "Industrial sensor" is a sensor installed in a factory or work site for collecting information on equipment and parts in real time.

[0285] "Emotion engine" is a technology or system for analyzing the expressions and voices of workers and recognizing their emotional states.

[0286] "Generated inventory schedule" is an inventory plan automatically generated based on the prediction results of an AI model.

[0287] "Receiving means" refers to technologies and methods for receiving and managing data transmitted from a terminal.

[0288] "Notifying means" refers to methods and technologies for conveying information from the system to workers.

[0289] The present invention relates to an asset management and work support system within a factory. This system retrieves asset data from an existing database and uses a generative AI model to predict the status of assets. The aim is to optimize the inventory and maintenance schedules of assets and reduce the burden on workers.

[0290] First, the server periodically retrieves asset data from an existing database. This data includes asset type, quantity, location, and status. This information is imported using SQL queries and then converted to CSV or JSON format. Data processing libraries such as Pandas are used for this process.

[0291] Subsequently, the server trains a generative AI model based on the imported data. Specifically, it splits the data into a training dataset and a validation dataset, and trains the model using machine learning libraries such as TensorFlow or PyTorch. Random forests and neural networks are suitable algorithms to use. The accuracy of the model is evaluated and improved through hyperparameter tuning.

[0292] Once the model training is complete, the server generates an asset inventory schedule based on the model's predictions. This schedule is designed to focus on assets at high risk of deterioration or those of high importance, and is then delivered to the terminals.

[0293] Furthermore, the server is equipped with an emotion engine that recognizes the worker's emotions in real time. When a worker performs inventory work, the emotion engine uses the camera and microphone on the terminal (e.g., smartphone or tablet) to identify emotions from their facial expressions and voice. This emotion data is transmitted to the server in real time, and the inventory schedule and notification methods are adjusted accordingly.

[0294] When a user performs tasks based on an inventory schedule, the results are entered into the terminal and sent to the server. The server receives these inventory results and updates the database. Similarly, when a user performs asset disposal tasks based on the disposal timing, the results are also entered into the terminal and sent to the server. The server updates the database again and uses this as feedback to improve the prediction accuracy of the model.

[0295] As a concrete example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predictions that older machinery and equipment are deteriorating and will need to be decommissioned the following month. Based on this information, the server generates an inventory schedule and decommissioning timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes their facial expressions and voice to analyze their emotions. For example, if fatigue is detected, the workload can be adjusted.

[0296] Example of a prompt:

[0297] "For training data for the asset management system within the factory, please enter the type, location, condition, quantity, and maintenance history of each piece of equipment. Also, please specify the features and target variables to be used for predicting the equipment's condition."

[0298] This system is expected to improve the efficiency of asset management and reduce labor costs and expenses. Furthermore, by utilizing an emotion engine, it is possible to reduce the burden on workers and improve the comfort of the work environment.

[0299] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0300] Step 1: Data Acquisition

[0301] The server retrieves asset data from an existing database. It uses SQL queries to access the database and retrieve information such as asset type, quantity, location, and status. The imported data is then converted to CSV or JSON format. Libraries such as Pandas are used for this conversion. The retrieved data is temporarily stored in storage for use in subsequent processing.

[0302] Input: Existing database

[0303] Output: Asset data in CSV or JSON format

[0304] Step 2: Training of the generative AI model

[0305] The server trains a generative AI model based on the imported asset data. First, the data is split into a training dataset and a validation dataset. Next, algorithms such as random forest and neural network are applied using machine learning libraries like TensorFlow and PyTorch to train the model. Hyperparameters are tuned to evaluate and improve the accuracy of the model.

[0306] Input: Asset data in CSV or JSON format

[0307] Output: Trained AI model

[0308] Step 3: Generation of an inventory schedule

[0309] The server generates an inventory schedule for assets based on the prediction results of the trained AI model. This schedule is assembled prioritizing assets with a high deterioration risk or high importance. The generated schedule is sent to the terminal.

[0310] Input: Trained AI model

[0311] Output: Inventory schedule

[0312] Step 4: Distribution of the schedule

[0313] The server distributes the generated inventory schedule to the terminal. The terminal displays this schedule as guidelines for workers when performing inventory work.

[0314] Input: Inventory schedule

[0315] Output: Schedule displayed on the terminal

[0316] Step 5: Receive inventory results and update the database.

[0317] After completing the task, the user enters the inventory results into a terminal. The terminal sends the results to the server, which updates the database. This updated data is then used to train the next AI model.

[0318] Input: Inventory results entered from the terminal.

[0319] Output: Updated database

[0320] Step 6: Emotion recognition by the emotion engine

[0321] The server recognizes workers' emotions in real time through an emotion engine. Using cameras and microphones installed on the terminals, it analyzes workers' facial expressions and voices to acquire emotion data. This data is sent to the server, and inventory schedules and notification methods are adjusted accordingly.

[0322] Input: Emotional data obtained from the device

[0323] Output: Adjusted inventory schedule and notification methods

[0324] Step 7: Notification of the optimal timing for asset disposal

[0325] The server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the worker. The user, upon receiving the notification, carries out the disposal work and enters the results into their terminal.

[0326] Input: AI model prediction results

[0327] Output: Notification of disposal timing

[0328] Step 8: Receive the removal results and update the database.

[0329] After the user completes the asset disposal process, they input the results into a terminal. The terminal sends the results to a server, which updates the database. This data will be used to train future AI models.

[0330] Input: Disposal results entered from the terminal

[0331] Output: Updated database

[0332] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0333] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0334] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0335] [Second Embodiment]

[0336] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0337] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0338] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0339] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0340] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0341] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0342] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0343] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0344] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0345] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0346] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0347] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0348] This invention provides a method for streamlining asset management in SB / WCP systems using generative AI, thereby reducing the time and costs associated with monthly inventory and asset disposal operations. The system's program processing is explained below in natural language, followed by a specific operational example.

[0349] First, the server retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. This data is periodically imported using SQL queries or other data retrieval methods. The imported data is converted to an internal data format (e.g., CSV or JSON) and processed efficiently.

[0350] Next, the server trains a generative AI model based on the imported data. This AI model is designed to predict asset usage frequency and degradation, and uses machine learning algorithms (e.g., random forests or neural networks). Training involves splitting the data into training and validation datasets and tuning hyperparameters to evaluate and improve the model's accuracy.

[0351] Once the AI ​​model is trained, the server generates an asset inventory schedule based on the model's predictions. The inventory aims to prioritize the identification of assets that are at risk of deterioration or are of high importance. The generated schedule is delivered to the terminal and serves as a guideline for the inventory work performed by the user.

[0352] Users use a terminal to check the inventory schedule and perform the verification of designated assets. During the inventory, users input the actual quantity and condition of assets into the terminal and send the inventory results to the server. The server receives these results, updates its database, and uses it to generate forecasts and schedules for the next inventory.

[0353] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions. By notifying users of the optimal disposal timing before critical assets deteriorate and become unusable, proactive management becomes possible. Users receive notifications, perform the actual disposal work, and input the results into their terminals. The server then receives the disposal results and updates its database.

[0354] Specific example:

[0355] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training predicts that older air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and distributes it to the terminal. The user checks the inventory schedule and performs the inventory work. The inventory results are entered into the terminal and sent to the server. The server reflects the results and notifies the user of the removal timing. Based on the notification, the user removes the old air conditioning equipment and enters the removal results. Finally, the server makes predictions and plans for the following month based on the updated data.

[0356] This entire process significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, repeated training and schedule generation improve the accuracy of the AI ​​model, enabling increasingly efficient asset management.

[0357] The following describes the processing flow.

[0358] Step 1:

[0359] The server retrieves asset data from an existing database. Specifically, the server periodically executes SQL queries to extract the latest asset data from the database and convert it into an internal data format (e.g., CSV or JSON).

[0360] Step 2:

[0361] The server trains a generative AI model based on the imported data. Specifically, the server splits the data into a training dataset and a validation dataset, and trains the model using a machine learning algorithm (e.g., random forest, neural network). If there are any deficiencies in the training results, it tunes the hyperparameters and retrains the model.

[0362] Step 3:

[0363] The server generates an asset inventory schedule based on a trained AI model. Specifically, it creates a schedule that prioritizes the inventory of assets with a high risk of deterioration or those of high importance, based on the model's predictions.

[0364] Step 4:

[0365] The server distributes the generated inventory schedule to the terminals. Specifically, it sends the schedule file, converted to a format usable by the terminals, over the network, making it accessible to the users.

[0366] Step 5:

[0367] Users perform asset verification tasks based on the inventory schedule. Specifically, they check the condition and quantity of assets on-site according to the inventory schedule and input the results into their handheld terminals.

[0368] Step 6:

[0369] The terminal sends the inventory results entered by the user to the server. Specifically, it sends the input data to the server via the network, and the server stores the received data in its internal database.

[0370] Step 7:

[0371] The server updates the database based on the inventory results. Specifically, it evaluates the received inventory results and updates the existing asset database with the latest information.

[0372] Step 8:

[0373] The server calculates the optimal timing for asset decommissioning based on the updated database and predictions from the AI ​​model. Specifically, it uses a trained AI model to predict deterioration and identify assets that need to be decommissioned and when.

[0374] Step 9:

[0375] The server notifies the user of the timing for disposal. Specifically, it generates a notification message that lists the assets that need to be disposed of and the timing for disposal, and delivers this message to the terminal.

[0376] Step 10:

[0377] The user receives a notification and carries out the removal work. Specifically, they remove the unwanted assets on-site according to the notification and enter the results (removal date, method, etc.) into the terminal.

[0378] Step 11:

[0379] The terminal sends the user-entered removal results to the server. Specifically, it sends the input data to the server via the network, and the server stores the received data in its internal database.

[0380] Step 12:

[0381] The server updates the database based on the disposal results. Specifically, it evaluates the received disposal results and updates the existing asset database with the latest information.

[0382] (Example 1)

[0383] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0384] Traditional asset management systems required manual inventory and disposal processes, consuming considerable effort and time. Furthermore, accurately predicting asset deterioration and usage frequency was difficult, leading to decreased asset management efficiency. This could potentially result in increased costs due to asset deterioration and improper management.

[0385] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0386] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for dividing the training into a training dataset and a validation dataset and adjusting hyperparameters, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, and means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database. This significantly improves the efficiency of asset management and reduces the man-hours and costs of the work.

[0387] An "existing database" is a collection of information that has already been built and is in operation, and is a system in which data related to assets is stored.

[0388] "Asset data" refers to information about physical or digital assets owned by a company or organization, including attributes such as type, quantity, location, and status.

[0389] A "generative AI model" is a program that uses machine learning algorithms to learn patterns and rules from given data, and is used to predict the deterioration status and usage frequency of assets.

[0390] A "training dataset" is a collection of data used by generative AI models for learning, and it serves as the foundational data for giving the model predictive capabilities.

[0391] A "validation dataset" is a collection of data used to evaluate the prediction accuracy of generative AI models, and is managed separately from the training dataset.

[0392] "Hyperparameters" are settings or adjustment items that affect the performance and results of a machine learning model, and are adjusted to optimize the model.

[0393] An "inventory schedule" is a plan or schedule for verifying assets, and is created by prioritizing tasks based on the condition of the assets and their risk of deterioration.

[0394] A "terminal" refers to a computer or mobile device used by a user, and is hardware used for sending and receiving data with a system.

[0395] "Decommissioning timing" refers to the optimal time to dispose of or transfer a particular asset before it becomes unusable, and is calculated for preventative management purposes.

[0396] This invention provides an efficient asset management system using a generative AI model. The overall system processing flow and specific operational examples are described below.

[0397] First, the server retrieves asset data from an existing database. This data includes attribute information such as asset type, quantity, location, and status. The data is periodically imported into the server using SQL queries or other data retrieval methods.

[0398] Next, the server converts the acquired data into an internal data format (e.g., CSV or JSON). This conversion is performed using libraries such as Python's Pandas library, and the converted data is stored in storage for efficient processing.

[0399] The server trains a generative AI model based on the imported data. Machine learning algorithms (e.g., random forests or neural networks) are used for training. The server splits the data into training and validation datasets and tunes the hyperparameters. This process improves the prediction accuracy of the AI ​​model.

[0400] Once the AI ​​model is trained, the server generates an asset inventory schedule based on the model's predictions. The generated schedule prioritizes the verification tasks based on the asset's deterioration risk and importance. The generated schedule is delivered to the terminal and serves as a guideline for the inventory work performed by the user.

[0401] Users use a terminal to check the inventory schedule and perform verification tasks for designated assets. During the process, users input the actual quantity and condition of the assets into the terminal and send the results to the server. The server updates its database based on the received inventory results and uses this information to generate forecasts and schedules for the next inventory.

[0402] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions. Before important assets deteriorate and become unusable, the server notifies the user of this information. Based on the notification, the user performs the disposal work and inputs the results into their terminal, sending them to the server. After receiving the disposal results, the server updates its database.

[0403] Specific example

[0404] At the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training predicts that older air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and distributes it to the terminal. The user checks the inventory schedule and performs the inventory work. The inventory results are entered into the terminal and sent to the server. The server reflects the results and notifies the user of the removal timing. Based on the notification, the user removes the old air conditioning equipment and enters the removal results. Finally, the server makes predictions and plans for the following month based on the updated data.

[0405] This process significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, repeated training and schedule generation improve the accuracy of the AI ​​model, enabling increasingly efficient asset management.

[0406] Example of a prompt

[0407] An example of a prompt statement to illustrate this series of processes is as follows:

[0408] "In the asset management system, prompts are used to generate specific inventory schedules and disposal timings:

[0409] The server retrieves data from the asset database, converts it to CSV format, and inputs it as training data into a generative AI model. This model uses a random forest algorithm to predict the deterioration status of assets. Based on the predictions, the server generates the inventory schedule and disposal timing for the current month and distributes it to the terminal. After the inventory is completed, the results are sent from the terminal to the server, and the database is updated.

[0410] This clarifies the specific embodiments of the invention, enabling implementers to use the invention appropriately and efficiently.

[0411] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0412] Step 1: Obtain asset data

[0413] The server periodically retrieves asset data from an existing database. The input includes attribute information such as asset type, quantity, location, and status. This data is retrieved using SQL queries or other means. Specifically, the server executes an SQL query like "SELECT FROM assets WHERE month='current_month'". The output of this query is imported into the server.

[0414] Step 2: Data conversion and saving

[0415] The server converts the acquired asset data into an internal standard format (e.g., CSV or JSON). The data acquired in step 1 is used as input. Specifically, the server uses the Python Pandas library to convert the data into a DataFrame and saves it to a file such as "assets_data.csv". As output, the converted data is saved to storage in a format that can be processed efficiently.

[0416] Step 3: Training the AI ​​model

[0417] The server trains a generative AI model based on the imported data. The data saved in step 2 is used as input. Specifically, the server uses the Scikit-learn library to split the data into a training dataset and a validation dataset, and then trains a random forest model. Hyperparameter tuning is also performed. The output is a trained AI model.

[0418] Step 4: Generate and distribute the inventory schedule.

[0419] The server generates an inventory schedule based on the predictions of the trained AI model. The input includes the AI ​​model and asset data obtained in step 3. Specifically, the server uses the model to predict asset deterioration and creates an inventory schedule based on the results. The output is the generated inventory schedule, which is then delivered to the terminal.

[0420] Step 5: Conduct inventory work

[0421] The user uses a terminal to check the inventory schedule and perform the verification of the specified assets. The input includes the inventory schedule distributed in step 4. Specifically, the user checks the quantity and condition of assets such as air conditioning equipment according to the schedule and inputs the results into the terminal. The output is the verification results. These results are sent from the terminal to the server.

[0422] Step 6: Update the database of inventory results

[0423] The server receives the inventory results sent from the terminal and updates the database. The input used is the confirmation result obtained by the user in step 5. Specifically, the server integrates the new data into the existing database and uses it to generate future predictions and schedules. The output is the updated database.

[0424] Step 7: Planning and Implementing Asset Disposal

[0425] The server calculates and notifies the optimal timing for asset decommissioning based on the prediction results of the AI ​​model. Inputs include the trained AI model obtained in step 3 and the latest data. Specifically, the server predicts the deterioration status of old air conditioning equipment and calculates the decommissioning timing. This information is notified to the terminal. An output is generated: a decommissioning notification.

[0426] Step 8: Implement asset disposal and enter the results.

[0427] The user performs the appropriate asset removal work based on notifications from the server. The input includes the removal timing notified in step 7. Specifically, the user removes the old air conditioning equipment and inputs the result into the terminal. The output is the removal result, which is then sent from the terminal to the server.

[0428] Step 9: Update the database of removal results

[0429] The server receives the decommissioning results sent from the terminal and updates the database. The input used is the decommissioning results obtained in step 8. Specifically, the server integrates the decommissioning data into the existing database and uses it for future forecasts and planning. The output is the updated database. This series of steps enables efficient asset management and significantly reduces labor and costs.

[0430] (Application Example 1)

[0431] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0432] In modern logistics centers, asset management and inventory management are extremely complex and time-consuming tasks. They require a great deal of manual work, are prone to errors, and lack preventative management leads to increased losses due to asset deterioration. Traditional systems struggle to accurately predict asset deterioration and usage frequency, making it difficult to efficiently generate inventory and disposal schedules. To address these challenges, a more efficient and accurate asset management system is needed.

[0433] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0434] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, and means including an application on the terminal that notifies the user of the predicted inventory schedule and disposal timing, and allows the user to confirm and input the work. This makes it possible to accurately predict the state and frequency of use of assets and efficiently generate and notify inventory and disposal schedules, significantly improving the efficiency and accuracy of asset management.

[0435] A "server" is a device that retrieves asset data from an existing database, trains an AI model based on that data, and generates and distributes inventory schedules and disposal timings based on the prediction results.

[0436] "Asset data" refers to data that includes information about the type, quantity, location, condition, and frequency of use of goods and equipment.

[0437] A "generative AI model" is a machine learning model that uses training data to predict the deterioration status and usage frequency of assets, and to calculate appropriate inventory schedules and disposal timings.

[0438] "Training" is the process of using existing data to train a generative AI model so that it can accurately predict the state of an asset.

[0439] An "inventory schedule" is a plan for conducting inventory work at specific times, based on the deterioration status and importance of the assets.

[0440] A "terminal" is a device used by users to check inventory schedules and disposal timings, and to input actual work results.

[0441] "Removal timing" refers to the optimal period for properly removing an asset before it deteriorates and becomes unusable.

[0442] "Inventory results" refer to information about the quantity and condition of assets obtained by the user through actual inventory work.

[0443] A "database" is a database management system that stores asset data and allows servers to retrieve the information they need.

[0444] "User notifications" is a system that communicates information such as inventory schedules and disposal timings to users via their devices.

[0445] This invention is a system for improving the efficiency of asset management in logistics centers. Specifically, it acquires asset data, predicts asset status using a generative AI model, generates and distributes inventory schedules, and notifies users of asset disposal timing. The main components of this system involve servers, terminals, and users.

[0446] The server first retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. The retrieved data is periodically imported and converted into an internal CSV or JSON format. The server then uses this data to train a generative AI model. This AI model is implemented using machine learning libraries such as Scikit-learn, and its hyperparameters are tuned by splitting it into training and validation datasets.

[0447] A trained AI model predicts the deterioration status and usage frequency of assets, and generates an inventory schedule based on these predictions. The generated schedule is delivered to the user's device, which may include a smartphone or tablet. The user uses this device to check the notified inventory schedule and perform the inventory work for the specified assets. The work results are entered into the device and sent to the server, updating the database.

[0448] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the user. The user receives the notification, enters the result of asset disposal into their terminal, and this data is sent to the server, updating the database again.

[0449] As a concrete example, at the beginning of the month, the server retrieves all asset data, and an AI model predicts the deterioration of older forklifts. Based on this prediction, an inventory schedule indicating the need to remove forklifts is generated and delivered to the terminal. The user performs the inventory work, enters the results into the terminal, and sends them to the server, updating the database. Removal is carried out in a similar manner.

[0450] Examples of prompts for a generative AI model:

[0451] Train an AI model to predict asset degradation and usage frequency using the following dataset. The data is provided in CSV format, with each row representing a single asset.

[0452] The column names are as follows: Asset ID, Asset Type, Quantity, Location, Status, Usage Frequency.

[0453] The model should use this data to identify assets that are expected to deteriorate in the following month and predict when they will need to be removed.

[0454] To implement this invention, the server side performs data import, AI model training, schedule generation, notification, and database updates, while the terminal side inputs inventory results and receives notifications. This series of operations significantly improves the efficiency and accuracy of asset management in logistics centers.

[0455] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0456] Step 1:

[0457] The server retrieves asset data from an existing database. It sends SQL queries to the database to extract asset data, which includes asset type, quantity, location, status, etc. It receives asset data from the database as input and obtains raw data from the database as output.

[0458] Step 2:

[0459] The server converts the acquired asset data into an internal format (CSV or JSON). This allows for efficient data processing. It receives raw data from the database as input and obtains converted data as output.

[0460] Step 3:

[0461] The server trains a generative AI model based on the transformed asset data. In this process, the data is split into training and validation datasets, and the model is trained using Scikit-learn's RandomForestRegressor. It accepts asset data in an internal format as input and produces a trained AI model as output.

[0462] Step 4:

[0463] The server uses a trained AI model to predict the deterioration status and usage frequency of assets. This results in the asset status being obtained as a prediction. It receives a trained AI model and asset data in an internal format as input and obtains the prediction result as output.

[0464] Step 5:

[0465] The server generates an asset inventory schedule based on the predictions of an AI model. This is intended to prioritize the identification of assets at risk of deterioration or those of high importance. It takes prediction results as input and obtains an inventory schedule as output.

[0466] Step 6:

[0467] The server distributes the generated inventory schedule to the terminal. This allows users to perform inventory tasks according to the schedule. It receives the inventory schedule as input and receives notifications to the terminal as output.

[0468] Step 7:

[0469] The user uses a terminal to check the inventory schedule and perform the verification of the specified assets. The terminal receives the inventory schedule as input and inputs the inventory results as output.

[0470] Step 8:

[0471] The server receives inventory results entered from the terminal and updates the database. This ensures that the latest asset data is reflected in the database. The system receives inventory results as input and obtains an updated database as output.

[0472] Step 9:

[0473] The server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the user. It receives prediction results as input and outputs a notification of the disposal timing.

[0474] Step 10:

[0475] The user receives a notification, enters the result of asset disposal into the terminal, and sends that data to the server. The user receives a notification of the disposal timing as input and inputs the disposal result into the terminal as output.

[0476] Step 11:

[0477] The server receives the disposal results entered from the terminal and updates the database. This provides the latest asset data to be used for future predictions and planning. It receives disposal results as input and obtains an updated database as output.

[0478] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0479] This invention provides a system that supports users' efficient and effective work by combining an emotion engine with asset management in SB / WCP systems that utilize generative AI. The following describes the program processing of this system in natural language, along with specific examples.

[0480] First, the server retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. This data is periodically imported using SQL queries, etc., and converted to CSV or JSON format for efficient internal processing.

[0481] Next, the server trains a generative AI model based on the imported data. This training involves splitting the data into a training dataset and a validation dataset, evaluating the model's accuracy, and tuning hyperparameters as needed. Possible machine learning algorithms used include random forests and neural networks.

[0482] Once the model training is complete, the server generates an asset inventory schedule based on the AI ​​model's predictions. This schedule is designed to focus on assets with high depreciation risk and those of high importance. The generated schedule is delivered to the terminal for the user to use.

[0483] The server also features an emotion engine to recognize user emotions. When a user performs inventory work, the emotion engine uses the camera and microphone to identify emotions from the user's facial expressions and voice. This emotion data is transmitted to the server in real time, and the server adjusts the way inventory schedules are presented and notifications are sent as needed.

[0484] Users perform asset verification tasks based on an inventory schedule. Once users have completed asset verification, they enter the results into their terminal and send them to the server. The server receives the inventory results, updates the database, and uses them as feedback to improve the prediction accuracy of the model.

[0485] Furthermore, the server calculates and notifies the user of the optimal timing for asset disposal based on the AI ​​model's predictions. When the user receives the notification, performs the asset disposal, and enters the results into their terminal, the server receives the disposal results and updates the database. This entire process ensures that the user can always manage their assets with the latest information.

[0486] Specific example:

[0487] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predicting that old air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes the user's facial expressions and voice to analyze their emotions. For example, if the user is tired, the workload can be adjusted to reduce it.

[0488] The inventory results are entered into a terminal, and the data is sent to the server. The server reflects the results and notifies the user when the time for removal approaches. Based on the notification, the user removes old air conditioning equipment, enters the results into the terminal, and sends them to the server. The server updates the database and uses it for future predictions and planning.

[0489] This system significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, the use of an emotional engine reduces the user's workload, providing a more comfortable work environment.

[0490] The following describes the processing flow.

[0491] Step 1:

[0492] The server retrieves asset data from an existing database. This involves periodically collecting the latest asset information using SQL queries and other data retrieval methods. This data includes asset type, quantity, location, and status.

[0493] Step 2:

[0494] The server trains a generative AI model based on imported asset data. First, it splits the data into a training dataset and a validation dataset. Then, it trains the model using machine learning algorithms such as random forests and neural networks. It evaluates the model's accuracy and tunes hyperparameters as needed.

[0495] Step 3:

[0496] The server generates an asset inventory schedule based on a trained AI model. This includes a process of creating a schedule that prioritizes checking assets at high risk of deterioration or those of high importance.

[0497] Step 4:

[0498] The server distributes the generated inventory schedule to the terminals. It sends the schedule file, converted to a format usable by the terminals, over the network, making it accessible to the users.

[0499] Step 5:

[0500] The device uses an emotion engine to recognize the user's emotions while they are performing inventory tasks. It analyzes the user's facial expressions and voice through the camera and microphone to acquire emotional data.

[0501] Step 6:

[0502] The server receives emotional data sent from the emotion engine and adjusts inventory schedules and notification methods. If the user is fatigued, measures such as reducing the workload are taken.

[0503] Step 7:

[0504] Users perform asset verification according to the inventory schedule. They inspect assets on-site and input their condition and quantity into a terminal.

[0505] Step 8:

[0506] The terminal sends the inventory results entered by the user to the server. The collected data is sent to the server via the network, and the server receives it.

[0507] Step 9:

[0508] The server updates the database based on the received inventory results. The latest asset information is reflected in the database, which is then used for future predictions and planning.

[0509] Step 10:

[0510] The server calculates the optimal timing for asset decommissioning based on the predictions of the AI ​​model. This includes a process that uses a deterioration prediction model to identify which assets need to be decommissioned and when.

[0511] Step 11:

[0512] The server notifies the user of the timing for removal. It generates a notification message and delivers it to the terminal. This helps the user perform the removal process at the appropriate time.

[0513] Step 12:

[0514] The user receives a notification and carries out the removal process. Following the notification, they physically remove the unwanted assets and enter the results (removal date, method, etc.) into the terminal.

[0515] Step 13:

[0516] The terminal sends the removal results entered by the user to the server. The collected data is sent to the server via the network, and the server receives it.

[0517] Step 14:

[0518] The server updates the database based on the received disposal results. The database is updated with the latest information, which is then reflected in future forecasts and plans. This cycle continuously improves asset management, leading to greater accuracy and efficiency.

[0519] (Example 2)

[0520] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0521] Traditional asset management systems often involve manual asset data management and inventory processes, resulting in low efficiency and a high risk of human error. Furthermore, they fail to consider user emotional states, leading to uneven workload distribution and an unoptimized user work environment.

[0522] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0523] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, and means for analyzing user emotion data acquired from the terminal's camera and microphone and adjusting the inventory schedule and notification method based on the results. This improves the efficiency of asset management, reduces human error, and provides an optimal work environment that takes into account the user's emotional state.

[0524] An "existing database" is a data storage system that has already been built and is in operation by a company or organization.

[0525] "Asset data" refers to information about physical or digital assets owned by a company or organization, including attribute information such as type, quantity, location, and status.

[0526] A "generative AI model" is a type of artificial intelligence that learns from imported data and makes predictions and classifications about new data.

[0527] "Training methods" refer to the process of using training data to improve the performance of a generative AI model.

[0528] An "inventory schedule" is a plan for checking the status and location of assets and managing them appropriately.

[0529] "Terminal" refers to input and output devices such as personal computers, tablets, and smartphones that are operated by the user.

[0530] "Emotional data" refers to information that indicates a user's psychological state, obtained from their facial expressions, voice, and other similar data.

[0531] "Means of analysis" refers to methods for analyzing specific data and interpreting its meaning and trends.

[0532] "Hardware" refers to the physical components that make up computer systems and electronic devices.

[0533] "Software" refers to a set of programs and procedures that run on computers and electronic devices.

[0534] "Methods for updating a database" refer to methods for adding new information to an existing database or modifying existing information.

[0535] "Means of notification" refers to a communication method used to transmit specified information to the user.

[0536] "Optimal disposal timing" is a criterion for measuring the time when an asset is most cost-effective or before it loses its functionality.

[0537] Modes for carrying out the invention

[0538] This invention relates to a system that combines generative AI models and emotion recognition technology to streamline asset management. The following describes the program processing of this system in natural language, along with specific examples.

[0539] The server first retrieves asset data from an existing database. This data includes asset type, quantity, location, status, etc. The data is retrieved using SQL queries and converted to CSV or JSON format using the Python pandas library. For example, asset data is extracted using the following SQL query: "SELECT FROM assets WHERE status = 'active';".

[0540] Next, the server trains a generative AI model based on the imported data. This training uses Scikit-learn's Random Forest or Tensorflow's Neural Network. The accuracy is evaluated by splitting the dataset into training and validation datasets, and hyperparameters are tuned as needed. For example, the following command is used for training: `RandomForestClassifier.fit(training_data, training_labels)`.

[0541] Once training is complete, the server generates an asset inventory schedule based on the predictions of the generative AI model. This schedule is designed to focus on assets with high depreciation risk and high importance. The generated schedule is saved in JSON format and delivered to the terminal. For example, a schedule saved as "schedule.json" is delivered to the terminal.

[0542] Furthermore, the device collects user emotion data through its camera and microphone while the user is performing inventory tasks. It uses the OpenCV library for facial recognition and Google's Speech-to-Text API for speech emotion analysis. The acquired emotion data is sent to the server in real time and analyzed by an emotion engine.

[0543] Users verify assets based on an inventory schedule. The verification results are entered from a terminal and sent to the server. For example, a user might fill out a form on a web application and press the "submit" button. This data is received by the server, and the database is updated.

[0544] Furthermore, the server calculates and notifies the user of the optimal timing for asset disposal based on the AI ​​model's predictions. Upon receiving the notification, the user performs the asset disposal task, inputs the results into their terminal, and sends them to the server. The server receives the disposal results and updates its database. This information is then used for future predictions and planning.

[0545] Specific example

[0546] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predicting that old air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes the user's facial expressions and voice to analyze their emotions. For example, if the user is tired, the workload can be adjusted to reduce it.

[0547] Inventory results are entered into a terminal and the data is sent to the server. The server reflects the results and notifies the user when the time for decommissioning is approaching. Based on the notification, the user decommissions old air conditioning equipment, enters the results into the terminal, and sends them to the server. The server updates the database and uses it for future predictions and planning. This system significantly improves the efficiency of asset management and reduces labor and costs. In addition, the use of an emotional engine reduces the workload on the user and provides a more comfortable working environment.

[0548] Examples of prompts to input into a generative AI model

[0549] 1. "Predict the deterioration state of the air conditioning equipment."

[0550] 2. "Please generate the schedule for the next inventory count."

[0551] 3. "Adjust the workload based on the user's fatigue level."

[0552] By leveraging such generative AI models and emotion engines, asset management processes are streamlined, and the user's work environment is optimized. This allows companies and organizations to achieve digitized asset management while minimizing asset degradation and operational burden.

[0553] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0554] Step 1:

[0555] First, the server retrieves asset data from an existing database. The input here is an SQL query and database connection information, and the output is asset data (type, quantity, location, status, etc.). This allows the server to obtain the necessary data for use in the next step. Specifically, it executes the SQL query "SELECT FROM assets WHERE status = 'active';".

[0556] Step 2:

[0557] Next, the server imports the acquired asset data and converts it to CSV or JSON format. The input here is the asset data acquired in step 1, and the output is a data file converted to CSV or JSON format. Specifically, this is done by executing a command like "df.to_csv('assets.csv')" using the Python pandas library.

[0558] Step 3:

[0559] The server trains a generative AI model based on the imported data. The input here is asset data in CSV or JSON format, and the output is the trained generative AI model. Specifically, it executes the command "RandomForestClassifier.fit(training_data, training_labels)" using Scikit-learn's random forest algorithm. The training data and validation data are also split in this step.

[0560] Step 4:

[0561] Once the model training is complete, the server generates an asset inventory schedule based on the AI ​​model's predictions. The input here is the trained generative AI model and imported asset data, and the output is the generated inventory schedule. Specifically, it executes "prediction = model.predict(asset_data)" and saves the result to a JSON file.

[0562] Step 5:

[0563] The generated inventory schedule is delivered from the server to the terminal. The input here is the inventory schedule in JSON format, and the output is the completion of schedule delivery to the terminal. Specifically, the data is sent to the terminal using an HTTP request.

[0564] Step 6:

[0565] The device collects emotional data through the camera and microphone while the user performs inventory tasks. The input here is the user's facial expressions and voice, and the output is analyzed emotional data. Specifically, it uses the OpenCV library and Google's Speech-to-Text API. Camera video is acquired using "cv2.VideoCapture(0)", and audio data is analyzed in real time.

[0566] Step 7:

[0567] The device sends the acquired emotion data to the server in real time. The input here is the analyzed emotion data, and the output is the completion of data transmission to the server. Specifically, it sends the emotion data to the server using an HTTP POST request.

[0568] Step 8:

[0569] The user verifies assets based on the inventory schedule. Inputs here are the inventory schedule and the actual asset status, while output is the verification results. The user checks the status and quantity of each asset and enters this information into a dedicated form on the terminal.

[0570] Step 9:

[0571] The terminal sends the inventory results to the server. The input here is the inventory results entered by the user, and the output is the completion of the data transmission to the server. Specifically, the inventory results are sent using an HTTP POST request.

[0572] Step 10:

[0573] The server reflects the received inventory results in the database. The input here is the inventory results, and the output is the updated database. Specifically, it executes an SQL query such as "UPDATE assets SET status = 'checked' WHERE asset_id = ?".

[0574] Step 11:

[0575] The server calculates and notifies the user of the optimal timing for asset removal based on the prediction results. The input here is the prediction result of a generative AI model, and the output is a notification to the user. Specifically, the function "notify(user_id, 'Asset ID: 123 needs to be removed')" is used.

[0576] Step 12:

[0577] The user performs asset disposal work based on the notification, inputs the results into the terminal, and sends them to the server. The input here is the user's disposal result, and the output is the completion of data transmission to the server.

[0578] Step 13:

[0579] The server updates the database with the received removal results. The input here is the removal results, and the output is the database with the removals reflected. Specifically, it executes the SQL query "DELETE FROM assets WHERE asset_id = ?".

[0580] This series of steps will improve the efficiency of asset management and reduce the workload on users.

[0581] (Application Example 2)

[0582] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0583] Traditional asset management systems have problems such as difficulty in efficiently predicting asset deterioration and disposal timing, resulting in a high burden on workers. Furthermore, there is a lack of means to grasp asset status in real time, hindering improvements in the efficiency and accuracy of asset management. Additionally, scheduling and notification adjustments that do not take into account the emotional state of workers are not taken into account, further hindering efforts to reduce the workload.

[0584] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0585] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, means for collecting information on equipment and parts acquired by industrial sensors in real time and transmitting it to the server, and means for recognizing the emotions of workers using an emotion engine and adjusting the inventory schedule and notification method based on that data. This improves the efficiency and accuracy of asset management and reduces the burden on workers.

[0586] An "existing database" refers to a database already in operation within the system that holds information about assets.

[0587] "Asset data" refers to a collection of data that includes detailed information about assets, such as type, quantity, location, and condition.

[0588] A "generative AI model" is an artificial intelligence model used to predict the state of an asset based on imported data.

[0589] "Training methods" refer to the techniques and methods used to perform the processes necessary to improve the accuracy of generative AI models using data.

[0590] An "inventory schedule" is a plan that outlines the timing and sequence for verifying and inspecting assets.

[0591] A "terminal" refers to a device or equipment used by workers that can receive schedules and input results.

[0592] "Imported data" refers to asset data retrieved from an existing database and incorporated into the system.

[0593] "CSV or JSON" refers to a type of file format used for data exchange and storage.

[0594] "Data splitting methods" refer to methods and techniques for dividing data into training datasets and validation datasets.

[0595] An "industrial sensor" is a sensor installed in factories and work sites to collect information about equipment and parts in real time.

[0596] An "emotion engine" is a technology or system that analyzes a worker's facial expressions and voice to recognize their emotional state.

[0597] The "generated inventory schedule" is an inventory plan that is automatically generated based on the prediction results of the AI ​​model.

[0598] "Means of receiving" refers to the technologies and methods for receiving and managing data transmitted from a terminal.

[0599] "Means of notification" refers to the methods and technologies used to transmit information from a system to workers.

[0600] This invention relates to an asset management and work support system for a factory. This system acquires asset data from an existing database and predicts the asset status using a generative AI model. The aim is to optimize asset inventory and maintenance schedules, thereby reducing the burden on workers.

[0601] First, the server periodically retrieves asset data from an existing database. This data includes asset type, quantity, location, and status. This information is imported using SQL queries and then converted to CSV or JSON format. Data processing libraries such as Pandas are used for this process.

[0602] Subsequently, the server trains a generative AI model based on the imported data. Specifically, it splits the data into a training dataset and a validation dataset, and trains the model using machine learning libraries such as TensorFlow or PyTorch. Random forests and neural networks are suitable algorithms to use. The accuracy of the model is evaluated and improved through hyperparameter tuning.

[0603] Once the model training is complete, the server generates an asset inventory schedule based on the model's predictions. This schedule is designed to focus on assets at high risk of deterioration or those of high importance, and is then delivered to the terminals.

[0604] Furthermore, the server is equipped with an emotion engine that recognizes the worker's emotions in real time. When a worker performs inventory work, the emotion engine uses the camera and microphone on the terminal (e.g., smartphone or tablet) to identify emotions from their facial expressions and voice. This emotion data is transmitted to the server in real time, and the inventory schedule and notification methods are adjusted accordingly.

[0605] When a user performs tasks based on an inventory schedule, the results are entered into the terminal and sent to the server. The server receives these inventory results and updates the database. Similarly, when a user performs asset disposal tasks based on the disposal timing, the results are also entered into the terminal and sent to the server. The server updates the database again and uses this as feedback to improve the prediction accuracy of the model.

[0606] As a concrete example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predictions that older machinery and equipment are deteriorating and will need to be decommissioned the following month. Based on this information, the server generates an inventory schedule and decommissioning timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes their facial expressions and voice to analyze their emotions. For example, if fatigue is detected, the workload can be adjusted.

[0607] Example of a prompt:

[0608] "For training data for the asset management system within the factory, please enter the type, location, condition, quantity, and maintenance history of each piece of equipment. Also, please specify the features and target variables to be used for predicting the equipment's condition."

[0609] This system is expected to improve the efficiency of asset management and reduce labor costs and expenses. Furthermore, by utilizing an emotion engine, it is possible to reduce the burden on workers and improve the comfort of the work environment.

[0610] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0611] Step 1: Data Acquisition

[0612] The server retrieves asset data from an existing database. It uses SQL queries to access the database and retrieve information such as asset type, quantity, location, and status. The imported data is then converted to CSV or JSON format. Libraries such as Pandas are used for this conversion. The retrieved data is temporarily stored in storage for use in subsequent processing.

[0613] Input: Existing database

[0614] Output: Asset data in CSV or JSON format

[0615] Step 2: Training the generative AI model

[0616] The server trains generative AI models based on imported asset data. First, it splits the data into training and validation datasets. Next, it uses machine learning libraries such as TensorFlow and PyTorch to apply algorithms such as random forests and neural networks to train the models. Finally, it tunes hyperparameters to evaluate and improve the model's accuracy.

[0617] Input: Asset data in CSV or JSON format

[0618] Output: Trained AI model

[0619] Step 3: Generate an inventory schedule

[0620] The server generates an asset inventory schedule based on the predictions of a trained AI model. This schedule prioritizes assets with a high risk of deterioration or those of high importance. The generated schedule is then sent to the terminal.

[0621] Input: Trained AI model

[0622] Output: Inventory schedule

[0623] Step 4: Schedule Distribution

[0624] The server distributes the generated inventory schedule to the terminal. The terminal displays this schedule as a guideline for workers to carry out the inventory work.

[0625] Input: Inventory schedule

[0626] Output: Schedule displayed on the terminal

[0627] Step 5: Receive inventory results and update the database.

[0628] After completing the task, the user enters the inventory results into a terminal. The terminal sends the results to the server, which updates the database. This updated data is then used to train the next AI model.

[0629] Input: Inventory results entered from the terminal.

[0630] Output: Updated database

[0631] Step 6: Emotion recognition by the emotion engine

[0632] The server recognizes workers' emotions in real time through an emotion engine. Using cameras and microphones installed on the terminals, it analyzes workers' facial expressions and voices to acquire emotion data. This data is sent to the server, and inventory schedules and notification methods are adjusted accordingly.

[0633] Input: Emotional data obtained from the device

[0634] Output: Adjusted inventory schedule and notification methods

[0635] Step 7: Notification of the optimal timing for asset disposal

[0636] The server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the worker. The user, upon receiving the notification, carries out the disposal work and enters the results into their terminal.

[0637] Input: AI model prediction results

[0638] Output: Notification of disposal timing

[0639] Step 8: Receive the removal results and update the database.

[0640] After the user completes the asset disposal process, they input the results into a terminal. The terminal sends the results to a server, which updates the database. This data will be used to train future AI models.

[0641] Input: Disposal results entered from the terminal

[0642] Output: Updated database

[0643] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0644] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0645] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0646] [Third Embodiment]

[0647] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0648] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0649] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0650] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0651] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0652] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0653] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0654] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0655] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0656] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0657] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0658] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0659] This invention provides a method for streamlining asset management in SB / WCP systems using generative AI, thereby reducing the time and costs associated with monthly inventory and asset disposal operations. The system's program processing is explained below in natural language, followed by a specific operational example.

[0660] First, the server retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. This data is periodically imported using SQL queries or other data retrieval methods. The imported data is converted to an internal data format (e.g., CSV or JSON) and processed efficiently.

[0661] Next, the server trains a generative AI model based on the imported data. This AI model is designed to predict asset usage frequency and degradation, and uses machine learning algorithms (e.g., random forests or neural networks). Training involves splitting the data into training and validation datasets and tuning hyperparameters to evaluate and improve the model's accuracy.

[0662] Once the AI ​​model is trained, the server generates an asset inventory schedule based on the model's predictions. The inventory aims to prioritize the identification of assets that are at risk of deterioration or are of high importance. The generated schedule is delivered to the terminal and serves as a guideline for the inventory work performed by the user.

[0663] Users use a terminal to check the inventory schedule and perform the verification of designated assets. During the inventory, users input the actual quantity and condition of assets into the terminal and send the inventory results to the server. The server receives these results, updates its database, and uses it to generate forecasts and schedules for the next inventory.

[0664] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions. By notifying users of the optimal disposal timing before critical assets deteriorate and become unusable, proactive management becomes possible. Users receive notifications, perform the actual disposal work, and input the results into their terminals. The server then receives the disposal results and updates its database.

[0665] Specific example:

[0666] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training predicts that older air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and distributes it to the terminal. The user checks the inventory schedule and performs the inventory work. The inventory results are entered into the terminal and sent to the server. The server reflects the results and notifies the user of the removal timing. Based on the notification, the user removes the old air conditioning equipment and enters the removal results. Finally, the server makes predictions and plans for the following month based on the updated data.

[0667] This entire process significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, repeated training and schedule generation improve the accuracy of the AI ​​model, enabling increasingly efficient asset management.

[0668] The following describes the processing flow.

[0669] Step 1:

[0670] The server retrieves asset data from an existing database. Specifically, the server periodically executes SQL queries to extract the latest asset data from the database and convert it into an internal data format (e.g., CSV or JSON).

[0671] Step 2:

[0672] The server trains a generative AI model based on the imported data. Specifically, the server splits the data into a training dataset and a validation dataset, and trains the model using a machine learning algorithm (e.g., random forest, neural network). If there are any deficiencies in the training results, it tunes the hyperparameters and retrains the model.

[0673] Step 3:

[0674] The server generates an asset inventory schedule based on a trained AI model. Specifically, it creates a schedule that prioritizes the inventory of assets with a high risk of deterioration or those of high importance, based on the model's predictions.

[0675] Step 4:

[0676] The server distributes the generated inventory schedule to the terminals. Specifically, it sends the schedule file, converted to a format usable by the terminals, over the network, making it accessible to the users.

[0677] Step 5:

[0678] Users perform asset verification tasks based on the inventory schedule. Specifically, they check the condition and quantity of assets on-site according to the inventory schedule and input the results into their handheld terminals.

[0679] Step 6:

[0680] The terminal sends the inventory results entered by the user to the server. Specifically, it sends the input data to the server via the network, and the server stores the received data in its internal database.

[0681] Step 7:

[0682] The server updates the database based on the inventory results. Specifically, it evaluates the received inventory results and updates the existing asset database with the latest information.

[0683] Step 8:

[0684] The server calculates the optimal timing for asset decommissioning based on the updated database and predictions from the AI ​​model. Specifically, it uses a trained AI model to predict deterioration and identify assets that need to be decommissioned and when.

[0685] Step 9:

[0686] The server notifies the user of the timing for disposal. Specifically, it generates a notification message that lists the assets that need to be disposed of and the timing for disposal, and delivers this message to the terminal.

[0687] Step 10:

[0688] The user receives a notification and carries out the removal work. Specifically, they remove the unwanted assets on-site according to the notification and enter the results (removal date, method, etc.) into the terminal.

[0689] Step 11:

[0690] The terminal sends the user-entered removal results to the server. Specifically, it sends the input data to the server via the network, and the server stores the received data in its internal database.

[0691] Step 12:

[0692] The server updates the database based on the disposal results. Specifically, it evaluates the received disposal results and updates the existing asset database with the latest information.

[0693] (Example 1)

[0694] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0695] Traditional asset management systems required manual inventory and disposal processes, consuming considerable effort and time. Furthermore, accurately predicting asset deterioration and usage frequency was difficult, leading to decreased asset management efficiency. This could potentially result in increased costs due to asset deterioration and improper management.

[0696] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0697] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for dividing the training into a training dataset and a validation dataset and adjusting hyperparameters, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, and means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database. This significantly improves the efficiency of asset management and reduces the man-hours and costs of the work.

[0698] An "existing database" is a collection of information that has already been built and is in operation, and is a system in which data related to assets is stored.

[0699] "Asset data" refers to information about physical or digital assets owned by a company or organization, including attributes such as type, quantity, location, and status.

[0700] A "generative AI model" is a program that uses machine learning algorithms to learn patterns and rules from given data, and is used to predict the deterioration status and usage frequency of assets.

[0701] A "training dataset" is a collection of data used by generative AI models for learning, and it serves as the foundational data for giving the model predictive capabilities.

[0702] A "validation dataset" is a collection of data used to evaluate the prediction accuracy of generative AI models, and is managed separately from the training dataset.

[0703] "Hyperparameters" are settings or adjustment items that affect the performance and results of a machine learning model, and are adjusted to optimize the model.

[0704] An "inventory schedule" is a plan or schedule for verifying assets, and is created by prioritizing tasks based on the condition of the assets and their risk of deterioration.

[0705] A "terminal" refers to a computer or mobile device used by a user, and is hardware used for sending and receiving data with a system.

[0706] "Decommissioning timing" refers to the optimal time to dispose of or transfer a particular asset before it becomes unusable, and is calculated for preventative management purposes.

[0707] This invention provides an efficient asset management system using a generative AI model. The overall system processing flow and specific operational examples are described below.

[0708] First, the server retrieves asset data from an existing database. This data includes attribute information such as asset type, quantity, location, and status. The data is periodically imported into the server using SQL queries or other data retrieval methods.

[0709] Next, the server converts the acquired data into an internal data format (e.g., CSV or JSON). This conversion is performed using libraries such as Python's Pandas library, and the converted data is stored in storage for efficient processing.

[0710] The server trains a generative AI model based on the imported data. Machine learning algorithms (e.g., random forests or neural networks) are used for training. The server splits the data into training and validation datasets and tunes the hyperparameters. This process improves the prediction accuracy of the AI ​​model.

[0711] Once the AI ​​model is trained, the server generates an asset inventory schedule based on the model's predictions. The generated schedule prioritizes the verification tasks based on the asset's deterioration risk and importance. The generated schedule is delivered to the terminal and serves as a guideline for the inventory work performed by the user.

[0712] Users use a terminal to check the inventory schedule and perform verification tasks for designated assets. During the process, users input the actual quantity and condition of the assets into the terminal and send the results to the server. The server updates its database based on the received inventory results and uses this information to generate forecasts and schedules for the next inventory.

[0713] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions. Before important assets deteriorate and become unusable, the server notifies the user of this information. Based on the notification, the user performs the disposal work and inputs the results into their terminal, sending them to the server. After receiving the disposal results, the server updates its database.

[0714] Specific example

[0715] At the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training predicts that older air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and distributes it to the terminal. The user checks the inventory schedule and performs the inventory work. The inventory results are entered into the terminal and sent to the server. The server reflects the results and notifies the user of the removal timing. Based on the notification, the user removes the old air conditioning equipment and enters the removal results. Finally, the server makes predictions and plans for the following month based on the updated data.

[0716] This process significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, repeated training and schedule generation improve the accuracy of the AI ​​model, enabling increasingly efficient asset management.

[0717] Example of a prompt

[0718] An example of a prompt statement to illustrate this series of processes is as follows:

[0719] "In the asset management system, prompts are used to generate specific inventory schedules and disposal timings:

[0720] The server retrieves data from the asset database, converts it to CSV format, and inputs it as training data into a generative AI model. This model uses a random forest algorithm to predict the deterioration status of assets. Based on the predictions, the server generates the inventory schedule and disposal timing for the current month and distributes it to the terminal. After the inventory is completed, the results are sent from the terminal to the server, and the database is updated.

[0721] This clarifies the specific embodiments of the invention, enabling implementers to use the invention appropriately and efficiently.

[0722] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0723] Step 1: Obtain asset data

[0724] The server periodically retrieves asset data from an existing database. The input includes attribute information such as asset type, quantity, location, and status. This data is retrieved using SQL queries or other means. Specifically, the server executes an SQL query like "SELECT FROM assets WHERE month='current_month'". The output of this query is imported into the server.

[0725] Step 2: Data conversion and saving

[0726] The server converts the acquired asset data into an internal standard format (e.g., CSV or JSON). The data acquired in step 1 is used as input. Specifically, the server uses the Python Pandas library to convert the data into a DataFrame and saves it to a file such as "assets_data.csv". As output, the converted data is saved to storage in a format that can be processed efficiently.

[0727] Step 3: Training the AI ​​model

[0728] The server trains a generative AI model based on the imported data. The data saved in step 2 is used as input. Specifically, the server uses the Scikit-learn library to split the data into a training dataset and a validation dataset, and then trains a random forest model. Hyperparameter tuning is also performed. The output is a trained AI model.

[0729] Step 4: Generate and distribute the inventory schedule.

[0730] The server generates an inventory schedule based on the predictions of the trained AI model. The input includes the AI ​​model and asset data obtained in step 3. Specifically, the server uses the model to predict asset deterioration and creates an inventory schedule based on the results. The output is the generated inventory schedule, which is then delivered to the terminal.

[0731] Step 5: Conduct inventory work

[0732] The user uses a terminal to check the inventory schedule and perform the verification of the specified assets. The input includes the inventory schedule distributed in step 4. Specifically, the user checks the quantity and condition of assets such as air conditioning equipment according to the schedule and inputs the results into the terminal. The output is the verification results. These results are sent from the terminal to the server.

[0733] Step 6: Update the database of inventory results

[0734] The server receives the inventory results sent from the terminal and updates the database. The input used is the confirmation result obtained by the user in step 5. Specifically, the server integrates the new data into the existing database and uses it to generate future predictions and schedules. The output is the updated database.

[0735] Step 7: Planning and Implementing Asset Disposal

[0736] The server calculates and notifies the optimal timing for asset decommissioning based on the prediction results of the AI ​​model. Inputs include the trained AI model obtained in step 3 and the latest data. Specifically, the server predicts the deterioration status of old air conditioning equipment and calculates the decommissioning timing. This information is notified to the terminal. An output is generated: a decommissioning notification.

[0737] Step 8: Implement asset disposal and enter the results.

[0738] The user performs the appropriate asset removal work based on notifications from the server. The input includes the removal timing notified in step 7. Specifically, the user removes the old air conditioning equipment and inputs the result into the terminal. The output is the removal result, which is then sent from the terminal to the server.

[0739] Step 9: Update the database of removal results

[0740] The server receives the decommissioning results sent from the terminal and updates the database. The input used is the decommissioning results obtained in step 8. Specifically, the server integrates the decommissioning data into the existing database and uses it for future forecasts and planning. The output is the updated database. This series of steps enables efficient asset management and significantly reduces labor and costs.

[0741] (Application Example 1)

[0742] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0743] In modern logistics centers, asset management and inventory management are extremely complex and time-consuming tasks. They require a great deal of manual work, are prone to errors, and lack preventative management leads to increased losses due to asset deterioration. Traditional systems struggle to accurately predict asset deterioration and usage frequency, making it difficult to efficiently generate inventory and disposal schedules. To address these challenges, a more efficient and accurate asset management system is needed.

[0744] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0745] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, and means including an application on the terminal that notifies the user of the predicted inventory schedule and disposal timing, and allows the user to confirm and input the work. This makes it possible to accurately predict the state and frequency of use of assets and efficiently generate and notify inventory and disposal schedules, significantly improving the efficiency and accuracy of asset management.

[0746] A "server" is a device that retrieves asset data from an existing database, trains an AI model based on that data, and generates and distributes inventory schedules and disposal timings based on the prediction results.

[0747] "Asset data" refers to data that includes information about the type, quantity, location, condition, and frequency of use of goods and equipment.

[0748] A "generative AI model" is a machine learning model that uses training data to predict the deterioration status and usage frequency of assets, and to calculate appropriate inventory schedules and disposal timings.

[0749] "Training" is the process of using existing data to train a generative AI model so that it can accurately predict the state of an asset.

[0750] An "inventory schedule" is a plan for conducting inventory work at specific times, based on the deterioration status and importance of the assets.

[0751] A "terminal" is a device used by users to check inventory schedules and disposal timings, and to input actual work results.

[0752] "Removal timing" refers to the optimal period for properly removing an asset before it deteriorates and becomes unusable.

[0753] "Inventory results" refer to information about the quantity and condition of assets obtained by the user through actual inventory work.

[0754] A "database" is a database management system that stores asset data and allows servers to retrieve the information they need.

[0755] "User notifications" is a system that communicates information such as inventory schedules and disposal timings to users via their devices.

[0756] This invention is a system for improving the efficiency of asset management in logistics centers. Specifically, it acquires asset data, predicts asset status using a generative AI model, generates and distributes inventory schedules, and notifies users of asset disposal timing. The main components of this system involve servers, terminals, and users.

[0757] The server first retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. The retrieved data is periodically imported and converted into an internal CSV or JSON format. The server then uses this data to train a generative AI model. This AI model is implemented using machine learning libraries such as Scikit-learn, and its hyperparameters are tuned by splitting it into training and validation datasets.

[0758] A trained AI model predicts the deterioration status and usage frequency of assets, and generates an inventory schedule based on these predictions. The generated schedule is delivered to the user's device, which may include a smartphone or tablet. The user uses this device to check the notified inventory schedule and perform the inventory work for the specified assets. The work results are entered into the device and sent to the server, updating the database.

[0759] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the user. The user receives the notification, enters the result of asset disposal into their terminal, and this data is sent to the server, updating the database again.

[0760] As a concrete example, at the beginning of the month, the server retrieves all asset data, and an AI model predicts the deterioration of older forklifts. Based on this prediction, an inventory schedule indicating the need to remove forklifts is generated and delivered to the terminal. The user performs the inventory work, enters the results into the terminal, and sends them to the server, updating the database. Removal is carried out in a similar manner.

[0761] Examples of prompts for a generative AI model:

[0762] Train an AI model to predict asset degradation and usage frequency using the following dataset. The data is provided in CSV format, with each row representing a single asset.

[0763] The column names are as follows: Asset ID, Asset Type, Quantity, Location, Status, Usage Frequency.

[0764] The model should use this data to identify assets that are expected to deteriorate in the following month and predict when they will need to be removed.

[0765] To implement this invention, the server side performs data import, AI model training, schedule generation, notification, and database updates, while the terminal side inputs inventory results and receives notifications. This series of operations significantly improves the efficiency and accuracy of asset management in logistics centers.

[0766] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0767] Step 1:

[0768] The server retrieves asset data from an existing database. It sends SQL queries to the database to extract asset data, which includes asset type, quantity, location, status, etc. It receives asset data from the database as input and obtains raw data from the database as output.

[0769] Step 2:

[0770] The server converts the acquired asset data into an internal format (CSV or JSON). This allows for efficient data processing. It receives raw data from the database as input and obtains converted data as output.

[0771] Step 3:

[0772] The server trains a generative AI model based on the transformed asset data. In this process, the data is split into training and validation datasets, and the model is trained using Scikit-learn's RandomForestRegressor. It accepts asset data in an internal format as input and produces a trained AI model as output.

[0773] Step 4:

[0774] The server uses a trained AI model to predict the deterioration status and usage frequency of assets. This results in the asset status being obtained as a prediction. It receives a trained AI model and asset data in an internal format as input and obtains the prediction result as output.

[0775] Step 5:

[0776] The server generates an asset inventory schedule based on the predictions of an AI model. This is intended to prioritize the identification of assets at risk of deterioration or those of high importance. It takes prediction results as input and obtains an inventory schedule as output.

[0777] Step 6:

[0778] The server distributes the generated inventory schedule to the terminal. This allows users to perform inventory tasks according to the schedule. It receives the inventory schedule as input and receives notifications to the terminal as output.

[0779] Step 7:

[0780] The user uses a terminal to check the inventory schedule and perform the verification of the specified assets. The terminal receives the inventory schedule as input and inputs the inventory results as output.

[0781] Step 8:

[0782] The server receives inventory results entered from the terminal and updates the database. This ensures that the latest asset data is reflected in the database. The system receives inventory results as input and obtains an updated database as output.

[0783] Step 9:

[0784] The server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the user. It receives prediction results as input and outputs a notification of the disposal timing.

[0785] Step 10:

[0786] The user receives a notification, enters the result of asset disposal into the terminal, and sends that data to the server. The user receives a notification of the disposal timing as input and inputs the disposal result into the terminal as output.

[0787] Step 11:

[0788] The server receives the disposal results entered from the terminal and updates the database. This provides the latest asset data to be used for future predictions and planning. It receives disposal results as input and obtains an updated database as output.

[0789] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0790] This invention provides a system that supports users' efficient and effective work by combining an emotion engine with asset management in SB / WCP systems that utilize generative AI. The following describes the program processing of this system in natural language, along with specific examples.

[0791] First, the server retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. This data is periodically imported using SQL queries, etc., and converted to CSV or JSON format for efficient internal processing.

[0792] Next, the server trains a generative AI model based on the imported data. This training involves splitting the data into a training dataset and a validation dataset, evaluating the model's accuracy, and tuning hyperparameters as needed. Possible machine learning algorithms used include random forests and neural networks.

[0793] Once the model training is complete, the server generates an asset inventory schedule based on the AI ​​model's predictions. This schedule is designed to focus on assets with high depreciation risk and those of high importance. The generated schedule is delivered to the terminal for the user to use.

[0794] The server also features an emotion engine to recognize user emotions. When a user performs inventory work, the emotion engine uses the camera and microphone to identify emotions from the user's facial expressions and voice. This emotion data is transmitted to the server in real time, and the server adjusts the way inventory schedules are presented and notifications are sent as needed.

[0795] Users perform asset verification tasks based on an inventory schedule. Once users have completed asset verification, they enter the results into their terminal and send them to the server. The server receives the inventory results, updates the database, and uses them as feedback to improve the prediction accuracy of the model.

[0796] Furthermore, the server calculates and notifies the user of the optimal timing for asset disposal based on the AI ​​model's predictions. When the user receives the notification, performs the asset disposal, and enters the results into their terminal, the server receives the disposal results and updates the database. This entire process ensures that the user can always manage their assets with the latest information.

[0797] Specific example:

[0798] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predicting that old air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes the user's facial expressions and voice to analyze their emotions. For example, if the user is tired, the workload can be adjusted to reduce it.

[0799] The inventory results are entered into a terminal, and the data is sent to the server. The server reflects the results and notifies the user when the time for removal approaches. Based on the notification, the user removes old air conditioning equipment, enters the results into the terminal, and sends them to the server. The server updates the database and uses it for future predictions and planning.

[0800] This system significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, the use of an emotional engine reduces the user's workload, providing a more comfortable work environment.

[0801] The following describes the processing flow.

[0802] Step 1:

[0803] The server retrieves asset data from an existing database. This involves periodically collecting the latest asset information using SQL queries and other data retrieval methods. This data includes asset type, quantity, location, and status.

[0804] Step 2:

[0805] The server trains a generative AI model based on imported asset data. First, it splits the data into a training dataset and a validation dataset. Then, it trains the model using machine learning algorithms such as random forests and neural networks. It evaluates the model's accuracy and tunes hyperparameters as needed.

[0806] Step 3:

[0807] The server generates an asset inventory schedule based on a trained AI model. This includes a process of creating a schedule that prioritizes checking assets at high risk of deterioration or those of high importance.

[0808] Step 4:

[0809] The server distributes the generated inventory schedule to the terminals. It sends the schedule file, converted to a format usable by the terminals, over the network, making it accessible to the users.

[0810] Step 5:

[0811] The device uses an emotion engine to recognize the user's emotions while they are performing inventory tasks. It analyzes the user's facial expressions and voice through the camera and microphone to acquire emotional data.

[0812] Step 6:

[0813] The server receives emotional data sent from the emotion engine and adjusts inventory schedules and notification methods. If the user is fatigued, measures such as reducing the workload are taken.

[0814] Step 7:

[0815] Users perform asset verification according to the inventory schedule. They inspect assets on-site and input their condition and quantity into a terminal.

[0816] Step 8:

[0817] The terminal sends the inventory results entered by the user to the server. The collected data is sent to the server via the network, and the server receives it.

[0818] Step 9:

[0819] The server updates the database based on the received inventory results. The latest asset information is reflected in the database, which is then used for future predictions and planning.

[0820] Step 10:

[0821] The server calculates the optimal timing for asset decommissioning based on the predictions of the AI ​​model. This includes a process that uses a deterioration prediction model to identify which assets need to be decommissioned and when.

[0822] Step 11:

[0823] The server notifies the user of the timing for removal. It generates a notification message and delivers it to the terminal. This helps the user perform the removal process at the appropriate time.

[0824] Step 12:

[0825] The user receives a notification and carries out the removal process. Following the notification, they physically remove the unwanted assets and enter the results (removal date, method, etc.) into the terminal.

[0826] Step 13:

[0827] The terminal sends the removal results entered by the user to the server. The collected data is sent to the server via the network, and the server receives it.

[0828] Step 14:

[0829] The server updates the database based on the received disposal results. The database is updated with the latest information, which is then reflected in future forecasts and plans. This cycle continuously improves asset management, leading to greater accuracy and efficiency.

[0830] (Example 2)

[0831] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0832] Traditional asset management systems often involve manual asset data management and inventory processes, resulting in low efficiency and a high risk of human error. Furthermore, they fail to consider user emotional states, leading to uneven workload distribution and an unoptimized user work environment.

[0833] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0834] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, and means for analyzing user emotion data acquired from the terminal's camera and microphone and adjusting the inventory schedule and notification method based on the results. This improves the efficiency of asset management, reduces human error, and provides an optimal work environment that takes into account the user's emotional state.

[0835] An "existing database" is a data storage system that has already been built and is in operation by a company or organization.

[0836] "Asset data" refers to information about physical or digital assets owned by a company or organization, including attribute information such as type, quantity, location, and status.

[0837] A "generative AI model" is a type of artificial intelligence that learns from imported data and makes predictions and classifications about new data.

[0838] "Training methods" refer to the process of using training data to improve the performance of a generative AI model.

[0839] An "inventory schedule" is a plan for checking the status and location of assets and managing them appropriately.

[0840] "Terminal" refers to input and output devices such as personal computers, tablets, and smartphones that are operated by the user.

[0841] "Emotional data" refers to information that indicates a user's psychological state, obtained from their facial expressions, voice, and other similar data.

[0842] "Means of analysis" refers to methods for analyzing specific data and interpreting its meaning and trends.

[0843] "Hardware" refers to the physical components that make up computer systems and electronic devices.

[0844] "Software" refers to a set of programs and procedures that run on computers and electronic devices.

[0845] "Methods for updating a database" refer to methods for adding new information to an existing database or modifying existing information.

[0846] "Means of notification" refers to a communication method used to transmit specified information to the user.

[0847] "Optimal disposal timing" is a criterion for measuring the time when an asset is most cost-effective or before it loses its functionality.

[0848] Modes for carrying out the invention

[0849] This invention relates to a system that combines generative AI models and emotion recognition technology to streamline asset management. The following describes the program processing of this system in natural language, along with specific examples.

[0850] The server first retrieves asset data from an existing database. This data includes asset type, quantity, location, status, etc. The data is retrieved using SQL queries and converted to CSV or JSON format using the Python pandas library. For example, asset data is extracted using the following SQL query: "SELECT FROM assets WHERE status = 'active';".

[0851] Next, the server trains a generative AI model based on the imported data. This training uses Scikit-learn's Random Forest or Tensorflow's Neural Network. The accuracy is evaluated by splitting the dataset into training and validation datasets, and hyperparameters are tuned as needed. For example, the following command is used for training: `RandomForestClassifier.fit(training_data, training_labels)`.

[0852] Once training is complete, the server generates an asset inventory schedule based on the predictions of the generative AI model. This schedule is designed to focus on assets with high depreciation risk and high importance. The generated schedule is saved in JSON format and delivered to the terminal. For example, a schedule saved as "schedule.json" is delivered to the terminal.

[0853] Furthermore, the device collects user emotion data through its camera and microphone while the user is performing inventory tasks. It uses the OpenCV library for facial recognition and Google's Speech-to-Text API for speech emotion analysis. The acquired emotion data is sent to the server in real time and analyzed by an emotion engine.

[0854] Users verify assets based on an inventory schedule. The verification results are entered from a terminal and sent to the server. For example, a user might fill out a form on a web application and press the "submit" button. This data is received by the server, and the database is updated.

[0855] Furthermore, the server calculates and notifies the user of the optimal timing for asset disposal based on the AI ​​model's predictions. Upon receiving the notification, the user performs the asset disposal task, inputs the results into their terminal, and sends them to the server. The server receives the disposal results and updates its database. This information is then used for future predictions and planning.

[0856] Specific example

[0857] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predicting that old air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes the user's facial expressions and voice to analyze their emotions. For example, if the user is tired, the workload can be adjusted to reduce it.

[0858] Inventory results are entered into a terminal and the data is sent to the server. The server reflects the results and notifies the user when the time for decommissioning is approaching. Based on the notification, the user decommissions old air conditioning equipment, enters the results into the terminal, and sends them to the server. The server updates the database and uses it for future predictions and planning. This system significantly improves the efficiency of asset management and reduces labor and costs. In addition, the use of an emotional engine reduces the workload on the user and provides a more comfortable working environment.

[0859] Examples of prompts to input into a generative AI model

[0860] 1. "Predict the deterioration state of the air conditioning equipment."

[0861] 2. "Please generate the schedule for the next inventory count."

[0862] 3. "Adjust the workload based on the user's fatigue level."

[0863] By leveraging such generative AI models and emotion engines, asset management processes are streamlined, and the user's work environment is optimized. This allows companies and organizations to achieve digitized asset management while minimizing asset degradation and operational burden.

[0864] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0865] Step 1:

[0866] First, the server retrieves asset data from an existing database. The input here is an SQL query and database connection information, and the output is asset data (type, quantity, location, status, etc.). This allows the server to obtain the necessary data for use in the next step. Specifically, it executes the SQL query "SELECT FROM assets WHERE status = 'active';".

[0867] Step 2:

[0868] Next, the server imports the acquired asset data and converts it to CSV or JSON format. The input here is the asset data acquired in step 1, and the output is a data file converted to CSV or JSON format. Specifically, this is done by executing a command like "df.to_csv('assets.csv')" using the Python pandas library.

[0869] Step 3:

[0870] The server trains a generative AI model based on the imported data. The input here is asset data in CSV or JSON format, and the output is the trained generative AI model. Specifically, it executes the command "RandomForestClassifier.fit(training_data, training_labels)" using Scikit-learn's random forest algorithm. The training data and validation data are also split in this step.

[0871] Step 4:

[0872] Once the model training is complete, the server generates an asset inventory schedule based on the AI ​​model's predictions. The input here is the trained generative AI model and imported asset data, and the output is the generated inventory schedule. Specifically, it executes "prediction = model.predict(asset_data)" and saves the result to a JSON file.

[0873] Step 5:

[0874] The generated inventory schedule is delivered from the server to the terminal. The input here is the inventory schedule in JSON format, and the output is the completion of schedule delivery to the terminal. Specifically, the data is sent to the terminal using an HTTP request.

[0875] Step 6:

[0876] The device collects emotional data through the camera and microphone while the user performs inventory tasks. The input here is the user's facial expressions and voice, and the output is analyzed emotional data. Specifically, it uses the OpenCV library and Google's Speech-to-Text API. Camera video is acquired using "cv2.VideoCapture(0)", and audio data is analyzed in real time.

[0877] Step 7:

[0878] The device sends the acquired emotion data to the server in real time. The input here is the analyzed emotion data, and the output is the completion of data transmission to the server. Specifically, it sends the emotion data to the server using an HTTP POST request.

[0879] Step 8:

[0880] The user verifies assets based on the inventory schedule. Inputs here are the inventory schedule and the actual asset status, while output is the verification results. The user checks the status and quantity of each asset and enters this information into a dedicated form on the terminal.

[0881] Step 9:

[0882] The terminal sends the inventory results to the server. The input here is the inventory results entered by the user, and the output is the completion of the data transmission to the server. Specifically, the inventory results are sent using an HTTP POST request.

[0883] Step 10:

[0884] The server reflects the received inventory results in the database. The input here is the inventory results, and the output is the updated database. Specifically, it executes an SQL query such as "UPDATE assets SET status = 'checked' WHERE asset_id = ?".

[0885] Step 11:

[0886] The server calculates and notifies the user of the optimal timing for asset removal based on the prediction results. The input here is the prediction result of a generative AI model, and the output is a notification to the user. Specifically, the function "notify(user_id, 'Asset ID: 123 needs to be removed')" is used.

[0887] Step 12:

[0888] The user performs asset disposal work based on the notification, inputs the results into the terminal, and sends them to the server. The input here is the user's disposal result, and the output is the completion of data transmission to the server.

[0889] Step 13:

[0890] The server updates the database with the received removal results. The input here is the removal results, and the output is the database with the removals reflected. Specifically, it executes the SQL query "DELETE FROM assets WHERE asset_id = ?".

[0891] This series of steps will improve the efficiency of asset management and reduce the workload on users.

[0892] (Application Example 2)

[0893] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0894] Traditional asset management systems have problems such as difficulty in efficiently predicting asset deterioration and disposal timing, resulting in a high burden on workers. Furthermore, there is a lack of means to grasp asset status in real time, hindering improvements in the efficiency and accuracy of asset management. Additionally, scheduling and notification adjustments that do not take into account the emotional state of workers are not taken into account, further hindering efforts to reduce the workload.

[0895] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0896] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, means for collecting information on equipment and parts acquired by industrial sensors in real time and transmitting it to the server, and means for recognizing the emotions of workers using an emotion engine and adjusting the inventory schedule and notification method based on that data. This improves the efficiency and accuracy of asset management and reduces the burden on workers.

[0897] An "existing database" refers to a database already in operation within the system that holds information about assets.

[0898] "Asset data" refers to a collection of data that includes detailed information about assets, such as type, quantity, location, and condition.

[0899] A "generative AI model" is an artificial intelligence model used to predict the state of an asset based on imported data.

[0900] "Training methods" refer to the techniques and methods used to perform the processes necessary to improve the accuracy of generative AI models using data.

[0901] An "inventory schedule" is a plan that outlines the timing and sequence for verifying and inspecting assets.

[0902] A "terminal" refers to a device or equipment used by workers that can receive schedules and input results.

[0903] "Imported data" refers to asset data retrieved from an existing database and incorporated into the system.

[0904] "CSV or JSON" refers to a type of file format used for data exchange and storage.

[0905] "Data splitting methods" refer to methods and techniques for dividing data into training datasets and validation datasets.

[0906] An "industrial sensor" is a sensor installed in factories and work sites to collect information about equipment and parts in real time.

[0907] An "emotion engine" is a technology or system that analyzes a worker's facial expressions and voice to recognize their emotional state.

[0908] The "generated inventory schedule" is an inventory plan that is automatically generated based on the prediction results of the AI ​​model.

[0909] "Means of receiving" refers to the technologies and methods for receiving and managing data transmitted from a terminal.

[0910] "Means of notification" refers to the methods and technologies used to transmit information from a system to workers.

[0911] This invention relates to an asset management and work support system for a factory. This system acquires asset data from an existing database and predicts the asset status using a generative AI model. The aim is to optimize asset inventory and maintenance schedules, thereby reducing the burden on workers.

[0912] First, the server periodically retrieves asset data from an existing database. This data includes asset type, quantity, location, and status. This information is imported using SQL queries and then converted to CSV or JSON format. Data processing libraries such as Pandas are used for this process.

[0913] Subsequently, the server trains a generative AI model based on the imported data. Specifically, it splits the data into a training dataset and a validation dataset, and trains the model using machine learning libraries such as TensorFlow or PyTorch. Random forests and neural networks are suitable algorithms to use. The accuracy of the model is evaluated and improved through hyperparameter tuning.

[0914] Once the model training is complete, the server generates an asset inventory schedule based on the model's predictions. This schedule is designed to focus on assets at high risk of deterioration or those of high importance, and is then delivered to the terminals.

[0915] Furthermore, the server is equipped with an emotion engine that recognizes the worker's emotions in real time. When a worker performs inventory work, the emotion engine uses the camera and microphone on the terminal (e.g., smartphone or tablet) to identify emotions from their facial expressions and voice. This emotion data is transmitted to the server in real time, and the inventory schedule and notification methods are adjusted accordingly.

[0916] When a user performs tasks based on an inventory schedule, the results are entered into the terminal and sent to the server. The server receives these inventory results and updates the database. Similarly, when a user performs asset disposal tasks based on the disposal timing, the results are also entered into the terminal and sent to the server. The server updates the database again and uses this as feedback to improve the prediction accuracy of the model.

[0917] As a concrete example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predictions that older machinery and equipment are deteriorating and will need to be decommissioned the following month. Based on this information, the server generates an inventory schedule and decommissioning timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes their facial expressions and voice to analyze their emotions. For example, if fatigue is detected, the workload can be adjusted.

[0918] Example of a prompt:

[0919] "For training data for the asset management system within the factory, please enter the type, location, condition, quantity, and maintenance history of each piece of equipment. Also, please specify the features and target variables to be used for predicting the equipment's condition."

[0920] This system is expected to improve the efficiency of asset management and reduce labor costs and expenses. Furthermore, by utilizing an emotion engine, it is possible to reduce the burden on workers and improve the comfort of the work environment.

[0921] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0922] Step 1: Data Acquisition

[0923] The server retrieves asset data from an existing database. It uses SQL queries to access the database and retrieve information such as asset type, quantity, location, and status. The imported data is then converted to CSV or JSON format. Libraries such as Pandas are used for this conversion. The retrieved data is temporarily stored in storage for use in subsequent processing.

[0924] Input: Existing database

[0925] Output: Asset data in CSV or JSON format

[0926] Step 2: Training the generative AI model

[0927] The server trains generative AI models based on imported asset data. First, it splits the data into training and validation datasets. Next, it uses machine learning libraries such as TensorFlow and PyTorch to apply algorithms such as random forests and neural networks to train the models. Finally, it tunes hyperparameters to evaluate and improve the model's accuracy.

[0928] Input: Asset data in CSV or JSON format

[0929] Output: Trained AI model

[0930] Step 3: Generate an inventory schedule

[0931] The server generates an asset inventory schedule based on the predictions of a trained AI model. This schedule prioritizes assets with a high risk of deterioration or those of high importance. The generated schedule is then sent to the terminal.

[0932] Input: Trained AI model

[0933] Output: Inventory schedule

[0934] Step 4: Schedule Distribution

[0935] The server distributes the generated inventory schedule to the terminal. The terminal displays this schedule as a guideline for workers to carry out the inventory work.

[0936] Input: Inventory schedule

[0937] Output: Schedule displayed on the terminal

[0938] Step 5: Receive inventory results and update the database.

[0939] After completing the task, the user enters the inventory results into a terminal. The terminal sends the results to the server, which updates the database. This updated data is then used to train the next AI model.

[0940] Input: Inventory results entered from the terminal.

[0941] Output: Updated database

[0942] Step 6: Emotion recognition by the emotion engine

[0943] The server recognizes workers' emotions in real time through an emotion engine. Using cameras and microphones installed on the terminals, it analyzes workers' facial expressions and voices to acquire emotion data. This data is sent to the server, and inventory schedules and notification methods are adjusted accordingly.

[0944] Input: Emotional data obtained from the device

[0945] Output: Adjusted inventory schedule and notification methods

[0946] Step 7: Notification of the optimal timing for asset disposal

[0947] The server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the worker. The user, upon receiving the notification, carries out the disposal work and enters the results into their terminal.

[0948] Input: AI model prediction results

[0949] Output: Notification of disposal timing

[0950] Step 8: Receive the removal results and update the database.

[0951] After the user completes the asset disposal process, they input the results into a terminal. The terminal sends the results to a server, which updates the database. This data will be used to train future AI models.

[0952] Input: Disposal results entered from the terminal

[0953] Output: Updated database

[0954] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0955] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0956] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0957] [Fourth Embodiment]

[0958] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0959] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0960] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0961] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0962] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0963] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0964] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0965] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0966] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0967] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0968] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0969] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0970] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0971] This invention provides a method for streamlining asset management in SB / WCP systems using generative AI, thereby reducing the time and costs associated with monthly inventory and asset disposal operations. The system's program processing is explained below in natural language, followed by a specific operational example.

[0972] First, the server retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. This data is periodically imported using SQL queries or other data retrieval methods. The imported data is converted to an internal data format (e.g., CSV or JSON) and processed efficiently.

[0973] Next, the server trains a generative AI model based on the imported data. This AI model is designed to predict asset usage frequency and degradation, and uses machine learning algorithms (e.g., random forests or neural networks). Training involves splitting the data into training and validation datasets and tuning hyperparameters to evaluate and improve the model's accuracy.

[0974] Once the AI ​​model is trained, the server generates an asset inventory schedule based on the model's predictions. The inventory aims to prioritize the identification of assets that are at risk of deterioration or are of high importance. The generated schedule is delivered to the terminal and serves as a guideline for the inventory work performed by the user.

[0975] Users use a terminal to check the inventory schedule and perform the verification of designated assets. During the inventory, users input the actual quantity and condition of assets into the terminal and send the inventory results to the server. The server receives these results, updates its database, and uses it to generate forecasts and schedules for the next inventory.

[0976] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions. By notifying users of the optimal disposal timing before critical assets deteriorate and become unusable, proactive management becomes possible. Users receive notifications, perform the actual disposal work, and input the results into their terminals. The server then receives the disposal results and updates its database.

[0977] Specific example:

[0978] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training predicts that older air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and distributes it to the terminal. The user checks the inventory schedule and performs the inventory work. The inventory results are entered into the terminal and sent to the server. The server reflects the results and notifies the user of the removal timing. Based on the notification, the user removes the old air conditioning equipment and enters the removal results. Finally, the server makes predictions and plans for the following month based on the updated data.

[0979] This entire process significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, repeated training and schedule generation improve the accuracy of the AI ​​model, enabling increasingly efficient asset management.

[0980] The following describes the processing flow.

[0981] Step 1:

[0982] The server retrieves asset data from an existing database. Specifically, the server periodically executes SQL queries to extract the latest asset data from the database and convert it into an internal data format (e.g., CSV or JSON).

[0983] Step 2:

[0984] The server trains a generative AI model based on the imported data. Specifically, the server splits the data into a training dataset and a validation dataset, and trains the model using a machine learning algorithm (e.g., random forest, neural network). If there are any deficiencies in the training results, it tunes the hyperparameters and retrains the model.

[0985] Step 3:

[0986] The server generates an asset inventory schedule based on a trained AI model. Specifically, it creates a schedule that prioritizes the inventory of assets with a high risk of deterioration or those of high importance, based on the model's predictions.

[0987] Step 4:

[0988] The server distributes the generated inventory schedule to the terminals. Specifically, it sends the schedule file, converted to a format usable by the terminals, over the network, making it accessible to the users.

[0989] Step 5:

[0990] Users perform asset verification tasks based on the inventory schedule. Specifically, they check the condition and quantity of assets on-site according to the inventory schedule and input the results into their handheld terminals.

[0991] Step 6:

[0992] The terminal sends the inventory results entered by the user to the server. Specifically, it sends the input data to the server via the network, and the server stores the received data in its internal database.

[0993] Step 7:

[0994] The server updates the database based on the inventory results. Specifically, it evaluates the received inventory results and updates the existing asset database with the latest information.

[0995] Step 8:

[0996] The server calculates the optimal timing for asset decommissioning based on the updated database and predictions from the AI ​​model. Specifically, it uses a trained AI model to predict deterioration and identify assets that need to be decommissioned and when.

[0997] Step 9:

[0998] The server notifies the user of the timing for disposal. Specifically, it generates a notification message that lists the assets that need to be disposed of and the timing for disposal, and delivers this message to the terminal.

[0999] Step 10:

[1000] The user receives a notification and carries out the removal work. Specifically, they remove the unwanted assets on-site according to the notification and enter the results (removal date, method, etc.) into the terminal.

[1001] Step 11:

[1002] The terminal sends the user-entered removal results to the server. Specifically, it sends the input data to the server via the network, and the server stores the received data in its internal database.

[1003] Step 12:

[1004] The server updates the database based on the disposal results. Specifically, it evaluates the received disposal results and updates the existing asset database with the latest information.

[1005] (Example 1)

[1006] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1007] Traditional asset management systems required manual inventory and disposal processes, consuming considerable effort and time. Furthermore, accurately predicting asset deterioration and usage frequency was difficult, leading to decreased asset management efficiency. This could potentially result in increased costs due to asset deterioration and improper management.

[1008] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1009] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for dividing the training into a training dataset and a validation dataset and adjusting hyperparameters, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, and means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database. This significantly improves the efficiency of asset management and reduces the man-hours and costs of the work.

[1010] An "existing database" is a collection of information that has already been built and is in operation, and is a system in which data related to assets is stored.

[1011] "Asset data" refers to information about physical or digital assets owned by a company or organization, including attributes such as type, quantity, location, and status.

[1012] A "generative AI model" is a program that uses machine learning algorithms to learn patterns and rules from given data, and is used to predict the deterioration status and usage frequency of assets.

[1013] A "training dataset" is a collection of data used by generative AI models for learning, and it serves as the foundational data for giving the model predictive capabilities.

[1014] A "validation dataset" is a collection of data used to evaluate the prediction accuracy of generative AI models, and is managed separately from the training dataset.

[1015] "Hyperparameters" are settings or adjustment items that affect the performance and results of a machine learning model, and are adjusted to optimize the model.

[1016] An "inventory schedule" is a plan or schedule for verifying assets, and is created by prioritizing tasks based on the condition of the assets and their risk of deterioration.

[1017] A "terminal" refers to a computer or mobile device used by a user, and is hardware used for sending and receiving data with a system.

[1018] "Decommissioning timing" refers to the optimal time to dispose of or transfer a particular asset before it becomes unusable, and is calculated for preventative management purposes.

[1019] This invention provides an efficient asset management system using a generative AI model. The overall system processing flow and specific operational examples are described below.

[1020] First, the server retrieves asset data from an existing database. This data includes attribute information such as asset type, quantity, location, and status. The data is periodically imported into the server using SQL queries or other data retrieval methods.

[1021] Next, the server converts the acquired data into an internal data format (e.g., CSV or JSON). This conversion is performed using libraries such as Python's Pandas library, and the converted data is stored in storage for efficient processing.

[1022] The server trains a generative AI model based on the imported data. Machine learning algorithms (e.g., random forests or neural networks) are used for training. The server splits the data into training and validation datasets and tunes the hyperparameters. This process improves the prediction accuracy of the AI ​​model.

[1023] Once the AI ​​model is trained, the server generates an asset inventory schedule based on the model's predictions. The generated schedule prioritizes the verification tasks based on the asset's deterioration risk and importance. The generated schedule is delivered to the terminal and serves as a guideline for the inventory work performed by the user.

[1024] Users use a terminal to check the inventory schedule and perform verification tasks for designated assets. During the process, users input the actual quantity and condition of the assets into the terminal and send the results to the server. The server updates its database based on the received inventory results and uses this information to generate forecasts and schedules for the next inventory.

[1025] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions. Before important assets deteriorate and become unusable, the server notifies the user of this information. Based on the notification, the user performs the disposal work and inputs the results into their terminal, sending them to the server. After receiving the disposal results, the server updates its database.

[1026] Specific example

[1027] At the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training predicts that older air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and distributes it to the terminal. The user checks the inventory schedule and performs the inventory work. The inventory results are entered into the terminal and sent to the server. The server reflects the results and notifies the user of the removal timing. Based on the notification, the user removes the old air conditioning equipment and enters the removal results. Finally, the server makes predictions and plans for the following month based on the updated data.

[1028] This process significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, repeated training and schedule generation improve the accuracy of the AI ​​model, enabling increasingly efficient asset management.

[1029] Example of a prompt

[1030] An example of a prompt statement to illustrate this series of processes is as follows:

[1031] "In the asset management system, prompts are used to generate specific inventory schedules and disposal timings:

[1032] The server retrieves data from the asset database, converts it to CSV format, and inputs it as training data into a generative AI model. This model uses a random forest algorithm to predict the deterioration status of assets. Based on the predictions, the server generates the inventory schedule and disposal timing for the current month and distributes it to the terminal. After the inventory is completed, the results are sent from the terminal to the server, and the database is updated.

[1033] This clarifies the specific embodiments of the invention, enabling implementers to use the invention appropriately and efficiently.

[1034] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1035] Step 1: Obtain asset data

[1036] The server periodically retrieves asset data from an existing database. The input includes attribute information such as asset type, quantity, location, and status. This data is retrieved using SQL queries or other means. Specifically, the server executes an SQL query like "SELECT FROM assets WHERE month='current_month'". The output of this query is imported into the server.

[1037] Step 2: Data conversion and saving

[1038] The server converts the acquired asset data into an internal standard format (e.g., CSV or JSON). The data acquired in step 1 is used as input. Specifically, the server uses the Python Pandas library to convert the data into a DataFrame and saves it to a file such as "assets_data.csv". As output, the converted data is saved to storage in a format that can be processed efficiently.

[1039] Step 3: Training the AI ​​model

[1040] The server trains a generative AI model based on the imported data. The data saved in step 2 is used as input. Specifically, the server uses the Scikit-learn library to split the data into a training dataset and a validation dataset, and then trains a random forest model. Hyperparameter tuning is also performed. The output is a trained AI model.

[1041] Step 4: Generate and distribute the inventory schedule.

[1042] The server generates an inventory schedule based on the predictions of the trained AI model. The input includes the AI ​​model and asset data obtained in step 3. Specifically, the server uses the model to predict asset deterioration and creates an inventory schedule based on the results. The output is the generated inventory schedule, which is then delivered to the terminal.

[1043] Step 5: Conduct inventory work

[1044] The user uses a terminal to check the inventory schedule and perform the verification of the specified assets. The input includes the inventory schedule distributed in step 4. Specifically, the user checks the quantity and condition of assets such as air conditioning equipment according to the schedule and inputs the results into the terminal. The output is the verification results. These results are sent from the terminal to the server.

[1045] Step 6: Update the database of inventory results

[1046] The server receives the inventory results sent from the terminal and updates the database. The input used is the confirmation result obtained by the user in step 5. Specifically, the server integrates the new data into the existing database and uses it to generate future predictions and schedules. The output is the updated database.

[1047] Step 7: Planning and Implementing Asset Disposal

[1048] The server calculates and notifies the optimal timing for asset decommissioning based on the prediction results of the AI ​​model. Inputs include the trained AI model obtained in step 3 and the latest data. Specifically, the server predicts the deterioration status of old air conditioning equipment and calculates the decommissioning timing. This information is notified to the terminal. An output is generated: a decommissioning notification.

[1049] Step 8: Implement asset disposal and enter the results.

[1050] The user performs the appropriate asset removal work based on notifications from the server. The input includes the removal timing notified in step 7. Specifically, the user removes the old air conditioning equipment and inputs the result into the terminal. The output is the removal result, which is then sent from the terminal to the server.

[1051] Step 9: Update the database of removal results

[1052] The server receives the decommissioning results sent from the terminal and updates the database. The input used is the decommissioning results obtained in step 8. Specifically, the server integrates the decommissioning data into the existing database and uses it for future forecasts and planning. The output is the updated database. This series of steps enables efficient asset management and significantly reduces labor and costs.

[1053] (Application Example 1)

[1054] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1055] In modern logistics centers, asset management and inventory management are extremely complex and time-consuming tasks. They require a great deal of manual work, are prone to errors, and lack preventative management leads to increased losses due to asset deterioration. Traditional systems struggle to accurately predict asset deterioration and usage frequency, making it difficult to efficiently generate inventory and disposal schedules. To address these challenges, a more efficient and accurate asset management system is needed.

[1056] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1057] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, and means including an application on the terminal that notifies the user of the predicted inventory schedule and disposal timing, and allows the user to confirm and input the work. This makes it possible to accurately predict the state and frequency of use of assets and efficiently generate and notify inventory and disposal schedules, significantly improving the efficiency and accuracy of asset management.

[1058] A "server" is a device that retrieves asset data from an existing database, trains an AI model based on that data, and generates and distributes inventory schedules and disposal timings based on the prediction results.

[1059] "Asset data" refers to data that includes information about the type, quantity, location, condition, and frequency of use of goods and equipment.

[1060] A "generative AI model" is a machine learning model that uses training data to predict the deterioration status and usage frequency of assets, and to calculate appropriate inventory schedules and disposal timings.

[1061] "Training" is the process of using existing data to train a generative AI model so that it can accurately predict the state of an asset.

[1062] An "inventory schedule" is a plan for conducting inventory work at specific times, based on the deterioration status and importance of the assets.

[1063] A "terminal" is a device used by users to check inventory schedules and disposal timings, and to input actual work results.

[1064] "Removal timing" refers to the optimal period for properly removing an asset before it deteriorates and becomes unusable.

[1065] "Inventory results" refer to information about the quantity and condition of assets obtained by the user through actual inventory work.

[1066] A "database" is a database management system that stores asset data and allows servers to retrieve the information they need.

[1067] "User notifications" is a system that communicates information such as inventory schedules and disposal timings to users via their devices.

[1068] This invention is a system for improving the efficiency of asset management in logistics centers. Specifically, it acquires asset data, predicts asset status using a generative AI model, generates and distributes inventory schedules, and notifies users of asset disposal timing. The main components of this system involve servers, terminals, and users.

[1069] The server first retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. The retrieved data is periodically imported and converted into an internal CSV or JSON format. The server then uses this data to train a generative AI model. This AI model is implemented using machine learning libraries such as Scikit-learn, and its hyperparameters are tuned by splitting it into training and validation datasets.

[1070] A trained AI model predicts the deterioration status and usage frequency of assets, and generates an inventory schedule based on these predictions. The generated schedule is delivered to the user's device, which may include a smartphone or tablet. The user uses this device to check the notified inventory schedule and perform the inventory work for the specified assets. The work results are entered into the device and sent to the server, updating the database.

[1071] Furthermore, the server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the user. The user receives the notification, enters the result of asset disposal into their terminal, and this data is sent to the server, updating the database again.

[1072] As a concrete example, at the beginning of the month, the server retrieves all asset data, and an AI model predicts the deterioration of older forklifts. Based on this prediction, an inventory schedule indicating the need to remove forklifts is generated and delivered to the terminal. The user performs the inventory work, enters the results into the terminal, and sends them to the server, updating the database. Removal is carried out in a similar manner.

[1073] Examples of prompts for a generative AI model:

[1074] Train an AI model to predict asset degradation and usage frequency using the following dataset. The data is provided in CSV format, with each row representing a single asset.

[1075] The column names are as follows: Asset ID, Asset Type, Quantity, Location, Status, Usage Frequency.

[1076] The model should use this data to identify assets that are expected to deteriorate in the following month and predict when they will need to be removed.

[1077] To implement this invention, the server side performs data import, AI model training, schedule generation, notification, and database updates, while the terminal side inputs inventory results and receives notifications. This series of operations significantly improves the efficiency and accuracy of asset management in logistics centers.

[1078] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1079] Step 1:

[1080] The server retrieves asset data from an existing database. It sends SQL queries to the database to extract asset data, which includes asset type, quantity, location, status, etc. It receives asset data from the database as input and obtains raw data from the database as output.

[1081] Step 2:

[1082] The server converts the acquired asset data into an internal format (CSV or JSON). This allows for efficient data processing. It receives raw data from the database as input and obtains converted data as output.

[1083] Step 3:

[1084] The server trains a generative AI model based on the transformed asset data. In this process, the data is split into training and validation datasets, and the model is trained using Scikit-learn's RandomForestRegressor. It accepts asset data in an internal format as input and produces a trained AI model as output.

[1085] Step 4:

[1086] The server uses a trained AI model to predict the deterioration status and usage frequency of assets. This results in the asset status being obtained as a prediction. It receives a trained AI model and asset data in an internal format as input and obtains the prediction result as output.

[1087] Step 5:

[1088] The server generates an asset inventory schedule based on the predictions of an AI model. This is intended to prioritize the identification of assets at risk of deterioration or those of high importance. It takes prediction results as input and obtains an inventory schedule as output.

[1089] Step 6:

[1090] The server distributes the generated inventory schedule to the terminal. This allows users to perform inventory tasks according to the schedule. It receives the inventory schedule as input and receives notifications to the terminal as output.

[1091] Step 7:

[1092] The user uses a terminal to check the inventory schedule and perform the verification of the specified assets. The terminal receives the inventory schedule as input and inputs the inventory results as output.

[1093] Step 8:

[1094] The server receives inventory results entered from the terminal and updates the database. This ensures that the latest asset data is reflected in the database. The system receives inventory results as input and obtains an updated database as output.

[1095] Step 9:

[1096] The server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the user. It receives prediction results as input and outputs a notification of the disposal timing.

[1097] Step 10:

[1098] The user receives a notification, enters the result of asset disposal into the terminal, and sends that data to the server. The user receives a notification of the disposal timing as input and inputs the disposal result into the terminal as output.

[1099] Step 11:

[1100] The server receives the disposal results entered from the terminal and updates the database. This provides the latest asset data to be used for future predictions and planning. It receives disposal results as input and obtains an updated database as output.

[1101] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1102] This invention provides a system that supports users' efficient and effective work by combining an emotion engine with asset management in SB / WCP systems that utilize generative AI. The following describes the program processing of this system in natural language, along with specific examples.

[1103] First, the server retrieves asset data from an existing database. This data includes information such as asset type, quantity, location, and status. This data is periodically imported using SQL queries, etc., and converted to CSV or JSON format for efficient internal processing.

[1104] Next, the server trains a generative AI model based on the imported data. This training involves splitting the data into a training dataset and a validation dataset, evaluating the model's accuracy, and tuning hyperparameters as needed. Possible machine learning algorithms used include random forests and neural networks.

[1105] Once the model training is complete, the server generates an asset inventory schedule based on the AI ​​model's predictions. This schedule is designed to focus on assets with high depreciation risk and those of high importance. The generated schedule is delivered to the terminal for the user to use.

[1106] The server also features an emotion engine to recognize user emotions. When a user performs inventory work, the emotion engine uses the camera and microphone to identify emotions from the user's facial expressions and voice. This emotion data is transmitted to the server in real time, and the server adjusts the way inventory schedules are presented and notifications are sent as needed.

[1107] Users perform asset verification tasks based on an inventory schedule. Once users have completed asset verification, they enter the results into their terminal and send them to the server. The server receives the inventory results, updates the database, and uses them as feedback to improve the prediction accuracy of the model.

[1108] Furthermore, the server calculates and notifies the user of the optimal timing for asset disposal based on the AI ​​model's predictions. When the user receives the notification, performs the asset disposal, and enters the results into their terminal, the server receives the disposal results and updates the database. This entire process ensures that the user can always manage their assets with the latest information.

[1109] Specific example:

[1110] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predicting that old air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes the user's facial expressions and voice to analyze their emotions. For example, if the user is tired, the workload can be adjusted to reduce it.

[1111] The inventory results are entered into a terminal, and the data is sent to the server. The server reflects the results and notifies the user when the time for removal approaches. Based on the notification, the user removes old air conditioning equipment, enters the results into the terminal, and sends them to the server. The server updates the database and uses it for future predictions and planning.

[1112] This system significantly improves the efficiency of asset management, reducing workload and costs. Furthermore, the use of an emotional engine reduces the user's workload, providing a more comfortable work environment.

[1113] The following describes the processing flow.

[1114] Step 1:

[1115] The server retrieves asset data from an existing database. This involves periodically collecting the latest asset information using SQL queries and other data retrieval methods. This data includes asset type, quantity, location, and status.

[1116] Step 2:

[1117] The server trains a generative AI model based on imported asset data. First, it splits the data into a training dataset and a validation dataset. Then, it trains the model using machine learning algorithms such as random forests and neural networks. It evaluates the model's accuracy and tunes hyperparameters as needed.

[1118] Step 3:

[1119] The server generates an asset inventory schedule based on a trained AI model. This includes a process of creating a schedule that prioritizes checking assets at high risk of deterioration or those of high importance.

[1120] Step 4:

[1121] The server distributes the generated inventory schedule to the terminals. It sends the schedule file, converted to a format usable by the terminals, over the network, making it accessible to the users.

[1122] Step 5:

[1123] The device uses an emotion engine to recognize the user's emotions while they are performing inventory tasks. It analyzes the user's facial expressions and voice through the camera and microphone to acquire emotional data.

[1124] Step 6:

[1125] The server receives emotional data sent from the emotion engine and adjusts inventory schedules and notification methods. If the user is fatigued, measures such as reducing the workload are taken.

[1126] Step 7:

[1127] Users perform asset verification according to the inventory schedule. They inspect assets on-site and input their condition and quantity into a terminal.

[1128] Step 8:

[1129] The terminal sends the inventory results entered by the user to the server. The collected data is sent to the server via the network, and the server receives it.

[1130] Step 9:

[1131] The server updates the database based on the received inventory results. The latest asset information is reflected in the database, which is then used for future predictions and planning.

[1132] Step 10:

[1133] The server calculates the optimal timing for asset decommissioning based on the predictions of the AI ​​model. This includes a process that uses a deterioration prediction model to identify which assets need to be decommissioned and when.

[1134] Step 11:

[1135] The server notifies the user of the timing for removal. It generates a notification message and delivers it to the terminal. This helps the user perform the removal process at the appropriate time.

[1136] Step 12:

[1137] The user receives a notification and carries out the removal process. Following the notification, they physically remove the unwanted assets and enter the results (removal date, method, etc.) into the terminal.

[1138] Step 13:

[1139] The terminal sends the removal results entered by the user to the server. The collected data is sent to the server via the network, and the server receives it.

[1140] Step 14:

[1141] The server updates the database based on the received disposal results. The database is updated with the latest information, which is then reflected in future forecasts and plans. This cycle continuously improves asset management, leading to greater accuracy and efficiency.

[1142] (Example 2)

[1143] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1144] Traditional asset management systems often involve manual asset data management and inventory processes, resulting in low efficiency and a high risk of human error. Furthermore, they fail to consider user emotional states, leading to uneven workload distribution and an unoptimized user work environment.

[1145] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1146] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, and means for analyzing user emotion data acquired from the terminal's camera and microphone and adjusting the inventory schedule and notification method based on the results. This improves the efficiency of asset management, reduces human error, and provides an optimal work environment that takes into account the user's emotional state.

[1147] An "existing database" is a data storage system that has already been built and is in operation by a company or organization.

[1148] "Asset data" refers to information about physical or digital assets owned by a company or organization, including attribute information such as type, quantity, location, and status.

[1149] A "generative AI model" is a type of artificial intelligence that learns from imported data and makes predictions and classifications about new data.

[1150] "Training methods" refer to the process of using training data to improve the performance of a generative AI model.

[1151] An "inventory schedule" is a plan for checking the status and location of assets and managing them appropriately.

[1152] "Terminal" refers to input and output devices such as personal computers, tablets, and smartphones that are operated by the user.

[1153] "Emotional data" refers to information that indicates a user's psychological state, obtained from their facial expressions, voice, and other similar data.

[1154] "Means of analysis" refers to methods for analyzing specific data and interpreting its meaning and trends.

[1155] "Hardware" refers to the physical components that make up computer systems and electronic devices.

[1156] "Software" refers to a set of programs and procedures that run on computers and electronic devices.

[1157] "Methods for updating a database" refer to methods for adding new information to an existing database or modifying existing information.

[1158] "Means of notification" refers to a communication method used to transmit specified information to the user.

[1159] "Optimal disposal timing" is a criterion for measuring the time when an asset is most cost-effective or before it loses its functionality.

[1160] Modes for carrying out the invention

[1161] This invention relates to a system that combines generative AI models and emotion recognition technology to streamline asset management. The following describes the program processing of this system in natural language, along with specific examples.

[1162] The server first retrieves asset data from an existing database. This data includes asset type, quantity, location, status, etc. The data is retrieved using SQL queries and converted to CSV or JSON format using the Python pandas library. For example, asset data is extracted using the following SQL query: "SELECT FROM assets WHERE status = 'active';".

[1163] Next, the server trains a generative AI model based on the imported data. This training uses Scikit-learn's Random Forest or Tensorflow's Neural Network. The accuracy is evaluated by splitting the dataset into training and validation datasets, and hyperparameters are tuned as needed. For example, the following command is used for training: `RandomForestClassifier.fit(training_data, training_labels)`.

[1164] Once training is complete, the server generates an asset inventory schedule based on the predictions of the generative AI model. This schedule is designed to focus on assets with high depreciation risk and high importance. The generated schedule is saved in JSON format and delivered to the terminal. For example, a schedule saved as "schedule.json" is delivered to the terminal.

[1165] Furthermore, the device collects user emotion data through its camera and microphone while the user is performing inventory tasks. It uses the OpenCV library for facial recognition and Google's Speech-to-Text API for speech emotion analysis. The acquired emotion data is sent to the server in real time and analyzed by an emotion engine.

[1166] Users verify assets based on an inventory schedule. The verification results are entered from a terminal and sent to the server. For example, a user might fill out a form on a web application and press the "submit" button. This data is received by the server, and the database is updated.

[1167] Furthermore, the server calculates and notifies the user of the optimal timing for asset disposal based on the AI ​​model's predictions. Upon receiving the notification, the user performs the asset disposal task, inputs the results into their terminal, and sends them to the server. The server receives the disposal results and updates its database. This information is then used for future predictions and planning.

[1168] Specific example

[1169] For example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predicting that old air conditioning equipment is deteriorating and will need to be removed the following month. Based on this information, the server generates an inventory schedule and removal timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes the user's facial expressions and voice to analyze their emotions. For example, if the user is tired, the workload can be adjusted to reduce it.

[1170] Inventory results are entered into a terminal and the data is sent to the server. The server reflects the results and notifies the user when the time for decommissioning is approaching. Based on the notification, the user decommissions old air conditioning equipment, enters the results into the terminal, and sends them to the server. The server updates the database and uses it for future predictions and planning. This system significantly improves the efficiency of asset management and reduces labor and costs. In addition, the use of an emotional engine reduces the workload on the user and provides a more comfortable working environment.

[1171] Examples of prompts to input into a generative AI model

[1172] 1. "Predict the deterioration state of the air conditioning equipment."

[1173] 2. "Please generate the schedule for the next inventory count."

[1174] 3. "Adjust the workload based on the user's fatigue level."

[1175] By leveraging such generative AI models and emotion engines, asset management processes are streamlined, and the user's work environment is optimized. This allows companies and organizations to achieve digitized asset management while minimizing asset degradation and operational burden.

[1176] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1177] Step 1:

[1178] First, the server retrieves asset data from an existing database. The input here is an SQL query and database connection information, and the output is asset data (type, quantity, location, status, etc.). This allows the server to obtain the necessary data for use in the next step. Specifically, it executes the SQL query "SELECT FROM assets WHERE status = 'active';".

[1179] Step 2:

[1180] Next, the server imports the acquired asset data and converts it to CSV or JSON format. The input here is the asset data acquired in step 1, and the output is a data file converted to CSV or JSON format. Specifically, this is done by executing a command like "df.to_csv('assets.csv')" using the Python pandas library.

[1181] Step 3:

[1182] The server trains a generative AI model based on the imported data. The input here is asset data in CSV or JSON format, and the output is the trained generative AI model. Specifically, it executes the command "RandomForestClassifier.fit(training_data, training_labels)" using Scikit-learn's random forest algorithm. The training data and validation data are also split in this step.

[1183] Step 4:

[1184] Once the model training is complete, the server generates an asset inventory schedule based on the AI ​​model's predictions. The input here is the trained generative AI model and imported asset data, and the output is the generated inventory schedule. Specifically, it executes "prediction = model.predict(asset_data)" and saves the result to a JSON file.

[1185] Step 5:

[1186] The generated inventory schedule is delivered from the server to the terminal. The input here is the inventory schedule in JSON format, and the output is the completion of schedule delivery to the terminal. Specifically, the data is sent to the terminal using an HTTP request.

[1187] Step 6:

[1188] The device collects emotional data through the camera and microphone while the user performs inventory tasks. The input here is the user's facial expressions and voice, and the output is analyzed emotional data. Specifically, it uses the OpenCV library and Google's Speech-to-Text API. Camera video is acquired using "cv2.VideoCapture(0)", and audio data is analyzed in real time.

[1189] Step 7:

[1190] The device sends the acquired emotion data to the server in real time. The input here is the analyzed emotion data, and the output is the completion of data transmission to the server. Specifically, it sends the emotion data to the server using an HTTP POST request.

[1191] Step 8:

[1192] The user verifies assets based on the inventory schedule. Inputs here are the inventory schedule and the actual asset status, while output is the verification results. The user checks the status and quantity of each asset and enters this information into a dedicated form on the terminal.

[1193] Step 9:

[1194] The terminal sends the inventory results to the server. The input here is the inventory results entered by the user, and the output is the completion of the data transmission to the server. Specifically, the inventory results are sent using an HTTP POST request.

[1195] Step 10:

[1196] The server reflects the received inventory results in the database. The input here is the inventory results, and the output is the updated database. Specifically, it executes an SQL query such as "UPDATE assets SET status = 'checked' WHERE asset_id = ?".

[1197] Step 11:

[1198] The server calculates and notifies the user of the optimal timing for asset removal based on the prediction results. The input here is the prediction result of a generative AI model, and the output is a notification to the user. Specifically, the function "notify(user_id, 'Asset ID: 123 needs to be removed')" is used.

[1199] Step 12:

[1200] The user performs asset disposal work based on the notification, inputs the results into the terminal, and sends them to the server. The input here is the user's disposal result, and the output is the completion of data transmission to the server.

[1201] Step 13:

[1202] The server updates the database with the received removal results. The input here is the removal results, and the output is the database with the removals reflected. Specifically, it executes the SQL query "DELETE FROM assets WHERE asset_id = ?".

[1203] This series of steps will improve the efficiency of asset management and reduce the workload on users.

[1204] (Application Example 2)

[1205] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1206] Traditional asset management systems have problems such as difficulty in efficiently predicting asset deterioration and disposal timing, resulting in a high burden on workers. Furthermore, there is a lack of means to grasp asset status in real time, hindering improvements in the efficiency and accuracy of asset management. Additionally, scheduling and notification adjustments that do not take into account the emotional state of workers are not taken into account, further hindering efforts to reduce the workload.

[1207] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1208] In this invention, the server includes means for acquiring asset data from an existing database, means for training a generative AI model to predict the state of assets based on the imported asset data, means for generating an asset inventory schedule based on the prediction results of the AI ​​model, means for distributing the generated inventory schedule to a terminal, means for receiving inventory results entered from the terminal and updating the database, means for calculating and notifying the optimal timing for asset disposal based on the prediction results of the AI ​​model, means for receiving disposal results entered from the terminal based on the notified disposal timing and updating the database, means for collecting information on equipment and parts acquired by industrial sensors in real time and transmitting it to the server, and means for recognizing the emotions of workers using an emotion engine and adjusting the inventory schedule and notification method based on that data. This improves the efficiency and accuracy of asset management and reduces the burden on workers.

[1209] An "existing database" refers to a database already in operation within the system that holds information about assets.

[1210] "Asset data" refers to a collection of data that includes detailed information about assets, such as type, quantity, location, and condition.

[1211] A "generative AI model" is an artificial intelligence model used to predict the state of an asset based on imported data.

[1212] "Training methods" refer to the techniques and methods used to perform the processes necessary to improve the accuracy of generative AI models using data.

[1213] An "inventory schedule" is a plan that outlines the timing and sequence for verifying and inspecting assets.

[1214] A "terminal" refers to a device or equipment used by workers that can receive schedules and input results.

[1215] "Imported data" refers to asset data retrieved from an existing database and incorporated into the system.

[1216] "CSV or JSON" refers to a type of file format used for data exchange and storage.

[1217] "Data splitting methods" refer to methods and techniques for dividing data into training datasets and validation datasets.

[1218] An "industrial sensor" is a sensor installed in factories and work sites to collect information about equipment and parts in real time.

[1219] An "emotion engine" is a technology or system that analyzes a worker's facial expressions and voice to recognize their emotional state.

[1220] The "generated inventory schedule" is an inventory plan that is automatically generated based on the prediction results of the AI ​​model.

[1221] "Means of receiving" refers to the technologies and methods for receiving and managing data transmitted from a terminal.

[1222] "Means of notification" refers to the methods and technologies used to transmit information from a system to workers.

[1223] This invention relates to an asset management and work support system for a factory. This system acquires asset data from an existing database and predicts the asset status using a generative AI model. The aim is to optimize asset inventory and maintenance schedules, thereby reducing the burden on workers.

[1224] First, the server periodically retrieves asset data from an existing database. This data includes asset type, quantity, location, and status. This information is imported using SQL queries and then converted to CSV or JSON format. Data processing libraries such as Pandas are used for this process.

[1225] Subsequently, the server trains a generative AI model based on the imported data. Specifically, it splits the data into a training dataset and a validation dataset, and trains the model using machine learning libraries such as TensorFlow or PyTorch. Random forests and neural networks are suitable algorithms to use. The accuracy of the model is evaluated and improved through hyperparameter tuning.

[1226] Once the model training is complete, the server generates an asset inventory schedule based on the model's predictions. This schedule is designed to focus on assets at high risk of deterioration or those of high importance, and is then delivered to the terminals.

[1227] Furthermore, the server is equipped with an emotion engine that recognizes the worker's emotions in real time. When a worker performs inventory work, the emotion engine uses the camera and microphone on the terminal (e.g., smartphone or tablet) to identify emotions from their facial expressions and voice. This emotion data is transmitted to the server in real time, and the inventory schedule and notification methods are adjusted accordingly.

[1228] When a user performs tasks based on an inventory schedule, the results are entered into the terminal and sent to the server. The server receives these inventory results and updates the database. Similarly, when a user performs asset disposal tasks based on the disposal timing, the results are also entered into the terminal and sent to the server. The server updates the database again and uses this as feedback to improve the prediction accuracy of the model.

[1229] As a concrete example, at the beginning of the month, the server retrieves all asset data from the database and trains a generative AI model. This training includes predictions that older machinery and equipment are deteriorating and will need to be decommissioned the following month. Based on this information, the server generates an inventory schedule and decommissioning timing, and delivers it to the terminal. The user checks the inventory schedule, and as they perform the inventory work, the emotion engine recognizes their facial expressions and voice to analyze their emotions. For example, if fatigue is detected, the workload can be adjusted.

[1230] Example of a prompt:

[1231] "For training data for the asset management system within the factory, please enter the type, location, condition, quantity, and maintenance history of each piece of equipment. Also, please specify the features and target variables to be used for predicting the equipment's condition."

[1232] This system is expected to improve the efficiency of asset management and reduce labor costs and expenses. Furthermore, by utilizing an emotion engine, it is possible to reduce the burden on workers and improve the comfort of the work environment.

[1233] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1234] Step 1: Data Acquisition

[1235] The server retrieves asset data from an existing database. It uses SQL queries to access the database and retrieve information such as asset type, quantity, location, and status. The imported data is then converted to CSV or JSON format. Libraries such as Pandas are used for this conversion. The retrieved data is temporarily stored in storage for use in subsequent processing.

[1236] Input: Existing database

[1237] Output: Asset data in CSV or JSON format

[1238] Step 2: Training the generative AI model

[1239] The server trains generative AI models based on imported asset data. First, it splits the data into training and validation datasets. Next, it uses machine learning libraries such as TensorFlow and PyTorch to apply algorithms such as random forests and neural networks to train the models. Finally, it tunes hyperparameters to evaluate and improve the model's accuracy.

[1240] Input: Asset data in CSV or JSON format

[1241] Output: Trained AI model

[1242] Step 3: Generate an inventory schedule

[1243] The server generates an asset inventory schedule based on the predictions of a trained AI model. This schedule prioritizes assets with a high risk of deterioration or those of high importance. The generated schedule is then sent to the terminal.

[1244] Input: Trained AI model

[1245] Output: Inventory schedule

[1246] Step 4: Schedule Distribution

[1247] The server distributes the generated inventory schedule to the terminal. The terminal displays this schedule as a guideline for workers to carry out the inventory work.

[1248] Input: Inventory schedule

[1249] Output: Schedule displayed on the terminal

[1250] Step 5: Receive inventory results and update the database.

[1251] After completing the task, the user enters the inventory results into a terminal. The terminal sends the results to the server, which updates the database. This updated data is then used to train the next AI model.

[1252] Input: Inventory results entered from the terminal.

[1253] Output: Updated database

[1254] Step 6: Emotion recognition by the emotion engine

[1255] The server recognizes workers' emotions in real time through an emotion engine. Using cameras and microphones installed on the terminals, it analyzes workers' facial expressions and voices to acquire emotion data. This data is sent to the server, and inventory schedules and notification methods are adjusted accordingly.

[1256] Input: Emotional data obtained from the device

[1257] Output: Adjusted inventory schedule and notification methods

[1258] Step 7: Notification of the optimal timing for asset disposal

[1259] The server calculates the optimal timing for asset disposal based on the AI ​​model's predictions and notifies the worker. The user, upon receiving the notification, carries out the disposal work and enters the results into their terminal.

[1260] Input: AI model prediction results

[1261] Output: Notification of disposal timing

[1262] Step 8: Receive the removal results and update the database.

[1263] After the user completes the asset disposal process, they input the results into a terminal. The terminal sends the results to a server, which updates the database. This data will be used to train future AI models.

[1264] Input: Disposal results entered from the terminal

[1265] Output: Updated database

[1266] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1267] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1268] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1269] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1270] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1271] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1272] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1273] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1274] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1275] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1276] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1277] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1278] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1279] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1280] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1281] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1282] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1283] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1284] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1285] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1286] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1287] The following is further disclosed regarding the embodiments described above.

[1288] (Claim 1)

[1289] A means of obtaining asset data from an existing database,

[1290] A means for training a generative AI model to predict the state of assets based on imported asset data,

[1291] A means for generating an asset inventory schedule based on the prediction results of an AI model,

[1292] A means of distributing the generated inventory schedule to the terminal,

[1293] A means of receiving inventory results entered from a terminal and updating the database,

[1294] A means of calculating and notifying the optimal timing for asset disposal based on the prediction results of an AI model,

[1295] A means of receiving the removal result entered from the terminal based on the notified removal timing and updating the database,

[1296] A system that includes this.

[1297] (Claim 2)

[1298] In the system described in claim 1,

[1299] A system that further includes means for converting the format of imported data to CSV or JSON.

[1300] (Claim 3)

[1301] In the system described in claim 1,

[1302] A system further comprising a data partitioning mechanism for dividing the dataset into a training dataset and a validation dataset.

[1303] "Example 1"

[1304] (Claim 1)

[1305] A means of obtaining asset data from an existing database,

[1306] A means for training a generative AI model to predict the state of assets based on imported asset data,

[1307] Training involves splitting the dataset into a training dataset and a validation dataset, and adjusting the hyperparameters.

[1308] A means for generating an asset inventory schedule based on the prediction results of an AI model,

[1309] A means of distributing the generated inventory schedule to the terminal,

[1310] A means of receiving inventory results entered from a terminal and updating the database,

[1311] A means of calculating and notifying the optimal timing for asset disposal based on the prediction results of an AI model,

[1312] A means of receiving the removal result entered from the terminal based on the notified removal timing and updating the database,

[1313] A system that includes this.

[1314] (Claim 2)

[1315] Further includes means to convert the format of imported data to CSV or JSON.

[1316] The system according to claim 1.

[1317] (Claim 3)

[1318] Further includes a data splitting mechanism for dividing the dataset into a training dataset and a validation dataset.

[1319] The system according to claim 1.

[1320] "Application Example 1"

[1321] (Claim 1)

[1322] A means of obtaining asset data from an existing database,

[1323] A means for training a generative AI model to predict the state of assets based on imported asset data,

[1324] A means for generating an asset inventory schedule based on the prediction results of an AI model,

[1325] A means of distributing the generated inventory schedule to the terminal,

[1326] A means of receiving inventory results entered from a terminal and updating the database,

[1327] A means of calculating and notifying the optimal timing for asset disposal based on the prediction results of an AI model,

[1328] A means of receiving the removal result entered from the terminal based on the notified removal timing and updating the database,

[1329] The system includes a means, including an application, that notifies the user of the predicted inventory schedule and disposal timing on the terminal, and allows the user to confirm and input the work.

[1330] A system that includes this.

[1331] (Claim 2)

[1332] The system according to claim 1, further comprising means for converting the format of imported data to CSV or JSON.

[1333] (Claim 3)

[1334] The system according to claim 1, further comprising means for dividing the dataset into a training dataset and a validation dataset.

[1335] "Example 2 of combining an emotion engine"

[1336] (Claim 1)

[1337] A means of obtaining asset data from an existing database,

[1338] A means for training a generative AI model to predict the state of assets based on imported asset data,

[1339] A means for generating an asset inventory schedule based on the prediction results of an AI model,

[1340] A means of distributing the generated inventory schedule to the terminal,

[1341] A means of receiving inventory results entered from a terminal and updating the database,

[1342] A means of calculating and notifying the optimal timing for asset disposal based on the prediction results of an AI model,

[1343] A means of receiving the removal result entered from the terminal based on the notified removal timing and updating the database,

[1344] A means for analyzing user emotion data acquired from the device's camera and microphone, and adjusting the inventory schedule and notification method based on the results,

[1345] A system that includes this.

[1346] (Claim 2)

[1347] The system according to claim 1, further comprising means for converting the format of imported data to CSV or JSON.

[1348] (Claim 3)

[1349] The system according to claim 1, further comprising means for dividing the dataset into a training dataset and a validation dataset.

[1350] "Application example 2 when combining with an emotional engine"

[1351] (Claim 1)

[1352] A means of obtaining asset data from an existing database,

[1353] A means for training a generative AI model to predict the state of assets based on imported asset data,

[1354] A means for generating an asset inventory schedule based on the prediction results of an AI model,

[1355] A means of distributing the generated inventory schedule to the terminal,

[1356] A means of receiving inventory results entered from a terminal and updating the database,

[1357] A means of calculating and notifying the optimal timing for asset disposal based on the prediction results of an AI model,

[1358] A means of receiving the removal result entered from the terminal based on the notified removal timing and updating the database,

[1359] A means of collecting information on equipment and parts acquired by industrial sensors in real time and transmitting it to a server,

[1360] A means of using an emotion engine to recognize the emotions of workers and adjusting inventory schedules and notification methods based on that data,

[1361] A system that includes this.

[1362] (Claim 2)

[1363] The system according to claim 1, further comprising means for converting the format of imported data to CSV or JSON.

[1364] (Claim 3)

[1365] The system according to claim 1, further comprising means for dividing the dataset into a training dataset and a validation dataset. [Explanation of Symbols]

[1366] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of obtaining asset data from an existing database, A means for training a generative AI model to predict the state of assets based on imported asset data, A means for generating an asset inventory schedule based on the prediction results of an AI model, A means of distributing the generated inventory schedule to the terminal, A means of receiving inventory results entered from a terminal and updating the database, A means of calculating and notifying the optimal timing for asset disposal based on the prediction results of an AI model, A means of receiving the removal result entered from the terminal based on the notified removal timing and updating the database, A system that includes this.

2. In the system described in claim 1, A system that further includes means for converting the format of imported data to CSV or JSON.

3. In the system described in claim 1, A system further comprising a data partitioning mechanism for dividing the dataset into a training dataset and a validation dataset.

Images