system

Abstract

A system is provided.SOLUTION: The system includes a means for collecting attendance data of employees, a means for collecting PC use data of the employees, a means for preprocessing the collected data, a means for analyzing the preprocessed data to predict the busy / idle status of the work, a means for calculating a recommended timing for taking a leave of absence for each employee, a means for proposing how to spend time when taking a leave of absence based on the hobby and preference of the employee, a means for visualizing a busy / idle status and a means for notifying each employee terminal of the generated proposal contents and a report.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In today's work environment, it is important to maximize employee work efficiency and ensure that employees take their paid leave at the appropriate time. However, many companies lack a way to accurately predict their employees' busy and slow periods, and they also lack systems that suggest appropriate ways to spend their paid leave or how to use employee benefits. This can result in increased employee stress and reduced work efficiency. Furthermore, it is difficult for HR departments to grasp the overall business situation and paid leave usage status, making it difficult for them to properly manage labor. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means.

[0006] The system is equipped with a means of collecting employee attendance data and PC usage data from each employee's device, a means of preprocessing the collected data, and a means of analyzing the preprocessed data to predict business busy / quick times. It calculates the recommended timing for each employee to take paid leave, suggests ways to spend paid leave based on the employee's hobbies and preferences, and also suggests the optimal employee benefit system for when paid leave is taken. This system makes it possible to visualize the busy / quick times and paid leave usage status of each department, and notify each employee of the generated proposals and reports. This makes it possible to optimize the work efficiency of each employee and improve labor management across the company.

[0007] "Employee attendance data" refers to records of information about each employee's working hours, such as start and finish times, and break times.

[0008] "PC usage data" includes information about device usage, such as employee computer usage, program execution logs, browser history, and keystroke records.

[0009] "Preprocessing means" refers to the process of cleaning collected data, completing missing values, removing noise, and making it suitable for analysis.

[0010] "Data analytics" refers to the activity of recognizing patterns and making predictions based on collected and pre-processed data using specific algorithms and machine learning models.

[0011] The "business situation" refers to the state or level that indicates how busy or slow the business is during a specific period.

[0012] "Recommended timing for taking paid leave" refers to the time or period recommended for the most appropriate time for employees to take paid leave.

[0013] "Hobbies and preferences" refers to the personal interests and preferences, activities, or tendencies of individual employees.

[0014] A "welfare system" is a system that includes additional benefits and services provided to employees by the company, such as insurance, health checkups, and facility use.

[0015] "Visualization" is the process of visually representing the state of data and the results of analysis to make them easier to understand.

[0016] "Notification" refers to the act of informing employees of generated information or suggestions via their devices or email, etc. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a 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.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the 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.

[0031] 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 of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process 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" according to the technology of the present 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 process 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.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] This invention provides a system that analyzes employee attendance data, PC usage data, etc., predicts the busy and slow periods of work that vary for each employee and department, and suggests the best time to take paid leave, how to spend it when it is taken, and how to best use employee benefits. The program for this system is explained in detail below.

[0039] 1. Data Collection

[0040] First, attendance data and PC usage data are periodically collected from each employee's device. The attendance data includes start and end times of work, break times, etc., while the PC usage data includes program execution logs and application usage history.

[0041] Server: Automatically collects data from employee devices and stores it in a central database.

[0042] Terminal: Attendance data entry and PC usage data log collection are performed in the background.

[0043] 2. Data Preprocessing

[0044] Next, the collected data is preprocessed, which involves filling in missing data, correcting outliers, and standardizing the data format.

[0045] Server: Cleans and preprocesses the collected data. Missing data is filled in using past data and average values.

[0046] 3. Data Analysis

[0047] Predict business activity based on pre-processed data. Use machine learning models and statistical algorithms to analyze the busy and slow periods for each employee and department.

[0048] Server: Trains the machine learning model, performs analysis, and sets confidence intervals to improve prediction accuracy.

[0049] 4. Calculating the recommended timing for taking paid leave

[0050] Based on the analysis results, the system calculates the optimal timing for each employee to take paid leave, taking into account past paid leave usage patterns and the department's overall schedule.

[0051] Server: Calculates the optimal time to take paid leave based on each employee's busy / quick schedule.

[0052] 5. Suggestions for how to spend your paid leave

[0053] Furthermore, the system uses data on employees' hobbies and preferences to suggest optimal ways to spend their paid leave, and utilizes a recommender system to recommend activities and travel destinations that match individual interests.

[0054] Server: Based on past data and preference information, suggests the best way for each employee to spend their time.

[0055] 6. Assistance in optimal use of employee benefits

[0056] We also propose employee benefit systems that can be used when taking paid leave, and provide optimal ways to use employee benefits so that employees can spend their paid leave meaningfully.

[0057] Server: Extracts and proposes systems suitable for each employee from a database of employee benefits.

[0058] 7. Visualization of business and vacation schedules by department

[0059] Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized, enabling the HR department to carry out effective labor management.

[0060] Server: Uses visualization tools to display data in graphs and dashboards.

[0061] 8. User Notification and Interface

[0062] Finally, the generated proposals and reports are sent to each employee's device, and employees can use them to apply for paid leave and use employee benefits.

[0063] Server: Generates proposals and reports and notifies each employee via email or a dedicated app.

[0064] Terminal: The employee checks the received notification and takes the necessary steps.

[0065] Specific examples

[0066] For example, if Employee A is considering taking paid leave next week, the system will analyze Employee A's past attendance data and PC usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system will suggest the most suitable camping spot or outdoor activity. Furthermore, the system will recommend the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan their paid leave based on that information.

[0067] In this way, the system of the present invention maximizes employee work efficiency and supports appropriate use of paid leave and employee benefits.

[0068] The processing flow will be explained below.

[0069] Step 1: Data collection

[0070] Server: Periodically collects attendance data and PC usage data from each employee's device. Specifically, the data is sent via API or a dedicated client app and stored in a central database.

[0071] Terminal: Employees enter their work start time, end time, and break time. PC usage data (program execution logs, browser history, etc.) is collected in the background and sent to the server in real time.

[0072] Step 2: Data Preprocessing

[0073] Server: Cleans the collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[0074] Specific behavior:

[0075] 1. Imputing missing data: Filling in missing data with past data or average values.

[0076] 2. Outlier Correction: Correcting or removing abnormal data based on pre-set criteria.

[0077] 3. Data format unification: Convert all data into a consistent format.

[0078] Step 3: Data analysis

[0079] Server: Analyzes the pre-processed data and uses machine learning models to predict the busy and slow periods for each employee and department. It uses algorithms such as random forests and neural networks.

[0080] Specific behavior:

[0081] 1. Training a machine learning model: Train the model based on historical data.

[0082] 2. Predicting peak and trough periods: Use the trained model to predict future peak and trough periods.

[0083] 3. Setting confidence intervals: Calculate confidence intervals for the predicted results to ensure the reliability of the results.

[0084] Step 4: Calculate the best time to take paid leave

[0085] Server: Based on the results of data analysis, calculates the optimal timing for each employee to take paid leave, taking into account the department's overall schedule and past paid leave usage patterns.

[0086] Specific behavior:

[0087] 1. Check the busy / quiet periods: Check the busy / quiet periods of each employee and identify the quiet periods.

[0088] 2. Department-wide coordination: Consider the busy and slow times of the entire department and choose a time that will minimize the impact on other employees.

[0089] 3. Calculating the timing of taking paid leave: Calculate and recommend the optimal time to take leave.

[0090] Step 5: Suggestions for how to spend your paid leave

[0091] Server: Based on the user's hobby and preference data, the server proposes the best way to spend their vacation time. Using a recommender system, the server recommends activities and travel destinations suitable for each user.

[0092] Specific behavior:

[0093] 1. View user profiles: Check past activity and interests.

[0094] 2. Recommendation generation: Generate recommendations tailored to the user's preferences.

[0095] 3. Save suggestions: Save suggestions for each user.

[0096] Step 6: Assistance in optimizing employee benefits

[0097] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on the user profile.

[0098] Specific behavior:

[0099] 1. Browse the benefits database: See a list of available benefits.

[0100] 2. Matching with user profile: Selecting benefit plans that match the user profile.

[0101] 3. Proposal generation: Propose and record optimal ways to use benefits.

[0102] Step 7: Visualize business and leisure situations by department and paid leave usage status

[0103] Server: Based on the analysis results, the busy / slow status of each department and the status of paid leave taken are visualized. This is displayed in the form of a dashboard or report.

[0104] Specific behavior:

[0105] 1. Data aggregation: Aggregate data by department.

[0106] 2. Applying visualization tools: Representing data visually using graphs and heat maps.

[0107] 3. Report generation: Generate dashboards and PDF reports and provide them to the HR department.

[0108] Step 8: User Notification and Interface

[0109] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[0110] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[0111] Specific behavior:

[0112] 1. Notification Generation: Converting proposals and reports into notification format.

[0113] 2. Send notifications: Send notifications to users via email or app notifications.

[0114] 3. User interface operation: The user takes the necessary steps based on the notification received.

[0115] Example 1

[0116] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0117] Conventional labor management systems are insufficient to maximize work efficiency and improve employee satisfaction because they are unable to comprehensively analyze employees' attendance data and PC usage data to suggest the appropriate timing for taking paid leave and the optimal way to use employee benefits. Furthermore, previous systems did not suggest ways to spend paid leave that took into account the hobbies and preferences of individual employees. This invention is intended to solve these problems.

[0118] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0119] In this invention, the server includes a means for collecting employee work data, a means for collecting employee computer usage data, and a means for preprocessing the collected data. This enables efficient data collection and preprocessing. It also includes a means for predicting workloads based on the analyzed data, a means for calculating recommended vacation timings for each employee, a means for suggesting vacation time based on employee preferences, a means for suggesting optimal employee benefit plans for vacation time, a means for visualizing the workload and vacation status for each department, and a means for notifying each employee of the generated proposals and reports. This maximizes employee work efficiency and supports appropriate use of paid vacation time and employee benefits.

[0120] "Employee work data" refers to information such as start time, finish time, and break time that is recorded when an employee works.

[0121] "Employee computer usage data" refers to information such as execution logs of computer programs used by employees during work hours and application usage history.

[0122] "Preprocessing" refers to the process of preparing collected raw data in a format suitable for analysis by supplementing missing data, correcting outliers, and standardizing data formats.

[0123] "Analysis" refers to the process of applying statistical techniques and machine learning models to preprocessed data to predict business activity levels and individual trends.

[0124] "Calculating the recommended timing" refers to calculating the most appropriate time for each employee to take paid leave based on the analyzed data.

[0125] "Suggestions based on hobbies and preferences" refers to recommending ways and activities to spend paid leave based on information about each employee's hobbies and interests.

[0126] "Proposing a benefit system" refers to recommending the most appropriate benefit program that an employee can use during their paid leave.

[0127] "Visualization" refers to the process of displaying analysis results and various data in graph or dashboard format to make them easy to understand.

[0128] "Notifying" refers to the process of sending the proposals and reports generated by the server to each employee's terminal to inform them.

[0129] "System" refers to a set of processes and devices that comprehensively combine the above-mentioned measures to maximize employee work efficiency and support appropriate use of paid leave and employee benefits.

[0130] The present invention is a system that analyzes employee work data and computer usage data, predicts busy and slow times at work, and suggests optimal times and ways to use paid leave, as well as how to utilize employee benefits. The system of the present invention is operated in cooperation with employee terminals and a server.

[0131] Data collection

[0132] First, a collection program is installed on each employee's device. This program continuously collects work data and computer usage data in the background. The data collected by the device is sent to a server at regular intervals. A central database is installed on the server, where the sent data is efficiently stored.

[0133] Data Preprocessing

[0134] Collected data is often incomplete as it is, so it is necessary to supplement missing data, correct outliers, and standardize the data format. This is called preprocessing, and is performed by the server. The server uses a dedicated cleaning algorithm to prepare the data into the appropriate format.

[0135] Data analysis

[0136] Based on the pre-processed data, the server analyzes the workload status. For the analysis, a machine learning model (e.g., random forest or support vector machine) is used. The server uses the trained model to predict the workload status of each employee and department. Confidence intervals can also be set to improve the accuracy of the prediction.

[0137] Calculating the best time to take paid leave

[0138] Based on the analyzed data, the server calculates the optimal timing for each employee to take paid leave, taking into account past paid leave usage patterns and the department's overall schedule.

[0139] Suggestions for how to spend your paid leave

[0140] Using employee hobby and preference data, the server suggests optimal ways to spend paid leave. Utilizing a recommender system, it recommends appropriate activities and travel destinations for each employee.

[0141] Optimal use of employee benefits

[0142] The server also suggests employee benefit plans that can be used when taking paid leave, extracting the best plan for each employee from a database of employee benefits and supporting them in using their paid leave in a meaningful way.

[0143] Visualization of the situation by department

[0144] Based on the analysis results, the server visualizes the busy / unbusy status and paid leave usage status of each department, and displays the results in intuitive graphs and dashboards using visualization tools (e.g., Tableau or Power BI).

[0145] Notification of proposal details

[0146] The final proposals and reports are sent to each employee's device. The server then sends the proposals via email or a dedicated application, allowing employees to apply for paid leave or use employee benefits.

[0147] Specific examples

[0148] For example, if Employee A is considering taking paid leave next week, the system analyzes Employee A's past work data and computer usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system suggests the most suitable camping spot or outdoor activity. It also recommends the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan their paid leave based on that information.

[0149] Prompt Sentence Examples

[0150] "Based on employee A's attendance data and PC usage data from the past six months, please predict the busy and slow periods for next week and suggest the best time and way for him to take paid leave."

[0151] In this way, the system of the present invention maximizes employee work efficiency and supports appropriate use of paid leave and employee benefits.

[0152] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0153] Step 1:

[0154] Data collection

[0155] Device:

[0156] Input: Employee working hours, break times, finish times, program usage, etc.

[0157] Processing: Work data and computer usage data are collected in the background.

[0158] Output: The raw data collected.

[0159] Specific operation: The terminal software automatically records the time when employees start and finish work and collects computer usage history as a log.

[0160] Step 2:

[0161] Data Transfer

[0162] Device:

[0163] Input: Raw data collected.

[0164] Processing: Data collected on the device is transferred to the server at regular intervals.

[0165] Output: The data sent to the server.

[0166] Specific operation: Data is sent from the terminal to the server via the communication module.

[0167] Step 3:

[0168] Data Preprocessing

[0169] server:

[0170] Input: Raw data received from the terminal.

[0171] Processing: Impute missing data, correct outliers, and standardize data formats.

[0172] Output: A clean dataset.

[0173] Specific operation: Run a data cleaning algorithm (e.g., scikit-learn's SimpleImputer) on the server to standardize the data.

[0174] Step 4:

[0175] Data analysis

[0176] server:

[0177] Input: The preprocessed dataset.

[0178] Processing: Predicting peak and trough times using machine learning models.

[0179] Output: Predicted busy / quick times for each employee and department.

[0180] What you'll do: Use Python libraries to train and run machine learning models (e.g., random forests, support vector machines).

[0181] Step 5:

[0182] Calculating the timing of paid leave

[0183] server:

[0184] Input: Predicted busy / slow times, past paid leave data, department schedule.

[0185] Processing: Calculate the optimal timing for taking paid leave.

[0186] Output: A list of optimal vacation times for each employee.

[0187] Specific operation: Using past data, an algorithm is used to calculate the optimal timing for taking paid leave.

[0188] Step 6:

[0189] Suggestions for spending time

[0190] server:

[0191] Input: Employee hobby and preference data, information on the timing of taking paid leave.

[0192] Processing: Use a recommender system to suggest the best way to spend your time.

[0193] Output: Specific activities and travel destination suggestions for each employee.

[0194] How it works: The recommender system analyzes employee preference data and suggests activities that are best suited to them.

[0195] Step 7:

[0196] Assistance in optimal use of employee benefits

[0197] server:

[0198] Input: List of optimal vacation times, employee benefits database.

[0199] Processing: Extraction of benefit plans available when taking paid leave.

[0200] Output: Proposal of optimal employee benefits plan for each employee.

[0201] What it does: Searches the benefits database to extract and recommend the best options.

[0202] Step 8:

[0203] visualization

[0204] server:

[0205] Input: Data on business / slow periods by department and paid leave usage status.

[0206] Processing: Visualize analysis results in graphs and dashboards.

[0207] Output: Visualization report for HR department.

[0208] Specific behavior: Use visualization tools (e.g., Tableau, Power BI) to display analysis results.

[0209] Step 9:

[0210] notification

[0211] server:

[0212] Input: Proposal and report.

[0213] Processing: Proposals and reports are sent to each employee's device.

[0214] Output: Notification sent to each employee terminal.

[0215] Specific operation: The generated information is notified to each employee via email or a dedicated application.

[0216] Specific examples

[0217] For example, if employee A is considering taking paid leave next week, the system analyzes employee A's past work data and computer usage data to predict that next week will be a relatively quiet period. Then, based on employee A's interest in outdoor activities, the system suggests the best camping spots and outdoor activities. It also recommends the use of company-provided recreational facilities as part of the employee benefits program available during that period. This information is sent to employee A's device, allowing them to apply for and plan their paid leave based on that information.

[0218] Example prompt for a generative AI model:

[0219] "Based on employee A's attendance data and PC usage data from the past six months, please predict the busy and slow periods for next week and suggest the best time and way for him to take paid leave."

[0220] (Application example 1)

[0221] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0222] In many modern organizations, in addition to employee attendance management and PC usage status monitoring, the operation and management of robots used in factories is also a key issue. However, managing these data separately reduces overall efficiency and makes it difficult to create consistent operational plans. Furthermore, efficient proposals are often not made regarding the appropriate timing for employees to take paid leave or how to utilize employee benefits when doing so. Furthermore, with regard to factory robots, operational data is not fully utilized, and optimal maintenance timing can be overlooked. A system that can eliminate these management complexities and reduced efficiency is needed.

[0223] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0224] In this invention, the server includes means for collecting employee attendance data, means for collecting employee PC usage data, means for collecting factory robot operation data, means for preprocessing the collected data, means for analyzing the preprocessed data to predict business conditions and optimal robot maintenance times, means for calculating recommended paid leave acquisition times and efficient robot operation plans for each employee, means for suggesting ways to spend paid leave based on employee preferences, means for suggesting optimal employee benefit plans for when paid leave is taken, means for visualizing business conditions and paid leave acquisition status by department, and means for notifying each employee terminal and the robot management system of the generated proposals and reports. This enables maximization of employee labor efficiency and optimal robot operation.

[0225] "Employee attendance data" refers to information related to working hours, such as the hours worked by an employee, arrival and departure times, and break times.

[0226] "Employee PC usage data" refers to data that indicates the usage status of computers used by employees during work hours, and specifically includes program execution logs and application usage history.

[0227] "Operation data of robots in factories" refers to data related to the operating status and operational status of robots operating in factories, including the start and end times of operation, the number of tasks performed, and energy consumption.

[0228] "Preprocessing" refers to the process of preparing collected data in a form that is easy to analyze. Specifically, this involves filling in missing data, standardizing data formats, correcting outliers, etc.

[0229] "Analysis" is the processing of data to identify patterns and trends based on pre-processed data and make predictions or recommendations based on those patterns and trends.

[0230] The "business situation" indicates the busy and quiet periods of business, and reflects the level of workload.

[0231] The "optimal maintenance period" refers to the most suitable time to perform maintenance in order to maintain the robot's operating efficiency at its maximum.

[0232] "Recommended timing" is information that identifies the most appropriate time for each employee to take paid leave.

[0233] An "efficient operation plan" indicates the schedule and strategy for operating the robot most efficiently.

[0234] "Suggestions for how to spend time" are suggestions for the best way for employees to use their time when they take paid leave, and are based on individual hobbies and preferences.

[0235] "Employee benefit system" refers to the various employee benefit services and systems available to employees, including recreational facilities and health promotion programs.

[0236] "Visualization" is a technique that makes data easier to understand by visually displaying the results of data analysis in graphs, dashboards, etc.

[0237] "Notifications" are used to inform employees and operations managers of the results of analysis and proposals, and are sent via email or a dedicated application.

[0238] This invention is a system that analyzes employee attendance data, PC usage data, and factory robot operation data, predicts business activity and optimal maintenance timing for the robots, and proposes optimal operation plans and methods for using employee benefits. This system includes the following processing steps.

[0239] 1. Data Collection:

[0240] The server collects data from employee terminals and factory robots. Work attendance data includes start and end times, break times, etc., while PC usage data includes program execution logs and application usage history. Robot operation data includes start and end times, number of completed tasks, and energy consumption.

[0241] 2. Data Preprocessing:

[0242] The server preprocesses the collected data, specifically by completing missing data, correcting outliers, standardizing data formats, etc. The preprocessed data is then prepared for analysis.

[0243] 3. Data Analysis:

[0244] The server uses machine learning models, such as the University of Pittsburgh's RandomForestRegressor, to predict the best times for robot maintenance and the workload based on the pre-processed data.

[0245] 4. Calculating optimal timing and operational planning:

[0246] Based on the analysis results, the server calculates the recommended timing for each employee to take paid leave and the efficient operation plan for the robot, taking into account past paid leave usage patterns and the schedule of the entire department.

[0247] 5. Suggested working hours and benefits:

[0248] The server uses employee hobby and preference data to suggest optimal ways to spend paid leave. It also suggests employee benefits that can be used when taking paid leave. This allows employees to spend their paid leave meaningfully.

[0249] 6. Visualization:

[0250] Based on the analysis results, the server visualizes the busy / slow times and paid leave usage status for each department, enabling the HR department to carry out effective labor management. Graphs and dashboards are used as visualization tools.

[0251] 7. Notifications and Interface:

[0252] The server then sends the generated proposals and reports to each employee's device and the robot management system, allowing employees to apply for paid leave and use employee benefits.

[0253] As a concrete example, if Employee A is considering taking paid leave next week, the system analyzes Employee A's past attendance data and PC usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system suggests the most suitable camping spot or outdoor activity. Furthermore, the system recommends the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan his paid leave based on that information.

[0254] Examples of prompts to input into a generative AI model include:

[0255] Input prompt: "Based on the recent operation data of robot R1, please suggest the next optimal maintenance timing and efficient operation plan. Operation start time is 2023-09-01 08:00:00, operation end time is 2023-09-01 10:00:00, number of completed tasks is 20, and energy usage is 25.5."

[0256] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0257] Step 1:

[0258] The server collects data from employee terminals and factory robots. Specifically, it collects attendance data (start and end times of work, break times), PC usage data (program execution logs and application usage history), and robot operation data (start and end times of operation, number of tasks completed, energy consumption). It receives various data as input and stores the collected data in a central database as output.

[0259] Step 2:

[0260] The server preprocesses the collected data. Specific operations include filling in missing data, standardizing data formats, and correcting outliers. For example, missing data is filled in using past data or average values, and outliers are corrected to within an appropriate range. The server receives the collected data as input and generates preprocessed data as output.

[0261] Step 3:

[0262] The server uses a machine learning model based on the preprocessed data to predict the busy and slow times of operations and the optimal maintenance times for the robots. Specifically, it performs analysis using algorithms such as RandomForestRegressor. It receives the preprocessed data as input and obtains the predicted results as output.

[0263] Step 4:

[0264] Based on the analysis results, the server calculates the recommended timing for taking paid leave for each employee and an efficient robot operation plan. Specifically, it generates an optimal schedule by taking into account past paid leave usage patterns and the department's overall schedule. It receives the analysis results as input and obtains the optimal timing and operation plan as output.

[0265] Step 5:

[0266] The server uses employee hobby and preference data to suggest optimal ways to spend paid leave. It then uses a recommender system to make suggestions based on employee feedback data and hobby data. It also suggests optimal employee benefit programs. It receives hobby and preference data and past feedback data as input, and obtains suggestions for how to spend the time and employee benefit programs as output.

[0267] Step 6:

[0268] The server visualizes the busy / slow times and paid vacation status for each department based on the analysis results. Specifically, it uses a visualization tool that displays the data in graph and dashboard format. It receives the analysis results as input and generates visualized graphs and dashboards as output.

[0269] Step 7:

[0270] The server notifies each employee's device and the robot management system of the generated proposals and reports. Specifically, it uses a mechanism to notify via email or a dedicated app. It receives proposals and reports as input and sends notification information as output.

[0271] Through this series of steps, the system can maximize employee labor efficiency and optimize robot operation.

[0272] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0273] This invention provides a system that analyzes employee attendance data, PC usage data, emotional data, etc., predicts the busy and slow periods of work that vary for each employee and department, and suggests the best time to take paid leave, how to spend it when taken, and how to best use employee benefits. The program for this system is explained in detail below.

[0274] 1. Data Collection

[0275] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. Attendance data includes work start and end times, break times, etc., while PC usage data includes program execution logs and application usage history. Emotion data is extracted using an emotion engine from the user's facial expressions, voice tone, text input, etc.

[0276] Terminal: Employees enter their start and end times and break times, and PC usage data and emotional data (facial expression analysis, voice tone analysis, etc.) are collected in the background and sent to the server in real time.

[0277] 2. Data Preprocessing

[0278] Server: Cleans all collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[0279] 3. Data Analysis

[0280] Server: Based on the preprocessed data, it uses machine learning models to predict the busy and slow periods for each employee and department. It also analyzes emotional data and evaluates the user's mental state.

[0281] 4. Calculating the recommended timing for taking paid leave

[0282] Server: Based on the results of data analysis, calculates the optimal timing for each employee to take paid leave. This takes into account not only the department's overall schedule and past paid leave usage patterns, but also the employee's mental state based on emotional data.

[0283] 5. Suggestions for how to spend your paid leave

[0284] Server: Based on the user's hobby, preference, and emotional data, the server proposes the best way to spend their vacation time. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also takes into account the user's mental state and suggests ways to promote relaxation and refreshment.

[0285] 6. Assistance in optimal use of employee benefits

[0286] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on emotional data and user profile. For example, if there is a high need for relaxation, suggests spa facilities and counseling services.

[0287] 7. Visualization of business and vacation schedules by department

[0288] Server: Based on the analysis results, the busy / slow periods and paid leave usage status for each department are visualized, enabling the HR department to carry out effective labor management.

[0289] 8. User Notification and Interface

[0290] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[0291] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[0292] Specific examples

[0293] For example, suppose employee B plans to take paid leave soon. The system analyzes employee B's attendance data and PC usage data to predict that the following week will be a relatively quiet period. It also analyzes employee B's emotional data and recognizes that employee B's mental state has recently been one that makes him prone to stress. As a result, the system suggests the best time to take paid leave and notifies employee B of refreshing ways to spend the day that suit his hobbies and preferences (for example, suggestions for nature walks or relaxation spots). At the same time, it also recommends that employee B take advantage of employee benefits that have a relaxing effect (for example, vouchers for spa facilities). Employee B can use these suggestions to apply for paid leave and take advantage of employee benefits.

[0294] In this way, the system of the present invention maximizes employee work efficiency and supports employees in taking paid leave and using employee benefits while taking into account emotional data.

[0295] The processing flow will be explained below.

[0296] Step 1: Data collection

[0297] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. Specifically, the data is sent via API or a dedicated client app and stored in a central database.

[0298] Terminal: Employees enter their work start time, end time, and break time. In the background, PC usage data (program execution logs, browser history, etc.) and emotional data (facial expression analysis, voice tone analysis, etc.) are collected and sent to the server in real time.

[0299] Step 2: Data Preprocessing

[0300] Server: Cleans all collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[0301] Specific behavior:

[0302] 1. Imputing missing data: Filling in missing data with past data or average values.

[0303] 2. Outlier Correction: Correcting or removing abnormal data based on pre-set criteria.

[0304] 3. Data format unification: Convert all data into a consistent format.

[0305] Step 3: Analyze the emotion data

[0306] Server: Analyzes collected emotional data and evaluates the user's mental state using an emotion engine, including facial expression recognition, voice tone analysis, and text analysis.

[0307] Specific behavior:

[0308] 1. Facial expression data analysis: Facial expression data collected from camera footage is analyzed to determine the emotional state.

[0309] 2. Voice data analysis: Analyze voice tones and detect emotional fluctuations.

[0310] 3. Analysis of text data: Emotional words are extracted from chat logs and email content to evaluate the emotional state.

[0311] Step 4: Data analysis

[0312] Server: Based on preprocessed attendance data, PC usage data, and emotion data, machine learning models are used to predict the busy / slow periods for each employee and department.

[0313] Specific behavior:

[0314] 1. Training a machine learning model: Train the model based on historical data.

[0315] 2. Predicting peak and trough periods: Use the trained model to predict future peak and trough periods.

[0316] 3. Setting confidence intervals: Calculate confidence intervals for the predicted results to ensure the reliability of the results.

[0317] Step 5: Calculate the best time to take paid leave

[0318] Server: Based on the results of data analysis and the evaluated emotional data, the server calculates the optimal timing for each employee to take paid leave, taking into account the department's overall schedule, past patterns of taking paid leave, and the employee's mental state based on emotional data.

[0319] Specific behavior:

[0320] 1. Check the busy / quiet periods: Check the busy / quiet periods of each employee and identify the quiet periods.

[0321] 2. Department-wide coordination: Consider the busy and slow times of the entire department and choose a time that will minimize the impact on other employees.

[0322] 3. Calculating the timing of taking paid leave: Calculate and recommend the optimal time to take leave.

[0323] Step 6: Suggestions for how to spend your paid leave

[0324] Server: Based on the user's hobby, preference, and emotional data, the server proposes the best way to spend their vacation time. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also takes into account the user's mental state and suggests ways to promote relaxation and refreshment.

[0325] Specific behavior:

[0326] 1. View user profiles: Check past activity and interests.

[0327] 2. Recommendation generation: Generate suggestions tailored to the user's preferences and take into account sentiment data.

[0328] 3. Save suggestions: Save suggestions for each user.

[0329] Step 7: Assistance in optimizing employee benefits

[0330] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on emotional data and user profile. For example, if there is a high need for relaxation, suggests spa facilities and counseling services.

[0331] Specific behavior:

[0332] 1. Browse the benefits database: See a list of available benefits.

[0333] 2. Matching with user profile: Selecting benefit plans that match the user profile.

[0334] 3. Proposal generation: Propose and record optimal ways to use benefits.

[0335] Step 8: Visualize business and vacation status by department

[0336] Server: Based on the analysis results, the busy / slow status of each department and the status of paid leave taken are visualized. This is displayed in the form of a dashboard or report.

[0337] Specific behavior:

[0338] 1. Data aggregation: Aggregate data by department.

[0339] 2. Applying visualization tools: Representing data visually using graphs and heat maps.

[0340] 3. Report generation: Generate dashboards and PDF reports and provide them to the HR department.

[0341] Step 9: User Notification and Interface

[0342] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[0343] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[0344] Specific behavior:

[0345] 1. Notification Generation: Converting proposals and reports into notification format.

[0346] 2. Send notifications: Send notifications to users via email or app notifications.

[0347] 3. User interface operation: The user takes the necessary steps based on the notification received.

[0348] Example 2

[0349] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0350] In today's corporate environment, maximizing employee work efficiency and maintaining a healthy working environment are important. However, conventional systems only collect attendance data and PC usage data, and do not adequately consider employees' emotions and mental state. Furthermore, they are unable to provide appropriate suggestions regarding the timing of paid leave and how to spend it when taken, creating challenges in optimizing employee welfare benefits. This invention solves these issues by providing a system that suggests optimal work life for each employee.

[0351] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0352] In this invention, the server includes means for collecting attendance data, PC usage data, and emotion data from each employee's terminal, means for cleaning the collected data and for completing and correcting missing and outlier values, means for analyzing the preprocessed data and predicting business activity using a machine learning model, means for analyzing each employee's emotion data and evaluating their mental state, means for calculating the optimal timing for taking paid leave based on the business activity and emotion data, means for suggesting optimal ways to spend paid leave based on the employee's hobby and preference data and emotion data, means for suggesting employee benefit plans available when taking paid leave, means for visualizing business activity and paid leave usage status by department, and means for notifying each employee of the generated suggestions and reports. This maximizes employee work efficiency and enables the creation of a work environment that takes mental health into consideration.

[0353] "Work attendance data" is information related to an employee's working status, such as start time, finish time, and break time.

[0354] "PC usage data" refers to information such as program execution logs and application usage history for computers used by employees during work hours.

[0355] "Emotion data" is information about an employee's emotions and mental state extracted using an emotion engine from a user's facial expressions, tone of voice, text input, etc.

[0356] A "machine learning model" is a mathematical algorithm or statistical model that learns from collected data and predicts patterns and trends.

[0357] "Busy / slow status" refers to the busyness or slowness of business during a specific period.

[0358] "Hobbies and Preference Data" is information about a user's interests and preferences, including hobbies, interests, and favorite activities.

[0359] A "welfare system" refers to various services and support systems provided by companies with the aim of improving the welfare of their employees, and includes, for example, vouchers for use of spa facilities and counseling services.

[0360] "Cleanup" is the process of filling in missing values ​​and correcting outliers in order to improve the quality of data.

[0361] "Visualization" is the process of visually representing data and information, making them easier to understand using graphs, charts, etc.

[0362] A "recommender system" is a system that recommends optimal activities, travel destinations, etc. based on a user's hobby and preference data.

[0363] "Notifications" are a means of informing users of specific information or offers in real time, and are provided via email or a dedicated app.

[0364] The present invention provides a system that analyzes employee attendance data, PC usage data, emotional data, etc., predicts the busy and slow business conditions that vary for each employee and department, and suggests the appropriate timing for taking paid leave, the appropriate way to spend it when taken, and the optimal way to use employee benefits.

[0365] 1. Data Collection

[0366] server

[0367] The server periodically collects attendance data, PC usage data, and emotion data from each employee's device. Attendance data includes work start and end times and break times, while PC usage data includes program execution logs and application usage history. Emotion data is extracted from the user's facial expressions, voice tone, text input, etc. using an emotion engine.

[0368] Terminal

[0369] Employees enter their work start and end times, as well as break times, into the terminals, which collect PC usage data and emotional data (facial expression analysis, voice tone analysis, etc.) in the background and send it to the server in real time.

[0370] 2. Data Preprocessing

[0371] server

[0372] All collected data is cleaned. Missing values ​​are filled in using past data or average values, and outliers are corrected or removed. Data formats are also standardized to make them suitable for analysis.

[0373] 3. Data Analysis

[0374] server

[0375] Based on the preprocessed data, a machine learning model is used to predict the busy and slow periods for each employee and department. Emotional data is also analyzed to evaluate the user's mental state. Specific software used includes the Python libraries Tensorflow (registered trademark) and Scikit-learn.

[0376] 4. Calculating the recommended timing for taking paid leave

[0377] server

[0378] Based on the results of the data analysis, the system calculates the optimal timing for each employee to take paid leave, taking into account not only the department's overall schedule and past paid leave usage patterns, but also the employee's mental state based on emotional data.

[0379] 5. Suggestions for how to spend your paid leave

[0380] server

[0381] Based on the user's hobby, preference and emotional data, the system suggests the best way to spend their paid vacation. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also includes suggestions to promote relaxation and refreshment. Specific software used includes Apache Spark (registered trademark) and ElasticSearch (registered trademark).

[0382] 6. Assistance in optimal use of employee benefits

[0383] server

[0384] It suggests employee benefits that can be used when taking paid leave, and suggests spa facilities and counseling services if employees need to relax based on their emotional data and profile information.

[0385] 7. Visualization of business and vacation schedules by department

[0386] server

[0387] Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized in graphs and charts, enabling the HR department to carry out effective labor management. For example, visualization can be done using Tableau or Power BI.

[0388] 8. User Notification and Interface

[0389] server

[0390] The generated proposals and reports are sent to each user's device, and notifications are sent via email or a dedicated app.

[0391] Terminal

[0392] The user checks the received notification and applies for paid leave or takes advantage of employee benefits.

[0393] Specific examples

[0394] For example, suppose employee B plans to take paid leave soon. The system analyzes employee B's attendance data and PC usage data to predict that the following week will be a relatively quiet period. It also analyzes employee B's emotional data and recognizes that employee B's mental state has recently been one that makes him prone to stress. As a result, the system suggests the best time to take paid leave and notifies employee B of refreshing ways to spend the day that suit his hobbies and preferences (for example, suggestions for nature walks or relaxation spots). At the same time, it also recommends that employee B take advantage of employee benefits that have a relaxing effect (for example, vouchers for spa facilities). Employee B can use these suggestions to apply for paid leave and take advantage of employee benefits.

[0395] Prompt Sentence Examples

[0396] Attendance data analysis prompt

[0397] Analyze employee attendance data to predict the next week's peak and trough times. The dataset includes start and finish times and break times.

[0398] Emotional Data Analysis Prompt

[0399] Analyze employee emotional data and assess their mental state. Data sets include facial expression analysis, voice tone, and text input.

[0400] Prompt to calculate the optimal timing for taking paid leave

[0401] Calculate the optimal timing for employees to take paid leave based on attendance and sentiment data.

[0402] Prompts for suggestions on how to spend your time based on your hobbies and preferences

[0403] Based on the user's hobby, preference, and emotional data, please suggest the best way to spend their vacation time. Please include suggestions that promote relaxation and refreshment.

[0404] Benefit Plan Recommendation Prompts

[0405] Based on emotional data and user profiles, suggest benefits that can be used when taking paid time off. If relaxation is needed, suggest spa facilities or counseling services.

[0406] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0407] Step 1: Data collection

[0408] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. The server sends data collection requests to each device at set intervals, receives the data sent from each device, and stores it in a database. The input is data from each employee's device, and the output is an integrated database.

[0409] Terminal: Employees enter their work start time, end time, and break time into a dedicated app. The terminal collects PC usage information (application usage history, usage time, etc.) and emotional data (facial expression analysis, voice tone analysis) in the background and sends them to the server in real time. The input is the employee's operations and background data collection, and the output is the data sent to the server.

[0410] Step 2: Data Preprocessing

[0411] Server: Cleans up all collected data. Specifically, if missing values ​​are found, they are filled in using past data or average values. Also, outliers are detected and corrected or removed. Data formats are standardized and made suitable for analysis. The input is raw data, and the output is cleaned data.

[0412] Step 3: Data analysis

[0413] Server: Using the preprocessed data, a machine learning model predicts the busy / quick schedule for each employee and department. Emotional data is also analyzed to evaluate each employee's mental state. Specific software used includes the Python libraries TensorFlow and Scikit-learn. The input is the cleaned-up data, and the output is predicted data on busy / quick schedules and the results of an evaluation of mental state.

[0414] Step 4: Calculate the best time to take paid leave

[0415] Server: Based on the analysis results, calculates the optimal timing for each employee to take paid leave. Calculates the optimal date to take paid leave by taking into account the department's overall schedule, past paid leave usage patterns, and mental state. The input is the analysis results, and the output is the recommended timing for taking paid leave.

[0416] Step 5: Suggestions for how to spend your paid leave

[0417] Server: Based on the user's hobby, preference, and emotional data, the server proposes the optimal way to spend their paid vacation. A recommender system is used to recommend activities and travel destinations suitable for each user. Specific software used includes Apache Spark and Elasticsearch. The input is hobby, preference, and emotional data, and the output is a proposal for the optimal way to spend the day.

[0418] Step 6: Assistance in optimizing employee benefits

[0419] Server: Proposes employee benefit programs that can be used when taking paid leave. Based on emotional data and profile information, recommends spa facilities or counseling services if there is a high need for relaxation. The input is emotional data and profile information, and the output is a proposed employee benefit program.

[0420] Step 7: Visualize business and leisure situations by department and paid leave usage status

[0421] Server: Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized in graphs and charts. This makes it easier for the HR department to carry out effective labor management. For example, visualization can be done using Tableau or Power BI. The input is the analysis results, and the output is visualized data.

[0422] Step 8: User Notification and Interface

[0423] Server: Sends the generated proposals and reports to each user's device. Notifications are sent via email or a dedicated app. The input is the generated proposals and reports, and the output is a notification message.

[0424] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits. The input is the notification message, and the output is the user's action.

[0425] (Application example 2)

[0426] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0427] While the operation of robots in modern factories improves production efficiency, their maintenance and operational status management are also important. In particular, there is a need to efficiently collect and analyze robot operation data and maintenance history data to predict optimal maintenance times. Also important is the overall improvement of work efficiency and employee benefits, including the appropriate timing and use of employee paid leave and suggestions for utilizing employee benefit programs.

[0428] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting employee attendance data, means for collecting employee PC usage data, means for collecting factory robot operation data and maintenance history data, means for preprocessing the collected data, means for analyzing the preprocessed data to predict business activity, means for calculating recommended paid vacation time for each employee and optimal maintenance times for each robot, means for proposing paid vacation time and maintenance content based on employee preferences and robot operation status, means for proposing optimal employee benefit programs for paid vacation time and robot maintenance, means for visualizing business activity and robot operation status and paid vacation time for each department, and means for notifying each employee terminal and each engineer terminal of the generated proposals and reports. This enables optimal paid vacation time and employee benefit use for each employee and efficient robot maintenance scheduling.

[0429] "Employee attendance data" refers to work-related record information such as the start and end times of each employee's shift, and break times.

[0430] "PC usage data" refers to data such as program execution logs and application usage history for computers used by employees for work.

[0431] "Robot operation data" refers to information such as the operation records, operating hours, and error occurrence status of robots operating in a factory.

[0432] "Maintenance history data" is recorded information such as the content, date, and person who performed the maintenance performed on the robot in the past.

[0433] "Preprocessing" refers to the process of preparing collected data for analysis, such as filling in missing values, correcting outliers, and standardizing data formats.

[0434] "Business activity status" is data showing the busy and slow periods of business, and indicates the workload status of employees and departments.

[0435] "Recommended timing for taking paid leave" is the time when the optimal time to take paid leave is calculated and suggested for each employee.

[0436] Based on "hobbies and preferences" means customizing the proposals to suit the interests and preferences of each employee.

[0437] "How to spend your paid leave" is a content that suggests activities and ways for employees to rest during the period they take paid leave.

[0438] "Employee benefits system" refers to various services and support systems provided by a company that employees can use, and often includes relaxation facilities and counseling services.

[0439] "Visualization of busy and slow periods" refers to visually displaying the results of data analysis so that busy and slow periods can be understood at a glance.

[0440] The "generated proposals" are proposals for each employee and robot that are calculated and created based on data analysis.

[0441] "Business and quiet periods by department" is data showing the busy and quiet periods for work in each specific department.

[0442] "Each employee terminal" refers to an individual device (PC, smartphone, etc.) used by an employee.

[0443] "Technician's Terminal" refers to an individual device used by a technician in a factory or operational location.

[0444] A "machine learning model" is an algorithm used in data analysis, a system that learns patterns from the original data and makes predictions and classifications.

[0445] A "confidence interval" is a numerical range that indicates the reliability of a prediction result and provides a statistical indication of how accurate the prediction is.

[0446] This invention is a system that analyzes employee attendance data, PC usage data, factory robot operation data, and maintenance history data, and based on that, suggests optimal timing for taking paid leave and performing maintenance. This system is implemented using both a server and a terminal.

[0447] Data collection

[0448] The server collects employee attendance data, PC usage data, factory robot operation data, and maintenance history data. These data are obtained from each employee's terminal and the factory management system. Specifically, employee attendance data includes work start and end times and break times, while PC usage data includes program execution logs and application usage history. Robot operation data includes operating hours and error occurrence status, and maintenance history data includes the details and dates of past maintenance.

[0449] Data Preprocessing

[0450] The server preprocesses the collected data, which includes filling in missing values, correcting outliers, and standardizing data formats, making the data suitable for analysis.

[0451] Data analysis

[0452] The server uses a machine learning model based on the preprocessed data to predict the busy and slow times of operations and the optimal maintenance times for the robot. This analysis uses a machine learning algorithm (e.g., Random Forest Regressor). It also uses a generative AI model to generate recommendations based on the user's preferences and the robot's operating status.

[0453] Paid leave and maintenance scheduling

[0454] Based on the analysis results, the server calculates the recommended timing for each employee to take paid leave and the optimal maintenance period for each robot, including setting a confidence interval based on past data and current operating conditions.

[0455] Suggestions and Notifications

[0456] The server then sends the generated proposals and reports to each employee's and technician's device via email or a dedicated application. The employee can then review the proposals, apply for paid leave and employee benefits, and schedule robot maintenance.

[0457] Specific examples

[0458] For example, if factory robot A has recently been operating slowly, the system will suggest the optimal date and time for the next maintenance based on past maintenance history and operation data. The next maintenance for this robot is predicted to be required two weeks from now on August 15th, and the appropriate maintenance work will be notified to the maintenance team.

[0459] Prompt Sentence Examples

[0460] "Please create a model that predicts the optimal next maintenance timing for each robot based on the factory robot operation data and maintenance history data. Also, please tell me the steps to develop a system that notifies technicians of this in the form of a smartphone application."

[0461] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0462] Step 1: Data collection

[0463] The server collects data from each employee's terminal and the factory management system. Inputs include employee attendance data, PC usage data, robot operation data, and maintenance history data. Outputs store this data on the server and prepare it in a format that can be used for the next processing step. Specific operations include retrieving data from a database and streaming data in real time.

[0464] Step 2: Data Preprocessing

[0465] The server preprocesses the collected data. The input is the raw data collected in step 1. Specific operations include imputing missing values, correcting outliers, and standardizing data formats. The output is clean data that has been prepared for analysis.

[0466] Step 3: Data analysis

[0467] The server uses a machine learning model to analyze the preprocessed data. The input is clean data. Specifically, it uses machine learning algorithms such as RandomForestRegressor to predict the busy / unbusy schedule and the optimal maintenance timing for the robot. The output is the predicted busy / unbusy schedule and recommended maintenance timing.

[0468] Step 4: Generate proposals

[0469] The server generates proposals using a generative AI model based on the analysis results. The input is the prediction results from step 3. Specific actions include customizing how employees spend their paid leave and the maintenance details, taking into account the individual hobbies and preferences of employees and the operating status of the robot. The output is specific proposals provided to the user.

[0470] Step 5: Notification

[0471] The server notifies each employee's terminal and the engineer's terminal of the generated proposal and report. The input is the proposal generated in step 4. Specific actions include sending an email or sending a notification via a dedicated application. The output is the proposal displayed on the user's terminal, which the user can check and take the next action.

[0472] Step 6: User Actions

[0473] The user takes action based on the notified proposal. The input is the proposal received in step 5. Specific actions include applying for paid leave, applying for employee benefits, and scheduling robot maintenance. The output is a record of the application and scheduling as actions taken by the user.

[0474] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the 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.

[0475] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0476] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0477] [Second embodiment]

[0478] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0479] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0480] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0481] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0482] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0483] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0484] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0485] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0486] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0487] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0488] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0489] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0490] This invention provides a system that analyzes employee attendance data, PC usage data, etc., predicts the busy and slow periods of work that vary for each employee and department, and suggests the best time to take paid leave, how to spend it when it is taken, and how to best use employee benefits. The program for this system is explained in detail below.

[0491] 1. Data Collection

[0492] First, attendance data and PC usage data are periodically collected from each employee's device. The attendance data includes start and end times of work, break times, etc., while the PC usage data includes program execution logs and application usage history.

[0493] Server: Automatically collects data from employee devices and stores it in a central database.

[0494] Terminal: Attendance data entry and PC usage data log collection are performed in the background.

[0495] 2. Data Preprocessing

[0496] Next, the collected data is preprocessed, which involves filling in missing data, correcting outliers, and standardizing the data format.

[0497] Server: Cleans and preprocesses the collected data. Missing data is filled in using past data and average values.

[0498] 3. Data Analysis

[0499] Predict business activity based on pre-processed data. Use machine learning models and statistical algorithms to analyze the busy and slow periods for each employee and department.

[0500] Server: Trains the machine learning model, performs analysis, and sets confidence intervals to improve prediction accuracy.

[0501] 4. Calculating the recommended timing for taking paid leave

[0502] Based on the analysis results, the system calculates the optimal timing for each employee to take paid leave, taking into account past paid leave usage patterns and the department's overall schedule.

[0503] Server: Calculates the optimal time to take paid leave based on each employee's busy / quick schedule.

[0504] 5. Suggestions for how to spend your paid leave

[0505] Furthermore, the system uses data on employees' hobbies and preferences to suggest optimal ways to spend their paid leave, and utilizes a recommender system to recommend activities and travel destinations that match individual interests.

[0506] Server: Based on past data and preference information, suggests the best way for each employee to spend their time.

[0507] 6. Assistance in optimal use of employee benefits

[0508] We also propose employee benefit systems that can be used when taking paid leave, and provide optimal ways to use employee benefits so that employees can spend their paid leave meaningfully.

[0509] Server: Extracts and proposes systems suitable for each employee from a database of employee benefits.

[0510] 7. Visualization of business and vacation schedules by department

[0511] Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized, enabling the HR department to carry out effective labor management.

[0512] Server: Uses visualization tools to display data in graphs and dashboards.

[0513] 8. User Notification and Interface

[0514] Finally, the generated proposals and reports are sent to each employee's device, and employees can use them to apply for paid leave and use employee benefits.

[0515] Server: Generates proposals and reports and notifies each employee via email or a dedicated app.

[0516] Terminal: The employee checks the received notification and takes the necessary steps.

[0517] Specific examples

[0518] For example, if Employee A is considering taking paid leave next week, the system will analyze Employee A's past attendance data and PC usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system will suggest the most suitable camping spot or outdoor activity. Furthermore, the system will recommend the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan their paid leave based on that information.

[0519] In this way, the system of the present invention maximizes employee work efficiency and supports appropriate use of paid leave and employee benefits.

[0520] The processing flow will be explained below.

[0521] Step 1: Data collection

[0522] Server: Periodically collects attendance data and PC usage data from each employee's device. Specifically, the data is sent via API or a dedicated client app and stored in a central database.

[0523] Terminal: Employees enter their work start time, end time, and break time. PC usage data (program execution logs, browser history, etc.) is collected in the background and sent to the server in real time.

[0524] Step 2: Data Preprocessing

[0525] Server: Cleans the collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[0526] Specific behavior:

[0527] 1. Imputing missing data: Filling in missing data with past data or average values.

[0528] 2. Outlier Correction: Correcting or removing abnormal data based on pre-set criteria.

[0529] 3. Data format unification: Convert all data into a consistent format.

[0530] Step 3: Data analysis

[0531] Server: Analyzes the pre-processed data and uses machine learning models to predict the busy and slow periods for each employee and department. It uses algorithms such as random forests and neural networks.

[0532] Specific behavior:

[0533] 1. Training a machine learning model: Train the model based on historical data.

[0534] 2. Predicting peak and trough periods: Use the trained model to predict future peak and trough periods.

[0535] 3. Setting confidence intervals: Calculate confidence intervals for the predicted results to ensure the reliability of the results.

[0536] Step 4: Calculate the best time to take paid leave

[0537] Server: Based on the results of data analysis, calculates the optimal timing for each employee to take paid leave, taking into account the department's overall schedule and past paid leave usage patterns.

[0538] Specific behavior:

[0539] 1. Check the busy / quiet periods: Check the busy / quiet periods of each employee and identify the quiet periods.

[0540] 2. Department-wide coordination: Consider the busy and slow times of the entire department and choose a time that will minimize the impact on other employees.

[0541] 3. Calculating the timing of taking paid leave: Calculate and recommend the optimal time to take leave.

[0542] Step 5: Suggestions for how to spend your paid leave

[0543] Server: Based on the user's hobby and preference data, the server proposes the best way to spend their vacation time. Using a recommender system, the server recommends activities and travel destinations suitable for each user.

[0544] Specific behavior:

[0545] 1. View user profiles: Check past activity and interests.

[0546] 2. Recommendation generation: Generate recommendations tailored to the user's preferences.

[0547] 3. Save suggestions: Save suggestions for each user.

[0548] Step 6: Assistance in optimizing employee benefits

[0549] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on the user profile.

[0550] Specific behavior:

[0551] 1. Browse the benefits database: See a list of available benefits.

[0552] 2. Matching with user profile: Selecting benefit plans that match the user profile.

[0553] 3. Proposal generation: Propose and record optimal ways to use benefits.

[0554] Step 7: Visualize business and leisure situations by department and paid leave usage status

[0555] Server: Based on the analysis results, the busy / slow status of each department and the status of paid leave taken are visualized. This is displayed in the form of a dashboard or report.

[0556] Specific behavior:

[0557] 1. Data aggregation: Aggregate data by department.

[0558] 2. Applying visualization tools: Representing data visually using graphs and heat maps.

[0559] 3. Report generation: Generate dashboards and PDF reports and provide them to the HR department.

[0560] Step 8: User Notification and Interface

[0561] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[0562] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[0563] Specific behavior:

[0564] 1. Notification Generation: Converting proposals and reports into notification format.

[0565] 2. Send notifications: Send notifications to users via email or app notifications.

[0566] 3. User interface operation: The user takes the necessary steps based on the notification received.

[0567] Example 1

[0568] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0569] Conventional labor management systems are insufficient to maximize work efficiency and improve employee satisfaction because they are unable to comprehensively analyze employees' attendance data and PC usage data to suggest the appropriate timing for taking paid leave and the optimal way to use employee benefits. Furthermore, previous systems did not suggest ways to spend paid leave that took into account the hobbies and preferences of individual employees. This invention is intended to solve these problems.

[0570] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0571] In this invention, the server includes a means for collecting employee work data, a means for collecting employee computer usage data, and a means for preprocessing the collected data. This enables efficient data collection and preprocessing. It also includes a means for predicting workloads based on the analyzed data, a means for calculating recommended vacation timings for each employee, a means for suggesting vacation time based on employee preferences, a means for suggesting optimal employee benefit plans for vacation time, a means for visualizing the workload and vacation status for each department, and a means for notifying each employee of the generated proposals and reports. This maximizes employee work efficiency and supports appropriate use of paid vacation time and employee benefits.

[0572] "Employee work data" refers to information such as start time, finish time, and break time that is recorded when an employee works.

[0573] "Employee computer usage data" refers to information such as execution logs of computer programs used by employees during work hours and application usage history.

[0574] "Preprocessing" refers to the process of preparing collected raw data in a format suitable for analysis by supplementing missing data, correcting outliers, and standardizing data formats.

[0575] "Analysis" refers to the process of applying statistical techniques and machine learning models to preprocessed data to predict business activity levels and individual trends.

[0576] "Calculating the recommended timing" refers to calculating the most appropriate time for each employee to take paid leave based on the analyzed data.

[0577] "Suggestions based on hobbies and preferences" refers to recommending ways and activities to spend paid leave based on information about each employee's hobbies and interests.

[0578] "Proposing a benefit system" refers to recommending the most appropriate benefit program that an employee can use during their paid leave.

[0579] "Visualization" refers to the process of displaying analysis results and various data in graph or dashboard format to make them easy to understand.

[0580] "Notifying" refers to the process of sending the proposals and reports generated by the server to each employee's terminal to inform them.

[0581] "System" refers to a set of processes and devices that comprehensively combine the above-mentioned measures to maximize employee work efficiency and support appropriate use of paid leave and employee benefits.

[0582] The present invention is a system that analyzes employee work data and computer usage data, predicts busy and slow times at work, and suggests optimal times and ways to use paid leave, as well as how to utilize employee benefits. The system of the present invention is operated in cooperation with employee terminals and a server.

[0583] Data collection

[0584] First, a collection program is installed on each employee's device. This program continuously collects work data and computer usage data in the background. The data collected by the device is sent to a server at regular intervals. A central database is installed on the server, where the sent data is efficiently stored.

[0585] Data Preprocessing

[0586] Collected data is often incomplete as it is, so it is necessary to supplement missing data, correct outliers, and standardize the data format. This is called preprocessing, and is performed by the server. The server uses a dedicated cleaning algorithm to prepare the data into the appropriate format.

[0587] Data analysis

[0588] Based on the pre-processed data, the server analyzes the workload status. For the analysis, a machine learning model (e.g., random forest or support vector machine) is used. The server uses the trained model to predict the workload status of each employee and department. Confidence intervals can also be set to improve the accuracy of the prediction.

[0589] Calculating the best time to take paid leave

[0590] Based on the analyzed data, the server calculates the optimal timing for each employee to take paid leave, taking into account past paid leave usage patterns and the department's overall schedule.

[0591] Suggestions for how to spend your paid leave

[0592] Using employee hobby and preference data, the server suggests optimal ways to spend paid leave. Utilizing a recommender system, it recommends appropriate activities and travel destinations for each employee.

[0593] Optimal use of employee benefits

[0594] The server also suggests employee benefit plans that can be used when taking paid leave, extracting the best plan for each employee from a database of employee benefits and supporting them in using their paid leave in a meaningful way.

[0595] Visualization of the situation by department

[0596] Based on the analysis results, the server visualizes the busy / unbusy status and paid leave usage status of each department, and displays the results in intuitive graphs and dashboards using visualization tools (e.g., Tableau or Power BI).

[0597] Notification of proposal details

[0598] The final proposals and reports are sent to each employee's device. The server then sends the proposals via email or a dedicated application, allowing employees to apply for paid leave or use employee benefits.

[0599] Specific examples

[0600] For example, if Employee A is considering taking paid leave next week, the system analyzes Employee A's past work data and computer usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system suggests the most suitable camping spot or outdoor activity. It also recommends the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan their paid leave based on that information.

[0601] Prompt Sentence Examples

[0602] "Based on employee A's attendance data and PC usage data from the past six months, please predict the busy and slow periods for next week and suggest the best time and way for him to take paid leave."

[0603] In this way, the system of the present invention maximizes employee work efficiency and supports appropriate use of paid leave and employee benefits.

[0604] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0605] Step 1:

[0606] Data collection

[0607] Device:

[0608] Input: Employee working hours, break times, finish times, program usage, etc.

[0609] Processing: Work data and computer usage data are collected in the background.

[0610] Output: The raw data collected.

[0611] Specific operation: The terminal software automatically records the time when employees start and finish work and collects computer usage history as a log.

[0612] Step 2:

[0613] Data Transfer

[0614] Device:

[0615] Input: Raw data collected.

[0616] Processing: Data collected on the device is transferred to the server at regular intervals.

[0617] Output: The data sent to the server.

[0618] Specific operation: Data is sent from the terminal to the server via the communication module.

[0619] Step 3:

[0620] Data Preprocessing

[0621] server:

[0622] Input: Raw data received from the terminal.

[0623] Processing: Impute missing data, correct outliers, and standardize data formats.

[0624] Output: A clean dataset.

[0625] Specific operation: Run a data cleaning algorithm (e.g., scikit-learn's SimpleImputer) on the server to standardize the data.

[0626] Step 4:

[0627] Data analysis

[0628] server:

[0629] Input: The preprocessed dataset.

[0630] Processing: Predicting peak and trough times using machine learning models.

[0631] Output: Predicted busy / quick times for each employee and department.

[0632] What you'll do: Use Python libraries to train and run machine learning models (e.g., random forests, support vector machines).

[0633] Step 5:

[0634] Calculating the timing of paid leave

[0635] server:

[0636] Input: Predicted busy / slow times, past paid leave data, department schedule.

[0637] Processing: Calculate the optimal timing for taking paid leave.

[0638] Output: A list of optimal vacation times for each employee.

[0639] Specific operation: Using past data, an algorithm is used to calculate the optimal timing for taking paid leave.

[0640] Step 6:

[0641] Suggestions for spending time

[0642] server:

[0643] Input: Employee hobby and preference data, information on the timing of taking paid leave.

[0644] Processing: Use a recommender system to suggest the best way to spend your time.

[0645] Output: Specific activities and travel destination suggestions for each employee.

[0646] How it works: The recommender system analyzes employee preference data and suggests activities that are best suited to them.

[0647] Step 7:

[0648] Assistance in optimal use of employee benefits

[0649] server:

[0650] Input: List of optimal vacation times, employee benefits database.

[0651] Processing: Extraction of benefit plans available when taking paid leave.

[0652] Output: Proposal of optimal employee benefits plan for each employee.

[0653] What it does: Searches the benefits database to extract and recommend the best options.

[0654] Step 8:

[0655] visualization

[0656] server:

[0657] Input: Data on business / slow periods by department and paid leave usage status.

[0658] Processing: Visualize analysis results in graphs and dashboards.

[0659] Output: Visualization report for HR department.

[0660] Specific behavior: Use visualization tools (e.g., Tableau, Power BI) to display analysis results.

[0661] Step 9:

[0662] notification

[0663] server:

[0664] Input: Proposal and report.

[0665] Processing: Proposals and reports are sent to each employee's device.

[0666] Output: Notification sent to each employee terminal.

[0667] Specific operation: The generated information is notified to each employee via email or a dedicated application.

[0668] Specific examples

[0669] For example, if employee A is considering taking paid leave next week, the system analyzes employee A's past work data and computer usage data to predict that next week will be a relatively quiet period. Then, based on employee A's interest in outdoor activities, the system suggests the best camping spots and outdoor activities. It also recommends the use of company-provided recreational facilities as part of the employee benefits program available during that period. This information is sent to employee A's device, allowing them to apply for and plan their paid leave based on that information.

[0670] Example prompt for a generative AI model:

[0671] "Based on employee A's attendance data and PC usage data from the past six months, please predict the busy and slow periods for next week and suggest the best time and way for him to take paid leave."

[0672] (Application example 1)

[0673] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0674] In many modern organizations, in addition to employee attendance management and PC usage status monitoring, the operation and management of robots used in factories is also a key issue. However, managing these data separately reduces overall efficiency and makes it difficult to create consistent operational plans. Furthermore, efficient proposals are often not made regarding the appropriate timing for employees to take paid leave or how to utilize employee benefits when doing so. Furthermore, with regard to factory robots, operational data is not fully utilized, and optimal maintenance timing can be overlooked. A system that can eliminate these management complexities and reduced efficiency is needed.

[0675] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0676] In this invention, the server includes means for collecting employee attendance data, means for collecting employee PC usage data, means for collecting factory robot operation data, means for preprocessing the collected data, means for analyzing the preprocessed data to predict business conditions and optimal robot maintenance times, means for calculating recommended paid leave acquisition times and efficient robot operation plans for each employee, means for suggesting ways to spend paid leave based on employee preferences, means for suggesting optimal employee benefit plans for when paid leave is taken, means for visualizing business conditions and paid leave acquisition status by department, and means for notifying each employee terminal and the robot management system of the generated proposals and reports. This enables maximization of employee labor efficiency and optimal robot operation.

[0677] "Employee attendance data" refers to information related to working hours, such as the hours worked by an employee, arrival and departure times, and break times.

[0678] "Employee PC usage data" refers to data that indicates the usage status of computers used by employees during work hours, and specifically includes program execution logs and application usage history.

[0679] "Operation data of robots in factories" refers to data related to the operating status and operational status of robots operating in factories, including the start and end times of operation, the number of tasks performed, and energy consumption.

[0680] "Preprocessing" refers to the process of preparing collected data in a form that is easy to analyze. Specifically, this involves filling in missing data, standardizing data formats, correcting outliers, etc.

[0681] "Analysis" is the processing of data to identify patterns and trends based on pre-processed data and make predictions or recommendations based on those patterns and trends.

[0682] The "business situation" indicates the busy and quiet periods of business, and reflects the level of workload.

[0683] The "optimal maintenance period" refers to the most suitable time to perform maintenance in order to maintain the robot's operating efficiency at its maximum.

[0684] "Recommended timing" is information that identifies the most appropriate time for each employee to take paid leave.

[0685] An "efficient operation plan" indicates the schedule and strategy for operating the robot most efficiently.

[0686] "Suggestions for how to spend time" are suggestions for the best way for employees to use their time when they take paid leave, and are based on individual hobbies and preferences.

[0687] "Employee benefit system" refers to the various employee benefit services and systems available to employees, including recreational facilities and health promotion programs.

[0688] "Visualization" is a technique that makes data easier to understand by visually displaying the results of data analysis in graphs, dashboards, etc.

[0689] "Notifications" are used to inform employees and operations managers of the results of analysis and proposals, and are sent via email or a dedicated application.

[0690] This invention is a system that analyzes employee attendance data, PC usage data, and factory robot operation data, predicts business activity and optimal maintenance timing for the robots, and proposes optimal operation plans and methods for using employee benefits. This system includes the following processing steps.

[0691] 1. Data Collection:

[0692] The server collects data from employee terminals and factory robots. Work attendance data includes start and end times, break times, etc., while PC usage data includes program execution logs and application usage history. Robot operation data includes start and end times, number of completed tasks, and energy consumption.

[0693] 2. Data Preprocessing:

[0694] The server preprocesses the collected data, specifically by completing missing data, correcting outliers, standardizing data formats, etc. The preprocessed data is then prepared for analysis.

[0695] 3. Data Analysis:

[0696] The server uses machine learning models, such as the University of Pittsburgh's RandomForestRegressor, to predict the best times for robot maintenance and the workload based on the pre-processed data.

[0697] 4. Calculating optimal timing and operational planning:

[0698] Based on the analysis results, the server calculates the recommended timing for each employee to take paid leave and the efficient operation plan for the robot, taking into account past paid leave usage patterns and the schedule of the entire department.

[0699] 5. Suggested working hours and benefits:

[0700] The server uses employee hobby and preference data to suggest optimal ways to spend paid leave. It also suggests employee benefits that can be used when taking paid leave. This allows employees to spend their paid leave meaningfully.

[0701] 6. Visualization:

[0702] Based on the analysis results, the server visualizes the busy / slow times and paid leave usage status for each department, enabling the HR department to carry out effective labor management. Graphs and dashboards are used as visualization tools.

[0703] 7. Notifications and Interface:

[0704] The server then sends the generated proposals and reports to each employee's device and the robot management system, allowing employees to apply for paid leave and use employee benefits.

[0705] As a concrete example, if Employee A is considering taking paid leave next week, the system analyzes Employee A's past attendance data and PC usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system suggests the most suitable camping spot or outdoor activity. Furthermore, the system recommends the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan his paid leave based on that information.

[0706] Examples of prompts to input into a generative AI model include:

[0707] Input prompt: "Based on the recent operation data of robot R1, please suggest the next optimal maintenance timing and efficient operation plan. Operation start time is 2023-09-01 08:00:00, operation end time is 2023-09-01 10:00:00, number of completed tasks is 20, and energy usage is 25.5."

[0708] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0709] Step 1:

[0710] The server collects data from employee terminals and factory robots. Specifically, it collects attendance data (start and end times of work, break times), PC usage data (program execution logs and application usage history), and robot operation data (start and end times of operation, number of tasks completed, energy consumption). It receives various data as input and stores the collected data in a central database as output.

[0711] Step 2:

[0712] The server preprocesses the collected data. Specific operations include filling in missing data, standardizing data formats, and correcting outliers. For example, missing data is filled in using past data or average values, and outliers are corrected to within an appropriate range. The server receives the collected data as input and generates preprocessed data as output.

[0713] Step 3:

[0714] The server uses a machine learning model based on the preprocessed data to predict the busy and slow times of operations and the optimal maintenance times for the robots. Specifically, it performs analysis using algorithms such as RandomForestRegressor. It receives the preprocessed data as input and obtains the predicted results as output.

[0715] Step 4:

[0716] Based on the analysis results, the server calculates the recommended timing for taking paid leave for each employee and an efficient robot operation plan. Specifically, it generates an optimal schedule by taking into account past paid leave usage patterns and the department's overall schedule. It receives the analysis results as input and obtains the optimal timing and operation plan as output.

[0717] Step 5:

[0718] The server uses employee hobby and preference data to suggest optimal ways to spend paid leave. It then uses a recommender system to make suggestions based on employee feedback data and hobby data. It also suggests optimal employee benefit programs. It receives hobby and preference data and past feedback data as input, and obtains suggestions for how to spend the time and employee benefit programs as output.

[0719] Step 6:

[0720] The server visualizes the busy / slow times and paid vacation status for each department based on the analysis results. Specifically, it uses a visualization tool that displays the data in graph and dashboard format. It receives the analysis results as input and generates visualized graphs and dashboards as output.

[0721] Step 7:

[0722] The server notifies each employee's device and the robot management system of the generated proposals and reports. Specifically, it uses a mechanism to notify via email or a dedicated app. It receives proposals and reports as input and sends notification information as output.

[0723] Through this series of steps, the system can maximize employee labor efficiency and optimize robot operation.

[0724] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0725] This invention provides a system that analyzes employee attendance data, PC usage data, emotional data, etc., predicts the busy and slow periods of work that vary for each employee and department, and suggests the best time to take paid leave, how to spend it when taken, and how to best use employee benefits. The program for this system is explained in detail below.

[0726] 1. Data Collection

[0727] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. Attendance data includes work start and end times, break times, etc., while PC usage data includes program execution logs and application usage history. Emotion data is extracted using an emotion engine from the user's facial expressions, voice tone, text input, etc.

[0728] Terminal: Employees enter their start and end times and break times, and PC usage data and emotional data (facial expression analysis, voice tone analysis, etc.) are collected in the background and sent to the server in real time.

[0729] 2. Data Preprocessing

[0730] Server: Cleans all collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[0731] 3. Data Analysis

[0732] Server: Based on the preprocessed data, it uses machine learning models to predict the busy and slow periods for each employee and department. It also analyzes emotional data and evaluates the user's mental state.

[0733] 4. Calculating the recommended timing for taking paid leave

[0734] Server: Based on the results of data analysis, calculates the optimal timing for each employee to take paid leave. This takes into account not only the department's overall schedule and past paid leave usage patterns, but also the employee's mental state based on emotional data.

[0735] 5. Suggestions for how to spend your paid leave

[0736] Server: Based on the user's hobby, preference, and emotional data, the server proposes the best way to spend their vacation time. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also takes into account the user's mental state and suggests ways to promote relaxation and refreshment.

[0737] 6. Assistance in optimal use of employee benefits

[0738] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on emotional data and user profile. For example, if there is a high need for relaxation, suggests spa facilities and counseling services.

[0739] 7. Visualization of business and vacation schedules by department

[0740] Server: Based on the analysis results, the busy / slow periods and paid leave usage status for each department are visualized, enabling the HR department to carry out effective labor management.

[0741] 8. User Notification and Interface

[0742] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[0743] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[0744] Specific examples

[0745] For example, suppose employee B plans to take paid leave soon. The system analyzes employee B's attendance data and PC usage data to predict that the following week will be a relatively quiet period. It also analyzes employee B's emotional data and recognizes that employee B's mental state has recently been one that makes him prone to stress. As a result, the system suggests the best time to take paid leave and notifies employee B of refreshing ways to spend the day that suit his hobbies and preferences (for example, suggestions for nature walks or relaxation spots). At the same time, it also recommends that employee B take advantage of employee benefits that have a relaxing effect (for example, vouchers for spa facilities). Employee B can use these suggestions to apply for paid leave and take advantage of employee benefits.

[0746] In this way, the system of the present invention maximizes employee work efficiency and supports employees in taking paid leave and using employee benefits while taking into account emotional data.

[0747] The processing flow will be explained below.

[0748] Step 1: Data collection

[0749] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. Specifically, the data is sent via API or a dedicated client app and stored in a central database.

[0750] Terminal: Employees enter their work start time, end time, and break time. In the background, PC usage data (program execution logs, browser history, etc.) and emotional data (facial expression analysis, voice tone analysis, etc.) are collected and sent to the server in real time.

[0751] Step 2: Data Preprocessing

[0752] Server: Cleans all collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[0753] Specific behavior:

[0754] 1. Imputing missing data: Filling in missing data with past data or average values.

[0755] 2. Outlier Correction: Correcting or removing abnormal data based on pre-set criteria.

[0756] 3. Data format unification: Convert all data into a consistent format.

[0757] Step 3: Analyze the emotion data

[0758] Server: Analyzes collected emotional data and evaluates the user's mental state using an emotion engine, including facial expression recognition, voice tone analysis, and text analysis.

[0759] Specific behavior:

[0760] 1. Facial expression data analysis: Facial expression data collected from camera footage is analyzed to determine the emotional state.

[0761] 2. Voice data analysis: Analyze voice tones and detect emotional fluctuations.

[0762] 3. Analysis of text data: Emotional words are extracted from chat logs and email content to evaluate the emotional state.

[0763] Step 4: Data analysis

[0764] Server: Based on preprocessed attendance data, PC usage data, and emotion data, machine learning models are used to predict the busy / slow periods for each employee and department.

[0765] Specific behavior:

[0766] 1. Training a machine learning model: Train the model based on historical data.

[0767] 2. Predicting peak and trough periods: Use the trained model to predict future peak and trough periods.

[0768] 3. Setting confidence intervals: Calculate confidence intervals for the predicted results to ensure the reliability of the results.

[0769] Step 5: Calculate the best time to take paid leave

[0770] Server: Based on the results of data analysis and the evaluated emotional data, the server calculates the optimal timing for each employee to take paid leave, taking into account the department's overall schedule, past patterns of taking paid leave, and the employee's mental state based on emotional data.

[0771] Specific behavior:

[0772] 1. Check the busy / quiet periods: Check the busy / quiet periods of each employee and identify the quiet periods.

[0773] 2. Department-wide coordination: Consider the busy and slow times of the entire department and choose a time that will minimize the impact on other employees.

[0774] 3. Calculating the timing of taking paid leave: Calculate and recommend the optimal time to take leave.

[0775] Step 6: Suggestions for how to spend your paid leave

[0776] Server: Based on the user's hobby, preference, and emotional data, the server proposes the best way to spend their vacation time. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also takes into account the user's mental state and suggests ways to promote relaxation and refreshment.

[0777] Specific behavior:

[0778] 1. View user profiles: Check past activity and interests.

[0779] 2. Recommendation generation: Generate suggestions tailored to the user's preferences and take into account sentiment data.

[0780] 3. Save suggestions: Save suggestions for each user.

[0781] Step 7: Assistance in optimizing employee benefits

[0782] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on emotional data and user profile. For example, if there is a high need for relaxation, suggests spa facilities and counseling services.

[0783] Specific behavior:

[0784] 1. Browse the benefits database: See a list of available benefits.

[0785] 2. Matching with user profile: Selecting benefit plans that match the user profile.

[0786] 3. Proposal generation: Propose and record optimal ways to use benefits.

[0787] Step 8: Visualize business and vacation status by department

[0788] Server: Based on the analysis results, the busy / slow status of each department and the status of paid leave taken are visualized. This is displayed in the form of a dashboard or report.

[0789] Specific behavior:

[0790] 1. Data aggregation: Aggregate data by department.

[0791] 2. Applying visualization tools: Representing data visually using graphs and heat maps.

[0792] 3. Report generation: Generate dashboards and PDF reports and provide them to the HR department.

[0793] Step 9: User Notification and Interface

[0794] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[0795] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[0796] Specific behavior:

[0797] 1. Notification Generation: Converting proposals and reports into notification format.

[0798] 2. Send notifications: Send notifications to users via email or app notifications.

[0799] 3. User interface operation: The user takes the necessary steps based on the notification received.

[0800] Example 2

[0801] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0802] In today's corporate environment, maximizing employee work efficiency and maintaining a healthy working environment are important. However, conventional systems only collect attendance data and PC usage data, and do not adequately consider employees' emotions and mental state. Furthermore, they are unable to provide appropriate suggestions regarding the timing of paid leave and how to spend it when taken, creating challenges in optimizing employee welfare benefits. This invention solves these issues by providing a system that suggests optimal work life for each employee.

[0803] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0804] In this invention, the server includes means for collecting attendance data, PC usage data, and emotion data from each employee's terminal, means for cleaning the collected data and for completing and correcting missing and outlier values, means for analyzing the preprocessed data and predicting business activity using a machine learning model, means for analyzing each employee's emotion data and evaluating their mental state, means for calculating the optimal timing for taking paid leave based on the business activity and emotion data, means for suggesting optimal ways to spend paid leave based on the employee's hobby and preference data and emotion data, means for suggesting employee benefit plans available when taking paid leave, means for visualizing business activity and paid leave usage status by department, and means for notifying each employee of the generated suggestions and reports. This maximizes employee work efficiency and enables the creation of a work environment that takes mental health into consideration.

[0805] "Work attendance data" is information related to an employee's working status, such as start time, finish time, and break time.

[0806] "PC usage data" refers to information such as program execution logs and application usage history for computers used by employees during work hours.

[0807] "Emotion data" is information about an employee's emotions and mental state extracted using an emotion engine from a user's facial expressions, tone of voice, text input, etc.

[0808] A "machine learning model" is a mathematical algorithm or statistical model that learns from collected data and predicts patterns and trends.

[0809] "Busy / slow status" refers to the busyness or slowness of business during a specific period.

[0810] "Hobbies and Preference Data" is information about a user's interests and preferences, including hobbies, interests, and favorite activities.

[0811] A "welfare system" refers to various services and support systems provided by companies with the aim of improving the welfare of their employees, and includes, for example, vouchers for use of spa facilities and counseling services.

[0812] "Cleanup" is the process of filling in missing values ​​and correcting outliers in order to improve the quality of data.

[0813] "Visualization" is the process of visually representing data and information, making them easier to understand using graphs, charts, etc.

[0814] A "recommender system" is a system that recommends optimal activities, travel destinations, etc. based on a user's hobby and preference data.

[0815] "Notifications" are a means of informing users of specific information or offers in real time, and are provided via email or a dedicated app.

[0816] The present invention provides a system that analyzes employee attendance data, PC usage data, emotional data, etc., predicts the busy and slow business conditions that vary for each employee and department, and suggests the appropriate timing for taking paid leave, the appropriate way to spend it when taken, and the optimal way to use employee benefits.

[0817] 1. Data Collection

[0818] server

[0819] The server periodically collects attendance data, PC usage data, and emotion data from each employee's device. Attendance data includes work start and end times and break times, while PC usage data includes program execution logs and application usage history. Emotion data is extracted from the user's facial expressions, voice tone, text input, etc. using an emotion engine.

[0820] Terminal

[0821] Employees enter their work start and end times, as well as break times, into the terminals, which collect PC usage data and emotional data (facial expression analysis, voice tone analysis, etc.) in the background and send it to the server in real time.

[0822] 2. Data Preprocessing

[0823] server

[0824] All collected data is cleaned. Missing values ​​are filled in using past data or average values, and outliers are corrected or removed. Data formats are also standardized to make them suitable for analysis.

[0825] 3. Data Analysis

[0826] server

[0827] Based on the preprocessed data, machine learning models are used to predict the busy and slow periods for each employee and department. Emotional data is also analyzed to evaluate the user's mental state. Specific software used includes the Python libraries TensorFlow and Scikit-learn.

[0828] 4. Calculating the recommended timing for taking paid leave

[0829] server

[0830] Based on the results of the data analysis, the system calculates the optimal timing for each employee to take paid leave, taking into account not only the department's overall schedule and past paid leave usage patterns, but also the employee's mental state based on emotional data.

[0831] 5. Suggestions for how to spend your paid leave

[0832] server

[0833] The system suggests optimal ways to spend paid vacation time based on the user's hobby, preference, and emotional data. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also includes suggestions to promote relaxation and refreshment. Specific software used includes Apache Spark and Elasticsearch.

[0834] 6. Assistance in optimal use of employee benefits

[0835] server

[0836] It suggests employee benefits that can be used when taking paid leave, and suggests spa facilities and counseling services if employees need to relax based on their emotional data and profile information.

[0837] 7. Visualization of business and vacation schedules by department

[0838] server

[0839] Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized in graphs and charts, enabling the HR department to carry out effective labor management. For example, visualization can be done using Tableau or Power BI.

[0840] 8. User Notification and Interface

[0841] server

[0842] The generated proposals and reports are sent to each user's device, and notifications are sent via email or a dedicated app.

[0843] Terminal

[0844] The user checks the received notification and applies for paid leave or takes advantage of employee benefits.

[0845] Specific examples

[0846] For example, suppose employee B plans to take paid leave soon. The system analyzes employee B's attendance data and PC usage data to predict that the following week will be a relatively quiet period. It also analyzes employee B's emotional data and recognizes that employee B's mental state has recently been one that makes him prone to stress. As a result, the system suggests the best time to take paid leave and notifies employee B of refreshing ways to spend the day that suit his hobbies and preferences (for example, suggestions for nature walks or relaxation spots). At the same time, it also recommends that employee B take advantage of employee benefits that have a relaxing effect (for example, vouchers for spa facilities). Employee B can use these suggestions to apply for paid leave and take advantage of employee benefits.

[0847] Prompt Sentence Examples

[0848] Attendance data analysis prompt

[0849] Analyze employee attendance data to predict the next week's peak and trough times. The dataset includes start and finish times and break times.

[0850] Emotional Data Analysis Prompt

[0851] Analyze employee emotional data and assess their mental state. Data sets include facial expression analysis, voice tone, and text input.

[0852] Prompt to calculate the optimal timing for taking paid leave

[0853] Calculate the optimal timing for employees to take paid leave based on attendance and sentiment data.

[0854] Prompts for suggestions on how to spend your time based on your hobbies and preferences

[0855] Based on the user's hobby, preference, and emotional data, please suggest the best way to spend their vacation time. Please include suggestions that promote relaxation and refreshment.

[0856] Benefit Plan Recommendation Prompts

[0857] Based on emotional data and user profiles, suggest benefits that can be used when taking paid time off. If relaxation is needed, suggest spa facilities or counseling services.

[0858] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0859] Step 1: Data collection

[0860] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. The server sends data collection requests to each device at set intervals, receives the data sent from each device, and stores it in a database. The input is data from each employee's device, and the output is an integrated database.

[0861] Terminal: Employees enter their work start time, end time, and break time into a dedicated app. The terminal collects PC usage information (application usage history, usage time, etc.) and emotional data (facial expression analysis, voice tone analysis) in the background and sends them to the server in real time. The input is the employee's operations and background data collection, and the output is the data sent to the server.

[0862] Step 2: Data Preprocessing

[0863] Server: Cleans up all collected data. Specifically, if missing values ​​are found, they are filled in using past data or average values. Also, outliers are detected and corrected or removed. Data formats are standardized and made suitable for analysis. The input is raw data, and the output is cleaned data.

[0864] Step 3: Data analysis

[0865] Server: Using the preprocessed data, a machine learning model predicts the busy / quick schedule for each employee and department. Emotional data is also analyzed to evaluate each employee's mental state. Specific software used includes the Python libraries TensorFlow and Scikit-learn. The input is the cleaned-up data, and the output is predicted data on busy / quick schedules and the results of an evaluation of mental state.

[0866] Step 4: Calculate the best time to take paid leave

[0867] Server: Based on the analysis results, calculates the optimal timing for each employee to take paid leave. Calculates the optimal date to take paid leave by taking into account the department's overall schedule, past paid leave usage patterns, and mental state. The input is the analysis results, and the output is the recommended timing for taking paid leave.

[0868] Step 5: Suggestions for how to spend your paid leave

[0869] Server: Based on the user's hobby, preference, and emotional data, the server proposes the optimal way to spend their paid vacation. A recommender system is used to recommend activities and travel destinations suitable for each user. Specific software used includes Apache Spark and Elasticsearch. The input is hobby, preference, and emotional data, and the output is a proposal for the optimal way to spend the day.

[0870] Step 6: Assistance in optimizing employee benefits

[0871] Server: Proposes employee benefit programs that can be used when taking paid leave. Based on emotional data and profile information, recommends spa facilities or counseling services if there is a high need for relaxation. The input is emotional data and profile information, and the output is a proposed employee benefit program.

[0872] Step 7: Visualize business and leisure situations by department and paid leave usage status

[0873] Server: Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized in graphs and charts. This makes it easier for the HR department to carry out effective labor management. For example, visualization can be done using Tableau or Power BI. The input is the analysis results, and the output is visualized data.

[0874] Step 8: User Notification and Interface

[0875] Server: Sends the generated proposals and reports to each user's device. Notifications are sent via email or a dedicated app. The input is the generated proposals and reports, and the output is a notification message.

[0876] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits. The input is the notification message, and the output is the user's action.

[0877] (Application example 2)

[0878] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0879] While the operation of robots in modern factories improves production efficiency, their maintenance and operational status management are also important. In particular, there is a need to efficiently collect and analyze robot operation data and maintenance history data to predict optimal maintenance times. Also important is the overall improvement of work efficiency and employee benefits, including the appropriate timing and use of employee paid leave and suggestions for utilizing employee benefit programs.

[0880] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting employee attendance data, means for collecting employee PC usage data, means for collecting factory robot operation data and maintenance history data, means for preprocessing the collected data, means for analyzing the preprocessed data to predict business activity, means for calculating recommended paid vacation time for each employee and optimal maintenance times for each robot, means for proposing paid vacation time and maintenance content based on employee preferences and robot operation status, means for proposing optimal employee benefit programs for paid vacation time and robot maintenance, means for visualizing business activity and robot operation status and paid vacation time for each department, and means for notifying each employee terminal and each engineer terminal of the generated proposals and reports. This enables optimal paid vacation time and employee benefit use for each employee and efficient robot maintenance scheduling.

[0881] "Employee attendance data" refers to work-related record information such as the start and end times of each employee's shift, and break times.

[0882] "PC usage data" refers to data such as program execution logs and application usage history for computers used by employees for work.

[0883] "Robot operation data" refers to information such as the operation records, operating hours, and error occurrence status of robots operating in a factory.

[0884] "Maintenance history data" is recorded information such as the content, date, and person who performed the maintenance performed on the robot in the past.

[0885] "Preprocessing" refers to the process of preparing collected data for analysis, such as filling in missing values, correcting outliers, and standardizing data formats.

[0886] "Business activity status" is data showing the busy and slow periods of business, and indicates the workload status of employees and departments.

[0887] "Recommended timing for taking paid leave" is the time when the optimal time to take paid leave is calculated and suggested for each employee.

[0888] Based on "hobbies and preferences" means customizing the proposals to suit the interests and preferences of each employee.

[0889] "How to spend your paid leave" is a content that suggests activities and ways for employees to rest during the period they take paid leave.

[0890] "Employee benefits system" refers to various services and support systems provided by a company that employees can use, and often includes relaxation facilities and counseling services.

[0891] "Visualization of busy and slow periods" refers to visually displaying the results of data analysis so that busy and slow periods can be understood at a glance.

[0892] The "generated proposals" are proposals for each employee and robot that are calculated and created based on data analysis.

[0893] "Business and quiet periods by department" is data showing the busy and quiet periods for work in each specific department.

[0894] "Each employee terminal" refers to an individual device (PC, smartphone, etc.) used by an employee.

[0895] "Technician's Terminal" refers to an individual device used by a technician in a factory or operational location.

[0896] A "machine learning model" is an algorithm used in data analysis, a system that learns patterns from the original data and makes predictions and classifications.

[0897] A "confidence interval" is a numerical range that indicates the reliability of a prediction result and provides a statistical indication of how accurate the prediction is.

[0898] This invention is a system that analyzes employee attendance data, PC usage data, factory robot operation data, and maintenance history data, and based on that, suggests optimal timing for taking paid leave and performing maintenance. This system is implemented using both a server and a terminal.

[0899] Data collection

[0900] The server collects employee attendance data, PC usage data, factory robot operation data, and maintenance history data. These data are obtained from each employee's terminal and the factory management system. Specifically, employee attendance data includes work start and end times and break times, while PC usage data includes program execution logs and application usage history. Robot operation data includes operating hours and error occurrence status, and maintenance history data includes the details and dates of past maintenance.

[0901] Data Preprocessing

[0902] The server preprocesses the collected data, which includes filling in missing values, correcting outliers, and standardizing data formats, making the data suitable for analysis.

[0903] Data analysis

[0904] The server uses a machine learning model based on the preprocessed data to predict the busy and slow times of operations and the optimal maintenance times for the robot. This analysis uses a machine learning algorithm (e.g., Random Forest Regressor). It also uses a generative AI model to generate recommendations based on the user's preferences and the robot's operating status.

[0905] Paid leave and maintenance scheduling

[0906] Based on the analysis results, the server calculates the recommended timing for each employee to take paid leave and the optimal maintenance period for each robot, including setting a confidence interval based on past data and current operating conditions.

[0907] Suggestions and Notifications

[0908] The server then sends the generated proposals and reports to each employee's and technician's device via email or a dedicated application. The employee can then review the proposals, apply for paid leave and employee benefits, and schedule robot maintenance.

[0909] Specific examples

[0910] For example, if factory robot A has recently been operating slowly, the system will suggest the optimal date and time for the next maintenance based on past maintenance history and operation data. The next maintenance for this robot is predicted to be required two weeks from now on August 15th, and the appropriate maintenance work will be notified to the maintenance team.

[0911] Prompt Sentence Examples

[0912] "Please create a model that predicts the optimal next maintenance timing for each robot based on the factory robot operation data and maintenance history data. Also, please tell me the steps to develop a system that notifies technicians of this in the form of a smartphone application."

[0913] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0914] Step 1: Data collection

[0915] The server collects data from each employee's terminal and the factory management system. Inputs include employee attendance data, PC usage data, robot operation data, and maintenance history data. Outputs store this data on the server and prepare it in a format that can be used for the next processing step. Specific operations include retrieving data from a database and streaming data in real time.

[0916] Step 2: Data Preprocessing

[0917] The server preprocesses the collected data. The input is the raw data collected in step 1. Specific operations include imputing missing values, correcting outliers, and standardizing data formats. The output is clean data that has been prepared for analysis.

[0918] Step 3: Data analysis

[0919] The server uses a machine learning model to analyze the preprocessed data. The input is clean data. Specifically, it uses machine learning algorithms such as RandomForestRegressor to predict the busy / unbusy schedule and the optimal maintenance timing for the robot. The output is the predicted busy / unbusy schedule and recommended maintenance timing.

[0920] Step 4: Generate proposals

[0921] The server generates proposals using a generative AI model based on the analysis results. The input is the prediction results from step 3. Specific actions include customizing how employees spend their paid leave and the maintenance details, taking into account the individual hobbies and preferences of employees and the operating status of the robot. The output is specific proposals provided to the user.

[0922] Step 5: Notification

[0923] The server notifies each employee's terminal and the engineer's terminal of the generated proposal and report. The input is the proposal generated in step 4. Specific actions include sending an email or sending a notification via a dedicated application. The output is the proposal displayed on the user's terminal, which the user can check and take the next action.

[0924] Step 6: User Actions

[0925] The user takes action based on the notified proposal. The input is the proposal received in step 5. Specific actions include applying for paid leave, applying for employee benefits, and scheduling robot maintenance. The output is a record of the application and scheduling as actions taken by the user.

[0926] 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 a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0927] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0928] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0929] [Third embodiment]

[0930] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0931] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0932] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0933] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0934] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0935] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0936] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0937] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0938] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[0939] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0940] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0941] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0942] This invention provides a system that analyzes employee attendance data, PC usage data, etc., predicts the busy and slow periods of work that vary for each employee and department, and suggests the best time to take paid leave, how to spend it when it is taken, and how to best use employee benefits. The program for this system is explained in detail below.

[0943] 1. Data Collection

[0944] First, attendance data and PC usage data are periodically collected from each employee's device. The attendance data includes start and end times of work, break times, etc., while the PC usage data includes program execution logs and application usage history.

[0945] Server: Automatically collects data from employee devices and stores it in a central database.

[0946] Terminal: Attendance data entry and PC usage data log collection are performed in the background.

[0947] 2. Data Preprocessing

[0948] Next, the collected data is preprocessed, which involves filling in missing data, correcting outliers, and standardizing the data format.

[0949] Server: Cleans and preprocesses the collected data. Missing data is filled in using past data and average values.

[0950] 3. Data Analysis

[0951] Predict business activity based on pre-processed data. Use machine learning models and statistical algorithms to analyze the busy and slow periods for each employee and department.

[0952] Server: Trains the machine learning model, performs analysis, and sets confidence intervals to improve prediction accuracy.

[0953] 4. Calculating the recommended timing for taking paid leave

[0954] Based on the analysis results, the system calculates the optimal timing for each employee to take paid leave, taking into account past paid leave usage patterns and the department's overall schedule.

[0955] Server: Calculates the optimal time to take paid leave based on each employee's busy / quick schedule.

[0956] 5. Suggestions for how to spend your paid leave

[0957] Furthermore, the system uses data on employees' hobbies and preferences to suggest optimal ways to spend their paid leave, and utilizes a recommender system to recommend activities and travel destinations that match individual interests.

[0958] Server: Based on past data and preference information, suggests the best way for each employee to spend their time.

[0959] 6. Assistance in optimal use of employee benefits

[0960] We also propose employee benefit systems that can be used when taking paid leave, and provide optimal ways to use employee benefits so that employees can spend their paid leave meaningfully.

[0961] Server: Extracts and proposes systems suitable for each employee from a database of employee benefits.

[0962] 7. Visualization of business and vacation schedules by department

[0963] Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized, enabling the HR department to carry out effective labor management.

[0964] Server: Uses visualization tools to display data in graphs and dashboards.

[0965] 8. User Notification and Interface

[0966] Finally, the generated proposals and reports are sent to each employee's device, and employees can use them to apply for paid leave and use employee benefits.

[0967] Server: Generates proposals and reports and notifies each employee via email or a dedicated app.

[0968] Terminal: The employee checks the received notification and takes the necessary steps.

[0969] Specific examples

[0970] For example, if Employee A is considering taking paid leave next week, the system will analyze Employee A's past attendance data and PC usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system will suggest the most suitable camping spot or outdoor activity. Furthermore, the system will recommend the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan their paid leave based on that information.

[0971] In this way, the system of the present invention maximizes employee work efficiency and supports appropriate use of paid leave and employee benefits.

[0972] The processing flow will be explained below.

[0973] Step 1: Data collection

[0974] Server: Periodically collects attendance data and PC usage data from each employee's device. Specifically, the data is sent via API or a dedicated client app and stored in a central database.

[0975] Terminal: Employees enter their work start time, end time, and break time. PC usage data (program execution logs, browser history, etc.) is collected in the background and sent to the server in real time.

[0976] Step 2: Data Preprocessing

[0977] Server: Cleans the collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[0978] Specific behavior:

[0979] 1. Imputing missing data: Filling in missing data with past data or average values.

[0980] 2. Outlier Correction: Correcting or removing abnormal data based on pre-set criteria.

[0981] 3. Data format unification: Convert all data into a consistent format.

[0982] Step 3: Data analysis

[0983] Server: Analyzes the pre-processed data and uses machine learning models to predict the busy and slow periods for each employee and department. It uses algorithms such as random forests and neural networks.

[0984] Specific behavior:

[0985] 1. Training a machine learning model: Train the model based on historical data.

[0986] 2. Predicting peak and trough periods: Use the trained model to predict future peak and trough periods.

[0987] 3. Setting confidence intervals: Calculate confidence intervals for the predicted results to ensure the reliability of the results.

[0988] Step 4: Calculate the best time to take paid leave

[0989] Server: Based on the results of data analysis, calculates the optimal timing for each employee to take paid leave, taking into account the department's overall schedule and past paid leave usage patterns.

[0990] Specific behavior:

[0991] 1. Check the busy / quiet periods: Check the busy / quiet periods of each employee and identify the quiet periods.

[0992] 2. Department-wide coordination: Consider the busy and slow times of the entire department and choose a time that will minimize the impact on other employees.

[0993] 3. Calculating the timing of taking paid leave: Calculate and recommend the optimal time to take leave.

[0994] Step 5: Suggestions for how to spend your paid leave

[0995] Server: Based on the user's hobby and preference data, the server proposes the best way to spend their vacation time. Using a recommender system, the server recommends activities and travel destinations suitable for each user.

[0996] Specific behavior:

[0997] 1. View user profiles: Check past activity and interests.

[0998] 2. Recommendation generation: Generate recommendations tailored to the user's preferences.

[0999] 3. Save suggestions: Save suggestions for each user.

[1000] Step 6: Assistance in optimizing employee benefits

[1001] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on the user profile.

[1002] Specific behavior:

[1003] 1. Browse the benefits database: See a list of available benefits.

[1004] 2. Matching with user profile: Selecting benefit plans that match the user profile.

[1005] 3. Proposal generation: Propose and record optimal ways to use benefits.

[1006] Step 7: Visualize business and leisure situations by department and paid leave usage status

[1007] Server: Based on the analysis results, the busy / slow status of each department and the status of paid leave taken are visualized. This is displayed in the form of a dashboard or report.

[1008] Specific behavior:

[1009] 1. Data aggregation: Aggregate data by department.

[1010] 2. Applying visualization tools: Representing data visually using graphs and heat maps.

[1011] 3. Report generation: Generate dashboards and PDF reports and provide them to the HR department.

[1012] Step 8: User Notification and Interface

[1013] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[1014] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[1015] Specific behavior:

[1016] 1. Notification Generation: Converting proposals and reports into notification format.

[1017] 2. Send notifications: Send notifications to users via email or app notifications.

[1018] 3. User interface operation: The user takes the necessary steps based on the notification received.

[1019] Example 1

[1020] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1021] Conventional labor management systems are insufficient to maximize work efficiency and improve employee satisfaction because they are unable to comprehensively analyze employees' attendance data and PC usage data to suggest the appropriate timing for taking paid leave and the optimal way to use employee benefits. Furthermore, previous systems did not suggest ways to spend paid leave that took into account the hobbies and preferences of individual employees. This invention is intended to solve these problems.

[1022] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1023] In this invention, the server includes a means for collecting employee work data, a means for collecting employee computer usage data, and a means for preprocessing the collected data. This enables efficient data collection and preprocessing. It also includes a means for predicting workloads based on the analyzed data, a means for calculating recommended vacation timings for each employee, a means for suggesting vacation time based on employee preferences, a means for suggesting optimal employee benefit plans for vacation time, a means for visualizing the workload and vacation status for each department, and a means for notifying each employee of the generated proposals and reports. This maximizes employee work efficiency and supports appropriate use of paid vacation time and employee benefits.

[1024] "Employee work data" refers to information such as start time, finish time, and break time that is recorded when an employee works.

[1025] "Employee computer usage data" refers to information such as execution logs of computer programs used by employees during work hours and application usage history.

[1026] "Preprocessing" refers to the process of preparing collected raw data in a format suitable for analysis by supplementing missing data, correcting outliers, and standardizing data formats.

[1027] "Analysis" refers to the process of applying statistical techniques and machine learning models to preprocessed data to predict business activity levels and individual trends.

[1028] "Calculating the recommended timing" refers to calculating the most appropriate time for each employee to take paid leave based on the analyzed data.

[1029] "Suggestions based on hobbies and preferences" refers to recommending ways and activities to spend paid leave based on information about each employee's hobbies and interests.

[1030] "Proposing a benefit system" refers to recommending the most appropriate benefit program that an employee can use during their paid leave.

[1031] "Visualization" refers to the process of displaying analysis results and various data in graph or dashboard format to make them easy to understand.

[1032] "Notifying" refers to the process of sending the proposals and reports generated by the server to each employee's terminal to inform them.

[1033] "System" refers to a set of processes and devices that comprehensively combine the above-mentioned measures to maximize employee work efficiency and support appropriate use of paid leave and employee benefits.

[1034] The present invention is a system that analyzes employee work data and computer usage data, predicts busy and slow times at work, and suggests optimal times and ways to use paid leave, as well as how to utilize employee benefits. The system of the present invention is operated in cooperation with employee terminals and a server.

[1035] Data collection

[1036] First, a collection program is installed on each employee's device. This program continuously collects work data and computer usage data in the background. The data collected by the device is sent to a server at regular intervals. A central database is installed on the server, where the sent data is efficiently stored.

[1037] Data Preprocessing

[1038] Collected data is often incomplete as it is, so it is necessary to supplement missing data, correct outliers, and standardize the data format. This is called preprocessing, and is performed by the server. The server uses a dedicated cleaning algorithm to prepare the data into the appropriate format.

[1039] Data analysis

[1040] Based on the pre-processed data, the server analyzes the workload status. For the analysis, a machine learning model (e.g., random forest or support vector machine) is used. The server uses the trained model to predict the workload status of each employee and department. Confidence intervals can also be set to improve the accuracy of the prediction.

[1041] Calculating the best time to take paid leave

[1042] Based on the analyzed data, the server calculates the optimal timing for each employee to take paid leave, taking into account past paid leave usage patterns and the department's overall schedule.

[1043] Suggestions for how to spend your paid leave

[1044] Using employee hobby and preference data, the server suggests optimal ways to spend paid leave. Utilizing a recommender system, it recommends appropriate activities and travel destinations for each employee.

[1045] Optimal use of employee benefits

[1046] The server also suggests employee benefit plans that can be used when taking paid leave, extracting the best plan for each employee from a database of employee benefits and supporting them in using their paid leave in a meaningful way.

[1047] Visualization of the situation by department

[1048] Based on the analysis results, the server visualizes the busy / unbusy status and paid leave usage status of each department, and displays the results in intuitive graphs and dashboards using visualization tools (e.g., Tableau or Power BI).

[1049] Notification of proposal details

[1050] The final proposals and reports are sent to each employee's device. The server then sends the proposals via email or a dedicated application, allowing employees to apply for paid leave or use employee benefits.

[1051] Specific examples

[1052] For example, if Employee A is considering taking paid leave next week, the system analyzes Employee A's past work data and computer usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system suggests the most suitable camping spot or outdoor activity. It also recommends the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan their paid leave based on that information.

[1053] Prompt Sentence Examples

[1054] "Based on employee A's attendance data and PC usage data from the past six months, please predict the busy and slow periods for next week and suggest the best time and way for him to take paid leave."

[1055] In this way, the system of the present invention maximizes employee work efficiency and supports appropriate use of paid leave and employee benefits.

[1056] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1057] Step 1:

[1058] Data collection

[1059] Device:

[1060] Input: Employee working hours, break times, finish times, program usage, etc.

[1061] Processing: Work data and computer usage data are collected in the background.

[1062] Output: The raw data collected.

[1063] Specific operation: The terminal software automatically records the time when employees start and finish work and collects computer usage history as a log.

[1064] Step 2:

[1065] Data Transfer

[1066] Device:

[1067] Input: Raw data collected.

[1068] Processing: Data collected on the device is transferred to the server at regular intervals.

[1069] Output: The data sent to the server.

[1070] Specific operation: Data is sent from the terminal to the server via the communication module.

[1071] Step 3:

[1072] Data Preprocessing

[1073] server:

[1074] Input: Raw data received from the terminal.

[1075] Processing: Impute missing data, correct outliers, and standardize data formats.

[1076] Output: A clean dataset.

[1077] Specific operation: Run a data cleaning algorithm (e.g., scikit-learn's SimpleImputer) on the server to standardize the data.

[1078] Step 4:

[1079] Data analysis

[1080] server:

[1081] Input: The preprocessed dataset.

[1082] Processing: Predicting peak and trough times using machine learning models.

[1083] Output: Predicted busy / quick times for each employee and department.

[1084] What you'll do: Use Python libraries to train and run machine learning models (e.g., random forests, support vector machines).

[1085] Step 5:

[1086] Calculating the timing of paid leave

[1087] server:

[1088] Input: Predicted busy / slow times, past paid leave data, department schedule.

[1089] Processing: Calculate the optimal timing for taking paid leave.

[1090] Output: A list of optimal vacation times for each employee.

[1091] Specific operation: Using past data, an algorithm is used to calculate the optimal timing for taking paid leave.

[1092] Step 6:

[1093] Suggestions for spending time

[1094] server:

[1095] Input: Employee hobby and preference data, information on the timing of taking paid leave.

[1096] Processing: Use a recommender system to suggest the best way to spend your time.

[1097] Output: Specific activities and travel destination suggestions for each employee.

[1098] How it works: The recommender system analyzes employee preference data and suggests activities that are best suited to them.

[1099] Step 7:

[1100] Assistance in optimal use of employee benefits

[1101] server:

[1102] Input: List of optimal vacation times, employee benefits database.

[1103] Processing: Extraction of benefit plans available when taking paid leave.

[1104] Output: Proposal of optimal employee benefits plan for each employee.

[1105] What it does: Searches the benefits database to extract and recommend the best options.

[1106] Step 8:

[1107] visualization

[1108] server:

[1109] Input: Data on business / slow periods by department and paid leave usage status.

[1110] Processing: Visualize analysis results in graphs and dashboards.

[1111] Output: Visualization report for HR department.

[1112] Specific behavior: Use visualization tools (e.g., Tableau, Power BI) to display analysis results.

[1113] Step 9:

[1114] notification

[1115] server:

[1116] Input: Proposal and report.

[1117] Processing: Proposals and reports are sent to each employee's device.

[1118] Output: Notification sent to each employee terminal.

[1119] Specific operation: The generated information is notified to each employee via email or a dedicated application.

[1120] Specific examples

[1121] For example, if employee A is considering taking paid leave next week, the system analyzes employee A's past work data and computer usage data to predict that next week will be a relatively quiet period. Then, based on employee A's interest in outdoor activities, the system suggests the best camping spots and outdoor activities. It also recommends the use of company-provided recreational facilities as part of the employee benefits program available during that period. This information is sent to employee A's device, allowing them to apply for and plan their paid leave based on that information.

[1122] Example prompt for a generative AI model:

[1123] "Based on employee A's attendance data and PC usage data from the past six months, please predict the busy and slow periods for next week and suggest the best time and way for him to take paid leave."

[1124] (Application example 1)

[1125] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1126] In many modern organizations, in addition to employee attendance management and PC usage status monitoring, the operation and management of robots used in factories is also a key issue. However, managing these data separately reduces overall efficiency and makes it difficult to create consistent operational plans. Furthermore, efficient proposals are often not made regarding the appropriate timing for employees to take paid leave or how to utilize employee benefits when doing so. Furthermore, with regard to factory robots, operational data is not fully utilized, and optimal maintenance timing can be overlooked. A system that can eliminate these management complexities and reduced efficiency is needed.

[1127] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1128] In this invention, the server includes means for collecting employee attendance data, means for collecting employee PC usage data, means for collecting factory robot operation data, means for preprocessing the collected data, means for analyzing the preprocessed data to predict business conditions and optimal robot maintenance times, means for calculating recommended paid leave acquisition times and efficient robot operation plans for each employee, means for suggesting ways to spend paid leave based on employee preferences, means for suggesting optimal employee benefit plans for when paid leave is taken, means for visualizing business conditions and paid leave acquisition status by department, and means for notifying each employee terminal and the robot management system of the generated proposals and reports. This enables maximization of employee labor efficiency and optimal robot operation.

[1129] "Employee attendance data" refers to information related to working hours, such as the hours worked by an employee, arrival and departure times, and break times.

[1130] "Employee PC usage data" refers to data that indicates the usage status of computers used by employees during work hours, and specifically includes program execution logs and application usage history.

[1131] "Operation data of robots in factories" refers to data related to the operating status and operational status of robots operating in factories, including the start and end times of operation, the number of tasks performed, and energy consumption.

[1132] "Preprocessing" refers to the process of preparing collected data in a form that is easy to analyze. Specifically, this involves filling in missing data, standardizing data formats, correcting outliers, etc.

[1133] "Analysis" is the processing of data to identify patterns and trends based on pre-processed data and make predictions or recommendations based on those patterns and trends.

[1134] The "business situation" indicates the busy and quiet periods of business, and reflects the level of workload.

[1135] The "optimal maintenance period" refers to the most suitable time to perform maintenance in order to maintain the robot's operating efficiency at its maximum.

[1136] "Recommended timing" is information that identifies the most appropriate time for each employee to take paid leave.

[1137] An "efficient operation plan" indicates the schedule and strategy for operating the robot most efficiently.

[1138] "Suggestions for how to spend time" are suggestions for the best way for employees to use their time when they take paid leave, and are based on individual hobbies and preferences.

[1139] "Employee benefit system" refers to the various employee benefit services and systems available to employees, including recreational facilities and health promotion programs.

[1140] "Visualization" is a technique that makes data easier to understand by visually displaying the results of data analysis in graphs, dashboards, etc.

[1141] "Notifications" are used to inform employees and operations managers of the results of analysis and proposals, and are sent via email or a dedicated application.

[1142] This invention is a system that analyzes employee attendance data, PC usage data, and factory robot operation data, predicts business activity and optimal maintenance timing for the robots, and proposes optimal operation plans and methods for using employee benefits. This system includes the following processing steps.

[1143] 1. Data Collection:

[1144] The server collects data from employee terminals and factory robots. Work attendance data includes start and end times, break times, etc., while PC usage data includes program execution logs and application usage history. Robot operation data includes start and end times, number of completed tasks, and energy consumption.

[1145] 2. Data Preprocessing:

[1146] The server preprocesses the collected data, specifically by completing missing data, correcting outliers, standardizing data formats, etc. The preprocessed data is then prepared for analysis.

[1147] 3. Data Analysis:

[1148] The server uses machine learning models, such as the University of Pittsburgh's RandomForestRegressor, to predict the best times for robot maintenance and the workload based on the pre-processed data.

[1149] 4. Calculating optimal timing and operational planning:

[1150] Based on the analysis results, the server calculates the recommended timing for each employee to take paid leave and the efficient operation plan for the robot, taking into account past paid leave usage patterns and the schedule of the entire department.

[1151] 5. Suggested working hours and benefits:

[1152] The server uses employee hobby and preference data to suggest optimal ways to spend paid leave. It also suggests employee benefits that can be used when taking paid leave. This allows employees to spend their paid leave meaningfully.

[1153] 6. Visualization:

[1154] Based on the analysis results, the server visualizes the busy / slow times and paid leave usage status for each department, enabling the HR department to carry out effective labor management. Graphs and dashboards are used as visualization tools.

[1155] 7. Notifications and Interface:

[1156] The server then sends the generated proposals and reports to each employee's device and the robot management system, allowing employees to apply for paid leave and use employee benefits.

[1157] As a concrete example, if Employee A is considering taking paid leave next week, the system analyzes Employee A's past attendance data and PC usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system suggests the most suitable camping spot or outdoor activity. Furthermore, the system recommends the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan his paid leave based on that information.

[1158] Examples of prompts to input into a generative AI model include:

[1159] Input prompt: "Based on the recent operation data of robot R1, please suggest the next optimal maintenance timing and efficient operation plan. Operation start time is 2023-09-01 08:00:00, operation end time is 2023-09-01 10:00:00, number of completed tasks is 20, and energy usage is 25.5."

[1160] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1161] Step 1:

[1162] The server collects data from employee terminals and factory robots. Specifically, it collects attendance data (start and end times of work, break times), PC usage data (program execution logs and application usage history), and robot operation data (start and end times of operation, number of tasks completed, energy consumption). It receives various data as input and stores the collected data in a central database as output.

[1163] Step 2:

[1164] The server preprocesses the collected data. Specific operations include filling in missing data, standardizing data formats, and correcting outliers. For example, missing data is filled in using past data or average values, and outliers are corrected to within an appropriate range. The server receives the collected data as input and generates preprocessed data as output.

[1165] Step 3:

[1166] The server uses a machine learning model based on the preprocessed data to predict the busy and slow times of operations and the optimal maintenance times for the robots. Specifically, it performs analysis using algorithms such as RandomForestRegressor. It receives the preprocessed data as input and obtains the predicted results as output.

[1167] Step 4:

[1168] Based on the analysis results, the server calculates the recommended timing for taking paid leave for each employee and an efficient robot operation plan. Specifically, it generates an optimal schedule by taking into account past paid leave usage patterns and the department's overall schedule. It receives the analysis results as input and obtains the optimal timing and operation plan as output.

[1169] Step 5:

[1170] The server uses employee hobby and preference data to suggest optimal ways to spend paid leave. It then uses a recommender system to make suggestions based on employee feedback data and hobby data. It also suggests optimal employee benefit programs. It receives hobby and preference data and past feedback data as input, and obtains suggestions for how to spend the time and employee benefit programs as output.

[1171] Step 6:

[1172] The server visualizes the busy / slow times and paid vacation status for each department based on the analysis results. Specifically, it uses a visualization tool that displays the data in graph and dashboard format. It receives the analysis results as input and generates visualized graphs and dashboards as output.

[1173] Step 7:

[1174] The server notifies each employee's device and the robot management system of the generated proposals and reports. Specifically, it uses a mechanism to notify via email or a dedicated app. It receives proposals and reports as input and sends notification information as output.

[1175] Through this series of steps, the system can maximize employee labor efficiency and optimize robot operation.

[1176] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1177] This invention provides a system that analyzes employee attendance data, PC usage data, emotional data, etc., predicts the busy and slow periods of work that vary for each employee and department, and suggests the best time to take paid leave, how to spend it when taken, and how to best use employee benefits. The program for this system is explained in detail below.

[1178] 1. Data Collection

[1179] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. Attendance data includes work start and end times, break times, etc., while PC usage data includes program execution logs and application usage history. Emotion data is extracted using an emotion engine from the user's facial expressions, voice tone, text input, etc.

[1180] Terminal: Employees enter their start and end times and break times, and PC usage data and emotional data (facial expression analysis, voice tone analysis, etc.) are collected in the background and sent to the server in real time.

[1181] 2. Data Preprocessing

[1182] Server: Cleans all collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[1183] 3. Data Analysis

[1184] Server: Based on the preprocessed data, it uses machine learning models to predict the busy and slow periods for each employee and department. It also analyzes emotional data and evaluates the user's mental state.

[1185] 4. Calculating the recommended timing for taking paid leave

[1186] Server: Based on the results of data analysis, calculates the optimal timing for each employee to take paid leave. This takes into account not only the department's overall schedule and past paid leave usage patterns, but also the employee's mental state based on emotional data.

[1187] 5. Suggestions for how to spend your paid leave

[1188] Server: Based on the user's hobby, preference, and emotional data, the server proposes the best way to spend their vacation time. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also takes into account the user's mental state and suggests ways to promote relaxation and refreshment.

[1189] 6. Assistance in optimal use of employee benefits

[1190] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on emotional data and user profile. For example, if there is a high need for relaxation, suggests spa facilities and counseling services.

[1191] 7. Visualization of business and vacation schedules by department

[1192] Server: Based on the analysis results, the busy / slow periods and paid leave usage status for each department are visualized, enabling the HR department to carry out effective labor management.

[1193] 8. User Notification and Interface

[1194] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[1195] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[1196] Specific examples

[1197] For example, suppose employee B plans to take paid leave soon. The system analyzes employee B's attendance data and PC usage data to predict that the following week will be a relatively quiet period. It also analyzes employee B's emotional data and recognizes that employee B's mental state has recently been one that makes him prone to stress. As a result, the system suggests the best time to take paid leave and notifies employee B of refreshing ways to spend the day that suit his hobbies and preferences (for example, suggestions for nature walks or relaxation spots). At the same time, it also recommends that employee B take advantage of employee benefits that have a relaxing effect (for example, vouchers for spa facilities). Employee B can use these suggestions to apply for paid leave and take advantage of employee benefits.

[1198] In this way, the system of the present invention maximizes employee work efficiency and supports employees in taking paid leave and using employee benefits while taking into account emotional data.

[1199] The processing flow will be explained below.

[1200] Step 1: Data collection

[1201] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. Specifically, the data is sent via API or a dedicated client app and stored in a central database.

[1202] Terminal: Employees enter their work start time, end time, and break time. In the background, PC usage data (program execution logs, browser history, etc.) and emotional data (facial expression analysis, voice tone analysis, etc.) are collected and sent to the server in real time.

[1203] Step 2: Data Preprocessing

[1204] Server: Cleans all collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[1205] Specific behavior:

[1206] 1. Imputing missing data: Filling in missing data with past data or average values.

[1207] 2. Outlier Correction: Correcting or removing abnormal data based on pre-set criteria.

[1208] 3. Data format unification: Convert all data into a consistent format.

[1209] Step 3: Analyze the emotion data

[1210] Server: Analyzes collected emotional data and evaluates the user's mental state using an emotion engine, including facial expression recognition, voice tone analysis, and text analysis.

[1211] Specific behavior:

[1212] 1. Facial expression data analysis: Facial expression data collected from camera footage is analyzed to determine the emotional state.

[1213] 2. Voice data analysis: Analyze voice tones and detect emotional fluctuations.

[1214] 3. Analysis of text data: Emotional words are extracted from chat logs and email content to evaluate the emotional state.

[1215] Step 4: Data analysis

[1216] Server: Based on preprocessed attendance data, PC usage data, and emotion data, machine learning models are used to predict the busy / slow periods for each employee and department.

[1217] Specific behavior:

[1218] 1. Training a machine learning model: Train the model based on historical data.

[1219] 2. Predicting peak and trough periods: Use the trained model to predict future peak and trough periods.

[1220] 3. Setting confidence intervals: Calculate confidence intervals for the predicted results to ensure the reliability of the results.

[1221] Step 5: Calculate the best time to take paid leave

[1222] Server: Based on the results of data analysis and the evaluated emotional data, the server calculates the optimal timing for each employee to take paid leave, taking into account the department's overall schedule, past patterns of taking paid leave, and the employee's mental state based on emotional data.

[1223] Specific behavior:

[1224] 1. Check the busy / quiet periods: Check the busy / quiet periods of each employee and identify the quiet periods.

[1225] 2. Department-wide coordination: Consider the busy and slow times of the entire department and choose a time that will minimize the impact on other employees.

[1226] 3. Calculating the timing of taking paid leave: Calculate and recommend the optimal time to take leave.

[1227] Step 6: Suggestions for how to spend your paid leave

[1228] Server: Based on the user's hobby, preference, and emotional data, the server proposes the best way to spend their vacation time. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also takes into account the user's mental state and suggests ways to promote relaxation and refreshment.

[1229] Specific behavior:

[1230] 1. View user profiles: Check past activity and interests.

[1231] 2. Recommendation generation: Generate suggestions tailored to the user's preferences and take into account sentiment data.

[1232] 3. Save suggestions: Save suggestions for each user.

[1233] Step 7: Assistance in optimizing employee benefits

[1234] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on emotional data and user profile. For example, if there is a high need for relaxation, suggests spa facilities and counseling services.

[1235] Specific behavior:

[1236] 1. Browse the benefits database: See a list of available benefits.

[1237] 2. Matching with user profile: Selecting benefit plans that match the user profile.

[1238] 3. Proposal generation: Propose and record optimal ways to use benefits.

[1239] Step 8: Visualize business and vacation status by department

[1240] Server: Based on the analysis results, the busy / slow status of each department and the status of paid leave taken are visualized. This is displayed in the form of a dashboard or report.

[1241] Specific behavior:

[1242] 1. Data aggregation: Aggregate data by department.

[1243] 2. Applying visualization tools: Representing data visually using graphs and heat maps.

[1244] 3. Report generation: Generate dashboards and PDF reports and provide them to the HR department.

[1245] Step 9: User Notification and Interface

[1246] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[1247] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[1248] Specific behavior:

[1249] 1. Notification Generation: Converting proposals and reports into notification format.

[1250] 2. Send notifications: Send notifications to users via email or app notifications.

[1251] 3. User interface operation: The user takes the necessary steps based on the notification received.

[1252] Example 2

[1253] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1254] In today's corporate environment, maximizing employee work efficiency and maintaining a healthy working environment are important. However, conventional systems only collect attendance data and PC usage data, and do not adequately consider employees' emotions and mental state. Furthermore, they are unable to provide appropriate suggestions regarding the timing of paid leave and how to spend it when taken, creating challenges in optimizing employee welfare benefits. This invention solves these issues by providing a system that suggests optimal work life for each employee.

[1255] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1256] In this invention, the server includes means for collecting attendance data, PC usage data, and emotion data from each employee's terminal, means for cleaning the collected data and for completing and correcting missing and outlier values, means for analyzing the preprocessed data and predicting business activity using a machine learning model, means for analyzing each employee's emotion data and evaluating their mental state, means for calculating the optimal timing for taking paid leave based on the business activity and emotion data, means for suggesting optimal ways to spend paid leave based on the employee's hobby and preference data and emotion data, means for suggesting employee benefit plans available when taking paid leave, means for visualizing business activity and paid leave usage status by department, and means for notifying each employee of the generated suggestions and reports. This maximizes employee work efficiency and enables the creation of a work environment that takes mental health into consideration.

[1257] "Work attendance data" is information related to an employee's working status, such as start time, finish time, and break time.

[1258] "PC usage data" refers to information such as program execution logs and application usage history for computers used by employees during work hours.

[1259] "Emotion data" is information about an employee's emotions and mental state extracted using an emotion engine from a user's facial expressions, tone of voice, text input, etc.

[1260] A "machine learning model" is a mathematical algorithm or statistical model that learns from collected data and predicts patterns and trends.

[1261] "Busy / slow status" refers to the busyness or slowness of business during a specific period.

[1262] "Hobbies and Preference Data" is information about a user's interests and preferences, including hobbies, interests, and favorite activities.

[1263] A "welfare system" refers to various services and support systems provided by companies with the aim of improving the welfare of their employees, and includes, for example, vouchers for use of spa facilities and counseling services.

[1264] "Cleanup" is the process of filling in missing values ​​and correcting outliers in order to improve the quality of data.

[1265] "Visualization" is the process of visually representing data and information, making them easier to understand using graphs, charts, etc.

[1266] A "recommender system" is a system that recommends optimal activities, travel destinations, etc. based on a user's hobby and preference data.

[1267] "Notifications" are a means of informing users of specific information or offers in real time, and are provided via email or a dedicated app.

[1268] The present invention provides a system that analyzes employee attendance data, PC usage data, emotional data, etc., predicts the busy and slow business conditions that vary for each employee and department, and suggests the appropriate timing for taking paid leave, the appropriate way to spend it when taken, and the optimal way to use employee benefits.

[1269] 1. Data Collection

[1270] server

[1271] The server periodically collects attendance data, PC usage data, and emotion data from each employee's device. Attendance data includes work start and end times and break times, while PC usage data includes program execution logs and application usage history. Emotion data is extracted from the user's facial expressions, voice tone, text input, etc. using an emotion engine.

[1272] Terminal

[1273] Employees enter their work start and end times, as well as break times, into the terminals, which collect PC usage data and emotional data (facial expression analysis, voice tone analysis, etc.) in the background and send it to the server in real time.

[1274] 2. Data Preprocessing

[1275] server

[1276] All collected data is cleaned. Missing values ​​are filled in using past data or average values, and outliers are corrected or removed. Data formats are also standardized to make them suitable for analysis.

[1277] 3. Data Analysis

[1278] server

[1279] Based on the preprocessed data, machine learning models are used to predict the busy and slow periods for each employee and department. Emotional data is also analyzed to evaluate the user's mental state. Specific software used includes the Python libraries TensorFlow and Scikit-learn.

[1280] 4. Calculating the recommended timing for taking paid leave

[1281] server

[1282] Based on the results of the data analysis, the system calculates the optimal timing for each employee to take paid leave, taking into account not only the department's overall schedule and past paid leave usage patterns, but also the employee's mental state based on emotional data.

[1283] 5. Suggestions for how to spend your paid leave

[1284] server

[1285] The system suggests optimal ways to spend paid vacation time based on the user's hobby, preference, and emotional data. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also includes suggestions to promote relaxation and refreshment. Specific software used includes Apache Spark and Elasticsearch.

[1286] 6. Assistance in optimal use of employee benefits

[1287] server

[1288] It suggests employee benefits that can be used when taking paid leave, and suggests spa facilities and counseling services if employees need to relax based on their emotional data and profile information.

[1289] 7. Visualization of business and vacation schedules by department

[1290] server

[1291] Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized in graphs and charts, enabling the HR department to carry out effective labor management. For example, visualization can be done using Tableau or Power BI.

[1292] 8. User Notification and Interface

[1293] server

[1294] The generated proposals and reports are sent to each user's device, and notifications are sent via email or a dedicated app.

[1295] Terminal

[1296] The user checks the received notification and applies for paid leave or takes advantage of employee benefits.

[1297] Specific examples

[1298] For example, suppose employee B plans to take paid leave soon. The system analyzes employee B's attendance data and PC usage data to predict that the following week will be a relatively quiet period. It also analyzes employee B's emotional data and recognizes that employee B's mental state has recently been one that makes him prone to stress. As a result, the system suggests the best time to take paid leave and notifies employee B of refreshing ways to spend the day that suit his hobbies and preferences (for example, suggestions for nature walks or relaxation spots). At the same time, it also recommends that employee B take advantage of employee benefits that have a relaxing effect (for example, vouchers for spa facilities). Employee B can use these suggestions to apply for paid leave and take advantage of employee benefits.

[1299] Prompt Sentence Examples

[1300] Attendance data analysis prompt

[1301] Analyze employee attendance data to predict the next week's peak and trough times. The dataset includes start and finish times and break times.

[1302] Emotional Data Analysis Prompt

[1303] Analyze employee emotional data and assess their mental state. Data sets include facial expression analysis, voice tone, and text input.

[1304] Prompt to calculate the optimal timing for taking paid leave

[1305] Calculate the optimal timing for employees to take paid leave based on attendance and sentiment data.

[1306] Prompts for suggestions on how to spend your time based on your hobbies and preferences

[1307] Based on the user's hobby, preference, and emotional data, please suggest the best way to spend their vacation time. Please include suggestions that promote relaxation and refreshment.

[1308] Benefit Plan Recommendation Prompts

[1309] Based on emotional data and user profiles, suggest benefits that can be used when taking paid time off. If relaxation is needed, suggest spa facilities or counseling services.

[1310] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1311] Step 1: Data collection

[1312] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. The server sends data collection requests to each device at set intervals, receives the data sent from each device, and stores it in a database. The input is data from each employee's device, and the output is an integrated database.

[1313] Terminal: Employees enter their work start time, end time, and break time into a dedicated app. The terminal collects PC usage information (application usage history, usage time, etc.) and emotional data (facial expression analysis, voice tone analysis) in the background and sends them to the server in real time. The input is the employee's operations and background data collection, and the output is the data sent to the server.

[1314] Step 2: Data Preprocessing

[1315] Server: Cleans up all collected data. Specifically, if missing values ​​are found, they are filled in using past data or average values. Also, outliers are detected and corrected or removed. Data formats are standardized and made suitable for analysis. The input is raw data, and the output is cleaned data.

[1316] Step 3: Data analysis

[1317] Server: Using the preprocessed data, a machine learning model predicts the busy / quick schedule for each employee and department. Emotional data is also analyzed to evaluate each employee's mental state. Specific software used includes the Python libraries TensorFlow and Scikit-learn. The input is the cleaned-up data, and the output is predicted data on busy / quick schedules and the results of an evaluation of mental state.

[1318] Step 4: Calculate the best time to take paid leave

[1319] Server: Based on the analysis results, calculates the optimal timing for each employee to take paid leave. Calculates the optimal date to take paid leave by taking into account the department's overall schedule, past paid leave usage patterns, and mental state. The input is the analysis results, and the output is the recommended timing for taking paid leave.

[1320] Step 5: Suggestions for how to spend your paid leave

[1321] Server: Based on the user's hobby, preference, and emotional data, the server proposes the optimal way to spend their paid vacation. A recommender system is used to recommend activities and travel destinations suitable for each user. Specific software used includes Apache Spark and Elasticsearch. The input is hobby, preference, and emotional data, and the output is a proposal for the optimal way to spend the day.

[1322] Step 6: Assistance in optimizing employee benefits

[1323] Server: Proposes employee benefit programs that can be used when taking paid leave. Based on emotional data and profile information, recommends spa facilities or counseling services if there is a high need for relaxation. The input is emotional data and profile information, and the output is a proposed employee benefit program.

[1324] Step 7: Visualize business and leisure situations by department and paid leave usage status

[1325] Server: Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized in graphs and charts. This makes it easier for the HR department to carry out effective labor management. For example, visualization can be done using Tableau or Power BI. The input is the analysis results, and the output is visualized data.

[1326] Step 8: User Notification and Interface

[1327] Server: Sends the generated proposals and reports to each user's device. Notifications are sent via email or a dedicated app. The input is the generated proposals and reports, and the output is a notification message.

[1328] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits. The input is the notification message, and the output is the user's action.

[1329] (Application example 2)

[1330] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1331] While the operation of robots in modern factories improves production efficiency, their maintenance and operational status management are also important. In particular, there is a need to efficiently collect and analyze robot operation data and maintenance history data to predict optimal maintenance times. Also important is the overall improvement of work efficiency and employee benefits, including the appropriate timing and use of employee paid leave and suggestions for utilizing employee benefit programs.

[1332] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting employee attendance data, means for collecting employee PC usage data, means for collecting factory robot operation data and maintenance history data, means for preprocessing the collected data, means for analyzing the preprocessed data to predict business activity, means for calculating recommended paid vacation time for each employee and optimal maintenance times for each robot, means for proposing paid vacation time and maintenance content based on employee preferences and robot operation status, means for proposing optimal employee benefit programs for paid vacation time and robot maintenance, means for visualizing business activity and robot operation status and paid vacation time for each department, and means for notifying each employee terminal and each engineer terminal of the generated proposals and reports. This enables optimal paid vacation time and employee benefit use for each employee and efficient robot maintenance scheduling.

[1333] "Employee attendance data" refers to work-related record information such as the start and end times of each employee's shift, and break times.

[1334] "PC usage data" refers to data such as program execution logs and application usage history for computers used by employees for work.

[1335] "Robot operation data" refers to information such as the operation records, operating hours, and error occurrence status of robots operating in a factory.

[1336] "Maintenance history data" is recorded information such as the content, date, and person who performed the maintenance performed on the robot in the past.

[1337] "Preprocessing" refers to the process of preparing collected data for analysis, such as filling in missing values, correcting outliers, and standardizing data formats.

[1338] "Business activity status" is data showing the busy and slow periods of business, and indicates the workload status of employees and departments.

[1339] "Recommended timing for taking paid leave" is the time when the optimal time to take paid leave is calculated and suggested for each employee.

[1340] Based on "hobbies and preferences" means customizing the proposals to suit the interests and preferences of each employee.

[1341] "How to spend your paid leave" is a content that suggests activities and ways for employees to rest during the period they take paid leave.

[1342] "Employee benefits system" refers to various services and support systems provided by a company that employees can use, and often includes relaxation facilities and counseling services.

[1343] "Visualization of busy and slow periods" refers to visually displaying the results of data analysis so that busy and slow periods can be understood at a glance.

[1344] The "generated proposals" are proposals for each employee and robot that are calculated and created based on data analysis.

[1345] "Business and quiet periods by department" is data showing the busy and quiet periods for work in each specific department.

[1346] "Each employee terminal" refers to an individual device (PC, smartphone, etc.) used by an employee.

[1347] "Technician's Terminal" refers to an individual device used by a technician in a factory or operational location.

[1348] A "machine learning model" is an algorithm used in data analysis, a system that learns patterns from the original data and makes predictions and classifications.

[1349] A "confidence interval" is a numerical range that indicates the reliability of a prediction result and provides a statistical indication of how accurate the prediction is.

[1350] This invention is a system that analyzes employee attendance data, PC usage data, factory robot operation data, and maintenance history data, and based on that, suggests optimal timing for taking paid leave and performing maintenance. This system is implemented using both a server and a terminal.

[1351] Data collection

[1352] The server collects employee attendance data, PC usage data, factory robot operation data, and maintenance history data. These data are obtained from each employee's terminal and the factory management system. Specifically, employee attendance data includes work start and end times and break times, while PC usage data includes program execution logs and application usage history. Robot operation data includes operating hours and error occurrence status, and maintenance history data includes the details and dates of past maintenance.

[1353] Data Preprocessing

[1354] The server preprocesses the collected data, which includes filling in missing values, correcting outliers, and standardizing data formats, making the data suitable for analysis.

[1355] Data analysis

[1356] The server uses a machine learning model based on the preprocessed data to predict the busy and slow times of operations and the optimal maintenance times for the robot. This analysis uses a machine learning algorithm (e.g., Random Forest Regressor). It also uses a generative AI model to generate recommendations based on the user's preferences and the robot's operating status.

[1357] Paid leave and maintenance scheduling

[1358] Based on the analysis results, the server calculates the recommended timing for each employee to take paid leave and the optimal maintenance period for each robot, including setting a confidence interval based on past data and current operating conditions.

[1359] Suggestions and Notifications

[1360] The server then sends the generated proposals and reports to each employee's and technician's device via email or a dedicated application. The employee can then review the proposals, apply for paid leave and employee benefits, and schedule robot maintenance.

[1361] Specific examples

[1362] For example, if factory robot A has recently been operating slowly, the system will suggest the optimal date and time for the next maintenance based on past maintenance history and operation data. The next maintenance for this robot is predicted to be required two weeks from now on August 15th, and the appropriate maintenance work will be notified to the maintenance team.

[1363] Prompt Sentence Examples

[1364] "Please create a model that predicts the optimal next maintenance timing for each robot based on the factory robot operation data and maintenance history data. Also, please tell me the steps to develop a system that notifies technicians of this in the form of a smartphone application."

[1365] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1366] Step 1: Data collection

[1367] The server collects data from each employee's terminal and the factory management system. Inputs include employee attendance data, PC usage data, robot operation data, and maintenance history data. Outputs store this data on the server and prepare it in a format that can be used for the next processing step. Specific operations include retrieving data from a database and streaming data in real time.

[1368] Step 2: Data Preprocessing

[1369] The server preprocesses the collected data. The input is the raw data collected in step 1. Specific operations include imputing missing values, correcting outliers, and standardizing data formats. The output is clean data that has been prepared for analysis.

[1370] Step 3: Data analysis

[1371] The server uses a machine learning model to analyze the preprocessed data. The input is clean data. Specifically, it uses machine learning algorithms such as RandomForestRegressor to predict the busy / unbusy schedule and the optimal maintenance timing for the robot. The output is the predicted busy / unbusy schedule and recommended maintenance timing.

[1372] Step 4: Generate proposals

[1373] The server generates proposals using a generative AI model based on the analysis results. The input is the prediction results from step 3. Specific actions include customizing how employees spend their paid leave and the maintenance details, taking into account the individual hobbies and preferences of employees and the operating status of the robot. The output is specific proposals provided to the user.

[1374] Step 5: Notification

[1375] The server notifies each employee's terminal and the engineer's terminal of the generated proposal and report. The input is the proposal generated in step 4. Specific actions include sending an email or sending a notification via a dedicated application. The output is the proposal displayed on the user's terminal, which the user can check and take the next action.

[1376] Step 6: User Actions

[1377] The user takes action based on the notified proposal. The input is the proposal received in step 5. Specific actions include applying for paid leave, applying for employee benefits, and scheduling robot maintenance. The output is a record of the application and scheduling as actions taken by the user.

[1378] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1379] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1380] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1381] [Fourth embodiment]

[1382] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1383] 7, a 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.

[1384] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1385] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1386] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1387] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1388] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1389] The control object 443 includes a display device, LEDs in the eyes, and motors for driving 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 emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1390] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1391] The specific processing program 56 is an example of a "program" according to the technology of the present 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.

[1392] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1393] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1394] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1395] This invention provides a system that analyzes employee attendance data, PC usage data, etc., predicts the busy and slow periods of work that vary for each employee and department, and suggests the best time to take paid leave, how to spend it when it is taken, and how to best use employee benefits. The program for this system is explained in detail below.

[1396] 1. Data Collection

[1397] First, attendance data and PC usage data are periodically collected from each employee's device. The attendance data includes start and end times of work, break times, etc., while the PC usage data includes program execution logs and application usage history.

[1398] Server: Automatically collects data from employee devices and stores it in a central database.

[1399] Terminal: Attendance data entry and PC usage data log collection are performed in the background.

[1400] 2. Data Preprocessing

[1401] Next, the collected data is preprocessed, which involves filling in missing data, correcting outliers, and standardizing the data format.

[1402] Server: Cleans and preprocesses the collected data. Missing data is filled in using past data and average values.

[1403] 3. Data Analysis

[1404] Predict business activity based on pre-processed data. Use machine learning models and statistical algorithms to analyze the busy and slow periods for each employee and department.

[1405] Server: Trains the machine learning model, performs analysis, and sets confidence intervals to improve prediction accuracy.

[1406] 4. Calculating the recommended timing for taking paid leave

[1407] Based on the analysis results, the system calculates the optimal timing for each employee to take paid leave, taking into account past paid leave usage patterns and the department's overall schedule.

[1408] Server: Calculates the optimal time to take paid leave based on each employee's busy / quick schedule.

[1409] 5. Suggestions for how to spend your paid leave

[1410] Furthermore, the system uses data on employees' hobbies and preferences to suggest optimal ways to spend their paid leave, and utilizes a recommender system to recommend activities and travel destinations that match individual interests.

[1411] Server: Based on past data and preference information, suggests the best way for each employee to spend their time.

[1412] 6. Assistance in optimal use of employee benefits

[1413] We also propose employee benefit systems that can be used when taking paid leave, and provide optimal ways to use employee benefits so that employees can spend their paid leave meaningfully.

[1414] Server: Extracts and proposes systems suitable for each employee from a database of employee benefits.

[1415] 7. Visualization of business and vacation schedules by department

[1416] Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized, enabling the HR department to carry out effective labor management.

[1417] Server: Uses visualization tools to display data in graphs and dashboards.

[1418] 8. User Notification and Interface

[1419] Finally, the generated proposals and reports are sent to each employee's device, and employees can use them to apply for paid leave and use employee benefits.

[1420] Server: Generates proposals and reports and notifies each employee via email or a dedicated app.

[1421] Terminal: The employee checks the received notification and takes the necessary steps.

[1422] Specific examples

[1423] For example, if Employee A is considering taking paid leave next week, the system will analyze Employee A's past attendance data and PC usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system will suggest the most suitable camping spot or outdoor activity. Furthermore, the system will recommend the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan their paid leave based on that information.

[1424] In this way, the system of the present invention maximizes employee work efficiency and supports appropriate use of paid leave and employee benefits.

[1425] The processing flow will be explained below.

[1426] Step 1: Data collection

[1427] Server: Periodically collects attendance data and PC usage data from each employee's device. Specifically, the data is sent via API or a dedicated client app and stored in a central database.

[1428] Terminal: Employees enter their work start time, end time, and break time. PC usage data (program execution logs, browser history, etc.) is collected in the background and sent to the server in real time.

[1429] Step 2: Data Preprocessing

[1430] Server: Cleans the collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[1431] Specific behavior:

[1432] 1. Imputing missing data: Filling in missing data with past data or average values.

[1433] 2. Outlier Correction: Correcting or removing abnormal data based on pre-set criteria.

[1434] 3. Data format unification: Convert all data into a consistent format.

[1435] Step 3: Data analysis

[1436] Server: Analyzes the pre-processed data and uses machine learning models to predict the busy and slow periods for each employee and department. It uses algorithms such as random forests and neural networks.

[1437] Specific behavior:

[1438] 1. Training a machine learning model: Train the model based on historical data.

[1439] 2. Predicting peak and trough periods: Use the trained model to predict future peak and trough periods.

[1440] 3. Setting confidence intervals: Calculate confidence intervals for the predicted results to ensure the reliability of the results.

[1441] Step 4: Calculate the best time to take paid leave

[1442] Server: Based on the results of data analysis, calculates the optimal timing for each employee to take paid leave, taking into account the department's overall schedule and past paid leave usage patterns.

[1443] Specific behavior:

[1444] 1. Check the busy / quiet periods: Check the busy / quiet periods of each employee and identify the quiet periods.

[1445] 2. Department-wide coordination: Consider the busy and slow times of the entire department and choose a time that will minimize the impact on other employees.

[1446] 3. Calculating the timing of taking paid leave: Calculate and recommend the optimal time to take leave.

[1447] Step 5: Suggestions for how to spend your paid leave

[1448] Server: Based on the user's hobby and preference data, the server proposes the best way to spend their vacation time. Using a recommender system, the server recommends activities and travel destinations suitable for each user.

[1449] Specific behavior:

[1450] 1. View user profiles: Check past activity and interests.

[1451] 2. Recommendation generation: Generate recommendations tailored to the user's preferences.

[1452] 3. Save suggestions: Save suggestions for each user.

[1453] Step 6: Assistance in optimizing employee benefits

[1454] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on the user profile.

[1455] Specific behavior:

[1456] 1. Browse the benefits database: See a list of available benefits.

[1457] 2. Matching with user profile: Selecting benefit plans that match the user profile.

[1458] 3. Proposal generation: Propose and record optimal ways to use benefits.

[1459] Step 7: Visualize business and leisure situations by department and paid leave usage status

[1460] Server: Based on the analysis results, the busy / slow status of each department and the status of paid leave taken are visualized. This is displayed in the form of a dashboard or report.

[1461] Specific behavior:

[1462] 1. Data aggregation: Aggregate data by department.

[1463] 2. Applying visualization tools: Representing data visually using graphs and heat maps.

[1464] 3. Report generation: Generate dashboards and PDF reports and provide them to the HR department.

[1465] Step 8: User Notification and Interface

[1466] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[1467] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[1468] Specific behavior:

[1469] 1. Notification Generation: Converting proposals and reports into notification format.

[1470] 2. Send notifications: Send notifications to users via email or app notifications.

[1471] 3. User interface operation: The user takes the necessary steps based on the notification received.

[1472] Example 1

[1473] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1474] Conventional labor management systems are insufficient to maximize work efficiency and improve employee satisfaction because they are unable to comprehensively analyze employees' attendance data and PC usage data to suggest the appropriate timing for taking paid leave and the optimal way to use employee benefits. Furthermore, previous systems did not suggest ways to spend paid leave that took into account the hobbies and preferences of individual employees. This invention is intended to solve these problems.

[1475] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1476] In this invention, the server includes a means for collecting employee work data, a means for collecting employee computer usage data, and a means for preprocessing the collected data. This enables efficient data collection and preprocessing. It also includes a means for predicting workloads based on the analyzed data, a means for calculating recommended vacation timings for each employee, a means for suggesting vacation time based on employee preferences, a means for suggesting optimal employee benefit plans for vacation time, a means for visualizing the workload and vacation status for each department, and a means for notifying each employee of the generated proposals and reports. This maximizes employee work efficiency and supports appropriate use of paid vacation time and employee benefits.

[1477] "Employee work data" refers to information such as start time, finish time, and break time that is recorded when an employee works.

[1478] "Employee computer usage data" refers to information such as execution logs of computer programs used by employees during work hours and application usage history.

[1479] "Preprocessing" refers to the process of preparing collected raw data in a format suitable for analysis by supplementing missing data, correcting outliers, and standardizing data formats.

[1480] "Analysis" refers to the process of applying statistical techniques and machine learning models to preprocessed data to predict business activity levels and individual trends.

[1481] "Calculating the recommended timing" refers to calculating the most appropriate time for each employee to take paid leave based on the analyzed data.

[1482] "Suggestions based on hobbies and preferences" refers to recommending ways and activities to spend paid leave based on information about each employee's hobbies and interests.

[1483] "Proposing a benefit system" refers to recommending the most appropriate benefit program that an employee can use during their paid leave.

[1484] "Visualization" refers to the process of displaying analysis results and various data in graph or dashboard format to make them easy to understand.

[1485] "Notifying" refers to the process of sending the proposals and reports generated by the server to each employee's terminal to inform them.

[1486] "System" refers to a set of processes and devices that comprehensively combine the above-mentioned measures to maximize employee work efficiency and support appropriate use of paid leave and employee benefits.

[1487] The present invention is a system that analyzes employee work data and computer usage data, predicts busy and slow times at work, and suggests optimal times and ways to use paid leave, as well as how to utilize employee benefits. The system of the present invention is operated in cooperation with employee terminals and a server.

[1488] Data collection

[1489] First, a collection program is installed on each employee's device. This program continuously collects work data and computer usage data in the background. The data collected by the device is sent to a server at regular intervals. A central database is installed on the server, where the sent data is efficiently stored.

[1490] Data Preprocessing

[1491] Collected data is often incomplete as it is, so it is necessary to supplement missing data, correct outliers, and standardize the data format. This is called preprocessing, and is performed by the server. The server uses a dedicated cleaning algorithm to prepare the data into the appropriate format.

[1492] Data analysis

[1493] Based on the pre-processed data, the server analyzes the workload status. For the analysis, a machine learning model (e.g., random forest or support vector machine) is used. The server uses the trained model to predict the workload status of each employee and department. Confidence intervals can also be set to improve the accuracy of the prediction.

[1494] Calculating the best time to take paid leave

[1495] Based on the analyzed data, the server calculates the optimal timing for each employee to take paid leave, taking into account past paid leave usage patterns and the department's overall schedule.

[1496] Suggestions for how to spend your paid leave

[1497] Using employee hobby and preference data, the server suggests optimal ways to spend paid leave. Utilizing a recommender system, it recommends appropriate activities and travel destinations for each employee.

[1498] Optimal use of employee benefits

[1499] The server also suggests employee benefit plans that can be used when taking paid leave, extracting the best plan for each employee from a database of employee benefits and supporting them in using their paid leave in a meaningful way.

[1500] Visualization of the situation by department

[1501] Based on the analysis results, the server visualizes the busy / unbusy status and paid leave usage status of each department, and displays the results in intuitive graphs and dashboards using visualization tools (e.g., Tableau or Power BI).

[1502] Notification of proposal details

[1503] The final proposals and reports are sent to each employee's device. The server then sends the proposals via email or a dedicated application, allowing employees to apply for paid leave or use employee benefits.

[1504] Specific examples

[1505] For example, if Employee A is considering taking paid leave next week, the system analyzes Employee A's past work data and computer usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system suggests the most suitable camping spot or outdoor activity. It also recommends the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan their paid leave based on that information.

[1506] Prompt Sentence Examples

[1507] "Based on employee A's attendance data and PC usage data from the past six months, please predict the busy and slow periods for next week and suggest the best time and way for him to take paid leave."

[1508] In this way, the system of the present invention maximizes employee work efficiency and supports appropriate use of paid leave and employee benefits.

[1509] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1510] Step 1:

[1511] Data collection

[1512] Device:

[1513] Input: Employee working hours, break times, finish times, program usage, etc.

[1514] Processing: Work data and computer usage data are collected in the background.

[1515] Output: The raw data collected.

[1516] Specific operation: The terminal software automatically records the time when employees start and finish work and collects computer usage history as a log.

[1517] Step 2:

[1518] Data Transfer

[1519] Device:

[1520] Input: Raw data collected.

[1521] Processing: Data collected on the device is transferred to the server at regular intervals.

[1522] Output: The data sent to the server.

[1523] Specific operation: Data is sent from the terminal to the server via the communication module.

[1524] Step 3:

[1525] Data Preprocessing

[1526] server:

[1527] Input: Raw data received from the terminal.

[1528] Processing: Impute missing data, correct outliers, and standardize data formats.

[1529] Output: A clean dataset.

[1530] Specific operation: Run a data cleaning algorithm (e.g., scikit-learn's SimpleImputer) on the server to standardize the data.

[1531] Step 4:

[1532] Data analysis

[1533] server:

[1534] Input: The preprocessed dataset.

[1535] Processing: Predicting peak and trough times using machine learning models.

[1536] Output: Predicted busy / quick times for each employee and department.

[1537] What you'll do: Use Python libraries to train and run machine learning models (e.g., random forests, support vector machines).

[1538] Step 5:

[1539] Calculating the timing of paid leave

[1540] server:

[1541] Input: Predicted busy / slow times, past paid leave data, department schedule.

[1542] Processing: Calculate the optimal timing for taking paid leave.

[1543] Output: A list of optimal vacation times for each employee.

[1544] Specific operation: Using past data, an algorithm is used to calculate the optimal timing for taking paid leave.

[1545] Step 6:

[1546] Suggestions for spending time

[1547] server:

[1548] Input: Employee hobby and preference data, information on the timing of taking paid leave.

[1549] Processing: Use a recommender system to suggest the best way to spend your time.

[1550] Output: Specific activities and travel destination suggestions for each employee.

[1551] How it works: The recommender system analyzes employee preference data and suggests activities that are best suited to them.

[1552] Step 7:

[1553] Assistance in optimal use of employee benefits

[1554] server:

[1555] Input: List of optimal vacation times, employee benefits database.

[1556] Processing: Extraction of benefit plans available when taking paid leave.

[1557] Output: Proposal of optimal employee benefits plan for each employee.

[1558] What it does: Searches the benefits database to extract and recommend the best options.

[1559] Step 8:

[1560] visualization

[1561] server:

[1562] Input: Data on business / slow periods by department and paid leave usage status.

[1563] Processing: Visualize analysis results in graphs and dashboards.

[1564] Output: Visualization report for HR department.

[1565] Specific behavior: Use visualization tools (e.g., Tableau, Power BI) to display analysis results.

[1566] Step 9:

[1567] notification

[1568] server:

[1569] Input: Proposal and report.

[1570] Processing: Proposals and reports are sent to each employee's device.

[1571] Output: Notification sent to each employee terminal.

[1572] Specific operation: The generated information is notified to each employee via email or a dedicated application.

[1573] Specific examples

[1574] For example, if employee A is considering taking paid leave next week, the system analyzes employee A's past work data and computer usage data to predict that next week will be a relatively quiet period. Then, based on employee A's interest in outdoor activities, the system suggests the best camping spots and outdoor activities. It also recommends the use of company-provided recreational facilities as part of the employee benefits program available during that period. This information is sent to employee A's device, allowing them to apply for and plan their paid leave based on that information.

[1575] Example prompt for a generative AI model:

[1576] "Based on employee A's attendance data and PC usage data from the past six months, please predict the busy and slow periods for next week and suggest the best time and way for him to take paid leave."

[1577] (Application example 1)

[1578] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1579] In many modern organizations, in addition to employee attendance management and PC usage status monitoring, the operation and management of robots used in factories is also a key issue. However, managing these data separately reduces overall efficiency and makes it difficult to create consistent operational plans. Furthermore, efficient proposals are often not made regarding the appropriate timing for employees to take paid leave or how to utilize employee benefits when doing so. Furthermore, with regard to factory robots, operational data is not fully utilized, and optimal maintenance timing can be overlooked. A system that can eliminate these management complexities and reduced efficiency is needed.

[1580] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1581] In this invention, the server includes means for collecting employee attendance data, means for collecting employee PC usage data, means for collecting factory robot operation data, means for preprocessing the collected data, means for analyzing the preprocessed data to predict business conditions and optimal robot maintenance times, means for calculating recommended paid leave acquisition times and efficient robot operation plans for each employee, means for suggesting ways to spend paid leave based on employee preferences, means for suggesting optimal employee benefit plans for when paid leave is taken, means for visualizing business conditions and paid leave acquisition status by department, and means for notifying each employee terminal and the robot management system of the generated proposals and reports. This enables maximization of employee labor efficiency and optimal robot operation.

[1582] "Employee attendance data" refers to information related to working hours, such as the hours worked by an employee, arrival and departure times, and break times.

[1583] "Employee PC usage data" refers to data that indicates the usage status of computers used by employees during work hours, and specifically includes program execution logs and application usage history.

[1584] "Operation data of robots in factories" refers to data related to the operating status and operational status of robots operating in factories, including the start and end times of operation, the number of tasks performed, and energy consumption.

[1585] "Preprocessing" refers to the process of preparing collected data in a form that is easy to analyze. Specifically, this involves filling in missing data, standardizing data formats, correcting outliers, etc.

[1586] "Analysis" is the processing of data to identify patterns and trends based on pre-processed data and make predictions or recommendations based on those patterns and trends.

[1587] The "business situation" indicates the busy and quiet periods of business, and reflects the level of workload.

[1588] The "optimal maintenance period" refers to the most suitable time to perform maintenance in order to maintain the robot's operating efficiency at its maximum.

[1589] "Recommended timing" is information that identifies the most appropriate time for each employee to take paid leave.

[1590] An "efficient operation plan" indicates the schedule and strategy for operating the robot most efficiently.

[1591] "Suggestions for how to spend time" are suggestions for the best way for employees to use their time when they take paid leave, and are based on individual hobbies and preferences.

[1592] "Employee benefit system" refers to the various employee benefit services and systems available to employees, including recreational facilities and health promotion programs.

[1593] "Visualization" is a technique that makes data easier to understand by visually displaying the results of data analysis in graphs, dashboards, etc.

[1594] "Notifications" are used to inform employees and operations managers of the results of analysis and proposals, and are sent via email or a dedicated application.

[1595] This invention is a system that analyzes employee attendance data, PC usage data, and factory robot operation data, predicts business activity and optimal maintenance timing for the robots, and proposes optimal operation plans and methods for using employee benefits. This system includes the following processing steps.

[1596] 1. Data Collection:

[1597] The server collects data from employee terminals and factory robots. Work attendance data includes start and end times, break times, etc., while PC usage data includes program execution logs and application usage history. Robot operation data includes start and end times, number of completed tasks, and energy consumption.

[1598] 2. Data Preprocessing:

[1599] The server preprocesses the collected data, specifically by completing missing data, correcting outliers, standardizing data formats, etc. The preprocessed data is then prepared for analysis.

[1600] 3. Data Analysis:

[1601] The server uses machine learning models, such as the University of Pittsburgh's RandomForestRegressor, to predict the best times for robot maintenance and the workload based on the pre-processed data.

[1602] 4. Calculating optimal timing and operational planning:

[1603] Based on the analysis results, the server calculates the recommended timing for each employee to take paid leave and the efficient operation plan for the robot, taking into account past paid leave usage patterns and the schedule of the entire department.

[1604] 5. Suggested working hours and benefits:

[1605] The server uses employee hobby and preference data to suggest optimal ways to spend paid leave. It also suggests employee benefits that can be used when taking paid leave. This allows employees to spend their paid leave meaningfully.

[1606] 6. Visualization:

[1607] Based on the analysis results, the server visualizes the busy / slow times and paid leave usage status for each department, enabling the HR department to carry out effective labor management. Graphs and dashboards are used as visualization tools.

[1608] 7. Notifications and Interface:

[1609] The server then sends the generated proposals and reports to each employee's device and the robot management system, allowing employees to apply for paid leave and use employee benefits.

[1610] As a concrete example, if Employee A is considering taking paid leave next week, the system analyzes Employee A's past attendance data and PC usage data to predict that next week will be a relatively quiet period. Then, taking into account that Employee A's hobby is outdoor activities, the system suggests the most suitable camping spot or outdoor activity. Furthermore, the system recommends the use of a company-provided recreational facility as an employee benefit program available during that period. This information is sent to Employee A's device, allowing Employee A to apply for and plan his paid leave based on that information.

[1611] Examples of prompts to input into a generative AI model include:

[1612] Input prompt: "Based on the recent operation data of robot R1, please suggest the next optimal maintenance timing and efficient operation plan. Operation start time is 2023-09-01 08:00:00, operation end time is 2023-09-01 10:00:00, number of completed tasks is 20, and energy usage is 25.5."

[1613] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1614] Step 1:

[1615] The server collects data from employee terminals and factory robots. Specifically, it collects attendance data (start and end times of work, break times), PC usage data (program execution logs and application usage history), and robot operation data (start and end times of operation, number of tasks completed, energy consumption). It receives various data as input and stores the collected data in a central database as output.

[1616] Step 2:

[1617] The server preprocesses the collected data. Specific operations include filling in missing data, standardizing data formats, and correcting outliers. For example, missing data is filled in using past data or average values, and outliers are corrected to within an appropriate range. The server receives the collected data as input and generates preprocessed data as output.

[1618] Step 3:

[1619] The server uses a machine learning model based on the preprocessed data to predict the busy and slow times of operations and the optimal maintenance times for the robots. Specifically, it performs analysis using algorithms such as RandomForestRegressor. It receives the preprocessed data as input and obtains the predicted results as output.

[1620] Step 4:

[1621] Based on the analysis results, the server calculates the recommended timing for taking paid leave for each employee and an efficient robot operation plan. Specifically, it generates an optimal schedule by taking into account past paid leave usage patterns and the department's overall schedule. It receives the analysis results as input and obtains the optimal timing and operation plan as output.

[1622] Step 5:

[1623] The server uses employee hobby and preference data to suggest optimal ways to spend paid leave. It then uses a recommender system to make suggestions based on employee feedback data and hobby data. It also suggests optimal employee benefit programs. It receives hobby and preference data and past feedback data as input, and obtains suggestions for how to spend the time and employee benefit programs as output.

[1624] Step 6:

[1625] The server visualizes the busy / slow times and paid vacation status for each department based on the analysis results. Specifically, it uses a visualization tool that displays the data in graph and dashboard format. It receives the analysis results as input and generates visualized graphs and dashboards as output.

[1626] Step 7:

[1627] The server notifies each employee's device and the robot management system of the generated proposals and reports. Specifically, it uses a mechanism to notify via email or a dedicated app. It receives proposals and reports as input and sends notification information as output.

[1628] Through this series of steps, the system can maximize employee labor efficiency and optimize robot operation.

[1629] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1630] This invention provides a system that analyzes employee attendance data, PC usage data, emotional data, etc., predicts the busy and slow periods of work that vary for each employee and department, and suggests the best time to take paid leave, how to spend it when taken, and how to best use employee benefits. The program for this system is explained in detail below.

[1631] 1. Data Collection

[1632] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. Attendance data includes work start and end times, break times, etc., while PC usage data includes program execution logs and application usage history. Emotion data is extracted using an emotion engine from the user's facial expressions, voice tone, text input, etc.

[1633] Terminal: Employees enter their start and end times and break times, and PC usage data and emotional data (facial expression analysis, voice tone analysis, etc.) are collected in the background and sent to the server in real time.

[1634] 2. Data Preprocessing

[1635] Server: Cleans all collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[1636] 3. Data Analysis

[1637] Server: Based on the preprocessed data, it uses machine learning models to predict the busy and slow periods for each employee and department. It also analyzes emotional data and evaluates the user's mental state.

[1638] 4. Calculating the recommended timing for taking paid leave

[1639] Server: Based on the results of data analysis, calculates the optimal timing for each employee to take paid leave. This takes into account not only the department's overall schedule and past paid leave usage patterns, but also the employee's mental state based on emotional data.

[1640] 5. Suggestions for how to spend your paid leave

[1641] Server: Based on the user's hobby, preference, and emotional data, the server proposes the best way to spend their vacation time. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also takes into account the user's mental state and suggests ways to promote relaxation and refreshment.

[1642] 6. Assistance in optimal use of employee benefits

[1643] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on emotional data and user profile. For example, if there is a high need for relaxation, suggests spa facilities and counseling services.

[1644] 7. Visualization of business and vacation schedules by department

[1645] Server: Based on the analysis results, the busy / slow periods and paid leave usage status for each department are visualized, enabling the HR department to carry out effective labor management.

[1646] 8. User Notification and Interface

[1647] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[1648] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[1649] Specific examples

[1650] For example, suppose employee B plans to take paid leave soon. The system analyzes employee B's attendance data and PC usage data to predict that the following week will be a relatively quiet period. It also analyzes employee B's emotional data and recognizes that employee B's mental state has recently been one that makes him prone to stress. As a result, the system suggests the best time to take paid leave and notifies employee B of refreshing ways to spend the day that suit his hobbies and preferences (for example, suggestions for nature walks or relaxation spots). At the same time, it also recommends that employee B take advantage of employee benefits that have a relaxing effect (for example, vouchers for spa facilities). Employee B can use these suggestions to apply for paid leave and take advantage of employee benefits.

[1651] In this way, the system of the present invention maximizes employee work efficiency and supports employees in taking paid leave and using employee benefits while taking into account emotional data.

[1652] The processing flow will be explained below.

[1653] Step 1: Data collection

[1654] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. Specifically, the data is sent via API or a dedicated client app and stored in a central database.

[1655] Terminal: Employees enter their work start time, end time, and break time. In the background, PC usage data (program execution logs, browser history, etc.) and emotional data (facial expression analysis, voice tone analysis, etc.) are collected and sent to the server in real time.

[1656] Step 2: Data Preprocessing

[1657] Server: Cleans all collected data. If there are missing values, they are filled in using past data or average values, and outliers are corrected or removed. The data format is also standardized to make it suitable for analysis.

[1658] Specific behavior:

[1659] 1. Imputing missing data: Filling in missing data with past data or average values.

[1660] 2. Outlier Correction: Correcting or removing abnormal data based on pre-set criteria.

[1661] 3. Data format unification: Convert all data into a consistent format.

[1662] Step 3: Analyze the emotion data

[1663] Server: Analyzes collected emotional data and evaluates the user's mental state using an emotion engine, including facial expression recognition, voice tone analysis, and text analysis.

[1664] Specific behavior:

[1665] 1. Facial expression data analysis: Facial expression data collected from camera footage is analyzed to determine the emotional state.

[1666] 2. Voice data analysis: Analyze voice tones and detect emotional fluctuations.

[1667] 3. Analysis of text data: Emotional words are extracted from chat logs and email content to evaluate the emotional state.

[1668] Step 4: Data analysis

[1669] Server: Based on preprocessed attendance data, PC usage data, and emotion data, machine learning models are used to predict the busy / slow periods for each employee and department.

[1670] Specific behavior:

[1671] 1. Training a machine learning model: Train the model based on historical data.

[1672] 2. Predicting peak and trough periods: Use the trained model to predict future peak and trough periods.

[1673] 3. Setting confidence intervals: Calculate confidence intervals for the predicted results to ensure the reliability of the results.

[1674] Step 5: Calculate the best time to take paid leave

[1675] Server: Based on the results of data analysis and the evaluated emotional data, the server calculates the optimal timing for each employee to take paid leave, taking into account the department's overall schedule, past patterns of taking paid leave, and the employee's mental state based on emotional data.

[1676] Specific behavior:

[1677] 1. Check the busy / quiet periods: Check the busy / quiet periods of each employee and identify the quiet periods.

[1678] 2. Department-wide coordination: Consider the busy and slow times of the entire department and choose a time that will minimize the impact on other employees.

[1679] 3. Calculating the timing of taking paid leave: Calculate and recommend the optimal time to take leave.

[1680] Step 6: Suggestions for how to spend your paid leave

[1681] Server: Based on the user's hobby, preference, and emotional data, the server proposes the best way to spend their vacation time. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also takes into account the user's mental state and suggests ways to promote relaxation and refreshment.

[1682] Specific behavior:

[1683] 1. View user profiles: Check past activity and interests.

[1684] 2. Recommendation generation: Generate suggestions tailored to the user's preferences and take into account sentiment data.

[1685] 3. Save suggestions: Save suggestions for each user.

[1686] Step 7: Assistance in optimizing employee benefits

[1687] Server: Proposes employee benefit plans that can be used when taking paid leave. Recommends the optimal plan based on emotional data and user profile. For example, if there is a high need for relaxation, suggests spa facilities and counseling services.

[1688] Specific behavior:

[1689] 1. Browse the benefits database: See a list of available benefits.

[1690] 2. Matching with user profile: Selecting benefit plans that match the user profile.

[1691] 3. Proposal generation: Propose and record optimal ways to use benefits.

[1692] Step 8: Visualize business and vacation status by department

[1693] Server: Based on the analysis results, the busy / slow status of each department and the status of paid leave taken are visualized. This is displayed in the form of a dashboard or report.

[1694] Specific behavior:

[1695] 1. Data aggregation: Aggregate data by department.

[1696] 2. Applying visualization tools: Representing data visually using graphs and heat maps.

[1697] 3. Report generation: Generate dashboards and PDF reports and provide them to the HR department.

[1698] Step 9: User Notification and Interface

[1699] Server: Generated proposals and reports are sent to each user's device. Notifications are sent via email or a dedicated app.

[1700] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits.

[1701] Specific behavior:

[1702] 1. Notification Generation: Converting proposals and reports into notification format.

[1703] 2. Send notifications: Send notifications to users via email or app notifications.

[1704] 3. User interface operation: The user takes the necessary steps based on the notification received.

[1705] Example 2

[1706] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1707] In today's corporate environment, maximizing employee work efficiency and maintaining a healthy working environment are important. However, conventional systems only collect attendance data and PC usage data, and do not adequately consider employees' emotions and mental state. Furthermore, they are unable to provide appropriate suggestions regarding the timing of paid leave and how to spend it when taken, creating challenges in optimizing employee welfare benefits. This invention solves these issues by providing a system that suggests optimal work life for each employee.

[1708] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1709] In this invention, the server includes means for collecting attendance data, PC usage data, and emotion data from each employee's terminal, means for cleaning the collected data and for completing and correcting missing and outlier values, means for analyzing the preprocessed data and predicting business activity using a machine learning model, means for analyzing each employee's emotion data and evaluating their mental state, means for calculating the optimal timing for taking paid leave based on the business activity and emotion data, means for suggesting optimal ways to spend paid leave based on the employee's hobby and preference data and emotion data, means for suggesting employee benefit plans available when taking paid leave, means for visualizing business activity and paid leave usage status by department, and means for notifying each employee of the generated suggestions and reports. This maximizes employee work efficiency and enables the creation of a work environment that takes mental health into consideration.

[1710] "Work attendance data" is information related to an employee's working status, such as start time, finish time, and break time.

[1711] "PC usage data" refers to information such as program execution logs and application usage history for computers used by employees during work hours.

[1712] "Emotion data" is information about an employee's emotions and mental state extracted using an emotion engine from a user's facial expressions, tone of voice, text input, etc.

[1713] A "machine learning model" is a mathematical algorithm or statistical model that learns from collected data and predicts patterns and trends.

[1714] "Busy / slow status" refers to the busyness or slowness of business during a specific period.

[1715] "Hobbies and Preference Data" is information about a user's interests and preferences, including hobbies, interests, and favorite activities.

[1716] A "welfare system" refers to various services and support systems provided by companies with the aim of improving the welfare of their employees, and includes, for example, vouchers for use of spa facilities and counseling services.

[1717] "Cleanup" is the process of filling in missing values ​​and correcting outliers in order to improve the quality of data.

[1718] "Visualization" is the process of visually representing data and information, making them easier to understand using graphs, charts, etc.

[1719] A "recommender system" is a system that recommends optimal activities, travel destinations, etc. based on a user's hobby and preference data.

[1720] "Notifications" are a means of informing users of specific information or offers in real time, and are provided via email or a dedicated app.

[1721] The present invention provides a system that analyzes employee attendance data, PC usage data, emotional data, etc., predicts the busy and slow business conditions that vary for each employee and department, and suggests the appropriate timing for taking paid leave, the appropriate way to spend it when taken, and the optimal way to use employee benefits.

[1722] 1. Data Collection

[1723] server

[1724] The server periodically collects attendance data, PC usage data, and emotion data from each employee's device. Attendance data includes work start and end times and break times, while PC usage data includes program execution logs and application usage history. Emotion data is extracted from the user's facial expressions, voice tone, text input, etc. using an emotion engine.

[1725] Terminal

[1726] Employees enter their work start and end times, as well as break times, into the terminals, which collect PC usage data and emotional data (facial expression analysis, voice tone analysis, etc.) in the background and send it to the server in real time.

[1727] 2. Data Preprocessing

[1728] server

[1729] All collected data is cleaned. Missing values ​​are filled in using past data or average values, and outliers are corrected or removed. Data formats are also standardized to make them suitable for analysis.

[1730] 3. Data Analysis

[1731] server

[1732] Based on the preprocessed data, machine learning models are used to predict the busy and slow periods for each employee and department. Emotional data is also analyzed to evaluate the user's mental state. Specific software used includes the Python libraries TensorFlow and Scikit-learn.

[1733] 4. Calculating the recommended timing for taking paid leave

[1734] server

[1735] Based on the results of the data analysis, the system calculates the optimal timing for each employee to take paid leave, taking into account not only the department's overall schedule and past paid leave usage patterns, but also the employee's mental state based on emotional data.

[1736] 5. Suggestions for how to spend your paid leave

[1737] server

[1738] The system suggests optimal ways to spend paid vacation time based on the user's hobby, preference, and emotional data. It uses a recommender system to recommend activities and travel destinations suitable for each user. It also includes suggestions to promote relaxation and refreshment. Specific software used includes Apache Spark and Elasticsearch.

[1739] 6. Assistance in optimal use of employee benefits

[1740] server

[1741] It suggests employee benefits that can be used when taking paid leave, and suggests spa facilities and counseling services if employees need to relax based on their emotional data and profile information.

[1742] 7. Visualization of business and vacation schedules by department

[1743] server

[1744] Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized in graphs and charts, enabling the HR department to carry out effective labor management. For example, visualization can be done using Tableau or Power BI.

[1745] 8. User Notification and Interface

[1746] server

[1747] The generated proposals and reports are sent to each user's device, and notifications are sent via email or a dedicated app.

[1748] Terminal

[1749] The user checks the received notification and applies for paid leave or takes advantage of employee benefits.

[1750] Specific examples

[1751] For example, suppose employee B plans to take paid leave soon. The system analyzes employee B's attendance data and PC usage data to predict that the following week will be a relatively quiet period. It also analyzes employee B's emotional data and recognizes that employee B's mental state has recently been one that makes him prone to stress. As a result, the system suggests the best time to take paid leave and notifies employee B of refreshing ways to spend the day that suit his hobbies and preferences (for example, suggestions for nature walks or relaxation spots). At the same time, it also recommends that employee B take advantage of employee benefits that have a relaxing effect (for example, vouchers for spa facilities). Employee B can use these suggestions to apply for paid leave and take advantage of employee benefits.

[1752] Prompt Sentence Examples

[1753] Attendance data analysis prompt

[1754] Analyze employee attendance data to predict the next week's peak and trough times. The dataset includes start and finish times and break times.

[1755] Emotional Data Analysis Prompt

[1756] Analyze employee emotional data and assess their mental state. Data sets include facial expression analysis, voice tone, and text input.

[1757] Prompt to calculate the optimal timing for taking paid leave

[1758] Calculate the optimal timing for employees to take paid leave based on attendance and sentiment data.

[1759] Prompts for suggestions on how to spend your time based on your hobbies and preferences

[1760] Based on the user's hobby, preference, and emotional data, please suggest the best way to spend their vacation time. Please include suggestions that promote relaxation and refreshment.

[1761] Benefit Plan Recommendation Prompts

[1762] Based on emotional data and user profiles, suggest benefits that can be used when taking paid time off. If relaxation is needed, suggest spa facilities or counseling services.

[1763] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1764] Step 1: Data collection

[1765] Server: Periodically collects attendance data, PC usage data, and emotion data from each employee's device. The server sends data collection requests to each device at set intervals, receives the data sent from each device, and stores it in a database. The input is data from each employee's device, and the output is an integrated database.

[1766] Terminal: Employees enter their work start time, end time, and break time into a dedicated app. The terminal collects PC usage information (application usage history, usage time, etc.) and emotional data (facial expression analysis, voice tone analysis) in the background and sends them to the server in real time. The input is the employee's operations and background data collection, and the output is the data sent to the server.

[1767] Step 2: Data Preprocessing

[1768] Server: Cleans up all collected data. Specifically, if missing values ​​are found, they are filled in using past data or average values. Also, outliers are detected and corrected or removed. Data formats are standardized and made suitable for analysis. The input is raw data, and the output is cleaned data.

[1769] Step 3: Data analysis

[1770] Server: Using the preprocessed data, a machine learning model predicts the busy / quick schedule for each employee and department. Emotional data is also analyzed to evaluate each employee's mental state. Specific software used includes the Python libraries TensorFlow and Scikit-learn. The input is the cleaned-up data, and the output is predicted data on busy / quick schedules and the results of an evaluation of mental state.

[1771] Step 4: Calculate the best time to take paid leave

[1772] Server: Based on the analysis results, calculates the optimal timing for each employee to take paid leave. Calculates the optimal date to take paid leave by taking into account the department's overall schedule, past paid leave usage patterns, and mental state. The input is the analysis results, and the output is the recommended timing for taking paid leave.

[1773] Step 5: Suggestions for how to spend your paid leave

[1774] Server: Based on the user's hobby, preference, and emotional data, the server proposes the optimal way to spend their paid vacation. A recommender system is used to recommend activities and travel destinations suitable for each user. Specific software used includes Apache Spark and Elasticsearch. The input is hobby, preference, and emotional data, and the output is a proposal for the optimal way to spend the day.

[1775] Step 6: Assistance in optimizing employee benefits

[1776] Server: Proposes employee benefit programs that can be used when taking paid leave. Based on emotional data and profile information, recommends spa facilities or counseling services if there is a high need for relaxation. The input is emotional data and profile information, and the output is a proposed employee benefit program.

[1777] Step 7: Visualize business and leisure situations by department and paid leave usage status

[1778] Server: Based on the analysis results, the busy / slow times and paid leave usage status for each department are visualized in graphs and charts. This makes it easier for the HR department to carry out effective labor management. For example, visualization can be done using Tableau or Power BI. The input is the analysis results, and the output is visualized data.

[1779] Step 8: User Notification and Interface

[1780] Server: Sends the generated proposals and reports to each user's device. Notifications are sent via email or a dedicated app. The input is the generated proposals and reports, and the output is a notification message.

[1781] Terminal: The user checks the received notification and applies for paid leave or uses employee benefits. The input is the notification message, and the output is the user's action.

[1782] (Application example 2)

[1783] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1784] While the operation of robots in modern factories improves production efficiency, their maintenance and operational status management are also important. In particular, there is a need to efficiently collect and analyze robot operation data and maintenance history data to predict optimal maintenance times. Also important is the overall improvement of work efficiency and employee benefits, including the appropriate timing and use of employee paid leave and suggestions for utilizing employee benefit programs.

[1785] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting employee attendance data, means for collecting employee PC usage data, means for collecting factory robot operation data and maintenance history data, means for preprocessing the collected data, means for analyzing the preprocessed data to predict business activity, means for calculating recommended paid vacation time for each employee and optimal maintenance times for each robot, means for proposing paid vacation time and maintenance content based on employee preferences and robot operation status, means for proposing optimal employee benefit programs for paid vacation time and robot maintenance, means for visualizing business activity and robot operation status and paid vacation time for each department, and means for notifying each employee terminal and each engineer terminal of the generated proposals and reports. This enables optimal paid vacation time and employee benefit use for each employee and efficient robot maintenance scheduling.

[1786] "Employee attendance data" refers to work-related record information such as the start and end times of each employee's shift, and break times.

[1787] "PC usage data" refers to data such as program execution logs and application usage history for computers used by employees for work.

[1788] "Robot operation data" refers to information such as the operation records, operating hours, and error occurrence status of robots operating in a factory.

[1789] "Maintenance history data" is recorded information such as the content, date, and person who performed the maintenance performed on the robot in the past.

[1790] "Preprocessing" refers to the process of preparing collected data for analysis, such as filling in missing values, correcting outliers, and standardizing data formats.

[1791] "Business activity status" is data showing the busy and slow periods of business, and indicates the workload status of employees and departments.

[1792] "Recommended timing for taking paid leave" is the time when the optimal time to take paid leave is calculated and suggested for each employee.

[1793] Based on "hobbies and preferences" means customizing the proposals to suit the interests and preferences of each employee.

[1794] "How to spend your paid leave" is a content that suggests activities and ways for employees to rest during the period they take paid leave.

[1795] "Employee benefits system" refers to various services and support systems provided by a company that employees can use, and often includes relaxation facilities and counseling services.

[1796] "Visualization of busy and slow periods" refers to visually displaying the results of data analysis so that busy and slow periods can be understood at a glance.

[1797] The "generated proposals" are proposals for each employee and robot that are calculated and created based on data analysis.

[1798] "Business and quiet periods by department" is data showing the busy and quiet periods for work in each specific department.

[1799] "Each employee terminal" refers to an individual device (PC, smartphone, etc.) used by an employee.

[1800] "Technician's Terminal" refers to an individual device used by a technician in a factory or operational location.

[1801] A "machine learning model" is an algorithm used in data analysis, a system that learns patterns from the original data and makes predictions and classifications.

[1802] A "confidence interval" is a numerical range that indicates the reliability of a prediction result and provides a statistical indication of how accurate the prediction is.

[1803] This invention is a system that analyzes employee attendance data, PC usage data, factory robot operation data, and maintenance history data, and based on that, suggests optimal timing for taking paid leave and performing maintenance. This system is implemented using both a server and a terminal.

[1804] Data collection

[1805] The server collects employee attendance data, PC usage data, factory robot operation data, and maintenance history data. These data are obtained from each employee's terminal and the factory management system. Specifically, employee attendance data includes work start and end times and break times, while PC usage data includes program execution logs and application usage history. Robot operation data includes operating hours and error occurrence status, and maintenance history data includes the details and dates of past maintenance.

[1806] Data Preprocessing

[1807] The server preprocesses the collected data, which includes filling in missing values, correcting outliers, and standardizing data formats, making the data suitable for analysis.

[1808] Data analysis

[1809] The server uses a machine learning model based on the preprocessed data to predict the busy and slow times of operations and the optimal maintenance times for the robot. This analysis uses a machine learning algorithm (e.g., Random Forest Regressor). It also uses a generative AI model to generate recommendations based on the user's preferences and the robot's operating status.

[1810] Paid leave and maintenance scheduling

[1811] Based on the analysis results, the server calculates the recommended timing for each employee to take paid leave and the optimal maintenance period for each robot, including setting a confidence interval based on past data and current operating conditions.

[1812] Suggestions and Notifications

[1813] The server then sends the generated proposals and reports to each employee's and technician's device via email or a dedicated application. The employee can then review the proposals, apply for paid leave and employee benefits, and schedule robot maintenance.

[1814] Specific examples

[1815] For example, if factory robot A has recently been operating slowly, the system will suggest the optimal date and time for the next maintenance based on past maintenance history and operation data. The next maintenance for this robot is predicted to be required two weeks from now on August 15th, and the appropriate maintenance work will be notified to the maintenance team.

[1816] Prompt Sentence Examples

[1817] "Please create a model that predicts the optimal next maintenance timing for each robot based on the factory robot operation data and maintenance history data. Also, please tell me the steps to develop a system that notifies technicians of this in the form of a smartphone application."

[1818] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1819] Step 1: Data collection

[1820] The server collects data from each employee's terminal and the factory management system. Inputs include employee attendance data, PC usage data, robot operation data, and maintenance history data. Outputs store this data on the server and prepare it in a format that can be used for the next processing step. Specific operations include retrieving data from a database and streaming data in real time.

[1821] Step 2: Data Preprocessing

[1822] The server preprocesses the collected data. The input is the raw data collected in step 1. Specific operations include imputing missing values, correcting outliers, and standardizing data formats. The output is clean data that has been prepared for analysis.

[1823] Step 3: Data analysis

[1824] The server uses a machine learning model to analyze the preprocessed data. The input is clean data. Specifically, it uses machine learning algorithms such as RandomForestRegressor to predict the busy / unbusy schedule and the optimal maintenance timing for the robot. The output is the predicted busy / unbusy schedule and recommended maintenance timing.

[1825] Step 4: Generate proposals

[1826] The server generates proposals using a generative AI model based on the analysis results. The input is the prediction results from step 3. Specific actions include customizing how employees spend their paid leave and the maintenance details, taking into account the individual hobbies and preferences of employees and the operating status of the robot. The output is specific proposals provided to the user.

[1827] Step 5: Notification

[1828] The server notifies each employee's terminal and the engineer's terminal of the generated proposal and report. The input is the proposal generated in step 4. Specific actions include sending an email or sending a notification via a dedicated application. The output is the proposal displayed on the user's terminal, which the user can check and take the next action.

[1829] Step 6: User Actions

[1830] The user takes action based on the notified proposal. The input is the proposal received in step 5. Specific actions include applying for paid leave, applying for employee benefits, and scheduling robot maintenance. The output is a record of the application and scheduling as actions taken by the user.

[1831] The specific processing unit ...

Claims

1. A means of collecting employee attendance data; A means of collecting employee PC usage data; means for pre-processing the collected data; A means for analyzing the preprocessed data and predicting business activity; A way to calculate the recommended timing for taking paid leave for each employee, A way to suggest ways for employees to spend their paid leave based on their hobbies and preferences, A means to propose the optimal employee benefit system when taking paid leave, A means to visualize the busy / slow times and paid leave usage status of each department, A means to notify each employee of the generated proposals and reports, A system including:

2. 2. The system according to claim 1, further comprising means for collecting attendance data and PC usage data from each employee terminal.

3. The system according to claim 1, further comprising a means for predicting business busy / slow states using a machine learning model based on the analysis results and setting a confidence interval.

Images