System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting human flow; means for pre-processing the human flow; means for learning a generative AI for predicting demand using the pre-processed human flow; means for generating a new proposed route using the learned generative AI; means for performing a simulation based on the generated new proposed route; and means for visualizing and outputting a simulation result.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] Even in urban areas, efficient transportation methods to destinations are often unavailable, forcing people to choose circuitous routes. Such inefficient transportation infrastructure increases inconvenience for users, increases travel times, and causes traffic congestion. Furthermore, when planning and building new routes, it is difficult to realistically forecast demand and assess profitability, resulting in wasteful investments. [Means for solving the problem]

[0005] To solve the above-mentioned problems, the present invention provides the following means. Specifically, the present invention provides a system including means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, and means for visualizing and outputting the simulation results. This makes it possible to generate new, efficient public transportation routes that are in high demand and propose profitable operation plans, thereby improving user convenience and easing traffic congestion.

[0006] "People flow data" refers to data that includes information on the movements of people in specific regions or areas, and is a general term for smartphone location data, public transportation usage data, and commercial facility entry data.

[0007] "Means of collection" refers to the means by which people flow data is obtained from external data sources and imported into the system. Specifically, APIs are generally used.

[0008] "Preprocessing means" refers to the means of processing collected people flow data, such as removing missing values ​​and outliers, and organizing and formatting the data.

[0009] "Generative AI" is an artificial intelligence model for demand forecasting using people flow data, and includes deep learning models and clustering algorithms.

[0010] The "means of learning" refers to the means of inputting collected and preprocessed people flow data into the generation AI and training the AI ​​model based on that data.

[0011] The "means for generating new route proposals" is a means for using trained generative AI to conceive new public transportation route plans based on demand between specific regions.

[0012] The "means for performing simulation" is a means for simulating actual operation conditions, number of users, sales forecasts, peak congestion times, etc. based on the generated new route plan.

[0013] The "means for visualizing and outputting" is a means for displaying the simulation results in a text report or graph format, and outputting the results in a form that can be easily understood by the user. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] This invention relates to a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes.

[0036] Overall system overview

[0037] 1. Data Collection and Preprocessing

[0038] The server collects various people flow data via API, such as smartphone location information in the target area, public transportation usage data, and entrance data for commercial facilities.

[0039] The server then preprocesses this data to remove missing values ​​and outliers, which is essential for accurate model training and prediction.

[0040] 2. Learning with AI models

[0041] The server inputs preprocessed people flow data into an AI model, learning the relationship between origins and destinations, movement patterns, and usage trends by time of day.

[0042] Learning is carried out using deep learning models and clustering algorithms, and evaluation and adjustment using validation data is also carried out in parallel to improve the model's performance.

[0043] 3. Demand forecasting and analysis

[0044] The server uses a trained AI model to predict demand for each target area and time period.

[0045] Specifically, the system identifies combinations of departure and destination points that are expected to have a large number of users, and analyzes detailed data such as the number of users, peak congestion times, and travel distances.

[0046] 4. New Route Generation and Simulation

[0047] The server generates new transport route proposals based on the results of the demand forecast, taking profitability into account and calculating revenue forecasts based on the number of users and operation costs.

[0048] The generated route plan is then verified through simulation, which provides detailed forecasts of operation schedules, peak times, and daily passenger numbers, enabling the operation plan to be concretely implemented.

[0049] 5. Visualizing and outputting results

[0050] Finally, the server visualizes the simulation results as graphs and text reports, outputting them to the user's terminal in a format that can be easily understood by transportation planners.

[0051] The user (transportation planner) can consider introducing new routes based on the received proposal results.

[0052] Specific examples

[0053] Consider the case of conducting a detailed analysis of people's movement patterns based on smartphone location information from each area of ​​Tokyo.

[0054] 1. Data Collection

[0055] The server collects smartphone location data and public transport usage data for one year, including data broken down by time of day for each day.

[0056] 2. Pretreatment

[0057] The server removes missing values ​​and extreme outliers from the collected data and formats the data by time period and day of the week.

[0058] 3. Training the AI ​​model

[0059] The server performs a detailed analysis of travel data, for example, from Shinjuku to Shibuya, and inputs this data into an AI model for learning.

[0060] The model is evaluated using validation data and appropriate parameters are set.

[0061] 4. Demand forecasting and analysis

[0062] The server predicts the number of users and peak congestion times based on the time periods and days of the week with the highest demand between Shinjuku and Shibuya.

[0063] 5. New Route Generation and Simulation

[0064] The server generates a direct route from Shinjuku to Shibuya and simulates the number of passengers, peak congestion, and sales forecasts for this route.

[0065] Based on the simulation results, the operation schedule and economic efficiency are evaluated.

[0066] 6. Visualizing and outputting results

[0067] The server creates a proposal result report based on the simulation results and outputs it to the user terminal of the transportation planner.

[0068] Users can use this information to consider introducing new direct routes.

[0069] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves convenience for users, and makes it possible to provide highly profitable public transportation facilities.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] The server collects people flow data. Specifically, it obtains data such as smartphone location information, public transportation usage data, and commercial facility entrance data through various APIs. The data is saved in standard formats (CSV, JSON, etc.).

[0073] Step 2:

[0074] The server preprocesses the collected data, removing missing and outlier values ​​and standardizing the format. It also divides the data by time of day and day of the week and prepares it in a form suitable for analysis.

[0075] Step 3:

[0076] The server inputs the preprocessed people flow data into the AI ​​model, using deep learning models and clustering algorithms, and splits the data into training and validation sections.

[0077] Step 4:

[0078] The server trains the AI ​​model, analyzing the relationship between departure and destination, travel patterns, and usage trends by time of day. During the training process, the model's parameters are adjusted to improve accuracy.

[0079] Step 5:

[0080] The server uses a trained AI model to forecast demand. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, and other factors based on that.

[0081] Step 6:

[0082] The server generates new route proposals based on the predicted demand, and also calculates sales forecasts based on the number of passengers and operating costs to evaluate profitability.

[0083] Step 7:

[0084] The server then simulates the new route plan, predicting factors such as operation schedules, daily passenger numbers, and peak congestion times, and verifies the validity of the operation plan.

[0085] Step 8:

[0086] The server visualizes the simulation results, displaying them in the form of graphs and text reports, making them easy for users to understand.

[0087] Step 9:

[0088] The server outputs visualized simulation results to the user's device, allowing transportation planners to use this information to consider introducing new routes.

[0089] For example, the server collects people flow data within Tokyo and generates and simulates a direct route from Shinjuku to Shibuya based on this data. Based on the simulation results, transportation planners consider the profitability and operation schedule of the new direct route.

[0090] Example 1

[0091] 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."

[0092] In conventional public transportation route planning, it is difficult to properly grasp user movement patterns and fluctuations in demand, making it difficult to design efficient and profitable routes. Another problem is the lack of a means to accurately grasp predicted demand and peak congestion times, and to effectively propose and simulate new routes.

[0093] 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.

[0094] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, means for visualizing and outputting the simulation results, means for analyzing operation costs and sales forecasts based on the simulation results, and means for evaluating the economic efficiency of the generated route plans.This enables transportation planners to plan and consider new public transportation routes that are efficient and profitable in detail.

[0095] "People flow data" is data that shows the movement patterns and behavior of people in a specific area or period of time.

[0096] "Generative AI" is an artificial intelligence model that uses machine learning algorithms to perform a specific task (in this case, demand forecasting and generating new route proposals).

[0097] "Preprocessing" refers to the process of removing missing values ​​and outliers from collected data and preparing the data in a format suitable for analysis and learning.

[0098] "Simulation" is the process of using models to predict and evaluate outcomes or reactions in specific hypothetical environments or scenarios.

[0099] "Visualization" is a technique for expressing data and simulation results in visual formats such as graphs and charts to make the information easier to understand.

[0100] "Operating costs" refers to the various costs incurred when operating a public transportation route (e.g., fuel costs, labor costs, maintenance costs, etc.).

[0101] "Sales forecasting" is the process of calculating the expected future revenues based on the services or products offered.

[0102] "Economic efficiency" is an indicator that evaluates the degree of return on investment or effectiveness relative to costs, and aims to utilize resources efficiently.

[0103] This invention is a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes. This system operates in cooperation with the elements of a server, terminals, and users. The specific hardware and software used in each step, as well as the data processing method, are described in detail below.

[0104] Data collection and preprocessing

[0105] First, the server collects people flow data from multiple data sources via API. This data includes smartphone location information, public transportation usage data, and commercial facility entry data. Specifically, it periodically sends API requests using Python libraries (such as requests) to retrieve the data.

[0106] The server then preprocesses the collected data, using data processing libraries such as Pandas and NumPy to detect and remove missing and outlier values ​​and convert the data into a format that is easier to analyze. For example, if missing values ​​are found, they are imputed with the median or mean.

[0107] Learning with AI models

[0108] The server trains a generative AI model using preprocessed people flow data. The AI ​​model uses deep learning (TensorFlow or PyTorch) and clustering algorithms to learn the relationship between departure and destination, travel patterns, and usage trends by time of day. The dataset is split into training data and test data, and cross-validation is performed to evaluate and optimize the model's performance.

[0109] Demand forecasting and analysis

[0110] The server uses a trained generative AI model to forecast demand. Specifically, it predicts travel demand for specific regions and time periods, and analyzes detailed data for each combination of origin and destination. The results of this analysis are visualized in the form of heat maps, bar graphs, and other formats. For example, Matplotlib and Seaborn are used to create graphs showing high and low demand.

[0111] New route generation and simulation

[0112] The server generates new transport route proposals based on the results of the demand forecast. At the same time, it also takes profitability into account and calculates sales forecasts and operating costs based on the expected number of users. The generated route proposals are then verified in detail using simulation software (such as AnyLogic or Simul8). The simulations produce detailed predictions of operation schedules, peak congestion times, and daily user numbers, and the operation plan is then finalized.

[0113] Visualizing and outputting results

[0114] Finally, the server visualizes the simulation results as graphs and text reports and outputs them to the user's terminal. Libraries such as Matplotlib and ReportLab are used to visualize the results. For example, a line graph showing peak congestion times or a bar graph showing sales forecasts can be generated, presenting the results in a format that transportation planners can easily understand.

[0115] Specific examples

[0116] Consider a case in which people's travel patterns are analyzed in detail based on smartphone location information from various areas of Tokyo. The server collects smartphone location data and public transportation usage data for one year. The collected data is then reorganized by time of day and day of the week, removing missing values ​​and extreme outliers. The data on travel from Shinjuku to Shibuya is then analyzed in detail and input into an AI model for training. Based on the time of day and day of the week with the highest demand between Shinjuku and Shibuya, the number of passengers and peak congestion times are predicted, and a direct route from Shinjuku to Shibuya is generated and simulated. Finally, a proposal report based on the simulation results is output to the user device of the transportation planner, who can then consider introducing new direct routes.

[0117] Prompt Sentence Examples

[0118] "For a direct line from Shinjuku to Shibuya, what days of the week and times of day would be most popular? Also, estimate the expected operating costs and revenues for that line."

[0119] This invention enables transportation planners to plan and consider efficient and profitable public transportation routes in detail, overcoming traditional challenges.

[0120] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0121] Step 1:

[0122] Data collection

[0123] The server collects people flow data, such as smartphone location information, public transportation usage data, and commercial facility entry data, via API. Specifically, the server uses a Python library (such as requests) to send API requests and obtain data. Input data includes location information, transportation usage history, and facility entry and exit records. The output data is a raw dataset that compiles this information.

[0124] Step 2:

[0125] Data Preprocessing

[0126] The server preprocesses the collected data. Specifically, it uses Pandas and NumPy to detect missing values ​​and outliers in the data and remove or correct them. For example, if a missing value is found, it is filled in with the average value of that column. The input data is the original dataset, and the output data is the preprocessed, clean dataset.

[0127] Step 3:

[0128] Training an AI model

[0129] The server trains a generative AI model using preprocessed people flow data. It uses deep learning models (such as TensorFlow or PyTorch) and clustering algorithms to learn the relationship between origins and destinations, travel patterns, and usage trends by time of day. Specifically, it splits the dataset into training data and test data, and fits the model using the training data. The input data is the preprocessed dataset, and the output data is the trained model.

[0130] Step 4:

[0131] Demand forecasting

[0132] The server uses a trained generative AI model to perform demand forecasting. It predicts demand for specific regions and time periods and analyzes detailed data for each combination of origin and destination. Specifically, it inputs new data into the model and obtains the forecast results. For example, it supplies data to forecast travel demand between Shinjuku and Shibuya. The input data is a new dataset for the model, and the output data is the forecast results.

[0133] Step 5:

[0134] Creating a new route

[0135] The server generates new transport route proposals based on the results of the demand forecast. It designs new route proposals taking into account factors such as the number of users, peak congestion times, and operating costs. Specifically, it runs an algorithm to identify combinations of departure and destination points with high demand and proposes the optimal route. The input data is the demand forecast results, and the output data is the new route proposal.

[0136] Step 6:

[0137] simulation

[0138] The server simulates the generated route plan, forecasting in detail operation schedules, peak congestion times, and daily passenger numbers, and verifying the operation plan. Specifically, it uses simulation software (such as AnyLogic or Simul8) to reproduce the operation of the route plan in a virtual environment. The input data is the new route plan, and the output data is the simulation results.

[0139] Step 7:

[0140] Visualizing and outputting results

[0141] The server visualizes the simulation results as graphs and text reports. Using libraries such as Matplotlib and ReportLab, the results are visually represented and output in an easy-to-understand format. Specific operations include creating line graphs showing peak congestion times and bar graphs showing sales forecasts. The input data are the simulation results, and the output data is the visualized report.

[0142] Through these steps, the system suggests efficient and profitable new public transport routes for detailed consideration by transport planners.

[0143] (Application example 1)

[0144] 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."

[0145] Conventional public transportation route design relies on static analysis based on past data, making it difficult to forecast demand and derive new routes using real-time people flow data. Furthermore, it is difficult to perform simulations to ensure the realistic effectiveness and efficiency of generated new route proposals, making it difficult to make quick and accurate decisions based on such simulations. Furthermore, visualization of information is inefficient, and there is a lack of means to provide information in a format that is easy for people in charge to understand. Therefore, there is a need for a flexible and efficient system for generating new routes that can respond to dynamically changing people flow patterns in real time.

[0146] 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.

[0147] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generative AI model that performs demand forecasting using the preprocessed people flow data, means for visualizing demand data for display on smart glasses, means for generating new route plans using the trained generative AI model, means for performing simulations based on the generated new route plans, and means for visualizing and outputting the simulation results. This enables the collection and analysis of people flow data in real time, enabling the design of efficient and economical new routes based on the results. Furthermore, by using smart glasses or other display devices, traffic managers can receive information in an intuitively understandable format, supporting rapid decision-making.

[0148] "People flow data" refers to information that indicates human movement patterns and behavior, and includes location information, entry data, and transportation usage data.

[0149] "Preprocessing" is the process of removing missing values ​​and outliers from collected data and formatting it into a form suitable for analysis and learning.

[0150] A "generative AI model" is an artificial intelligence algorithm that generates demand forecasts and new route proposals based on people flow data, and often uses deep learning models and clustering algorithms.

[0151] "Smart glasses" are display devices used to visually present information and can display data in real time using augmented reality (AR) technology.

[0152] "Demand forecasting" refers to predicting future transportation demand based on people flow data, and includes analyzing the relationship between departure and destination points and usage trends by time of day.

[0153] A "New Route Proposal" is a new public transportation route plan proposed based on predicted travel demand using a generative AI model.

[0154] "Simulation" refers to simulating the operation of a new route plan in a virtual environment in order to verify its operational efficiency and economic viability.

[0155] "Visualization" refers to visually displaying the results of data analysis and simulations in a form that is easily understandable to traffic managers.

[0156] This invention relates to a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes. This system can be used with smart glasses, allowing transportation managers to receive information in an intuitive and easy-to-understand format, enabling them to make quick decisions.

[0157] System Configuration

[0158] 1. Hardware and software usage:

[0159] Hardware: smart glasses, servers, smartphones

[0160] Software: TensorFlow (AI model), Flask (API), Python, Keras

[0161] Data collection and preprocessing

[0162] The server collects location information from smartphones and other sensors in real time via an API, then preprocesses the collected data to remove missing values ​​and outliers and prepare it in a format suitable for training by a deep learning model.

[0163] Learning with AI models

[0164] The server trains a generative AI model based on the preprocessed people flow data. During this process, it uses deep learning models and clustering algorithms to learn the relationship between departure and destination, movement patterns, and usage trends by time of day. This makes it possible to predict people flow based on time of day and events.

[0165] Demand forecasting and route generation

[0166] The trained generative AI model is used to predict demand and generate new route proposals based on the most popular combinations of departure and destination points in a specific area. The server then performs simulations in a virtual environment to verify the operational efficiency and economic viability of the new route proposals. The simulations forecast ridership, sales, and peak congestion times in detail, and evaluate optimal operation schedules and economic efficiency.

[0167] Visualizing and outputting results

[0168] The server creates a report of proposed results based on the simulation results. This report is displayed in real time on the smart glasses and presented in an intuitive format that is easy for users to understand, enabling traffic managers to make quick and accurate decisions.

[0169] Specific examples

[0170] For example, the system collects real-time data on people flow around Shinjuku Station and predicts the optimal route from Shinjuku to Shibuya. The smart glasses' AR display allows users to visually understand the current traffic situation and the next route they should take.

[0171] Example prompt sentence:

[0172] Generate the optimal route from Shinjuku Station to Shibuya Station based on the following data: Data format: { "timestamp": "2023-10-04T09:30:00Z", "location": {"latitude":35.6895, "longitude":139.6917}, "destination": {"latitude":35.6580, "longitude":139.7016}, "people_count": 200}

[0173] As a result, the present invention makes it possible to improve the efficiency of new route planning for transportation facilities, improve convenience for users, and provide highly profitable public transportation facilities.

[0174] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0175] Step 1:

[0176] The server collects location information from smartphones and other sensors in real time via API. As input, it receives smartphone location data and public transport usage data, including time, latitude, longitude, and number of users. As output, it compiles these data into a single dataset.

[0177] Step 2:

[0178] The server preprocesses the collected data and removes missing values ​​and outliers. As input, it takes in the raw people flow data collected in step 1. It then performs data imputation, outlier removal, normalization, and other processes to format the data in a way that is suitable for training AI models. The output is a clean, preprocessed dataset.

[0179] Step 3:

[0180] The server trains a generative AI model based on the preprocessed people flow data. The preprocessed data obtained in step 2 is used as input. Using deep learning models and clustering algorithms, the server learns the relationship between departure and destination points, travel patterns, and usage trends by time of day. The output is a trained generative AI model.

[0181] Step 4:

[0182] The server uses the trained generative AI model to predict demand. It provides real-time and historical people flow data as input. The generative AI model then uses this data to predict demand and identify popular combinations of origins and destinations in a specific area. The output is predicted demand data.

[0183] Step 5:

[0184] The server generates new route proposals based on the results of the demand forecast. The demand data predicted in step 4 is used as input. The generative AI model generates optimal transportation routes and lists the candidates. The output is multiple route candidates proposed as new route proposals.

[0185] Step 6:

[0186] The server performs a simulation based on the new route plan it has generated. The new route plan generated in step 5 is used as input. The simulation predicts the number of passengers, sales, peak congestion times, etc., and verifies the operational efficiency and economy of the route. The output is a dataset of the simulation results.

[0187] Step 7:

[0188] The server visualizes and outputs the simulation results. It takes the simulation result data obtained in step 6 as input and visualizes it in the form of graphs and text reports. The generated visualization data is sent to smart glasses or other display devices so that the user can view it in real time. The output is a report containing the visualized information.

[0189] 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.

[0190] The present invention relates to a system that utilizes people flow data to generate new, efficient and profitable public transportation routes and makes proposals that also take into account user emotions.

[0191] Overall system overview

[0192] 1. Data Collection and Preprocessing

[0193] The server collects various people flow data via API, such as smartphone location information in the target area, public transportation usage data, and entrance data for commercial facilities.

[0194] The server then preprocesses the data, removing missing and outlier values, and splits it by time of day and day of the week to prepare it for analysis.

[0195] 2. Learning with AI models

[0196] The server inputs preprocessed people flow data into an AI model, which uses deep learning models and clustering algorithms to learn the relationship between departure and destination points, movement patterns, and usage trends by time of day.

[0197] During learning, the model parameters are adjusted to improve accuracy.

[0198] 3. Demand forecasting and analysis

[0199] The server uses a trained AI model to forecast demand. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, and other factors based on that.

[0200] 4. New Route Generation and Simulation

[0201] The server generates new transport route proposals based on the results of the demand forecast, and also calculates sales forecasts based on the number of users and operation costs to evaluate profitability.

[0202] The generated route plan is then verified through simulation, which involves making detailed predictions of operation schedules, daily passenger numbers, and peak congestion times to verify the validity of the operation plan.

[0203] 5. Sentiment analysis using an emotion engine

[0204] The server uses an emotion engine to analyze users' emotions based on collected pedestrian flow data and simulation results, quantitatively evaluating their emotional state and grasping their satisfaction or dissatisfaction with the proposed new route.

[0205] 6. Adjusting route plans based on emotional responses

[0206] The server simulates the user's emotional response to the generated new route plan and modifies it based on the results, for example, by improving areas that users dislike or adding new options.

[0207] 7. Visualizing and outputting results

[0208] Finally, the server visualizes the simulation results and the user sentiment analysis results as graphs and text reports, and outputs them to the user's terminal in a format that can be easily understood by transportation planners.

[0209] Users (transportation planners) can use this information to consider introducing new routes.

[0210] Specific examples

[0211] Consider a case where people's movement patterns and emotional responses are analyzed in detail based on smartphone location information from various areas of Tokyo.

[0212] 1. Data Collection

[0213] The server collects smartphone location data and public transport usage data for one year, including data broken down by time of day for each day.

[0214] 2. Pretreatment

[0215] The server removes missing values ​​and extreme outliers and splits the data by time of day and day of the week.

[0216] 3. Training the AI ​​model

[0217] The server performs a detailed analysis of the travel data from Shinjuku to Shibuya, inputs it into the AI ​​model, and the model is evaluated using validation data to set appropriate parameters.

[0218] 4. Demand forecasting and analysis

[0219] The server predicts the number of users and peak congestion times based on the time periods and days of the week with the highest demand between Shinjuku and Shibuya.

[0220] 5. New Route Generation and Simulation

[0221] The server generates a direct route from Shinjuku to Shibuya, and simulates the number of passengers, peak congestion, and sales forecasts for this route. Based on the simulation results, the operation schedule and economic efficiency are evaluated.

[0222] 6. Sentiment analysis

[0223] The server uses an emotion engine to analyze user emotions based on the collected data and simulation results, and evaluates reactions to new lines. For example, it visualizes users' expectations and concerns about a direct line from Shinjuku to Shibuya.

[0224] 7. Adjustment and Output

[0225] The server adjusts the proposed new routes based on the results of the sentiment analysis and generates a report with the final simulation results, which transportation planners can use to consider introducing new routes.

[0226] As described above, the present invention makes it possible to improve the efficiency of new transportation route planning, improve user convenience and satisfaction, and provide highly profitable public transportation services.

[0227] The processing flow will be explained below.

[0228] Step 1:

[0229] The server collects people flow data. Specifically, it obtains smartphone location information, public transportation usage data, and commercial facility entry data via API and stores it in a database in standard formats (CSV or JSON). The data collection period is generally one year, and includes information on weekdays, weekends, and time periods.

[0230] Step 2:

[0231] The server preprocesses the collected data. Specifically, it removes missing values ​​and outliers and standardizes the data format. It also divides the data by time period and day of the week and prepares it in a form suitable for analysis. This process uses data cleaning tools and programs.

[0232] Step 3:

[0233] The server inputs the preprocessed people flow data into an AI model, such as a deep learning model or a clustering algorithm. The data is split into training and validation sets, and the training data set is used to train the model.

[0234] Step 4:

[0235] The server trains the AI ​​model. Specifically, it uses a data set to analyze the relationship between origins and destinations, travel patterns, and usage trends by time of day. During the training process, the model's parameters are adjusted to improve accuracy.

[0236] Step 5:

[0237] The server uses the trained AI model to make demand forecasts. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, etc. based on those combinations. This allows it to grasp the demand distribution in a specific area.

[0238] Step 6:

[0239] The server generates new route proposals based on predicted demand. At the same time, it also calculates sales forecasts based on the number of passengers and operating costs, and evaluates profitability. Specifically, if demand between Shinjuku and Shibuya is high, it proposes a direct route.

[0240] Step 7:

[0241] The server then simulates the new route plan. The simulation makes detailed predictions of the operation schedule, daily passenger numbers, and peak congestion times, and verifies the validity of the operation plan. For example, it creates an operation schedule for a direct line between Shinjuku and Shibuya, and estimates the number of passengers and operation costs based on that schedule.

[0242] Step 8:

[0243] The server uses an emotion engine to analyze users' emotional states based on the collected pedestrian flow data and simulation results. Specifically, it analyzes user comments, social media data, survey results, etc., to quantitatively evaluate users' satisfaction or dissatisfaction with new route proposals.

[0244] Step 9:

[0245] The server adjusts new route proposals based on the results of sentiment analysis. Specifically, it improves areas where users are dissatisfied and considers additional options. For example, it proposes a service schedule that avoids peak hours, which users are likely to dissatisfy.

[0246] Step 10:

[0247] The server visualizes the simulation results and sentiment analysis results in the form of graphs and text reports, making them easy for transportation planners (users) to understand.

[0248] Step 11:

[0249] The server outputs visualized simulation results and adjusted route plans to the user's device. Based on this, transportation planners (users) can consider introducing new routes. For example, they can evaluate the profitability and operation schedule of a direct route between Shinjuku and Shibuya and make a final decision on its introduction.

[0250] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves user convenience and satisfaction, and enables the provision of highly profitable public transportation services.

[0251] Example 2

[0252] 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."

[0253] In conventional public transport route planning, demand forecasting and route proposal generation are often done manually, resulting in inefficiency and a lack of accuracy. Furthermore, it is not easy to adjust routes while taking into account user emotions and satisfaction, which can lead to dissatisfaction after the introduction of new routes. Therefore, there is a need for efficient and accurate proposals for new public transport routes and route adjustments that take into account user emotions.

[0254] 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.

[0255] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that uses the preprocessed people flow data to learn relationships between departure points and destinations, travel patterns, and usage trends by time period, means for performing demand forecasts using the trained generation AI and identifying high-demand combinations of departure points and destinations, means for generating new route plans based on the demand forecast results, means for simulating the generated new route plans to predict operation schedules, ridership, peak congestion times, and sales, means for analyzing users' emotional states using an emotion engine and evaluating their satisfaction or dissatisfaction with the new route plans, means for revising the route plans based on users' emotional reactions to the generated new route plans, and means for visualizing and outputting the simulation results and emotion analysis results. This enables the proposal of efficient and profitable new public transportation routes and route adjustments that take users' emotions into consideration.

[0256] "People flow data" refers to data on people's movements, including individual movement patterns and location information, public transportation usage data, and commercial facility entry data.

[0257] "Preprocessing" refers to the process of removing missing or outliers from collected people flow data and dividing the data by time of day or day of the week, among other things, to prepare the data in a form suitable for analysis.

[0258] "Generative AI" is artificial intelligence that uses machine learning or deep learning to learn from data and perform demand forecasts and generate new routes.

[0259] "Demand forecasting" involves using trained generative AI to predict combinations that will be in high demand based on the relationship between people's departure and destination locations and their movement patterns.

[0260] A "new route proposal" is a proposal for a new public transportation route based on demand forecasts.

[0261] "Simulation" is the process of virtually calculating operation schedules, number of users, peak congestion times, sales forecasts, etc. based on the generated new route proposal.

[0262] The "emotion engine" is an artificial intelligence that analyzes the user's emotional state from collected pedestrian flow data and simulation results, and evaluates their satisfaction or dissatisfaction with new route proposals.

[0263] "Emotional response" refers to the emotional state, such as expectations or concerns, that a user shows toward a new route proposal.

[0264] "Visualization" is a method of displaying simulation results and sentiment analysis results as graphs or text reports to communicate them to people in an easy-to-understand manner.

[0265] The present invention relates to a system that utilizes people flow data to generate new, efficient and profitable public transportation routes and makes proposals that take user emotions into consideration. An embodiment of this system is described in detail below.

[0266] Overall system overview

[0267] This system consists of the following main hardware and software:

[0268] Server: Executes functions such as people flow data collection, preprocessing, AI model training, demand forecasting, new route plan generation, simulation, sentiment analysis, visualization and output.

[0269] User terminal: A terminal that receives and displays the simulation results and sentiment analysis results sent from the server.

[0270] Network: Infrastructure for data communication between servers and user devices.

[0271] Data collection and preprocessing

[0272] The server collects various people flow data via API, such as smartphone location information for the target area, public transportation usage data, and entrance data for commercial facilities. At this stage, the specific software used is a library for handling HTTP requests (e.g., requests for Python).

[0273] The server then preprocesses the collected data to remove missing and outlier values, using Python's Pandas library to manipulate data frames and NumPy for numerical data processing.

[0274] Learning with AI models

[0275] The server inputs the preprocessed people flow data into an AI model, which learns the relationship between origins and destinations, travel patterns, and usage trends by time of day. The specific software used is a deep learning framework such as TensorFlow or PyTorch.

[0276] During training, the model parameters are adjusted to find optimal hyperparameters (learning rate, number of epochs, batch size, etc.).

[0277] Demand forecasting and analysis

[0278] The server uses a trained AI model to forecast demand, particularly by identifying combinations of origins and destinations with high demand, and then predicting factors such as the number of passengers, peak congestion times, and travel distances.

[0279] This analysis uses graph drawing libraries such as Matplotlib and Seaborn to visually display the prediction results.

[0280] New route generation and simulation

[0281] The server generates new transportation route proposals based on the demand forecast results, which include information such as departure points, destinations, departure times, and arrival times.

[0282] The server then performs further simulations to generate detailed data such as operation schedules, daily ridership, peak congestion times, sales forecasts, etc. Simulation software such as AnyLogic can be used for this simulation.

[0283] Sentiment analysis with emotion engine

[0284] The server uses an emotion engine to analyze user emotions based on collected pedestrian flow data and simulation results. Specifically, it calculates emotion scores using natural language processing (NLP) libraries (e.g., SpaCy and BERT).

[0285] Adjusting route plans based on emotional responses

[0286] The server simulates the user's reaction to the generated new route plan based on the results of the emotion engine, and modifies the route plan based on the results, thereby providing a new route plan that will improve user satisfaction.

[0287] Visualizing and outputting results

[0288] The server visualizes the simulation results and sentiment analysis results as graphs and text reports, and sends them to the user's device, allowing the user to consider introducing new routes based on the results.

[0289] The specific tools used are Plotly and Matplotlib for visualization.

[0290] Specific examples

[0291] As a case study of Tokyo, we use one year's worth of smartphone location data and public transportation usage data. The server collects and preprocesses this data, then performs a detailed analysis of the travel data from Shinjuku to Shibuya and trains the AI ​​model. Demand forecasting identifies the time periods and days of the week with the highest demand between Shinjuku and Shibuya, and predicts the number of users and peak congestion times. The server then generates a direct route from Shinjuku to Shibuya and simulates the number of users, peak congestion, and sales forecasts for this route. Finally, an emotion engine is used to analyze user emotions, visualize expectations and concerns about the new route, and provide optimal route suggestions.

[0292] Examples of prompt statements

[0293] For example, the prompt we use for our generative AI model is:

[0294] "Using one year of smartphone location data and public transportation usage data in Tokyo, please predict travel patterns and demand from Shinjuku to Shibuya. Based on the results, please generate direct route plans from Shinjuku to Shibuya and analyze user responses using an emotion engine."

[0295] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves user convenience and satisfaction, and enables the provision of highly profitable public transportation services.

[0296] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0297] Step 1: Data collection

[0298] The server collects people flow data, such as smartphone location information, public transportation usage data, and commercial facility entrance data, for the target area via API. Specifically, the server sends an HTTP request to the API and retrieves data in JSON format. The input is location information and usage data from the API, and the output is raw data stored on the server.

[0299] Step 2: Data Preprocessing

[0300] The server preprocesses the collected data. Specifically, it uses Python's Pandas library to create a data frame and remove missing values ​​and outliers. Rows with missing values ​​are dropped or imputed with the median. Next, the data is split by time of day and day of the week, and processes such as standardization and normalization are performed. The input is the raw data from step 1, and the output is the preprocessed data.

[0301] Step 3: Training the AI ​​model

[0302] The server trains an AI model using the preprocessed data. The specific software used is TensorFlow or PyTorch. The data is divided into training data and validation data, a deep learning model is defined, and training begins with the model.fit() function. Hyperparameters (learning rate, number of epochs, batch size, etc.) are adjusted, and performance is evaluated after training using model.evaluate(). The input is the preprocessed data, and the output is the trained model.

[0303] Step 4: Demand forecasting and analysis

[0304] The server uses the trained AI model to perform demand forecasting. It executes the forecast using the model.predict() function to identify combinations of origins and destinations with high demand. Matplotlib and Seaborn are used to visually display the forecast results as heat maps and time series graphs. The input is the trained model and preprocessed data, and the output is the demand forecast results.

[0305] Step 5: Create a new route

[0306] The server generates new transport route proposals based on the demand forecast results. Route proposals are created that include departure points, destinations, departure times, arrival times, etc. Routes with high demand are selected from the forecast results, and an optimal route proposal is created based on these. The input is the demand forecast results, and the output is a new route proposal.

[0307] Step 6: Simulation

[0308] The server runs a simulation based on the generated new route plan. Specifically, it uses simulation software such as AnyLogic to calculate operation schedules, daily passenger numbers, peak congestion times, and sales forecasts. The simulation results are exported as Excel or CSV files. The input is the new route plan, and the output is the simulation results.

[0309] Step 7: Sentiment analysis

[0310] The server analyzes the user's emotions using an emotion engine based on the collected data and simulation results. Specifically, it uses NLP libraries such as SpaCy and BERT to calculate the emotion score using get_sentiment_score(). The input is the simulation result data, and the output is the emotion analysis result.

[0311] Step 8: Adjust the route plan

[0312] The server adjusts the new route proposal based on the results of the sentiment analysis. Specifically, it considers user satisfaction and dissatisfaction from the sentiment score and improves the operation schedule and routes. It then performs a simulation again to find the optimal route proposal. The input is the sentiment analysis results and the original route proposal, and the output is the adjusted new route proposal.

[0313] Step 9: Visualizing and outputting results

[0314] The server visualizes the simulation results and sentiment analysis results as graphs and text reports. It uses Matplotlib and Plotly to create visually easy-to-understand graphs and charts. The generated report is sent to the user's device. The input is the adjusted new route plan and the simulation results, and the output is a visualized report.

[0315] These are the specific processing steps of the program of this system. This series of steps makes it possible to propose new, efficient and profitable public transportation routes and adjust routes while taking user emotions into consideration.

[0316] (Application example 2)

[0317] 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."

[0318] Existing public transport routes face difficulties in efficiently responding to demand and optimizing operating costs, resulting in low user satisfaction. Furthermore, route planning does not take into account user emotions and reactions, which often leads to user dissatisfaction when new routes are introduced. There is a need to solve these issues and plan and operate public transport routes that are efficient and provide high user satisfaction.

[0319] The identification processing by the identification 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 people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, means for evaluating reactions to the new route plans using an emotion engine that analyzes user emotions, means for adjusting the new route plans based on the emotion analysis results, and means for visualizing and outputting the simulation and emotion analysis results. This enables efficient route planning based on demand and optimal operation plans that take user emotions and reactions into consideration.

[0320] (Definitions of important words)

[0321] "People flow data" is data that represents people's movements and includes a variety of data sources, such as location information, public transportation usage data, and entrance data for commercial facilities.

[0322] "Preprocessing" refers to the process performed to prepare raw data in an analyzable format, such as removing missing values ​​and outliers, and standardizing and splitting the data.

[0323] "Generative AI" is a system that uses artificial intelligence technology and has the learning capabilities to predict demand and recognize patterns from large amounts of data.

[0324] A "new route proposal" is a newly proposed public transportation route that differs from existing transportation routes.

[0325] "Simulation" is the act of virtually predicting and verifying various factors in actual operation (number of users, sales, peak congestion times, etc.) based on the generated new route plan.

[0326] An "emotion engine" is a technology for analyzing a user's emotions and reactions, and quantitatively evaluates the emotional state based on collected data and simulation results.

[0327] "Emotion analysis" involves using an emotion engine to analyze users' emotions and reactions, and is carried out to understand their satisfaction or dissatisfaction with new route proposals.

[0328] "Visualization" is the act of visually displaying analysis or simulation results as graphs, charts, or text reports, providing information in an easy-to-understand format.

[0329] To implement the present invention, the following configuration and procedures are used.

[0330] First, the server collects people flow data. Specifically, it collects location information obtained from smartphones, public transportation usage data, and commercial facility entry data via API. Standard API access methods are used to collect data, with appropriate authentication and access control for each data source.

[0331] The collected data is preprocessed on the server using the Pandas library to remove missing values ​​and outliers and extract only the necessary information, preparing the data in a form suitable for analysis.

[0332] Next, the preprocessed data is input into the generative AI, which trains it to make demand forecasts. This training uses the Scikit-learn library, which performs clustering and regression analysis. By learning from large amounts of data, the generative AI model is able to understand the relationship between departure and destination points and usage trends by time of day, allowing it to make more accurate predictions.

[0333] The trained generative AI model generates new route proposals by applying a clustering algorithm using KMeans to identify routes between points with high demand, while also calculating operating costs and predicted revenue.

[0334] The generated new route plan is then simulated. The simulation predicts the number of passengers, sales, peak congestion times, etc., and aims to optimize operation. The simulation results are visualized as graphs using the Matplotlib library.

[0335] Furthermore, an emotion engine is used to analyze user emotions. The emotion engine analyzes user emotions based on pre-collected data and simulation results, and evaluates their satisfaction or dissatisfaction with the new route plan. The emotion analysis results are then used to further adjust the route plan.

[0336] Finally, all data and analysis results are visualized and provided to transportation planners in a report detailing proposed new routes, simulation results, and user sentiment analysis, enabling planners to consider new routes that are efficient and offer high user satisfaction.

[0337] As a concrete example, "The following prompt sentence is input into the generative AI model: 'Last night, it's early morning. How can we improve convenience? How can we change the main travel route? Based on the traffic volume forecast results, please suggest a route for a new autonomous vehicle.'" Through such prompt sentences, specific questions are posed to the generative AI model.

[0338] As described above, the present invention makes it possible to provide a system that generates efficient and profitable public transportation routes by combining demand prediction based on people flow data with user emotion analysis.

[0339] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0340] Step 1:

[0341] The server collects people flow data. Smartphone location information, public transportation usage data, and commercial facility entrance data are obtained via API. Various data sources are used as input, and the collected data is stored on the server.

[0342] Step 2:

[0343] The data collected by the server is preprocessed. Using the Pandas library, missing values ​​and outliers are removed, and the data is split and reshaped by time of day and day of the week. The input is the people flow data collected in Step 1, and the output is preprocessed data in a format suitable for analysis.

[0344] Step 3:

[0345] The server uses the preprocessed data to train the generative AI. It uses the Scikit-learn library to perform clustering and regression analysis to learn the relationship between departure and destination locations and travel patterns. The input is the preprocessed data, and the output is a trained generative AI model.

[0346] Step 4:

[0347] The server generates new route proposals using a trained generative AI model. A clustering algorithm using KMeans is used to identify routes between points with high demand, and calculates operating costs and predicted sales. The input is the trained generative AI model and analysis data, and the output is new route proposals.

[0348] Step 5:

[0349] The server performs a simulation based on the generated new route plan. The simulation predicts the number of passengers, sales, peak congestion times, etc., and aims to optimize operations. The input is the new route plan, and the output is the simulation results.

[0350] Step 6:

[0351] The server uses an emotion engine to analyze the user's emotions. Based on the collected data and simulation results, the user's emotions are quantitatively evaluated. The input is the collected data and simulation results, and the output is the emotion analysis results.

[0352] Step 7:

[0353] The server adjusts the new route proposal based on the results of the sentiment analysis, correcting areas of user dissatisfaction and adding new options. The input is the sentiment analysis results, and the output is the adjusted new route proposal.

[0354] Step 8:

[0355] The server visualizes the simulation and sentiment analysis results and provides them to transportation planners as a report. Collected data, analysis results, and new route proposals are displayed in graphs, charts, and text. The input is the adjusted new route proposal and simulation results, and the output is a visualized report.

[0356] Step 9:

[0357] The user (transportation planner) considers the introduction of new routes based on the visualized report. This makes it possible to create new routes that are efficient and have high user satisfaction. The only input required is the visualized report.

[0358] 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.

[0359] 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.

[0360] 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.

[0361] [Second embodiment]

[0362] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0363] 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.

[0364] 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).

[0365] 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.

[0366] 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.

[0367] 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).

[0368] 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.

[0369] 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.

[0370] 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.

[0371] 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.

[0372] 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.

[0373] 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."

[0374] This invention relates to a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes.

[0375] Overall system overview

[0376] 1. Data Collection and Preprocessing

[0377] The server collects various people flow data via API, such as smartphone location information in the target area, public transportation usage data, and entrance data for commercial facilities.

[0378] The server then preprocesses this data to remove missing values ​​and outliers, which is essential for accurate model training and prediction.

[0379] 2. Learning with AI models

[0380] The server inputs preprocessed people flow data into an AI model, learning the relationship between origins and destinations, movement patterns, and usage trends by time of day.

[0381] Learning is carried out using deep learning models and clustering algorithms, and evaluation and adjustment using validation data is also carried out in parallel to improve the model's performance.

[0382] 3. Demand forecasting and analysis

[0383] The server uses a trained AI model to predict demand for each target area and time period.

[0384] Specifically, the system identifies combinations of departure and destination points that are expected to have a large number of users, and analyzes detailed data such as the number of users, peak congestion times, and travel distances.

[0385] 4. New Route Generation and Simulation

[0386] The server generates new transport route proposals based on the results of the demand forecast, taking profitability into account and calculating revenue forecasts based on the number of users and operation costs.

[0387] The generated route plan is then verified through simulation, which provides detailed forecasts of operation schedules, peak times, and daily passenger numbers, enabling the operation plan to be concretely implemented.

[0388] 5. Visualizing and outputting results

[0389] Finally, the server visualizes the simulation results as graphs and text reports, outputting them to the user's terminal in a format that can be easily understood by transportation planners.

[0390] The user (transportation planner) can consider introducing new routes based on the received proposal results.

[0391] Specific examples

[0392] Consider the case of conducting a detailed analysis of people's movement patterns based on smartphone location information from each area of ​​Tokyo.

[0393] 1. Data Collection

[0394] The server collects smartphone location data and public transport usage data for one year, including data broken down by time of day for each day.

[0395] 2. Pretreatment

[0396] The server removes missing values ​​and extreme outliers from the collected data and formats the data by time period and day of the week.

[0397] 3. Training the AI ​​model

[0398] The server performs a detailed analysis of travel data, for example, from Shinjuku to Shibuya, and inputs this data into an AI model for learning.

[0399] The model is evaluated using validation data and appropriate parameters are set.

[0400] 4. Demand forecasting and analysis

[0401] The server predicts the number of users and peak congestion times based on the time periods and days of the week with the highest demand between Shinjuku and Shibuya.

[0402] 5. New Route Generation and Simulation

[0403] The server generates a direct route from Shinjuku to Shibuya and simulates the number of passengers, peak congestion, and sales forecasts for this route.

[0404] Based on the simulation results, the operation schedule and economic efficiency are evaluated.

[0405] 6. Visualizing and outputting results

[0406] The server creates a proposal result report based on the simulation results and outputs it to the user terminal of the transportation planner.

[0407] Users can use this information to consider introducing new direct routes.

[0408] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves convenience for users, and makes it possible to provide highly profitable public transportation facilities.

[0409] The processing flow will be explained below.

[0410] Step 1:

[0411] The server collects people flow data. Specifically, it obtains data such as smartphone location information, public transportation usage data, and commercial facility entrance data through various APIs. The data is saved in standard formats (CSV, JSON, etc.).

[0412] Step 2:

[0413] The server preprocesses the collected data, removing missing and outlier values ​​and standardizing the format. It also divides the data by time of day and day of the week and prepares it in a form suitable for analysis.

[0414] Step 3:

[0415] The server inputs the preprocessed people flow data into the AI ​​model, using deep learning models and clustering algorithms, and splits the data into training and validation sections.

[0416] Step 4:

[0417] The server trains the AI ​​model, analyzing the relationship between departure and destination, travel patterns, and usage trends by time of day. During the training process, the model's parameters are adjusted to improve accuracy.

[0418] Step 5:

[0419] The server uses a trained AI model to forecast demand. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, and other factors based on that.

[0420] Step 6:

[0421] The server generates new route proposals based on the predicted demand, and also calculates sales forecasts based on the number of passengers and operating costs to evaluate profitability.

[0422] Step 7:

[0423] The server then simulates the new route plan, predicting factors such as operation schedules, daily passenger numbers, and peak congestion times, and verifies the validity of the operation plan.

[0424] Step 8:

[0425] The server visualizes the simulation results, displaying them in the form of graphs and text reports, making them easy for users to understand.

[0426] Step 9:

[0427] The server outputs visualized simulation results to the user's device, allowing transportation planners to use this information to consider introducing new routes.

[0428] For example, the server collects people flow data within Tokyo and generates and simulates a direct route from Shinjuku to Shibuya based on this data. Based on the simulation results, transportation planners consider the profitability and operation schedule of the new direct route.

[0429] Example 1

[0430] 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."

[0431] In conventional public transportation route planning, it is difficult to properly grasp user movement patterns and fluctuations in demand, making it difficult to design efficient and profitable routes. Another problem is the lack of a means to accurately grasp predicted demand and peak congestion times, and to effectively propose and simulate new routes.

[0432] 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.

[0433] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, means for visualizing and outputting the simulation results, means for analyzing operation costs and sales forecasts based on the simulation results, and means for evaluating the economic efficiency of the generated route plans.This enables transportation planners to plan and consider new public transportation routes that are efficient and profitable in detail.

[0434] "People flow data" is data that shows the movement patterns and behavior of people in a specific area or period of time.

[0435] "Generative AI" is an artificial intelligence model that uses machine learning algorithms to perform a specific task (in this case, demand forecasting and generating new route proposals).

[0436] "Preprocessing" refers to the process of removing missing values ​​and outliers from collected data and preparing the data in a format suitable for analysis and learning.

[0437] "Simulation" is the process of using models to predict and evaluate outcomes or reactions in specific hypothetical environments or scenarios.

[0438] "Visualization" is a technique for expressing data and simulation results in visual formats such as graphs and charts to make the information easier to understand.

[0439] "Operating costs" refers to the various costs incurred when operating a public transportation route (e.g., fuel costs, labor costs, maintenance costs, etc.).

[0440] "Sales forecasting" is the process of calculating the expected future revenues based on the services or products offered.

[0441] "Economic efficiency" is an indicator that evaluates the degree of return on investment or effectiveness relative to costs, and aims to utilize resources efficiently.

[0442] This invention is a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes. This system operates in cooperation with the elements of a server, terminals, and users. The specific hardware and software used in each step, as well as the data processing method, are described in detail below.

[0443] Data collection and preprocessing

[0444] First, the server collects people flow data from multiple data sources via API. This data includes smartphone location information, public transportation usage data, and commercial facility entry data. Specifically, it periodically sends API requests using Python libraries (such as requests) to retrieve the data.

[0445] The server then preprocesses the collected data, using data processing libraries such as Pandas and NumPy to detect and remove missing and outlier values ​​and convert the data into a format that is easier to analyze. For example, if missing values ​​are found, they are imputed with the median or mean.

[0446] Learning with AI models

[0447] The server trains a generative AI model using preprocessed people flow data. The AI ​​model uses deep learning (TensorFlow or PyTorch) and clustering algorithms to learn the relationship between departure and destination, travel patterns, and usage trends by time of day. The dataset is split into training data and test data, and cross-validation is performed to evaluate and optimize the model's performance.

[0448] Demand forecasting and analysis

[0449] The server uses a trained generative AI model to forecast demand. Specifically, it predicts travel demand for specific regions and time periods, and analyzes detailed data for each combination of origin and destination. The results of this analysis are visualized in the form of heat maps, bar graphs, and other formats. For example, Matplotlib and Seaborn are used to create graphs showing high and low demand.

[0450] New route generation and simulation

[0451] The server generates new transport route proposals based on the results of the demand forecast. At the same time, it also takes profitability into account and calculates sales forecasts and operating costs based on the expected number of users. The generated route proposals are then verified in detail using simulation software (such as AnyLogic or Simul8). The simulations produce detailed predictions of operation schedules, peak congestion times, and daily user numbers, and the operation plan is then finalized.

[0452] Visualizing and outputting results

[0453] Finally, the server visualizes the simulation results as graphs and text reports and outputs them to the user's terminal. Libraries such as Matplotlib and ReportLab are used to visualize the results. For example, a line graph showing peak congestion times or a bar graph showing sales forecasts can be generated, presenting the results in a format that transportation planners can easily understand.

[0454] Specific examples

[0455] Consider a case in which people's travel patterns are analyzed in detail based on smartphone location information from various areas of Tokyo. The server collects smartphone location data and public transportation usage data for one year. The collected data is then reorganized by time of day and day of the week, removing missing values ​​and extreme outliers. The data on travel from Shinjuku to Shibuya is then analyzed in detail and input into an AI model for training. Based on the time of day and day of the week with the highest demand between Shinjuku and Shibuya, the number of passengers and peak congestion times are predicted, and a direct route from Shinjuku to Shibuya is generated and simulated. Finally, a proposal report based on the simulation results is output to the user device of the transportation planner, who can then consider introducing new direct routes.

[0456] Prompt Sentence Examples

[0457] "For a direct line from Shinjuku to Shibuya, what days of the week and times of day would be most popular? Also, estimate the expected operating costs and revenues for that line."

[0458] This invention enables transportation planners to plan and consider efficient and profitable public transportation routes in detail, overcoming traditional challenges.

[0459] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0460] Step 1:

[0461] Data collection

[0462] The server collects people flow data, such as smartphone location information, public transportation usage data, and commercial facility entry data, via API. Specifically, the server uses a Python library (such as requests) to send API requests and obtain data. Input data includes location information, transportation usage history, and facility entry and exit records. The output data is a raw dataset that compiles this information.

[0463] Step 2:

[0464] Data Preprocessing

[0465] The server preprocesses the collected data. Specifically, it uses Pandas and NumPy to detect missing values ​​and outliers in the data and remove or correct them. For example, if a missing value is found, it is filled in with the average value of that column. The input data is the original dataset, and the output data is the preprocessed, clean dataset.

[0466] Step 3:

[0467] Training an AI model

[0468] The server trains a generative AI model using preprocessed people flow data. It uses deep learning models (such as TensorFlow or PyTorch) and clustering algorithms to learn the relationship between origins and destinations, travel patterns, and usage trends by time of day. Specifically, it splits the dataset into training data and test data, and fits the model using the training data. The input data is the preprocessed dataset, and the output data is the trained model.

[0469] Step 4:

[0470] Demand forecasting

[0471] The server uses a trained generative AI model to perform demand forecasting. It predicts demand for specific regions and time periods and analyzes detailed data for each combination of origin and destination. Specifically, it inputs new data into the model and obtains the forecast results. For example, it supplies data to forecast travel demand between Shinjuku and Shibuya. The input data is a new dataset for the model, and the output data is the forecast results.

[0472] Step 5:

[0473] Creating a new route

[0474] The server generates new transport route proposals based on the results of the demand forecast. It designs new route proposals taking into account factors such as the number of users, peak congestion times, and operating costs. Specifically, it runs an algorithm to identify combinations of departure and destination points with high demand and proposes the optimal route. The input data is the demand forecast results, and the output data is the new route proposal.

[0475] Step 6:

[0476] simulation

[0477] The server simulates the generated route plan, forecasting in detail operation schedules, peak congestion times, and daily passenger numbers, and verifying the operation plan. Specifically, it uses simulation software (such as AnyLogic or Simul8) to reproduce the operation of the route plan in a virtual environment. The input data is the new route plan, and the output data is the simulation results.

[0478] Step 7:

[0479] Visualizing and outputting results

[0480] The server visualizes the simulation results as graphs and text reports. Using libraries such as Matplotlib and ReportLab, the results are visually represented and output in an easy-to-understand format. Specific operations include creating line graphs showing peak congestion times and bar graphs showing sales forecasts. The input data are the simulation results, and the output data is the visualized report.

[0481] Through these steps, the system suggests efficient and profitable new public transport routes for detailed consideration by transport planners.

[0482] (Application example 1)

[0483] 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."

[0484] Conventional public transportation route design relies on static analysis based on past data, making it difficult to forecast demand and derive new routes using real-time people flow data. Furthermore, it is difficult to perform simulations to ensure the realistic effectiveness and efficiency of generated new route proposals, making it difficult to make quick and accurate decisions based on such simulations. Furthermore, visualization of information is inefficient, and there is a lack of means to provide information in a format that is easy for people in charge to understand. Therefore, there is a need for a flexible and efficient system for generating new routes that can respond to dynamically changing people flow patterns in real time.

[0485] 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.

[0486] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generative AI model that performs demand forecasting using the preprocessed people flow data, means for visualizing demand data for display on smart glasses, means for generating new route plans using the trained generative AI model, means for performing simulations based on the generated new route plans, and means for visualizing and outputting the simulation results. This enables the collection and analysis of people flow data in real time, enabling the design of efficient and economical new routes based on the results. Furthermore, by using smart glasses or other display devices, traffic managers can receive information in an intuitively understandable format, supporting rapid decision-making.

[0487] "People flow data" refers to information that indicates human movement patterns and behavior, and includes location information, entry data, and transportation usage data.

[0488] "Preprocessing" is the process of removing missing values ​​and outliers from collected data and formatting it into a form suitable for analysis and learning.

[0489] A "generative AI model" is an artificial intelligence algorithm that generates demand forecasts and new route proposals based on people flow data, and often uses deep learning models and clustering algorithms.

[0490] "Smart glasses" are display devices used to visually present information and can display data in real time using augmented reality (AR) technology.

[0491] "Demand forecasting" refers to predicting future transportation demand based on people flow data, and includes analyzing the relationship between departure and destination points and usage trends by time of day.

[0492] A "New Route Proposal" is a new public transportation route plan proposed based on predicted travel demand using a generative AI model.

[0493] "Simulation" refers to simulating the operation of a new route plan in a virtual environment in order to verify its operational efficiency and economic viability.

[0494] "Visualization" refers to visually displaying the results of data analysis and simulations in a form that is easily understandable to traffic managers.

[0495] This invention relates to a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes. This system can be used with smart glasses, allowing transportation managers to receive information in an intuitive and easy-to-understand format, enabling them to make quick decisions.

[0496] System Configuration

[0497] 1. Hardware and software usage:

[0498] Hardware: smart glasses, servers, smartphones

[0499] Software: TensorFlow (AI model), Flask (API), Python, Keras

[0500] Data collection and preprocessing

[0501] The server collects location information from smartphones and other sensors in real time via an API, then preprocesses the collected data to remove missing values ​​and outliers and prepare it in a format suitable for training by a deep learning model.

[0502] Learning with AI models

[0503] The server trains a generative AI model based on the preprocessed people flow data. During this process, it uses deep learning models and clustering algorithms to learn the relationship between departure and destination, movement patterns, and usage trends by time of day. This makes it possible to predict people flow based on time of day and events.

[0504] Demand forecasting and route generation

[0505] The trained generative AI model is used to predict demand and generate new route proposals based on the most popular combinations of departure and destination points in a specific area. The server then performs simulations in a virtual environment to verify the operational efficiency and economic viability of the new route proposals. The simulations forecast ridership, sales, and peak congestion times in detail, and evaluate optimal operation schedules and economic efficiency.

[0506] Visualizing and outputting results

[0507] The server creates a report of proposed results based on the simulation results. This report is displayed in real time on the smart glasses and presented in an intuitive format that is easy for users to understand, enabling traffic managers to make quick and accurate decisions.

[0508] Specific examples

[0509] For example, the system collects real-time data on people flow around Shinjuku Station and predicts the optimal route from Shinjuku to Shibuya. The smart glasses' AR display allows users to visually understand the current traffic situation and the next route they should take.

[0510] Example prompt sentence:

[0511] Generate the optimal route from Shinjuku Station to Shibuya Station based on the following data: Data format: { "timestamp": "2023-10-04T09:30:00Z", "location": {"latitude":35.6895, "longitude":139.6917}, "destination": {"latitude":35.6580, "longitude":139.7016}, "people_count": 200}

[0512] As a result, the present invention makes it possible to improve the efficiency of new route planning for transportation facilities, improve convenience for users, and provide highly profitable public transportation facilities.

[0513] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0514] Step 1:

[0515] The server collects location information from smartphones and other sensors in real time via API. As input, it receives smartphone location data and public transport usage data, including time, latitude, longitude, and number of users. As output, it compiles these data into a single dataset.

[0516] Step 2:

[0517] The server preprocesses the collected data and removes missing values ​​and outliers. As input, it takes in the raw people flow data collected in step 1. It then performs data imputation, outlier removal, normalization, and other processes to format the data in a way that is suitable for training AI models. The output is a clean, preprocessed dataset.

[0518] Step 3:

[0519] The server trains a generative AI model based on the preprocessed people flow data. The preprocessed data obtained in step 2 is used as input. Using deep learning models and clustering algorithms, the server learns the relationship between departure and destination points, travel patterns, and usage trends by time of day. The output is a trained generative AI model.

[0520] Step 4:

[0521] The server uses the trained generative AI model to predict demand. It provides real-time and historical people flow data as input. The generative AI model then uses this data to predict demand and identify popular combinations of origins and destinations in a specific area. The output is predicted demand data.

[0522] Step 5:

[0523] The server generates new route proposals based on the results of the demand forecast. The demand data predicted in step 4 is used as input. The generative AI model generates optimal transportation routes and lists the candidates. The output is multiple route candidates proposed as new route proposals.

[0524] Step 6:

[0525] The server performs a simulation based on the new route plan it has generated. The new route plan generated in step 5 is used as input. The simulation predicts the number of passengers, sales, peak congestion times, etc., and verifies the operational efficiency and economy of the route. The output is a dataset of the simulation results.

[0526] Step 7:

[0527] The server visualizes and outputs the simulation results. It takes the simulation result data obtained in step 6 as input and visualizes it in the form of graphs and text reports. The generated visualization data is sent to smart glasses or other display devices so that the user can view it in real time. The output is a report containing the visualized information.

[0528] 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.

[0529] The present invention relates to a system that utilizes people flow data to generate new, efficient and profitable public transportation routes and makes proposals that also take into account user emotions.

[0530] Overall system overview

[0531] 1. Data Collection and Preprocessing

[0532] The server collects various people flow data via API, such as smartphone location information in the target area, public transportation usage data, and entrance data for commercial facilities.

[0533] The server then preprocesses the data, removing missing and outlier values, and splits it by time of day and day of the week to prepare it for analysis.

[0534] 2. Learning with AI models

[0535] The server inputs preprocessed people flow data into an AI model, which uses deep learning models and clustering algorithms to learn the relationship between departure and destination points, movement patterns, and usage trends by time of day.

[0536] During learning, the model parameters are adjusted to improve accuracy.

[0537] 3. Demand forecasting and analysis

[0538] The server uses a trained AI model to forecast demand. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, and other factors based on that.

[0539] 4. New Route Generation and Simulation

[0540] The server generates new transport route proposals based on the results of the demand forecast, and also calculates sales forecasts based on the number of users and operation costs to evaluate profitability.

[0541] The generated route plan is then verified through simulation, which involves making detailed predictions of operation schedules, daily passenger numbers, and peak congestion times to verify the validity of the operation plan.

[0542] 5. Sentiment analysis using an emotion engine

[0543] The server uses an emotion engine to analyze users' emotions based on collected pedestrian flow data and simulation results, quantitatively evaluating their emotional state and grasping their satisfaction or dissatisfaction with the proposed new route.

[0544] 6. Adjusting route plans based on emotional responses

[0545] The server simulates the user's emotional response to the generated new route plan and modifies it based on the results, for example, by improving areas that users dislike or adding new options.

[0546] 7. Visualizing and outputting results

[0547] Finally, the server visualizes the simulation results and the user sentiment analysis results as graphs and text reports, and outputs them to the user's terminal in a format that can be easily understood by transportation planners.

[0548] Users (transportation planners) can use this information to consider introducing new routes.

[0549] Specific examples

[0550] Consider a case where people's movement patterns and emotional responses are analyzed in detail based on smartphone location information from various areas of Tokyo.

[0551] 1. Data Collection

[0552] The server collects smartphone location data and public transport usage data for one year, including data broken down by time of day for each day.

[0553] 2. Pretreatment

[0554] The server removes missing values ​​and extreme outliers and splits the data by time of day and day of the week.

[0555] 3. Training the AI ​​model

[0556] The server performs a detailed analysis of the travel data from Shinjuku to Shibuya, inputs it into the AI ​​model, and the model is evaluated using validation data to set appropriate parameters.

[0557] 4. Demand forecasting and analysis

[0558] The server predicts the number of users and peak congestion times based on the time periods and days of the week with the highest demand between Shinjuku and Shibuya.

[0559] 5. New Route Generation and Simulation

[0560] The server generates a direct route from Shinjuku to Shibuya, and simulates the number of passengers, peak congestion, and sales forecasts for this route. Based on the simulation results, the operation schedule and economic efficiency are evaluated.

[0561] 6. Sentiment analysis

[0562] The server uses an emotion engine to analyze user emotions based on the collected data and simulation results, and evaluates reactions to new lines. For example, it visualizes users' expectations and concerns about a direct line from Shinjuku to Shibuya.

[0563] 7. Adjustment and Output

[0564] The server adjusts the proposed new routes based on the results of the sentiment analysis and generates a report with the final simulation results, which transportation planners can use to consider introducing new routes.

[0565] As described above, the present invention makes it possible to improve the efficiency of new transportation route planning, improve user convenience and satisfaction, and provide highly profitable public transportation services.

[0566] The processing flow will be explained below.

[0567] Step 1:

[0568] The server collects people flow data. Specifically, it obtains smartphone location information, public transportation usage data, and commercial facility entry data via API and stores it in a database in standard formats (CSV or JSON). The data collection period is generally one year, and includes information on weekdays, weekends, and time periods.

[0569] Step 2:

[0570] The server preprocesses the collected data. Specifically, it removes missing values ​​and outliers and standardizes the data format. It also divides the data by time period and day of the week and prepares it in a form suitable for analysis. This process uses data cleaning tools and programs.

[0571] Step 3:

[0572] The server inputs the preprocessed people flow data into an AI model, such as a deep learning model or a clustering algorithm. The data is split into training and validation sets, and the training data set is used to train the model.

[0573] Step 4:

[0574] The server trains the AI ​​model. Specifically, it uses a data set to analyze the relationship between origins and destinations, travel patterns, and usage trends by time of day. During the training process, the model's parameters are adjusted to improve accuracy.

[0575] Step 5:

[0576] The server uses the trained AI model to make demand forecasts. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, etc. based on those combinations. This allows it to grasp the demand distribution in a specific area.

[0577] Step 6:

[0578] The server generates new route proposals based on predicted demand. At the same time, it also calculates sales forecasts based on the number of passengers and operating costs, and evaluates profitability. Specifically, if demand between Shinjuku and Shibuya is high, it proposes a direct route.

[0579] Step 7:

[0580] The server then simulates the new route plan. The simulation makes detailed predictions of the operation schedule, daily passenger numbers, and peak congestion times, and verifies the validity of the operation plan. For example, it creates an operation schedule for a direct line between Shinjuku and Shibuya, and estimates the number of passengers and operation costs based on that schedule.

[0581] Step 8:

[0582] The server uses an emotion engine to analyze users' emotional states based on the collected pedestrian flow data and simulation results. Specifically, it analyzes user comments, social media data, survey results, etc., to quantitatively evaluate users' satisfaction or dissatisfaction with new route proposals.

[0583] Step 9:

[0584] The server adjusts new route proposals based on the results of sentiment analysis. Specifically, it improves areas where users are dissatisfied and considers additional options. For example, it proposes a service schedule that avoids peak hours, which users are likely to dissatisfy.

[0585] Step 10:

[0586] The server visualizes the simulation results and sentiment analysis results in the form of graphs and text reports, making them easy for transportation planners (users) to understand.

[0587] Step 11:

[0588] The server outputs visualized simulation results and adjusted route plans to the user's device. Based on this, transportation planners (users) can consider introducing new routes. For example, they can evaluate the profitability and operation schedule of a direct route between Shinjuku and Shibuya and make a final decision on its introduction.

[0589] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves user convenience and satisfaction, and enables the provision of highly profitable public transportation services.

[0590] Example 2

[0591] 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."

[0592] In conventional public transport route planning, demand forecasting and route proposal generation are often done manually, resulting in inefficiency and a lack of accuracy. Furthermore, it is not easy to adjust routes while taking into account user emotions and satisfaction, which can lead to dissatisfaction after the introduction of new routes. Therefore, there is a need for efficient and accurate proposals for new public transport routes and route adjustments that take into account user emotions.

[0593] 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.

[0594] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that uses the preprocessed people flow data to learn relationships between departure points and destinations, travel patterns, and usage trends by time period, means for performing demand forecasts using the trained generation AI and identifying high-demand combinations of departure points and destinations, means for generating new route plans based on the demand forecast results, means for simulating the generated new route plans to predict operation schedules, ridership, peak congestion times, and sales, means for analyzing users' emotional states using an emotion engine and evaluating their satisfaction or dissatisfaction with the new route plans, means for revising the route plans based on users' emotional reactions to the generated new route plans, and means for visualizing and outputting the simulation results and emotion analysis results. This enables the proposal of efficient and profitable new public transportation routes and route adjustments that take users' emotions into consideration.

[0595] "People flow data" refers to data on people's movements, including individual movement patterns and location information, public transportation usage data, and commercial facility entry data.

[0596] "Preprocessing" refers to the process of removing missing or outliers from collected people flow data and dividing the data by time of day or day of the week, among other things, to prepare the data in a form suitable for analysis.

[0597] "Generative AI" is artificial intelligence that uses machine learning or deep learning to learn from data and perform demand forecasts and generate new routes.

[0598] "Demand forecasting" involves using trained generative AI to predict combinations that will be in high demand based on the relationship between people's departure and destination locations and their movement patterns.

[0599] A "new route proposal" is a proposal for a new public transportation route based on demand forecasts.

[0600] "Simulation" is the process of virtually calculating operation schedules, number of users, peak congestion times, sales forecasts, etc. based on the generated new route proposal.

[0601] The "emotion engine" is an artificial intelligence that analyzes the user's emotional state from collected pedestrian flow data and simulation results, and evaluates their satisfaction or dissatisfaction with new route proposals.

[0602] "Emotional response" refers to the emotional state, such as expectations or concerns, that a user shows toward a new route proposal.

[0603] "Visualization" is a method of displaying simulation results and sentiment analysis results as graphs or text reports to communicate them to people in an easy-to-understand manner.

[0604] The present invention relates to a system that utilizes people flow data to generate new, efficient and profitable public transportation routes and makes proposals that take user emotions into consideration. An embodiment of this system is described in detail below.

[0605] Overall system overview

[0606] This system consists of the following main hardware and software:

[0607] Server: Executes functions such as people flow data collection, preprocessing, AI model training, demand forecasting, new route plan generation, simulation, sentiment analysis, visualization and output.

[0608] User terminal: A terminal that receives and displays the simulation results and sentiment analysis results sent from the server.

[0609] Network: Infrastructure for data communication between servers and user devices.

[0610] Data collection and preprocessing

[0611] The server collects various people flow data via API, such as smartphone location information for the target area, public transportation usage data, and entrance data for commercial facilities. At this stage, the specific software used is a library for handling HTTP requests (e.g., requests for Python).

[0612] The server then preprocesses the collected data to remove missing and outlier values, using Python's Pandas library to manipulate data frames and NumPy for numerical data processing.

[0613] Learning with AI models

[0614] The server inputs the preprocessed people flow data into an AI model, which learns the relationship between origins and destinations, travel patterns, and usage trends by time of day. The specific software used is a deep learning framework such as TensorFlow or PyTorch.

[0615] During training, the model parameters are adjusted to find optimal hyperparameters (learning rate, number of epochs, batch size, etc.).

[0616] Demand forecasting and analysis

[0617] The server uses a trained AI model to forecast demand, particularly by identifying combinations of origins and destinations with high demand, and then predicting factors such as the number of passengers, peak congestion times, and travel distances.

[0618] This analysis uses graph drawing libraries such as Matplotlib and Seaborn to visually display the prediction results.

[0619] New route generation and simulation

[0620] The server generates new transportation route proposals based on the demand forecast results, which include information such as departure points, destinations, departure times, and arrival times.

[0621] The server then performs further simulations to generate detailed data such as operation schedules, daily ridership, peak congestion times, sales forecasts, etc. Simulation software such as AnyLogic can be used for this simulation.

[0622] Sentiment analysis with emotion engine

[0623] The server uses an emotion engine to analyze user emotions based on collected pedestrian flow data and simulation results. Specifically, it calculates emotion scores using natural language processing (NLP) libraries (e.g., SpaCy and BERT).

[0624] Adjusting route plans based on emotional responses

[0625] The server simulates the user's reaction to the generated new route plan based on the results of the emotion engine, and modifies the route plan based on the results, thereby providing a new route plan that will improve user satisfaction.

[0626] Visualizing and outputting results

[0627] The server visualizes the simulation results and sentiment analysis results as graphs and text reports, and sends them to the user's device, allowing the user to consider introducing new routes based on the results.

[0628] The specific tools used are Plotly and Matplotlib for visualization.

[0629] Specific examples

[0630] As a case study of Tokyo, we use one year's worth of smartphone location data and public transportation usage data. The server collects and preprocesses this data, then performs a detailed analysis of the travel data from Shinjuku to Shibuya and trains the AI ​​model. Demand forecasting identifies the time periods and days of the week with the highest demand between Shinjuku and Shibuya, and predicts the number of users and peak congestion times. The server then generates a direct route from Shinjuku to Shibuya and simulates the number of users, peak congestion, and sales forecasts for this route. Finally, an emotion engine is used to analyze user emotions, visualize expectations and concerns about the new route, and provide optimal route suggestions.

[0631] Examples of prompt statements

[0632] For example, the prompt we use for our generative AI model is:

[0633] "Using one year of smartphone location data and public transport usage data in Tokyo, please predict travel patterns and demand from Shinjuku to Shibuya. Based on the results, please generate direct route plans from Shinjuku to Shibuya and analyze user responses using an emotion engine."

[0634] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves user convenience and satisfaction, and enables the provision of highly profitable public transportation services.

[0635] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0636] Step 1: Data collection

[0637] The server collects people flow data, such as smartphone location information, public transportation usage data, and commercial facility entrance data, for the target area via API. Specifically, the server sends an HTTP request to the API and retrieves data in JSON format. The input is location information and usage data from the API, and the output is raw data stored on the server.

[0638] Step 2: Data Preprocessing

[0639] The server preprocesses the collected data. Specifically, it uses Python's Pandas library to create a data frame and remove missing values ​​and outliers. Rows with missing values ​​are dropped or imputed with the median. Next, the data is split by time of day and day of the week, and processes such as standardization and normalization are performed. The input is the raw data from step 1, and the output is the preprocessed data.

[0640] Step 3: Training the AI ​​model

[0641] The server trains an AI model using the preprocessed data. The specific software used is TensorFlow or PyTorch. The data is divided into training data and validation data, a deep learning model is defined, and training begins with the model.fit() function. Hyperparameters (learning rate, number of epochs, batch size, etc.) are adjusted, and performance is evaluated after training using model.evaluate(). The input is the preprocessed data, and the output is the trained model.

[0642] Step 4: Demand forecasting and analysis

[0643] The server uses the trained AI model to perform demand forecasting. It executes the forecast using the model.predict() function to identify combinations of origins and destinations with high demand. Matplotlib and Seaborn are used to visually display the forecast results as heat maps and time series graphs. The input is the trained model and preprocessed data, and the output is the demand forecast results.

[0644] Step 5: Create a new route

[0645] The server generates new transport route proposals based on the demand forecast results. Route proposals are created that include departure points, destinations, departure times, arrival times, etc. Routes with high demand are selected from the forecast results, and an optimal route proposal is created based on these. The input is the demand forecast results, and the output is a new route proposal.

[0646] Step 6: Simulation

[0647] The server runs a simulation based on the generated new route plan. Specifically, it uses simulation software such as AnyLogic to calculate operation schedules, daily passenger numbers, peak congestion times, and sales forecasts. The simulation results are exported as Excel or CSV files. The input is the new route plan, and the output is the simulation results.

[0648] Step 7: Sentiment analysis

[0649] The server analyzes the user's emotions using an emotion engine based on the collected data and simulation results. Specifically, it uses NLP libraries such as SpaCy and BERT to calculate the emotion score using get_sentiment_score(). The input is the simulation result data, and the output is the emotion analysis result.

[0650] Step 8: Adjust the route plan

[0651] The server adjusts the new route proposal based on the results of the sentiment analysis. Specifically, it considers user satisfaction and dissatisfaction from the sentiment score and improves the operation schedule and routes. It then performs a simulation again to find the optimal route proposal. The input is the sentiment analysis results and the original route proposal, and the output is the adjusted new route proposal.

[0652] Step 9: Visualizing and outputting results

[0653] The server visualizes the simulation results and sentiment analysis results as graphs and text reports. It uses Matplotlib and Plotly to create visually easy-to-understand graphs and charts. The generated report is sent to the user's device. The input is the adjusted new route plan and the simulation results, and the output is a visualized report.

[0654] These are the specific processing steps of the program of this system. This series of steps makes it possible to propose new, efficient and profitable public transportation routes and adjust routes while taking user emotions into consideration.

[0655] (Application example 2)

[0656] 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."

[0657] Existing public transport routes face difficulties in efficiently responding to demand and optimizing operating costs, resulting in low user satisfaction. Furthermore, route planning does not take into account user emotions and reactions, which often leads to user dissatisfaction when new routes are introduced. There is a need to solve these issues and plan and operate public transport routes that are efficient and provide high user satisfaction.

[0658] The identification processing by the identification 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 people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, means for evaluating reactions to the new route plans using an emotion engine that analyzes user emotions, means for adjusting the new route plans based on the emotion analysis results, and means for visualizing and outputting the simulation and emotion analysis results. This enables efficient route planning based on demand and optimal operation plans that take user emotions and reactions into consideration.

[0659] (Definitions of important words)

[0660] "People flow data" is data that represents people's movements and includes a variety of data sources, such as location information, public transportation usage data, and entrance data for commercial facilities.

[0661] "Preprocessing" refers to the process performed to prepare raw data in an analyzable format, such as removing missing values ​​and outliers, and standardizing and splitting the data.

[0662] "Generative AI" is a system that uses artificial intelligence technology and has the learning capabilities to predict demand and recognize patterns from large amounts of data.

[0663] A "new route proposal" is a newly proposed public transportation route that differs from existing transportation routes.

[0664] "Simulation" is the act of virtually predicting and verifying various factors in actual operation (number of users, sales, peak congestion times, etc.) based on the generated new route plan.

[0665] An "emotion engine" is a technology for analyzing a user's emotions and reactions, and quantitatively evaluates the emotional state based on collected data and simulation results.

[0666] "Emotion analysis" involves using an emotion engine to analyze users' emotions and reactions, and is carried out to understand their satisfaction or dissatisfaction with new route proposals.

[0667] "Visualization" is the act of visually displaying analysis or simulation results as graphs, charts, or text reports, providing information in an easy-to-understand format.

[0668] To implement the present invention, the following configuration and procedures are used.

[0669] First, the server collects people flow data. Specifically, it collects location information obtained from smartphones, public transportation usage data, and commercial facility entry data via API. Standard API access methods are used to collect data, with appropriate authentication and access control for each data source.

[0670] The collected data is preprocessed on the server using the Pandas library to remove missing values ​​and outliers and extract only the necessary information, preparing the data in a form suitable for analysis.

[0671] Next, the preprocessed data is input into the generative AI, which trains it to make demand forecasts. This training uses the Scikit-learn library, which performs clustering and regression analysis. By learning from large amounts of data, the generative AI model is able to understand the relationship between departure and destination points and usage trends by time of day, allowing it to make more accurate predictions.

[0672] The trained generative AI model generates new route proposals by applying a clustering algorithm using KMeans to identify routes between points with high demand, while also calculating operating costs and predicted revenue.

[0673] The generated new route plan is then simulated. The simulation predicts the number of passengers, sales, peak congestion times, etc., and aims to optimize operation. The simulation results are visualized as graphs using the Matplotlib library.

[0674] Furthermore, an emotion engine is used to analyze user emotions. The emotion engine analyzes user emotions based on pre-collected data and simulation results, and evaluates their satisfaction or dissatisfaction with the new route plan. The emotion analysis results are then used to further adjust the route plan.

[0675] Finally, all data and analysis results are visualized and provided to transportation planners in a report detailing proposed new routes, simulation results, and user sentiment analysis, enabling planners to consider new routes that are efficient and offer high user satisfaction.

[0676] As a concrete example, "The following prompt sentence is input into the generative AI model: 'Last night, it's early morning. How can we improve convenience? How can we change the main travel route? Based on the traffic volume forecast results, please suggest a route for a new autonomous vehicle.'" Through such prompt sentences, specific questions are posed to the generative AI model.

[0677] As described above, the present invention makes it possible to provide a system that generates efficient and profitable public transportation routes by combining demand prediction based on people flow data with user emotion analysis.

[0678] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0679] Step 1:

[0680] The server collects people flow data. Smartphone location information, public transportation usage data, and commercial facility entrance data are obtained via API. Various data sources are used as input, and the collected data is stored on the server.

[0681] Step 2:

[0682] The data collected by the server is preprocessed. Using the Pandas library, missing values ​​and outliers are removed, and the data is split and reshaped by time of day and day of the week. The input is the people flow data collected in Step 1, and the output is preprocessed data in a format suitable for analysis.

[0683] Step 3:

[0684] The server uses the preprocessed data to train the generative AI. It uses the Scikit-learn library to perform clustering and regression analysis to learn the relationship between departure and destination locations and travel patterns. The input is the preprocessed data, and the output is a trained generative AI model.

[0685] Step 4:

[0686] The server generates new route proposals using a trained generative AI model. A clustering algorithm using KMeans is used to identify routes between points with high demand, and calculates operating costs and predicted sales. The input is the trained generative AI model and analysis data, and the output is new route proposals.

[0687] Step 5:

[0688] The server performs a simulation based on the generated new route plan. The simulation predicts the number of passengers, sales, peak congestion times, etc., and aims to optimize operations. The input is the new route plan, and the output is the simulation results.

[0689] Step 6:

[0690] The server uses an emotion engine to analyze the user's emotions. Based on the collected data and simulation results, the user's emotions are quantitatively evaluated. The input is the collected data and simulation results, and the output is the emotion analysis results.

[0691] Step 7:

[0692] The server adjusts the new route proposal based on the results of the sentiment analysis, correcting areas of user dissatisfaction and adding new options. The input is the sentiment analysis results, and the output is the adjusted new route proposal.

[0693] Step 8:

[0694] The server visualizes the simulation and sentiment analysis results and provides them to transportation planners as a report. Collected data, analysis results, and new route proposals are displayed in graphs, charts, and text. The input is the adjusted new route proposal and simulation results, and the output is a visualized report.

[0695] Step 9:

[0696] The user (transportation planner) considers the introduction of new routes based on the visualized report. This makes it possible to create new routes that are efficient and have high user satisfaction. The only input required is the visualized report.

[0697] 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.

[0698] 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.

[0699] 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.

[0700] [Third embodiment]

[0701] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0702] 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.

[0703] 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).

[0704] 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.

[0705] 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.

[0706] 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).

[0707] 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.

[0708] 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.

[0709] 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.

[0710] 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.

[0711] 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.

[0712] 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."

[0713] This invention relates to a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes.

[0714] Overall system overview

[0715] 1. Data Collection and Preprocessing

[0716] The server collects various people flow data via API, such as smartphone location information in the target area, public transportation usage data, and entrance data for commercial facilities.

[0717] The server then preprocesses this data to remove missing values ​​and outliers, which is essential for accurate model training and prediction.

[0718] 2. Learning with AI models

[0719] The server inputs preprocessed people flow data into an AI model, learning the relationship between origins and destinations, movement patterns, and usage trends by time of day.

[0720] Learning is carried out using deep learning models and clustering algorithms, and evaluation and adjustment using validation data is also carried out in parallel to improve the model's performance.

[0721] 3. Demand forecasting and analysis

[0722] The server uses a trained AI model to predict demand for each target area and time period.

[0723] Specifically, the system identifies combinations of departure and destination points that are expected to have a large number of users, and analyzes detailed data such as the number of users, peak congestion times, and travel distances.

[0724] 4. New Route Generation and Simulation

[0725] The server generates new transport route proposals based on the results of the demand forecast, taking profitability into account and calculating revenue forecasts based on the number of users and operation costs.

[0726] The generated route plan is then verified through simulation, which provides detailed forecasts of operation schedules, peak times, and daily passenger numbers, enabling the operation plan to be concretely implemented.

[0727] 5. Visualizing and outputting results

[0728] Finally, the server visualizes the simulation results as graphs and text reports, outputting them to the user's terminal in a format that can be easily understood by transportation planners.

[0729] The user (transportation planner) can consider introducing new routes based on the received proposal results.

[0730] Specific examples

[0731] Consider the case of conducting a detailed analysis of people's movement patterns based on smartphone location information from each area of ​​Tokyo.

[0732] 1. Data Collection

[0733] The server collects smartphone location data and public transport usage data for one year, including data broken down by time of day for each day.

[0734] 2. Pretreatment

[0735] The server removes missing values ​​and extreme outliers from the collected data and formats the data by time period and day of the week.

[0736] 3. Training the AI ​​model

[0737] The server performs a detailed analysis of travel data, for example, from Shinjuku to Shibuya, and inputs this data into an AI model for learning.

[0738] The model is evaluated using validation data and appropriate parameters are set.

[0739] 4. Demand forecasting and analysis

[0740] The server predicts the number of users and peak congestion times based on the time periods and days of the week with the highest demand between Shinjuku and Shibuya.

[0741] 5. New Route Generation and Simulation

[0742] The server generates a direct route from Shinjuku to Shibuya and simulates the number of passengers, peak congestion, and sales forecasts for this route.

[0743] Based on the simulation results, the operation schedule and economic efficiency are evaluated.

[0744] 6. Visualizing and outputting results

[0745] The server creates a proposal result report based on the simulation results and outputs it to the user terminal of the transportation planner.

[0746] Users can use this information to consider introducing new direct routes.

[0747] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves convenience for users, and makes it possible to provide highly profitable public transportation facilities.

[0748] The processing flow will be explained below.

[0749] Step 1:

[0750] The server collects people flow data. Specifically, it obtains data such as smartphone location information, public transportation usage data, and commercial facility entrance data through various APIs. The data is saved in standard formats (CSV, JSON, etc.).

[0751] Step 2:

[0752] The server preprocesses the collected data, removing missing and outlier values ​​and standardizing the format. It also divides the data by time of day and day of the week and prepares it in a form suitable for analysis.

[0753] Step 3:

[0754] The server inputs the preprocessed people flow data into the AI ​​model, using deep learning models and clustering algorithms, and splits the data into training and validation sections.

[0755] Step 4:

[0756] The server trains the AI ​​model, analyzing the relationship between departure and destination, travel patterns, and usage trends by time of day. During the training process, the model's parameters are adjusted to improve accuracy.

[0757] Step 5:

[0758] The server uses a trained AI model to forecast demand. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, and other factors based on that.

[0759] Step 6:

[0760] The server generates new route proposals based on the predicted demand, and also calculates sales forecasts based on the number of passengers and operating costs to evaluate profitability.

[0761] Step 7:

[0762] The server then simulates the new route plan, predicting factors such as operation schedules, daily passenger numbers, and peak congestion times, and verifies the validity of the operation plan.

[0763] Step 8:

[0764] The server visualizes the simulation results, displaying them in the form of graphs and text reports, making them easy for users to understand.

[0765] Step 9:

[0766] The server outputs visualized simulation results to the user's device, allowing transportation planners to use this information to consider introducing new routes.

[0767] For example, the server collects people flow data within Tokyo and generates and simulates a direct route from Shinjuku to Shibuya based on this data. Based on the simulation results, transportation planners consider the profitability and operation schedule of the new direct route.

[0768] Example 1

[0769] 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."

[0770] In conventional public transportation route planning, it is difficult to properly grasp user movement patterns and fluctuations in demand, making it difficult to design efficient and profitable routes. Another problem is the lack of a means to accurately grasp predicted demand and peak congestion times, and to effectively propose and simulate new routes.

[0771] 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.

[0772] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, means for visualizing and outputting the simulation results, means for analyzing operation costs and sales forecasts based on the simulation results, and means for evaluating the economic efficiency of the generated route plans.This enables transportation planners to plan and consider new public transportation routes that are efficient and profitable in detail.

[0773] "People flow data" is data that shows the movement patterns and behavior of people in a specific area or period of time.

[0774] "Generative AI" is an artificial intelligence model that uses machine learning algorithms to perform a specific task (in this case, demand forecasting and generating new route proposals).

[0775] "Preprocessing" refers to the process of removing missing values ​​and outliers from collected data and preparing the data in a format suitable for analysis and learning.

[0776] "Simulation" is the process of using models to predict and evaluate outcomes or reactions in specific hypothetical environments or scenarios.

[0777] "Visualization" is a technique for expressing data and simulation results in visual formats such as graphs and charts to make the information easier to understand.

[0778] "Operating costs" refers to the various costs incurred when operating a public transportation route (e.g., fuel costs, labor costs, maintenance costs, etc.).

[0779] "Sales forecasting" is the process of calculating the expected future revenues based on the services or products offered.

[0780] "Economic efficiency" is an indicator that evaluates the degree of return on investment or effectiveness relative to costs, and aims to utilize resources efficiently.

[0781] This invention is a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes. This system operates in cooperation with the elements of a server, terminals, and users. The specific hardware and software used in each step, as well as the data processing method, are described in detail below.

[0782] Data collection and preprocessing

[0783] First, the server collects people flow data from multiple data sources via API. This data includes smartphone location information, public transportation usage data, and commercial facility entry data. Specifically, it periodically sends API requests using Python libraries (such as requests) to retrieve the data.

[0784] The server then preprocesses the collected data, using data processing libraries such as Pandas and NumPy to detect and remove missing and outlier values ​​and convert the data into a format that is easier to analyze. For example, if missing values ​​are found, they are imputed with the median or mean.

[0785] Learning with AI models

[0786] The server trains a generative AI model using preprocessed people flow data. The AI ​​model uses deep learning (TensorFlow or PyTorch) and clustering algorithms to learn the relationship between departure and destination, travel patterns, and usage trends by time of day. The dataset is split into training data and test data, and cross-validation is performed to evaluate and optimize the model's performance.

[0787] Demand forecasting and analysis

[0788] The server uses a trained generative AI model to forecast demand. Specifically, it predicts travel demand for specific regions and time periods, and analyzes detailed data for each combination of origin and destination. The results of this analysis are visualized in the form of heat maps, bar graphs, and other formats. For example, Matplotlib and Seaborn are used to create graphs showing high and low demand.

[0789] New route generation and simulation

[0790] The server generates new transport route proposals based on the results of the demand forecast. At the same time, it also takes profitability into account and calculates sales forecasts and operating costs based on the expected number of users. The generated route proposals are then verified in detail using simulation software (such as AnyLogic or Simul8). The simulations produce detailed predictions of operation schedules, peak congestion times, and daily user numbers, and the operation plan is then finalized.

[0791] Visualizing and outputting results

[0792] Finally, the server visualizes the simulation results as graphs and text reports and outputs them to the user's terminal. Libraries such as Matplotlib and ReportLab are used to visualize the results. For example, a line graph showing peak congestion times or a bar graph showing sales forecasts can be generated, presenting the results in a format that transportation planners can easily understand.

[0793] Specific examples

[0794] Consider a case in which people's travel patterns are analyzed in detail based on smartphone location information from various areas of Tokyo. The server collects smartphone location data and public transportation usage data for one year. The collected data is then reorganized by time of day and day of the week, removing missing values ​​and extreme outliers. The data on travel from Shinjuku to Shibuya is then analyzed in detail and input into an AI model for training. Based on the time of day and day of the week with the highest demand between Shinjuku and Shibuya, the number of passengers and peak congestion times are predicted, and a direct route from Shinjuku to Shibuya is generated and simulated. Finally, a proposal report based on the simulation results is output to the user device of the transportation planner, who can then consider introducing new direct routes.

[0795] Prompt Sentence Examples

[0796] "For a direct line from Shinjuku to Shibuya, what days of the week and times of day would be most popular? Also, estimate the expected operating costs and revenues for that line."

[0797] This invention enables transportation planners to plan and consider efficient and profitable public transportation routes in detail, overcoming traditional challenges.

[0798] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0799] Step 1:

[0800] Data collection

[0801] The server collects people flow data, such as smartphone location information, public transportation usage data, and commercial facility entry data, via API. Specifically, the server uses a Python library (such as requests) to send API requests and obtain data. Input data includes location information, transportation usage history, and facility entry and exit records. The output data is a raw dataset that compiles this information.

[0802] Step 2:

[0803] Data Preprocessing

[0804] The server preprocesses the collected data. Specifically, it uses Pandas and NumPy to detect missing values ​​and outliers in the data and remove or correct them. For example, if a missing value is found, it is filled in with the average value of that column. The input data is the original dataset, and the output data is the preprocessed, clean dataset.

[0805] Step 3:

[0806] Training an AI model

[0807] The server trains a generative AI model using preprocessed people flow data. It uses deep learning models (such as TensorFlow or PyTorch) and clustering algorithms to learn the relationship between origins and destinations, travel patterns, and usage trends by time of day. Specifically, it splits the dataset into training data and test data, and fits the model using the training data. The input data is the preprocessed dataset, and the output data is the trained model.

[0808] Step 4:

[0809] Demand forecasting

[0810] The server uses a trained generative AI model to perform demand forecasting. It predicts demand for specific regions and time periods and analyzes detailed data for each combination of origin and destination. Specifically, it inputs new data into the model and obtains the forecast results. For example, it supplies data to forecast travel demand between Shinjuku and Shibuya. The input data is a new dataset for the model, and the output data is the forecast results.

[0811] Step 5:

[0812] Creating a new route

[0813] The server generates new transport route proposals based on the results of the demand forecast. It designs new route proposals taking into account factors such as the number of users, peak congestion times, and operating costs. Specifically, it runs an algorithm to identify combinations of departure and destination points with high demand and proposes the optimal route. The input data is the demand forecast results, and the output data is the new route proposal.

[0814] Step 6:

[0815] simulation

[0816] The server simulates the generated route plan, forecasting in detail operation schedules, peak congestion times, and daily passenger numbers, and verifying the operation plan. Specifically, it uses simulation software (such as AnyLogic or Simul8) to reproduce the operation of the route plan in a virtual environment. The input data is the new route plan, and the output data is the simulation results.

[0817] Step 7:

[0818] Visualizing and outputting results

[0819] The server visualizes the simulation results as graphs and text reports. Using libraries such as Matplotlib and ReportLab, the results are visually represented and output in an easy-to-understand format. Specific operations include creating line graphs showing peak congestion times and bar graphs showing sales forecasts. The input data are the simulation results, and the output data is the visualized report.

[0820] Through these steps, the system suggests efficient and profitable new public transport routes for detailed consideration by transport planners.

[0821] (Application example 1)

[0822] 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."

[0823] Conventional public transportation route design relies on static analysis based on past data, making it difficult to forecast demand and derive new routes using real-time people flow data. Furthermore, it is difficult to perform simulations to ensure the realistic effectiveness and efficiency of generated new route proposals, making it difficult to make quick and accurate decisions based on such simulations. Furthermore, visualization of information is inefficient, and there is a lack of means to provide information in a format that is easy for people in charge to understand. Therefore, there is a need for a flexible and efficient system for generating new routes that can respond to dynamically changing people flow patterns in real time.

[0824] 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.

[0825] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generative AI model that performs demand forecasting using the preprocessed people flow data, means for visualizing demand data for display on smart glasses, means for generating new route plans using the trained generative AI model, means for performing simulations based on the generated new route plans, and means for visualizing and outputting the simulation results. This enables the collection and analysis of people flow data in real time, enabling the design of efficient and economical new routes based on the results. Furthermore, by using smart glasses or other display devices, traffic managers can receive information in an intuitively understandable format, supporting rapid decision-making.

[0826] "People flow data" refers to information that indicates human movement patterns and behavior, and includes location information, entry data, and transportation usage data.

[0827] "Preprocessing" is the process of removing missing values ​​and outliers from collected data and formatting it into a form suitable for analysis and learning.

[0828] A "generative AI model" is an artificial intelligence algorithm that generates demand forecasts and new route proposals based on people flow data, and often uses deep learning models and clustering algorithms.

[0829] "Smart glasses" are display devices used to visually present information and can display data in real time using augmented reality (AR) technology.

[0830] "Demand forecasting" refers to predicting future transportation demand based on people flow data, and includes analyzing the relationship between departure and destination points and usage trends by time of day.

[0831] A "New Route Proposal" is a new public transportation route plan proposed based on predicted travel demand using a generative AI model.

[0832] "Simulation" refers to simulating the operation of a new route plan in a virtual environment in order to verify its operational efficiency and economic viability.

[0833] "Visualization" refers to visually displaying the results of data analysis and simulations in a form that is easily understandable to traffic managers.

[0834] This invention relates to a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes. This system can be used with smart glasses, allowing transportation managers to receive information in an intuitive and easy-to-understand format, enabling them to make quick decisions.

[0835] System Configuration

[0836] 1. Hardware and software usage:

[0837] Hardware: smart glasses, servers, smartphones

[0838] Software: TensorFlow (AI model), Flask (API), Python, Keras

[0839] Data collection and preprocessing

[0840] The server collects location information from smartphones and other sensors in real time via an API, then preprocesses the collected data to remove missing values ​​and outliers and prepare it in a format suitable for training by a deep learning model.

[0841] Learning with AI models

[0842] The server trains a generative AI model based on the preprocessed people flow data. During this process, it uses deep learning models and clustering algorithms to learn the relationship between departure and destination, movement patterns, and usage trends by time of day. This makes it possible to predict people flow based on time of day and events.

[0843] Demand forecasting and route generation

[0844] The trained generative AI model is used to predict demand and generate new route proposals based on the most popular combinations of departure and destination points in a specific area. The server then performs simulations in a virtual environment to verify the operational efficiency and economic viability of the new route proposals. The simulations forecast ridership, sales, and peak congestion times in detail, and evaluate optimal operation schedules and economic efficiency.

[0845] Visualizing and outputting results

[0846] The server creates a report of proposed results based on the simulation results. This report is displayed in real time on the smart glasses and presented in an intuitive format that is easy for users to understand, enabling traffic managers to make quick and accurate decisions.

[0847] Specific examples

[0848] For example, the system collects real-time data on people flow around Shinjuku Station and predicts the optimal route from Shinjuku to Shibuya. The smart glasses' AR display allows users to visually understand the current traffic situation and the next route they should take.

[0849] Example prompt sentence:

[0850] Generate the optimal route from Shinjuku Station to Shibuya Station based on the following data: Data format: { "timestamp": "2023-10-04T09:30:00Z", "location": {"latitude":35.6895, "longitude":139.6917}, "destination": {"latitude":35.6580, "longitude":139.7016}, "people_count": 200}

[0851] As a result, the present invention makes it possible to improve the efficiency of new route planning for transportation facilities, improve convenience for users, and provide highly profitable public transportation facilities.

[0852] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0853] Step 1:

[0854] The server collects location information from smartphones and other sensors in real time via API. As input, it receives smartphone location data and public transport usage data, including time, latitude, longitude, and number of users. As output, it compiles these data into a single dataset.

[0855] Step 2:

[0856] The server preprocesses the collected data and removes missing values ​​and outliers. As input, it takes in the raw people flow data collected in step 1. It then performs data imputation, outlier removal, normalization, and other processes to format the data in a way that is suitable for training AI models. The output is a clean, preprocessed dataset.

[0857] Step 3:

[0858] The server trains a generative AI model based on the preprocessed people flow data. The preprocessed data obtained in step 2 is used as input. Using deep learning models and clustering algorithms, the server learns the relationship between departure and destination points, travel patterns, and usage trends by time of day. The output is a trained generative AI model.

[0859] Step 4:

[0860] The server uses the trained generative AI model to predict demand. It provides real-time and historical people flow data as input. The generative AI model then uses this data to predict demand and identify popular combinations of origins and destinations in a specific area. The output is predicted demand data.

[0861] Step 5:

[0862] The server generates new route proposals based on the results of the demand forecast. The demand data predicted in step 4 is used as input. The generative AI model generates optimal transportation routes and lists the candidates. The output is multiple route candidates proposed as new route proposals.

[0863] Step 6:

[0864] The server performs a simulation based on the new route plan it has generated. The new route plan generated in step 5 is used as input. The simulation predicts the number of passengers, sales, peak congestion times, etc., and verifies the operational efficiency and economy of the route. The output is a dataset of the simulation results.

[0865] Step 7:

[0866] The server visualizes and outputs the simulation results. It takes the simulation result data obtained in step 6 as input and visualizes it in the form of graphs and text reports. The generated visualization data is sent to smart glasses or other display devices so that the user can view it in real time. The output is a report containing the visualized information.

[0867] 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.

[0868] The present invention relates to a system that utilizes people flow data to generate new, efficient and profitable public transportation routes and makes proposals that also take into account user emotions.

[0869] Overall system overview

[0870] 1. Data Collection and Preprocessing

[0871] The server collects various people flow data via API, such as smartphone location information in the target area, public transportation usage data, and entrance data for commercial facilities.

[0872] The server then preprocesses the data, removing missing and outlier values, and splits it by time of day and day of the week to prepare it for analysis.

[0873] 2. Learning with AI models

[0874] The server inputs preprocessed people flow data into an AI model, which uses deep learning models and clustering algorithms to learn the relationship between departure and destination points, movement patterns, and usage trends by time of day.

[0875] During learning, the model parameters are adjusted to improve accuracy.

[0876] 3. Demand forecasting and analysis

[0877] The server uses a trained AI model to forecast demand. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, and other factors based on that.

[0878] 4. New Route Generation and Simulation

[0879] The server generates new transport route proposals based on the results of the demand forecast, and also calculates sales forecasts based on the number of users and operation costs to evaluate profitability.

[0880] The generated route plan is then verified through simulation, which involves making detailed predictions of operation schedules, daily passenger numbers, and peak congestion times to verify the validity of the operation plan.

[0881] 5. Sentiment analysis using an emotion engine

[0882] The server uses an emotion engine to analyze users' emotions based on collected pedestrian flow data and simulation results, quantitatively evaluating their emotional state and grasping their satisfaction or dissatisfaction with the proposed new route.

[0883] 6. Adjusting route plans based on emotional responses

[0884] The server simulates the user's emotional response to the generated new route plan and modifies it based on the results, for example, by improving areas that users dislike or adding new options.

[0885] 7. Visualizing and outputting results

[0886] Finally, the server visualizes the simulation results and the user sentiment analysis results as graphs and text reports, and outputs them to the user's terminal in a format that can be easily understood by transportation planners.

[0887] Users (transportation planners) can use this information to consider introducing new routes.

[0888] Specific examples

[0889] Consider a case where people's movement patterns and emotional responses are analyzed in detail based on smartphone location information from various areas of Tokyo.

[0890] 1. Data Collection

[0891] The server collects smartphone location data and public transport usage data for one year, including data broken down by time of day for each day.

[0892] 2. Pretreatment

[0893] The server removes missing values ​​and extreme outliers and splits the data by time of day and day of the week.

[0894] 3. Training the AI ​​model

[0895] The server performs a detailed analysis of the travel data from Shinjuku to Shibuya, inputs it into the AI ​​model, and the model is evaluated using validation data to set appropriate parameters.

[0896] 4. Demand forecasting and analysis

[0897] The server predicts the number of users and peak congestion times based on the time periods and days of the week with the highest demand between Shinjuku and Shibuya.

[0898] 5. New Route Generation and Simulation

[0899] The server generates a direct route from Shinjuku to Shibuya, and simulates the number of passengers, peak congestion, and sales forecasts for this route. Based on the simulation results, the operation schedule and economic efficiency are evaluated.

[0900] 6. Sentiment analysis

[0901] The server uses an emotion engine to analyze user emotions based on the collected data and simulation results, and evaluates reactions to new lines. For example, it visualizes users' expectations and concerns about a direct line from Shinjuku to Shibuya.

[0902] 7. Adjustment and Output

[0903] The server adjusts the proposed new routes based on the results of the sentiment analysis and generates a report with the final simulation results, which transportation planners can use to consider introducing new routes.

[0904] As described above, the present invention makes it possible to improve the efficiency of new transportation route planning, improve user convenience and satisfaction, and provide highly profitable public transportation services.

[0905] The processing flow will be explained below.

[0906] Step 1:

[0907] The server collects people flow data. Specifically, it obtains smartphone location information, public transportation usage data, and commercial facility entry data via API and stores it in a database in standard formats (CSV or JSON). The data collection period is generally one year, and includes information on weekdays, weekends, and time periods.

[0908] Step 2:

[0909] The server preprocesses the collected data. Specifically, it removes missing values ​​and outliers and standardizes the data format. It also divides the data by time period and day of the week and prepares it in a form suitable for analysis. This process uses data cleaning tools and programs.

[0910] Step 3:

[0911] The server inputs the preprocessed people flow data into an AI model, such as a deep learning model or a clustering algorithm. The data is split into training and validation sets, and the training data set is used to train the model.

[0912] Step 4:

[0913] The server trains the AI ​​model. Specifically, it uses a data set to analyze the relationship between origins and destinations, travel patterns, and usage trends by time of day. During the training process, the model's parameters are adjusted to improve accuracy.

[0914] Step 5:

[0915] The server uses the trained AI model to make demand forecasts. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, etc. based on those combinations. This allows it to grasp the demand distribution in a specific area.

[0916] Step 6:

[0917] The server generates new route proposals based on predicted demand. At the same time, it also calculates sales forecasts based on the number of passengers and operating costs, and evaluates profitability. Specifically, if demand between Shinjuku and Shibuya is high, it proposes a direct route.

[0918] Step 7:

[0919] The server then simulates the new route plan. The simulation makes detailed predictions of the operation schedule, daily passenger numbers, and peak congestion times, and verifies the validity of the operation plan. For example, it creates an operation schedule for a direct line between Shinjuku and Shibuya, and estimates the number of passengers and operation costs based on that schedule.

[0920] Step 8:

[0921] The server uses an emotion engine to analyze users' emotional states based on the collected pedestrian flow data and simulation results. Specifically, it analyzes user comments, social media data, survey results, etc., to quantitatively evaluate users' satisfaction or dissatisfaction with new route proposals.

[0922] Step 9:

[0923] The server adjusts new route proposals based on the results of sentiment analysis. Specifically, it improves areas where users are dissatisfied and considers additional options. For example, it proposes a service schedule that avoids peak hours, which users are likely to dissatisfy.

[0924] Step 10:

[0925] The server visualizes the simulation results and sentiment analysis results in the form of graphs and text reports, making them easy for transportation planners (users) to understand.

[0926] Step 11:

[0927] The server outputs visualized simulation results and adjusted route plans to the user's device. Based on this, transportation planners (users) can consider introducing new routes. For example, they can evaluate the profitability and operation schedule of a direct route between Shinjuku and Shibuya and make a final decision on its introduction.

[0928] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves user convenience and satisfaction, and enables the provision of highly profitable public transportation services.

[0929] Example 2

[0930] 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."

[0931] In conventional public transport route planning, demand forecasting and route proposal generation are often done manually, resulting in inefficiency and a lack of accuracy. Furthermore, it is not easy to adjust routes while taking into account user emotions and satisfaction, which can lead to dissatisfaction after the introduction of new routes. Therefore, there is a need for efficient and accurate proposals for new public transport routes and route adjustments that take into account user emotions.

[0932] 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.

[0933] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that uses the preprocessed people flow data to learn relationships between departure points and destinations, travel patterns, and usage trends by time period, means for performing demand forecasts using the trained generation AI and identifying high-demand combinations of departure points and destinations, means for generating new route plans based on the demand forecast results, means for simulating the generated new route plans to predict operation schedules, ridership, peak congestion times, and sales, means for analyzing users' emotional states using an emotion engine and evaluating their satisfaction or dissatisfaction with the new route plans, means for revising the route plans based on users' emotional reactions to the generated new route plans, and means for visualizing and outputting the simulation results and emotion analysis results. This enables the proposal of efficient and profitable new public transportation routes and route adjustments that take users' emotions into consideration.

[0934] "People flow data" refers to data on people's movements, including individual movement patterns and location information, public transportation usage data, and commercial facility entry data.

[0935] "Preprocessing" refers to the process of removing missing or outliers from collected people flow data and dividing the data by time of day or day of the week, among other things, to prepare the data in a form suitable for analysis.

[0936] "Generative AI" is artificial intelligence that uses machine learning or deep learning to learn from data and perform demand forecasts and generate new routes.

[0937] "Demand forecasting" involves using trained generative AI to predict combinations that will be in high demand based on the relationship between people's departure and destination locations and their movement patterns.

[0938] A "new route proposal" is a proposal for a new public transportation route based on demand forecasts.

[0939] "Simulation" is the process of virtually calculating operation schedules, number of users, peak congestion times, sales forecasts, etc. based on the generated new route proposal.

[0940] The "emotion engine" is an artificial intelligence that analyzes the user's emotional state from collected pedestrian flow data and simulation results, and evaluates their satisfaction or dissatisfaction with new route proposals.

[0941] "Emotional response" refers to the emotional state, such as expectations or concerns, that a user shows toward a new route proposal.

[0942] "Visualization" is a method of displaying simulation results and sentiment analysis results as graphs or text reports to communicate them to people in an easy-to-understand manner.

[0943] The present invention relates to a system that utilizes people flow data to generate new, efficient and profitable public transportation routes and makes proposals that take user emotions into consideration. An embodiment of this system is described in detail below.

[0944] Overall system overview

[0945] This system consists of the following main hardware and software:

[0946] Server: Executes functions such as people flow data collection, preprocessing, AI model training, demand forecasting, new route plan generation, simulation, sentiment analysis, visualization and output.

[0947] User terminal: A terminal that receives and displays the simulation results and sentiment analysis results sent from the server.

[0948] Network: Infrastructure for data communication between servers and user devices.

[0949] Data collection and preprocessing

[0950] The server collects various people flow data via API, such as smartphone location information for the target area, public transportation usage data, and entrance data for commercial facilities. At this stage, the specific software used is a library for handling HTTP requests (e.g., requests for Python).

[0951] The server then preprocesses the collected data to remove missing and outlier values, using Python's Pandas library to manipulate data frames and NumPy for numerical data processing.

[0952] Learning with AI models

[0953] The server inputs the preprocessed people flow data into an AI model, which learns the relationship between origins and destinations, travel patterns, and usage trends by time of day. The specific software used is a deep learning framework such as TensorFlow or PyTorch.

[0954] During training, the model parameters are adjusted to find optimal hyperparameters (learning rate, number of epochs, batch size, etc.).

[0955] Demand forecasting and analysis

[0956] The server uses a trained AI model to forecast demand, particularly by identifying combinations of origins and destinations with high demand, and then predicting factors such as the number of passengers, peak congestion times, and travel distances.

[0957] This analysis uses graph drawing libraries such as Matplotlib and Seaborn to visually display the prediction results.

[0958] New route generation and simulation

[0959] The server generates new transportation route proposals based on the demand forecast results, which include information such as departure points, destinations, departure times, and arrival times.

[0960] The server then performs further simulations to generate detailed data such as operation schedules, daily ridership, peak congestion times, sales forecasts, etc. Simulation software such as AnyLogic can be used for this simulation.

[0961] Sentiment analysis with emotion engine

[0962] The server uses an emotion engine to analyze user emotions based on collected pedestrian flow data and simulation results. Specifically, it calculates emotion scores using natural language processing (NLP) libraries (e.g., SpaCy and BERT).

[0963] Adjusting route plans based on emotional responses

[0964] The server simulates the user's reaction to the generated new route plan based on the results of the emotion engine, and modifies the route plan based on the results, thereby providing a new route plan that will improve user satisfaction.

[0965] Visualizing and outputting results

[0966] The server visualizes the simulation results and sentiment analysis results as graphs and text reports, and sends them to the user's device, allowing the user to consider introducing new routes based on the results.

[0967] The specific tools used are Plotly and Matplotlib for visualization.

[0968] Specific examples

[0969] As a case study of Tokyo, we use one year's worth of smartphone location data and public transportation usage data. The server collects and preprocesses this data, then performs a detailed analysis of the travel data from Shinjuku to Shibuya and trains the AI ​​model. Demand forecasting identifies the time periods and days of the week with the highest demand between Shinjuku and Shibuya, and predicts the number of users and peak congestion times. The server then generates a direct route from Shinjuku to Shibuya and simulates the number of users, peak congestion, and sales forecasts for this route. Finally, an emotion engine is used to analyze user emotions, visualize expectations and concerns about the new route, and provide optimal route suggestions.

[0970] Examples of prompt statements

[0971] For example, the prompt we use for our generative AI model is:

[0972] "Using one year of smartphone location data and public transport usage data in Tokyo, please predict travel patterns and demand from Shinjuku to Shibuya. Based on the results, please generate direct route plans from Shinjuku to Shibuya and analyze user responses using an emotion engine."

[0973] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves user convenience and satisfaction, and enables the provision of highly profitable public transportation services.

[0974] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0975] Step 1: Data collection

[0976] The server collects people flow data, such as smartphone location information, public transportation usage data, and commercial facility entrance data, for the target area via API. Specifically, the server sends an HTTP request to the API and retrieves data in JSON format. The input is location information and usage data from the API, and the output is raw data stored on the server.

[0977] Step 2: Data Preprocessing

[0978] The server preprocesses the collected data. Specifically, it uses Python's Pandas library to create a data frame and remove missing values ​​and outliers. Rows with missing values ​​are dropped or imputed with the median. Next, the data is split by time of day and day of the week, and processes such as standardization and normalization are performed. The input is the raw data from step 1, and the output is the preprocessed data.

[0979] Step 3: Training the AI ​​model

[0980] The server trains an AI model using the preprocessed data. The specific software used is TensorFlow or PyTorch. The data is divided into training data and validation data, a deep learning model is defined, and training begins with the model.fit() function. Hyperparameters (learning rate, number of epochs, batch size, etc.) are adjusted, and performance is evaluated after training using model.evaluate(). The input is the preprocessed data, and the output is the trained model.

[0981] Step 4: Demand forecasting and analysis

[0982] The server uses the trained AI model to perform demand forecasting. It executes the forecast using the model.predict() function to identify combinations of origins and destinations with high demand. Matplotlib and Seaborn are used to visually display the forecast results as heat maps and time series graphs. The input is the trained model and preprocessed data, and the output is the demand forecast results.

[0983] Step 5: Create a new route

[0984] The server generates new transport route proposals based on the demand forecast results. Route proposals are created that include departure points, destinations, departure times, arrival times, etc. Routes with high demand are selected from the forecast results, and an optimal route proposal is created based on these. The input is the demand forecast results, and the output is a new route proposal.

[0985] Step 6: Simulation

[0986] The server runs a simulation based on the generated new route plan. Specifically, it uses simulation software such as AnyLogic to calculate operation schedules, daily passenger numbers, peak congestion times, and sales forecasts. The simulation results are exported as Excel or CSV files. The input is the new route plan, and the output is the simulation results.

[0987] Step 7: Sentiment analysis

[0988] The server analyzes the user's emotions using an emotion engine based on the collected data and simulation results. Specifically, it uses NLP libraries such as SpaCy and BERT to calculate the emotion score using get_sentiment_score(). The input is the simulation result data, and the output is the emotion analysis result.

[0989] Step 8: Adjust the route plan

[0990] The server adjusts the new route proposal based on the results of the sentiment analysis. Specifically, it considers user satisfaction and dissatisfaction from the sentiment score and improves the operation schedule and routes. It then performs a simulation again to find the optimal route proposal. The input is the sentiment analysis results and the original route proposal, and the output is the adjusted new route proposal.

[0991] Step 9: Visualizing and outputting results

[0992] The server visualizes the simulation results and sentiment analysis results as graphs and text reports. It uses Matplotlib and Plotly to create visually easy-to-understand graphs and charts. The generated report is sent to the user's device. The input is the adjusted new route plan and the simulation results, and the output is a visualized report.

[0993] These are the specific processing steps of the program of this system. This series of steps makes it possible to propose new, efficient and profitable public transportation routes and adjust routes while taking user emotions into consideration.

[0994] (Application example 2)

[0995] 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."

[0996] Existing public transport routes face difficulties in efficiently responding to demand and optimizing operating costs, resulting in low user satisfaction. Furthermore, route planning does not take into account user emotions and reactions, which often leads to user dissatisfaction when new routes are introduced. There is a need to solve these issues and plan and operate public transport routes that are efficient and provide high user satisfaction.

[0997] The identification processing by the identification 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 people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, means for evaluating reactions to the new route plans using an emotion engine that analyzes user emotions, means for adjusting the new route plans based on the emotion analysis results, and means for visualizing and outputting the simulation and emotion analysis results. This enables efficient route planning based on demand and optimal operation plans that take user emotions and reactions into consideration.

[0998] (Definitions of important words)

[0999] "People flow data" is data that represents people's movements and includes a variety of data sources, such as location information, public transportation usage data, and entrance data for commercial facilities.

[1000] "Preprocessing" refers to the process performed to prepare raw data in an analyzable format, such as removing missing values ​​and outliers, and standardizing and splitting the data.

[1001] "Generative AI" is a system that uses artificial intelligence technology and has the learning capabilities to predict demand and recognize patterns from large amounts of data.

[1002] A "new route proposal" is a newly proposed public transportation route that differs from existing transportation routes.

[1003] "Simulation" is the act of virtually predicting and verifying various factors in actual operation (number of users, sales, peak congestion times, etc.) based on the generated new route plan.

[1004] An "emotion engine" is a technology for analyzing a user's emotions and reactions, and quantitatively evaluates the emotional state based on collected data and simulation results.

[1005] "Emotion analysis" involves using an emotion engine to analyze users' emotions and reactions, and is carried out to understand their satisfaction or dissatisfaction with new route proposals.

[1006] "Visualization" is the act of visually displaying analysis or simulation results as graphs, charts, or text reports, providing information in an easy-to-understand format.

[1007] To implement the present invention, the following configuration and procedures are used.

[1008] First, the server collects people flow data. Specifically, it collects location information obtained from smartphones, public transportation usage data, and commercial facility entry data via API. Standard API access methods are used to collect data, with appropriate authentication and access control for each data source.

[1009] The collected data is preprocessed on the server using the Pandas library to remove missing values ​​and outliers and extract only the necessary information, preparing the data in a form suitable for analysis.

[1010] Next, the preprocessed data is input into the generative AI, which trains it to make demand forecasts. This training uses the Scikit-learn library, which performs clustering and regression analysis. By learning from large amounts of data, the generative AI model is able to understand the relationship between departure and destination points and usage trends by time of day, allowing it to make more accurate predictions.

[1011] The trained generative AI model generates new route proposals by applying a clustering algorithm using KMeans to identify routes between points with high demand, while also calculating operating costs and predicted revenue.

[1012] The generated new route plan is then simulated. The simulation predicts the number of passengers, sales, peak congestion times, etc., and aims to optimize operation. The simulation results are visualized as graphs using the Matplotlib library.

[1013] Furthermore, an emotion engine is used to analyze user emotions. The emotion engine analyzes user emotions based on pre-collected data and simulation results, and evaluates their satisfaction or dissatisfaction with the new route plan. The emotion analysis results are then used to further adjust the route plan.

[1014] Finally, all data and analysis results are visualized and provided to transportation planners in a report detailing proposed new routes, simulation results, and user sentiment analysis, enabling planners to consider new routes that are efficient and offer high user satisfaction.

[1015] As a concrete example, "The following prompt sentence is input into the generative AI model: 'Last night, it's early morning. How can we improve convenience? How can we change the main travel route? Based on the traffic volume forecast results, please suggest a route for a new autonomous vehicle.'" Through such prompt sentences, specific questions are posed to the generative AI model.

[1016] As described above, the present invention makes it possible to provide a system that generates efficient and profitable public transportation routes by combining demand prediction based on people flow data with user emotion analysis.

[1017] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1018] Step 1:

[1019] The server collects people flow data. Smartphone location information, public transportation usage data, and commercial facility entrance data are obtained via API. Various data sources are used as input, and the collected data is stored on the server.

[1020] Step 2:

[1021] The data collected by the server is preprocessed. Using the Pandas library, missing values ​​and outliers are removed, and the data is split and reshaped by time of day and day of the week. The input is the people flow data collected in Step 1, and the output is preprocessed data in a format suitable for analysis.

[1022] Step 3:

[1023] The server uses the preprocessed data to train the generative AI. It uses the Scikit-learn library to perform clustering and regression analysis to learn the relationship between departure and destination locations and travel patterns. The input is the preprocessed data, and the output is a trained generative AI model.

[1024] Step 4:

[1025] The server generates new route proposals using a trained generative AI model. A clustering algorithm using KMeans is used to identify routes between points with high demand, and calculates operating costs and predicted sales. The input is the trained generative AI model and analysis data, and the output is new route proposals.

[1026] Step 5:

[1027] The server performs a simulation based on the generated new route plan. The simulation predicts the number of passengers, sales, peak congestion times, etc., and aims to optimize operations. The input is the new route plan, and the output is the simulation results.

[1028] Step 6:

[1029] The server uses an emotion engine to analyze the user's emotions. Based on the collected data and simulation results, the user's emotions are quantitatively evaluated. The input is the collected data and simulation results, and the output is the emotion analysis results.

[1030] Step 7:

[1031] The server adjusts the new route proposal based on the results of the sentiment analysis, correcting areas of user dissatisfaction and adding new options. The input is the sentiment analysis results, and the output is the adjusted new route proposal.

[1032] Step 8:

[1033] The server visualizes the simulation and sentiment analysis results and provides them to transportation planners as a report. Collected data, analysis results, and new route proposals are displayed in graphs, charts, and text. The input is the adjusted new route proposal and simulation results, and the output is a visualized report.

[1034] Step 9:

[1035] The user (transportation planner) considers the introduction of new routes based on the visualized report. This makes it possible to create new routes that are efficient and have high user satisfaction. The only input required is the visualized report.

[1036] 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.

[1037] 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.

[1038] 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.

[1039] [Fourth embodiment]

[1040] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1041] 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.

[1042] 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).

[1043] 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.

[1044] 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.

[1045] 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).

[1046] 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.

[1047] 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.

[1048] 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.

[1049] 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.

[1050] 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.

[1051] 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.

[1052] 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."

[1053] This invention relates to a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes.

[1054] Overall system overview

[1055] 1. Data Collection and Preprocessing

[1056] The server collects various people flow data via API, such as smartphone location information in the target area, public transportation usage data, and entrance data for commercial facilities.

[1057] The server then preprocesses this data to remove missing values ​​and outliers, which is essential for accurate model training and prediction.

[1058] 2. Learning with AI models

[1059] The server inputs preprocessed people flow data into an AI model, learning the relationship between origins and destinations, movement patterns, and usage trends by time of day.

[1060] Learning is carried out using deep learning models and clustering algorithms, and evaluation and adjustment using validation data is also carried out in parallel to improve the model's performance.

[1061] 3. Demand forecasting and analysis

[1062] The server uses a trained AI model to predict demand for each target area and time period.

[1063] Specifically, the system identifies combinations of departure and destination points that are expected to have a large number of users, and analyzes detailed data such as the number of users, peak congestion times, and travel distances.

[1064] 4. New Route Generation and Simulation

[1065] The server generates new transport route proposals based on the results of the demand forecast, taking profitability into account and calculating revenue forecasts based on the number of users and operation costs.

[1066] The generated route plan is then verified through simulation, which provides detailed forecasts of operation schedules, peak times, and daily passenger numbers, enabling the operation plan to be concretely implemented.

[1067] 5. Visualizing and outputting results

[1068] Finally, the server visualizes the simulation results as graphs and text reports, outputting them to the user's terminal in a format that can be easily understood by transportation planners.

[1069] The user (transportation planner) can consider introducing new routes based on the received proposal results.

[1070] Specific examples

[1071] Consider the case of conducting a detailed analysis of people's movement patterns based on smartphone location information from each area of ​​Tokyo.

[1072] 1. Data Collection

[1073] The server collects smartphone location data and public transport usage data for one year, including data broken down by time of day for each day.

[1074] 2. Pretreatment

[1075] The server removes missing values ​​and extreme outliers from the collected data and formats the data by time period and day of the week.

[1076] 3. Training the AI ​​model

[1077] The server performs a detailed analysis of travel data, for example, from Shinjuku to Shibuya, and inputs this data into an AI model for learning.

[1078] The model is evaluated using validation data and appropriate parameters are set.

[1079] 4. Demand forecasting and analysis

[1080] The server predicts the number of users and peak congestion times based on the time periods and days of the week with the highest demand between Shinjuku and Shibuya.

[1081] 5. New Route Generation and Simulation

[1082] The server generates a direct route from Shinjuku to Shibuya and simulates the number of passengers, peak congestion, and sales forecasts for this route.

[1083] Based on the simulation results, the operation schedule and economic efficiency are evaluated.

[1084] 6. Visualizing and outputting results

[1085] The server creates a proposal result report based on the simulation results and outputs it to the user terminal of the transportation planner.

[1086] Users can use this information to consider introducing new direct routes.

[1087] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves convenience for users, and makes it possible to provide highly profitable public transportation facilities.

[1088] The processing flow will be explained below.

[1089] Step 1:

[1090] The server collects people flow data. Specifically, it obtains data such as smartphone location information, public transportation usage data, and commercial facility entrance data through various APIs. The data is saved in standard formats (CSV, JSON, etc.).

[1091] Step 2:

[1092] The server preprocesses the collected data, removing missing and outlier values ​​and standardizing the format. It also divides the data by time of day and day of the week and prepares it in a form suitable for analysis.

[1093] Step 3:

[1094] The server inputs the preprocessed people flow data into the AI ​​model, using deep learning models and clustering algorithms, and splits the data into training and validation sections.

[1095] Step 4:

[1096] The server trains the AI ​​model, analyzing the relationship between departure and destination, travel patterns, and usage trends by time of day. During the training process, the model's parameters are adjusted to improve accuracy.

[1097] Step 5:

[1098] The server uses a trained AI model to forecast demand. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, and other factors based on that.

[1099] Step 6:

[1100] The server generates new route proposals based on the predicted demand, and also calculates sales forecasts based on the number of passengers and operating costs to evaluate profitability.

[1101] Step 7:

[1102] The server then simulates the new route plan, predicting factors such as operation schedules, daily passenger numbers, and peak congestion times, and verifies the validity of the operation plan.

[1103] Step 8:

[1104] The server visualizes the simulation results, displaying them in the form of graphs and text reports, making them easy for users to understand.

[1105] Step 9:

[1106] The server outputs visualized simulation results to the user's device, allowing transportation planners to use this information to consider introducing new routes.

[1107] For example, the server collects people flow data within Tokyo and generates and simulates a direct route from Shinjuku to Shibuya based on this data. Based on the simulation results, transportation planners consider the profitability and operation schedule of the new direct route.

[1108] Example 1

[1109] 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."

[1110] In conventional public transportation route planning, it is difficult to properly grasp user movement patterns and fluctuations in demand, making it difficult to design efficient and profitable routes. Another problem is the lack of a means to accurately grasp predicted demand and peak congestion times, and to effectively propose and simulate new routes.

[1111] 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.

[1112] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, means for visualizing and outputting the simulation results, means for analyzing operation costs and sales forecasts based on the simulation results, and means for evaluating the economic efficiency of the generated route plans.This enables transportation planners to plan and consider new public transportation routes that are efficient and profitable in detail.

[1113] "People flow data" is data that shows the movement patterns and behavior of people in a specific area or period of time.

[1114] "Generative AI" is an artificial intelligence model that uses machine learning algorithms to perform a specific task (in this case, demand forecasting and generating new route proposals).

[1115] "Preprocessing" refers to the process of removing missing values ​​and outliers from collected data and preparing the data in a format suitable for analysis and learning.

[1116] "Simulation" is the process of using models to predict and evaluate outcomes or reactions in specific hypothetical environments or scenarios.

[1117] "Visualization" is a technique for expressing data and simulation results in visual formats such as graphs and charts to make the information easier to understand.

[1118] "Operating costs" refers to the various costs incurred when operating a public transportation route (e.g., fuel costs, labor costs, maintenance costs, etc.).

[1119] "Sales forecasting" is the process of calculating the expected future revenues based on the services or products offered.

[1120] "Economic efficiency" is an indicator that evaluates the degree of return on investment or effectiveness relative to costs, and aims to utilize resources efficiently.

[1121] This invention is a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes. This system operates in cooperation with the elements of a server, terminals, and users. The specific hardware and software used in each step, as well as the data processing method, are described in detail below.

[1122] Data collection and preprocessing

[1123] First, the server collects people flow data from multiple data sources via API. This data includes smartphone location information, public transportation usage data, and commercial facility entry data. Specifically, it periodically sends API requests using Python libraries (such as requests) to retrieve the data.

[1124] The server then preprocesses the collected data, using data processing libraries such as Pandas and NumPy to detect and remove missing and outlier values ​​and convert the data into a format that is easier to analyze. For example, if missing values ​​are found, they are imputed with the median or mean.

[1125] Learning with AI models

[1126] The server trains a generative AI model using preprocessed people flow data. The AI ​​model uses deep learning (TensorFlow or PyTorch) and clustering algorithms to learn the relationship between departure and destination, travel patterns, and usage trends by time of day. The dataset is split into training data and test data, and cross-validation is performed to evaluate and optimize the model's performance.

[1127] Demand forecasting and analysis

[1128] The server uses a trained generative AI model to forecast demand. Specifically, it predicts travel demand for specific regions and time periods, and analyzes detailed data for each combination of origin and destination. The results of this analysis are visualized in the form of heat maps, bar graphs, and other formats. For example, Matplotlib and Seaborn are used to create graphs showing high and low demand.

[1129] New route generation and simulation

[1130] The server generates new transport route proposals based on the results of the demand forecast. At the same time, it also takes profitability into account and calculates sales forecasts and operating costs based on the expected number of users. The generated route proposals are then verified in detail using simulation software (such as AnyLogic or Simul8). The simulations produce detailed predictions of operation schedules, peak congestion times, and daily user numbers, and the operation plan is then finalized.

[1131] Visualizing and outputting results

[1132] Finally, the server visualizes the simulation results as graphs and text reports and outputs them to the user's terminal. Libraries such as Matplotlib and ReportLab are used to visualize the results. For example, a line graph showing peak congestion times or a bar graph showing sales forecasts can be generated, presenting the results in a format that transportation planners can easily understand.

[1133] Specific examples

[1134] Consider a case in which people's travel patterns are analyzed in detail based on smartphone location information from various areas of Tokyo. The server collects smartphone location data and public transportation usage data for one year. The collected data is then reorganized by time of day and day of the week, removing missing values ​​and extreme outliers. The data on travel from Shinjuku to Shibuya is then analyzed in detail and input into an AI model for training. Based on the time of day and day of the week with the highest demand between Shinjuku and Shibuya, the number of passengers and peak congestion times are predicted, and a direct route from Shinjuku to Shibuya is generated and simulated. Finally, a proposal report based on the simulation results is output to the user device of the transportation planner, who can then consider introducing new direct routes.

[1135] Prompt Sentence Examples

[1136] "For a direct line from Shinjuku to Shibuya, what days of the week and times of day would be most popular? Also, estimate the expected operating costs and revenues for that line."

[1137] This invention enables transportation planners to plan and consider efficient and profitable public transportation routes in detail, overcoming traditional challenges.

[1138] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1139] Step 1:

[1140] Data collection

[1141] The server collects people flow data, such as smartphone location information, public transportation usage data, and commercial facility entry data, via API. Specifically, the server uses a Python library (such as requests) to send API requests and obtain data. Input data includes location information, transportation usage history, and facility entry and exit records. The output data is a raw dataset that compiles this information.

[1142] Step 2:

[1143] Data Preprocessing

[1144] The server preprocesses the collected data. Specifically, it uses Pandas and NumPy to detect missing values ​​and outliers in the data and remove or correct them. For example, if a missing value is found, it is filled in with the average value of that column. The input data is the original dataset, and the output data is the preprocessed, clean dataset.

[1145] Step 3:

[1146] Training an AI model

[1147] The server trains a generative AI model using preprocessed people flow data. It uses deep learning models (such as TensorFlow or PyTorch) and clustering algorithms to learn the relationship between origins and destinations, travel patterns, and usage trends by time of day. Specifically, it splits the dataset into training data and test data, and fits the model using the training data. The input data is the preprocessed dataset, and the output data is the trained model.

[1148] Step 4:

[1149] Demand forecasting

[1150] The server uses a trained generative AI model to perform demand forecasting. It predicts demand for specific regions and time periods and analyzes detailed data for each combination of origin and destination. Specifically, it inputs new data into the model and obtains the forecast results. For example, it supplies data to forecast travel demand between Shinjuku and Shibuya. The input data is a new dataset for the model, and the output data is the forecast results.

[1151] Step 5:

[1152] Creating a new route

[1153] The server generates new transport route proposals based on the results of the demand forecast. It designs new route proposals taking into account factors such as the number of users, peak congestion times, and operating costs. Specifically, it runs an algorithm to identify combinations of departure and destination points with high demand and proposes the optimal route. The input data is the demand forecast results, and the output data is the new route proposal.

[1154] Step 6:

[1155] simulation

[1156] The server simulates the generated route plan, forecasting in detail operation schedules, peak congestion times, and daily passenger numbers, and verifying the operation plan. Specifically, it uses simulation software (such as AnyLogic or Simul8) to reproduce the operation of the route plan in a virtual environment. The input data is the new route plan, and the output data is the simulation results.

[1157] Step 7:

[1158] Visualizing and outputting results

[1159] The server visualizes the simulation results as graphs and text reports. Using libraries such as Matplotlib and ReportLab, the results are visually represented and output in an easy-to-understand format. Specific operations include creating line graphs showing peak congestion times and bar graphs showing sales forecasts. The input data are the simulation results, and the output data is the visualized report.

[1160] Through these steps, the system suggests efficient and profitable new public transport routes for detailed consideration by transport planners.

[1161] (Application example 1)

[1162] 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."

[1163] Conventional public transportation route design relies on static analysis based on past data, making it difficult to forecast demand and derive new routes using real-time people flow data. Furthermore, it is difficult to perform simulations to ensure the realistic effectiveness and efficiency of generated new route proposals, making it difficult to make quick and accurate decisions based on such simulations. Furthermore, visualization of information is inefficient, and there is a lack of means to provide information in a format that is easy for people in charge to understand. Therefore, there is a need for a flexible and efficient system for generating new routes that can respond to dynamically changing people flow patterns in real time.

[1164] 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.

[1165] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generative AI model that performs demand forecasting using the preprocessed people flow data, means for visualizing demand data for display on smart glasses, means for generating new route plans using the trained generative AI model, means for performing simulations based on the generated new route plans, and means for visualizing and outputting the simulation results. This enables the collection and analysis of people flow data in real time, enabling the design of efficient and economical new routes based on the results. Furthermore, by using smart glasses or other display devices, traffic managers can receive information in an intuitively understandable format, supporting rapid decision-making.

[1166] "People flow data" refers to information that indicates human movement patterns and behavior, and includes location information, entry data, and transportation usage data.

[1167] "Preprocessing" is the process of removing missing values ​​and outliers from collected data and formatting it into a form suitable for analysis and learning.

[1168] A "generative AI model" is an artificial intelligence algorithm that generates demand forecasts and new route proposals based on people flow data, and often uses deep learning models and clustering algorithms.

[1169] "Smart glasses" are display devices used to visually present information and can display data in real time using augmented reality (AR) technology.

[1170] "Demand forecasting" refers to predicting future transportation demand based on people flow data, and includes analyzing the relationship between departure and destination points and usage trends by time of day.

[1171] A "New Route Proposal" is a new public transportation route plan proposed based on predicted travel demand using a generative AI model.

[1172] "Simulation" refers to simulating the operation of a new route plan in a virtual environment in order to verify its operational efficiency and economic viability.

[1173] "Visualization" refers to visually displaying the results of data analysis and simulations in a form that is easily understandable to traffic managers.

[1174] This invention relates to a system that uses people flow data to generate and propose new, efficient and profitable public transportation routes. This system can be used with smart glasses, allowing transportation managers to receive information in an intuitive and easy-to-understand format, enabling them to make quick decisions.

[1175] System Configuration

[1176] 1. Hardware and software usage:

[1177] Hardware: smart glasses, servers, smartphones

[1178] Software: TensorFlow (AI model), Flask (API), Python, Keras

[1179] Data collection and preprocessing

[1180] The server collects location information from smartphones and other sensors in real time via an API, then preprocesses the collected data to remove missing values ​​and outliers and prepare it in a format suitable for training by a deep learning model.

[1181] Learning with AI models

[1182] The server trains a generative AI model based on the preprocessed people flow data. During this process, it uses deep learning models and clustering algorithms to learn the relationship between departure and destination, movement patterns, and usage trends by time of day. This makes it possible to predict people flow based on time of day and events.

[1183] Demand forecasting and route generation

[1184] The trained generative AI model is used to predict demand and generate new route proposals based on the most popular combinations of departure and destination points in a specific area. The server then performs simulations in a virtual environment to verify the operational efficiency and economic viability of the new route proposals. The simulations forecast ridership, sales, and peak congestion times in detail, and evaluate optimal operation schedules and economic efficiency.

[1185] Visualizing and outputting results

[1186] The server creates a report of proposed results based on the simulation results. This report is displayed in real time on the smart glasses and presented in an intuitive format that is easy for users to understand, enabling traffic managers to make quick and accurate decisions.

[1187] Specific examples

[1188] For example, the system collects real-time data on people flow around Shinjuku Station and predicts the optimal route from Shinjuku to Shibuya. The smart glasses' AR display allows users to visually understand the current traffic situation and the next route they should take.

[1189] Example prompt sentence:

[1190] Generate the optimal route from Shinjuku Station to Shibuya Station based on the following data: Data format: { "timestamp": "2023-10-04T09:30:00Z", "location": {"latitude":35.6895, "longitude":139.6917}, "destination": {"latitude":35.6580, "longitude":139.7016}, "people_count": 200}

[1191] As a result, the present invention makes it possible to improve the efficiency of new route planning for transportation facilities, improve convenience for users, and provide highly profitable public transportation facilities.

[1192] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1193] Step 1:

[1194] The server collects location information from smartphones and other sensors in real time via API. As input, it receives smartphone location data and public transport usage data, including time, latitude, longitude, and number of users. As output, it compiles these data into a single dataset.

[1195] Step 2:

[1196] The server preprocesses the collected data and removes missing values ​​and outliers. As input, it takes in the raw people flow data collected in step 1. It then performs data imputation, outlier removal, normalization, and other processes to format the data in a way that is suitable for training AI models. The output is a clean, preprocessed dataset.

[1197] Step 3:

[1198] The server trains a generative AI model based on the preprocessed people flow data. The preprocessed data obtained in step 2 is used as input. Using deep learning models and clustering algorithms, the server learns the relationship between departure and destination points, travel patterns, and usage trends by time of day. The output is a trained generative AI model.

[1199] Step 4:

[1200] The server uses the trained generative AI model to predict demand. It provides real-time and historical people flow data as input. The generative AI model then uses this data to predict demand and identify popular combinations of origins and destinations in a specific area. The output is predicted demand data.

[1201] Step 5:

[1202] The server generates new route proposals based on the results of the demand forecast. The demand data predicted in step 4 is used as input. The generative AI model generates optimal transportation routes and lists the candidates. The output is multiple route candidates proposed as new route proposals.

[1203] Step 6:

[1204] The server performs a simulation based on the new route plan it has generated. The new route plan generated in step 5 is used as input. The simulation predicts the number of passengers, sales, peak congestion times, etc., and verifies the operational efficiency and economy of the route. The output is a dataset of the simulation results.

[1205] Step 7:

[1206] The server visualizes and outputs the simulation results. It takes the simulation result data obtained in step 6 as input and visualizes it in the form of graphs and text reports. The generated visualization data is sent to smart glasses or other display devices so that the user can view it in real time. The output is a report containing the visualized information.

[1207] 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.

[1208] The present invention relates to a system that utilizes people flow data to generate new, efficient and profitable public transportation routes and makes proposals that also take into account user emotions.

[1209] Overall system overview

[1210] 1. Data Collection and Preprocessing

[1211] The server collects various people flow data via API, such as smartphone location information in the target area, public transportation usage data, and entrance data for commercial facilities.

[1212] The server then preprocesses the data, removing missing and outlier values, and splits it by time of day and day of the week to prepare it for analysis.

[1213] 2. Learning with AI models

[1214] The server inputs preprocessed people flow data into an AI model, which uses deep learning models and clustering algorithms to learn the relationship between departure and destination points, movement patterns, and usage trends by time of day.

[1215] During learning, the model parameters are adjusted to improve accuracy.

[1216] 3. Demand forecasting and analysis

[1217] The server uses a trained AI model to forecast demand. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, and other factors based on that.

[1218] 4. New Route Generation and Simulation

[1219] The server generates new transport route proposals based on the results of the demand forecast, and also calculates sales forecasts based on the number of users and operation costs to evaluate profitability.

[1220] The generated route plan is then verified through simulation, which involves making detailed predictions of operation schedules, daily passenger numbers, and peak congestion times to verify the validity of the operation plan.

[1221] 5. Sentiment analysis using an emotion engine

[1222] The server uses an emotion engine to analyze users' emotions based on collected pedestrian flow data and simulation results, quantitatively evaluating their emotional state and grasping their satisfaction or dissatisfaction with the proposed new route.

[1223] 6. Adjusting route plans based on emotional responses

[1224] The server simulates the user's emotional response to the generated new route plan and modifies it based on the results, for example, by improving areas that users dislike or adding new options.

[1225] 7. Visualizing and outputting results

[1226] Finally, the server visualizes the simulation results and the user sentiment analysis results as graphs and text reports, and outputs them to the user's terminal in a format that can be easily understood by transportation planners.

[1227] Users (transportation planners) can use this information to consider introducing new routes.

[1228] Specific examples

[1229] Consider a case where people's movement patterns and emotional responses are analyzed in detail based on smartphone location information from various areas of Tokyo.

[1230] 1. Data Collection

[1231] The server collects smartphone location data and public transport usage data for one year, including data broken down by time of day for each day.

[1232] 2. Pretreatment

[1233] The server removes missing values ​​and extreme outliers and splits the data by time of day and day of the week.

[1234] 3. Training the AI ​​model

[1235] The server performs a detailed analysis of the travel data from Shinjuku to Shibuya, inputs it into the AI ​​model, and the model is evaluated using validation data to set appropriate parameters.

[1236] 4. Demand forecasting and analysis

[1237] The server predicts the number of users and peak congestion times based on the time periods and days of the week with the highest demand between Shinjuku and Shibuya.

[1238] 5. New Route Generation and Simulation

[1239] The server generates a direct route from Shinjuku to Shibuya, and simulates the number of passengers, peak congestion, and sales forecasts for this route. Based on the simulation results, the operation schedule and economic efficiency are evaluated.

[1240] 6. Sentiment analysis

[1241] The server uses an emotion engine to analyze user emotions based on the collected data and simulation results, and evaluates reactions to new lines. For example, it visualizes users' expectations and concerns about a direct line from Shinjuku to Shibuya.

[1242] 7. Adjustment and Output

[1243] The server adjusts the proposed new routes based on the results of the sentiment analysis and generates a report with the final simulation results, which transportation planners can use to consider introducing new routes.

[1244] As described above, the present invention makes it possible to improve the efficiency of new transportation route planning, improve user convenience and satisfaction, and provide highly profitable public transportation services.

[1245] The processing flow will be explained below.

[1246] Step 1:

[1247] The server collects people flow data. Specifically, it obtains smartphone location information, public transportation usage data, and commercial facility entry data via API and stores it in a database in standard formats (CSV or JSON). The data collection period is generally one year, and includes information on weekdays, weekends, and time periods.

[1248] Step 2:

[1249] The server preprocesses the collected data. Specifically, it removes missing values ​​and outliers and standardizes the data format. It also divides the data by time period and day of the week and prepares it in a form suitable for analysis. This process uses data cleaning tools and programs.

[1250] Step 3:

[1251] The server inputs the preprocessed people flow data into an AI model, such as a deep learning model or a clustering algorithm. The data is split into training and validation sets, and the training data set is used to train the model.

[1252] Step 4:

[1253] The server trains the AI ​​model. Specifically, it uses a data set to analyze the relationship between origins and destinations, travel patterns, and usage trends by time of day. During the training process, the model's parameters are adjusted to improve accuracy.

[1254] Step 5:

[1255] The server uses the trained AI model to make demand forecasts. Specifically, it identifies combinations of departure and destination points with high demand, and then predicts the number of users, peak congestion times, travel distances, etc. based on those combinations. This allows it to grasp the demand distribution in a specific area.

[1256] Step 6:

[1257] The server generates new route proposals based on predicted demand. At the same time, it also calculates sales forecasts based on the number of passengers and operating costs, and evaluates profitability. Specifically, if demand between Shinjuku and Shibuya is high, it proposes a direct route.

[1258] Step 7:

[1259] The server then simulates the new route plan. The simulation makes detailed predictions of the operation schedule, daily passenger numbers, and peak congestion times, and verifies the validity of the operation plan. For example, it creates an operation schedule for a direct line between Shinjuku and Shibuya, and estimates the number of passengers and operation costs based on that schedule.

[1260] Step 8:

[1261] The server uses an emotion engine to analyze users' emotional states based on the collected pedestrian flow data and simulation results. Specifically, it analyzes user comments, social media data, survey results, etc., to quantitatively evaluate users' satisfaction or dissatisfaction with new route proposals.

[1262] Step 9:

[1263] The server adjusts new route proposals based on the results of sentiment analysis. Specifically, it improves areas where users are dissatisfied and considers additional options. For example, it proposes a service schedule that avoids peak hours, which users are likely to dissatisfy.

[1264] Step 10:

[1265] The server visualizes the simulation results and sentiment analysis results in the form of graphs and text reports, making them easy for transportation planners (users) to understand.

[1266] Step 11:

[1267] The server outputs visualized simulation results and adjusted route plans to the user's device. Based on this, transportation planners (users) can consider introducing new routes. For example, they can evaluate the profitability and operation schedule of a direct route between Shinjuku and Shibuya and make a final decision on its introduction.

[1268] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves user convenience and satisfaction, and enables the provision of highly profitable public transportation services.

[1269] Example 2

[1270] 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."

[1271] In conventional public transport route planning, demand forecasting and route proposal generation are often done manually, resulting in inefficiency and a lack of accuracy. Furthermore, it is not easy to adjust routes while taking into account user emotions and satisfaction, which can lead to dissatisfaction after the introduction of new routes. Therefore, there is a need for efficient and accurate proposals for new public transport routes and route adjustments that take into account user emotions.

[1272] 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.

[1273] In this invention, the server includes means for collecting people flow data, means for preprocessing the people flow data, means for training a generation AI that uses the preprocessed people flow data to learn relationships between departure points and destinations, travel patterns, and usage trends by time period, means for performing demand forecasts using the trained generation AI and identifying high-demand combinations of departure points and destinations, means for generating new route plans based on the demand forecast results, means for simulating the generated new route plans to predict operation schedules, ridership, peak congestion times, and sales, means for analyzing users' emotional states using an emotion engine and evaluating their satisfaction or dissatisfaction with the new route plans, means for revising the route plans based on users' emotional reactions to the generated new route plans, and means for visualizing and outputting the simulation results and emotion analysis results. This enables the proposal of efficient and profitable new public transportation routes and route adjustments that take users' emotions into consideration.

[1274] "People flow data" refers to data on people's movements, including individual movement patterns and location information, public transportation usage data, and commercial facility entry data.

[1275] "Preprocessing" refers to the process of removing missing or outliers from collected people flow data and dividing the data by time of day or day of the week, among other things, to prepare the data in a form suitable for analysis.

[1276] "Generative AI" is artificial intelligence that uses machine learning or deep learning to learn from data and perform demand forecasts and generate new routes.

[1277] "Demand forecasting" involves using trained generative AI to predict combinations that will be in high demand based on the relationship between people's departure and destination locations and their movement patterns.

[1278] A "new route proposal" is a proposal for a new public transportation route based on demand forecasts.

[1279] "Simulation" is the process of virtually calculating operation schedules, number of users, peak congestion times, sales forecasts, etc. based on the generated new route proposal.

[1280] The "emotion engine" is an artificial intelligence that analyzes the user's emotional state from collected pedestrian flow data and simulation results, and evaluates their satisfaction or dissatisfaction with new route proposals.

[1281] "Emotional response" refers to the emotional state, such as expectations or concerns, that a user shows toward a new route proposal.

[1282] "Visualization" is a method of displaying simulation results and sentiment analysis results as graphs or text reports to communicate them to people in an easy-to-understand manner.

[1283] The present invention relates to a system that utilizes people flow data to generate new, efficient and profitable public transportation routes and makes proposals that take user emotions into consideration. An embodiment of this system is described in detail below.

[1284] Overall system overview

[1285] This system consists of the following main hardware and software:

[1286] Server: Executes functions such as people flow data collection, preprocessing, AI model training, demand forecasting, new route plan generation, simulation, sentiment analysis, visualization and output.

[1287] User terminal: A terminal that receives and displays the simulation results and sentiment analysis results sent from the server.

[1288] Network: Infrastructure for data communication between servers and user devices.

[1289] Data collection and preprocessing

[1290] The server collects various people flow data via API, such as smartphone location information for the target area, public transportation usage data, and entrance data for commercial facilities. At this stage, the specific software used is a library for handling HTTP requests (e.g., requests for Python).

[1291] The server then preprocesses the collected data to remove missing and outlier values, using Python's Pandas library to manipulate data frames and NumPy for numerical data processing.

[1292] Learning with AI models

[1293] The server inputs the preprocessed people flow data into an AI model, which learns the relationship between origins and destinations, travel patterns, and usage trends by time of day. The specific software used is a deep learning framework such as TensorFlow or PyTorch.

[1294] During training, the model parameters are adjusted to find optimal hyperparameters (learning rate, number of epochs, batch size, etc.).

[1295] Demand forecasting and analysis

[1296] The server uses a trained AI model to forecast demand, particularly by identifying combinations of origins and destinations with high demand, and then predicting factors such as the number of passengers, peak congestion times, and travel distances.

[1297] This analysis uses graph drawing libraries such as Matplotlib and Seaborn to visually display the prediction results.

[1298] New route generation and simulation

[1299] The server generates new transportation route proposals based on the demand forecast results, which include information such as departure points, destinations, departure times, and arrival times.

[1300] The server then performs further simulations to generate detailed data such as operation schedules, daily ridership, peak congestion times, sales forecasts, etc. Simulation software such as AnyLogic can be used for this simulation.

[1301] Sentiment analysis with emotion engine

[1302] The server uses an emotion engine to analyze user emotions based on collected pedestrian flow data and simulation results. Specifically, it calculates emotion scores using natural language processing (NLP) libraries (e.g., SpaCy and BERT).

[1303] Adjusting route plans based on emotional responses

[1304] The server simulates the user's reaction to the generated new route plan based on the results of the emotion engine, and modifies the route plan based on the results, thereby providing a new route plan that will improve user satisfaction.

[1305] Visualizing and outputting results

[1306] The server visualizes the simulation results and sentiment analysis results as graphs and text reports, and sends them to the user's device, allowing the user to consider introducing new routes based on the results.

[1307] The specific tools used are Plotly and Matplotlib for visualization.

[1308] Specific examples

[1309] As a case study of Tokyo, we use one year's worth of smartphone location data and public transportation usage data. The server collects and preprocesses this data, then performs a detailed analysis of the travel data from Shinjuku to Shibuya and trains the AI ​​model. Demand forecasting identifies the time periods and days of the week with the highest demand between Shinjuku and Shibuya, and predicts the number of users and peak congestion times. The server then generates a direct route from Shinjuku to Shibuya and simulates the number of users, peak congestion, and sales forecasts for this route. Finally, an emotion engine is used to analyze user emotions, visualize expectations and concerns about the new route, and provide optimal route suggestions.

[1310] Examples of prompt statements

[1311] For example, the prompt we use for our generative AI model is:

[1312] "Using one year of smartphone location data and public transport usage data in Tokyo, please predict travel patterns and demand from Shinjuku to Shibuya. Based on the results, please generate direct route plans from Shinjuku to Shibuya and analyze user responses using an emotion engine."

[1313] As described above, the present invention improves the efficiency of new route planning for transportation facilities, improves user convenience and satisfaction, and enables the provision of highly profitable public transportation services.

[1314] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1315] Step 1: Data collection

[1316] The server collects people flow data, such as smartphone location information, public transportation usage data, and commercial facility entrance data, for the target area via API. Specifically, the server sends an HTTP request to the API and retrieves data in JSON format. The input is location information and usage data from the API, and the output is raw data stored on the server.

[1317] Step 2: Data Preprocessing

[1318] The server preprocesses the collected data. Specifically, it uses Python's Pandas library to create a data frame and remove missing values ​​and outliers. Rows with missing values ​​are dropped or imputed with the median. Next, the data is split by time of day and day of the week, and processes such as standardization and normalization are performed. The input is the raw data from step 1, and the output is the preprocessed data.

[1319] Step 3: Training the AI ​​model

[1320] The server trains an AI model using the preprocessed data. The specific software used is TensorFlow or PyTorch. The data is divided into training data and validation data, a deep learning model is defined, and training begins with the model.fit() function. Hyperparameters (learning rate, number of epochs, batch size, etc.) are adjusted, and performance is evaluated after training using model.evaluate(). The input is the preprocessed data, and the output is the trained model.

[1321] Step 4: Demand forecasting and analysis

[1322] The server uses the trained AI model to perform demand forecasting. It executes the forecast using the model.predict() function to identify combinations of origins and destinations with high demand. Matplotlib and Seaborn are used to visually display the forecast results as heat maps and time series graphs. The input is the trained model and preprocessed data, and the output is the demand forecast results.

[1323] Step 5: Create a new route

[1324] The server generates new transport route proposals based on the demand forecast results. Route proposals are created that include departure points, destinations, departure times, arrival times, etc. Routes with high demand are selected from the forecast results, and an optimal route proposal is created based on these. The input is the demand forecast results, and the output is a new route proposal.

[1325] Step 6: Simulation

[1326] The server runs a simulation based on the generated new route plan. Specifically, it uses simulation software such as AnyLogic to calculate operation schedules, daily passenger numbers, peak congestion times, and sales forecasts. The simulation results are exported as Excel or CSV files. The input is the new route plan, and the output is the simulation results.

[1327] Step 7: Sentiment analysis

[1328] The server analyzes the user's emotions using an emotion engine based on the collected data and simulation results. Specifically, it uses NLP libraries such as SpaCy and BERT to calculate the emotion score using get_sentiment_score(). The input is the simulation result data, and the output is the emotion analysis result.

[1329] Step 8: Adjust the route plan

[1330] The server adjusts the new route proposal based on the results of the sentiment analysis. Specifically, it considers user satisfaction and dissatisfaction from the sentiment score and improves the operation schedule and routes. It then performs a simulation again to find the optimal route proposal. The input is the sentiment analysis results and the original route proposal, and the output is the adjusted new route proposal.

[1331] Step 9: Visualizing and outputting results

[1332] The server visualizes the simulation results and sentiment analysis results as graphs and text reports. It uses Matplotlib and Plotly to create visually easy-to-understand graphs and charts. The generated report is sent to the user's device. The input is the adjusted new route plan and the simulation results, and the output is a visualized report.

[1333] These are the specific processing steps of the program of this system. This series of steps makes it possible to propose new, efficient and profitable public transportation routes and adjust routes while taking user emotions into consideration.

[1334] (Application example 2)

[1335] 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."

[1336] Existing public transport routes face difficulties in efficiently responding to demand and optimizing operating costs, resulting in low user satisfaction. Furthermore, route planning does not take into account user emotions and reactions, which often leads to user dissatisfaction when new routes are introduced. There is a need to solve these issues and plan and operate public transport routes that are efficient and provide high user satisfaction.

[1337] The identification processing by the identification 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 people flow data, means for preprocessing the people flow data, means for training a generation AI that performs demand forecasting using the preprocessed people flow data, means for generating new route plans using the trained generation AI, means for performing simulations based on the generated new route plans, means for evaluating reactions to the new route plans using an emotion engine that analyzes user emotions, means for adjusting the new route plans based on the emotion analysis results, and means for visualizing and outputting the simulation and emotion analysis results. This enables efficient route planning based on demand and optimal operation plans that take user emotions and reactions into consideration.

[1338] (Definitions of important words)

[1339] "People flow data" is data that represents people's movements and includes a variety of data sources, such as location information, public transportation usage data, and entrance data for commercial facilities.

[1340] "Preprocessing" refers to the process performed to prepare raw data in an analyzable format, such as removing missing values ​​and outliers, and standardizing and splitting the data.

[1341] "Generative AI" is a system that uses artificial intelligence technology and has the learning capabilities to predict demand and recognize patterns from large amounts of data.

[1342] A "new route proposal" is a newly proposed public transportation route that differs from existing transportation routes.

[1343] "Simulation" is the act of virtually predicting and verifying various factors in actual operation (number of users, sales, peak congestion times, etc.) based on the generated new route plan.

[1344] An "emotion engine" is a technology for analyzing a user's emotions and reactions, and quantitatively evaluates the emotional state based on collected data and simulation results.

[1345] "Emotion analysis" involves using an emotion engine to analyze users' emotions and reactions, and is carried out to understand their satisfaction or dissatisfaction with new route proposals.

[1346] "Visualization" is the act of visually displaying analysis or simulation results as graphs, charts, or text reports, providing information in an easy-to-understand format.

[1347] To implement the present invention, the following configuration and procedures are used.

[1348] First, the server collects people flow data. Specifically, it collects location information obtained from smartphones, public transportation usage data, and commercial facility entry data via API. Standard API access methods are used to collect data, with appropriate authentication and access control for each data source.

[1349] The collected data is preprocessed on the server using the Pandas library to remove missing values ​​and outliers and extract only the necessary information, preparing the data in a form suitable for analysis.

[1350] Next, the preprocessed data is input into the generative AI, which trains it to make demand forecasts. This training uses the Scikit-learn library, which performs clustering and regression analysis. By learning from large amounts of data, the generative AI model is able to understand the relationship between departure and destination points and usage trends by time of day, allowing it to make more accurate predictions.

[1351] The trained generative AI model generates new route proposals by applying a clustering algorithm using KMeans to identify routes between points with high demand, while also calculating operating costs and predicted revenue.

[1352] The generated new route plan is then simulated. The simulation predicts the number of passengers, sales, peak congestion times, etc., and aims to optimize operation. The simulation results are visualized as graphs using the Matplotlib library.

[1353] Furthermore, an emotion engine is used to analyze user emotions. The emotion engine analyzes user emotions based on pre-collected data and simulation results, and evaluates their satisfaction or dissatisfaction with the new route plan. The emotion analysis results are then used to further adjust the route plan.

[1354] Finally, all data and analysis results are visualized and provided to transportation planners in a report detailing proposed new routes, simulation results, and user sentiment analysis, enabling planners to consider new routes that are efficient and offer high user satisfaction.

[1355] As a concrete example, "The following prompt sentence is input into the generative AI model: 'Last night, it's early morning. How can we improve convenience? How can we change the main travel route? Based on the traffic volume forecast results, please suggest a route for a new autonomous vehicle.'" Through such prompt sentences, specific questions are posed to the generative AI model.

[1356] As described above, the present invention makes it possible to provide a system that generates efficient and profitable public transportation routes by combining demand prediction based on people flow data with user emotion analysis.

[1357] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1358] Step 1:

[1359] The server collects people flow data. Smartphone location information, public transportation usage data, and commercial facility entrance data are obtained via API. Various data sources are used as input, and the collected data is stored on the server.

[1360] Step 2:

[1361] The data collected by the server is preprocessed. Using the Pandas library, missing values ​​and outliers are removed, and the data is split and reshaped by time of day and day of the week. The input is the people flow data collected in Step 1, and the output is preprocessed data in a format suitable for analysis.

[1362] Step 3:

[1363] The server uses the preprocessed data to train the generative AI. It uses the Scikit-learn library to perform clustering and regression analysis to learn the relationship between departure and destination locations and travel patterns. The input is the preprocessed data, and the output is a trained generative AI model.

[1364] Step 4:

[1365] The server generates new route proposals using a trained generative AI model. A clustering algorithm using KMeans is used to identify routes between points with high demand, and calculates operating costs and predicted sales. The input is the trained generative AI model and analysis data, and the output is new route proposals.

[1366] Step 5:

[1367] The server performs a simulation based on the generated new route plan. The simulation predicts the number of passengers, sales, peak congestion times, etc., and aims to optimize operations. The input is the new route plan, and the output is the simulation results.

[1368] Step 6:

[1369] The server uses an emotion engine to analyze the user's emotions. Based on the collected data and simulation results, the user's emotions are quantitatively evaluated. The input is the collected data and simulation results, and the output is the emotion analysis results.

[1370] Step 7:

[1371] The server adjusts the new route proposal based on the results of the sentiment analysis, correcting areas of user dissatisfaction and adding new options. The input is the sentiment analysis results, and the output is the adjusted new route proposal.

[1372] Step 8:

[1373] The server visualizes the simulation and sentiment analysis results and provides them to transportation planners as a report. Collected data, analysis results, and new route proposals are displayed in graphs, charts, and text. The input is the adjusted new route proposal and simulation results, and the output is a visualized report.

[1374] Step 9:

[1375] The user (transportation planner) considers the introduction of new routes based on the visualized report. This makes it possible to create new routes that are efficient and have high user satisfaction. The only input required is the visualized report.

[1376] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice 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 voice data.

[1377] 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.

[1378] 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 robot 414.

[1379] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1380] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1381] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1382] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1383] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1384] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1385] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1386] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1387] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1388] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1389] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1390] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1391] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1392] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1393] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1394] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1395] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1396] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1397] The following is further disclosed regarding the above embodiment.

[1398] (Claim 1)

[1399] a means of collecting people flow data;

[1400] a means for preprocessing people flow data;

[1401] A means of training a generation AI that uses preprocessed people flow data to make demand forecasts;

[1402] A means for generating new route plans using the learned generation AI;

[1403] A means for performing a simulation based on the generated new route plan;

[1404] A means for visualizing and outputting the simulation results;

[1405] A system including:

[1406] (Claim 2)

[1407] The system of claim 1 generates new route proposals based on the most common combinations of departure points and destinations in a specific area.

[1408] (Claim 3)

[1409] The system according to claim 1, wherein the simulation predicts the number of users, sales, and peak congestion times.

[1410] "Example 1"

[1411] (Claim 1)

[1412] a means of collecting people flow data;

[1413] a means for preprocessing people flow data;

[1414] A means of training a generation AI that uses preprocessed people flow data to make demand forecasts;

[1415] A means for generating new route plans using the learned generation AI;

[1416] A means for performing a simulation based on the generated new route plan;

[1417] A means for visualizing and outputting the simulation results;

[1418] A means of analyzing operational costs and sales forecasts based on simulation results,

[1419] A means for evaluating the economic efficiency of the generated route proposal;

[1420] A system including:

[1421] (Claim 2)

[1422] The system according to claim 1, further comprising means for generating new route proposals based on the most common combinations of origins and destinations in a specific area.

[1423] (Claim 3)

[1424] 2. The system according to claim 1, further comprising means for predicting the number of users, operation costs, sales, and peak congestion times in the simulation.

[1425] "Application Example 1"

[1426] (Claim 1)

[1427] a means of collecting people flow data;

[1428] a means for preprocessing people flow data;

[1429] A means for training a generative AI model that performs demand forecasting using preprocessed people flow data;

[1430] a means for visualizing the demand data for display on the smart glasses;

[1431] A means for generating new route plans using the learned generative AI model; and

[1432] A means for performing a simulation based on the generated new route plan;

[1433] A means for visualizing and outputting the simulation results;

[1434] A system including:

[1435] (Claim 2)

[1436] The system of claim 1 generates new route proposals based on the most common combinations of departure points and destinations in a specific area.

[1437] (Claim 3)

[1438] The system according to claim 1, wherein the simulation predicts the number of users, sales, and peak congestion times.

[1439] "Example 2: Combining Emotion Engines"

[1440] (Claim 1)

[1441] a means of collecting people flow data;

[1442] a means for preprocessing people flow data;

[1443] A means of training a generative AI that uses preprocessed people flow data to learn the relationship between departure and destination, movement patterns, and usage trends by time of day.

[1444] A means to use the learned generative AI to forecast demand and identify high-demand origin-destination combinations; and

[1445] A means for generating new route proposals based on the demand forecast results;

[1446] A simulation means for predicting operation schedules, number of users, peak congestion times, and sales based on the generated new route proposal;

[1447] a means for analyzing a user's emotional state using an emotion engine to assess their satisfaction or dissatisfaction with the proposed new route;

[1448] a means for modifying the route plan based on the user's emotional response to the generated new route plan;

[1449] a means for visualizing and outputting the simulation results and the sentiment analysis results;

[1450] A system including:

[1451] (Claim 2)

[1452] The system of claim 1 generates new route proposals based on the most common combinations of departure points and destinations in a specific area.

[1453] (Claim 3)

[1454] The system according to claim 1, wherein the simulation predicts the number of users, sales, and peak congestion times.

[1455] "Application example 2 when combining emotion engines"

[1456] (Claim 1)

[1457] a means of collecting people flow data;

[1458] a means for preprocessing people flow data;

[1459] A means of training a generation AI that uses preprocessed people flow data to make demand forecasts;

[1460] A means for generating new route plans using the learned generation AI;

[1461] A means for performing a simulation based on the generated new route plan;

[1462] A means for evaluating reactions to new route proposals using an emotion engine that analyzes user emotions;

[1463] a means for adjusting new route proposals based on the results of sentiment analysis;

[1464] A means to visualize and output the results of the simulation and emotion analysis;

[1465] A system including:

[1466] (Claim 2)

[1467] The system according to claim 1, wherein new route proposals are generated based on the most common combinations of departure points and destinations in a specific area, and the route proposals are further adjusted taking into account the user's emotional evaluation.

[1468] (Claim 3)

[1469] 2. The system according to claim 1, wherein the simulation predicts the number of users, sales, and peak congestion times, and visualizes and outputs these results together with the results of user sentiment analysis. [Explanation of symbols]

[1470] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means of collecting people flow data; a means for preprocessing people flow data; A means of training a generation AI that uses preprocessed people flow data to make demand forecasts; A means for generating new route plans using the learned generation AI; A means for performing a simulation based on the generated new route plan; A means for visualizing and outputting the simulation results; A system including:

2. The system according to claim 1, wherein new route proposals are generated based on the most common combinations of departure points and destinations in a specific area.

3. The system according to claim 1, wherein the simulation predicts the number of users, sales, and peak congestion times.

Images