System

Abstract

A system is provided.SOLUTION: The system according to claim 1, further comprising: means for calculating an optimum delivery route using the generated AI; means for transmitting the calculated delivery route to the terminals; means for recalculating the route in real time and transmitting the route to the terminals again when the route needs to be changed; and means for receiving feedback after the completion of the delivery and reflecting the feedback on the next route generation.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] The logistics industry is facing increasingly serious challenges, including an increase in redeliveries, a driver shortage, and the 2024 problem. These challenges are reducing the efficiency of logistics networks and increasing the workload of drivers. Furthermore, an increase in redeliveries increases delivery costs and the environmental burden. The present invention aims to solve these challenges, improve logistics efficiency, and reduce the workload of drivers. [Means for solving the problem]

[0005] The present invention solves the above problems by the following means. First, it provides a means for acquiring delivery performance data, delivery locker availability, traffic congestion, and weather data. Next, it provides a means for preprocessing the acquired data and calculating the rate at which delivery destinations are at home based on past delivery performance. It also provides a means for calculating the optimal delivery route using generation AI. The calculated delivery route is sent to a terminal and displayed on the terminal to the driver. If a change to the route is necessary, it also includes a means for recalculating a new route in real time and sending it again to the terminal. Finally, it has a means for receiving feedback after delivery is completed and reflecting this in the generation of the next route. This series of systems aims to improve delivery efficiency, reduce redeliveries, and solve logistics issues.

[0006] "Delivery performance data" refers to data that includes information on the date and time, address, and success or failure of past deliveries.

[0007] "Delivery box availability" is information relating to the usage status of the delivery box, and specifically, is data indicating whether the delivery box is available or in use.

[0008] "Traffic congestion status" is information indicating the degree of congestion of traffic flow on a road, and is data reflecting the difficulty of travelling on a road during a particular time period.

[0009] "Weather data" is information about the weather in a specified area, including weather conditions such as sunny, rainy, and snowy, and forecasts of those conditions.

[0010] "Data preprocessing" is the process of standardizing the format of acquired data and removing incomplete and duplicate data.

[0011] The "at-home rate" is an index that indicates the probability that a delivery recipient will be at home during a specific time period, calculated from past delivery performance data.

[0012] "Generative AI" refers to artificial intelligence systems that use techniques such as machine learning and deep learning to perform specific tasks.

[0013] An "optimal delivery route" is a route calculated to efficiently deliver to a specified delivery destination, and is generated taking into account factors such as time, cost, the rate at which people are at home, and traffic congestion.

[0014] "Terminal" means an electronic device used by a delivery driver, including a smartphone, tablet, or other device.

[0015] "Feedback" refers to information and opinions provided by delivery drivers after completing a delivery, and is data that will be useful when generating the next delivery route. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

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

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

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

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0031] 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.

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

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] This invention is a system for improving the efficiency of logistics, which uses generative AI to calculate optimal delivery routes and provide them to truck drivers. This system is based mainly on interactions between a server, terminals, and users, and detailed embodiments are described below.

[0038] Data collection and preprocessing

[0039] First, the server collects delivery history data, delivery box availability data, traffic congestion data, and weather data. Delivery history data includes the date, time, address, and delivery success / failure information for past deliveries. Delivery box availability data is obtained in real time and reflects the location and usage status of each box. Traffic congestion and weather data are collected using an external API.

[0040] The server then pre-processes this data. From the delivery performance data, the rate of at-home delivery during a specific time period is calculated. Data pre-processing cleans incomplete and duplicate data.

[0041] Calculating the best route

[0042] The server inputs the preprocessed data into the generation AI to calculate the optimal delivery route. The generation AI takes into account past performance, the current state of the delivery lockers, traffic congestion, and weather data to generate an efficient route.

[0043] For example, based on past delivery data, the system can prioritize deliveries to areas with a high rate of people at home in the morning, and calculate routes that avoid traffic jams using real-time traffic information.In addition, if the weather is bad, it can select a route that uses many delivery boxes.

[0044] Route distribution and display

[0045] The server sends the calculated route to each driver's device. The device then displays the received route information to the user. The display format is a map or list, and includes delivery destinations, routes, and important points. For example, if the next delivery destination is "XXX Building," specific instructions such as "Turn right -> Go straight -> Turn left" are displayed.

[0046] Real-time updates and feedback

[0047] If the user (driver) needs to change the route during a delivery, the information is sent from the device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the user provides feedback, which is sent from the device to the server. The server analyzes the feedback and uses it to generate the next route.

[0048] Specific examples

[0049] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses with a high rate of people being at home around 10:00 AM when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C." The device that receives the information displays detailed route information to the driver along with a map.

[0050] If the driver encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and submits a new route. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to generate future routes.

[0051] In this way, the system of the present invention improves delivery efficiency and reduces redelivery, thereby contributing to solving logistics issues.

[0052] The processing flow will be explained below.

[0053] Step 1:

[0054] When the server logs in, it retrieves delivery performance data for the delivery area over the past year from the database. This data includes the address of each delivery destination, delivery time, and delivery success / failure information.

[0055] Step 2:

[0056] The server retrieves current availability data from the delivery locker API in real time. The dataset includes the ID, location, and usage status (available / in use) of each delivery locker.

[0057] Step 3:

[0058] The server retrieves traffic and weather data for the day from an external API (e.g., Google Maps API or Japan Meteorological Agency API), including the degree of traffic congestion on major roads and weather conditions (sunny, rainy, snowy, etc.).

[0059] Step 4:

[0060] The server cleanses the data it receives, removes incomplete and duplicate data, completes it, and standardizes the format of all data.

[0061] Step 5:

[0062] The server calculates the percentage of delivery destinations at home during a specific time period based on past delivery performance data. For example, it identifies delivery destinations with a high percentage of delivery destinations at home between 9:00 AM and 11:00 AM.

[0063] Step 6:

[0064] The server receives a list of packages to be delivered and delivery area information, including the delivery address and desired delivery time for each package.

[0065] Step 7:

[0066] The server uses the generated AI to calculate the optimal delivery route based on the pre-processed data. The AI ​​model is input with collected past data, the current availability of delivery boxes, traffic congestion information, weather conditions, etc.

[0067] Step 8:

[0068] The server sends the optimal route calculated as a result to each driver's device.

[0069] Step 9:

[0070] The device displays detailed route information received to the user (driver) in map or list format, showing the next delivery destination, specific route, and important points to note.

[0071] Step 10:

[0072] If the user (driver) needs to change the route during a delivery, the device sends that information to the server, which recalculates the route in real time based on the new data and sends the new route back to the device.

[0073] Step 11:

[0074] After completing a delivery, the user (driver) enters the reason for the delivery delay or any problems that occurred as feedback into the terminal.

[0075] Step 12:

[0076] The device sends the received feedback data to the server, which analyzes the collected feedback and reflects it in the next route generation.

[0077] These processing steps create a system that improves delivery efficiency and reduces redelivery.

[0078] Example 1

[0079] 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."

[0080] Conventional logistics systems have problems with insufficient optimization of delivery routes, making it difficult to achieve efficient deliveries. It is also difficult to respond in real time to traffic congestion and weather changes, leading to redeliveries and delivery delays. Furthermore, there is an insufficient system for utilizing feedback from drivers in the next delivery, making it difficult to improve delivery efficiency.

[0081] 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.

[0082] In this invention, the server includes means for acquiring delivery-related data, means for acquiring the usage status of delivery boxes, means for externally acquiring traffic and weather information, means for preprocessing the acquired data, means for calculating an optimal delivery route using generative artificial intelligence, means for transmitting the calculated delivery route to the terminal, means for recalculating the route in real time and transmitting it again to the terminal if a route change is necessary, means for receiving feedback after delivery is completed and reflecting it in the generation of the next route, means for providing the driver with route information displayed on the terminal, and means for receiving and processing route change requests from the terminal during delivery. This enables efficient delivery route optimization and real-time response. Furthermore, by reflecting driver feedback in the next delivery, continuous improvement in delivery efficiency can be achieved.

[0083] "Delivery-related data" means data containing information about past and current deliveries, including delivery dates and times, addresses, and delivery success and failure information.

[0084] "Delivery box usage status" is data that indicates the current usage status of each delivery box, and includes the location of the box, availability status, usage history, etc.

[0085] "Traffic information" is data that indicates the current traffic situation, including road congestion and accident information.

[0086] "Weather information" refers to data that indicates current and forecast weather conditions, including precipitation, temperature, wind speed, and the like.

[0087] "Preprocessing" refers to the process of preparing acquired data so that it can be analyzed, and includes the deletion of incomplete data, the integration of duplicate data, and the extraction of necessary data.

[0088] "Generative artificial intelligence" is a system that uses artificial intelligence techniques to generate new information based on data, which is then used to calculate optimal delivery routes.

[0089] An "optimal delivery route" is a delivery sequence and route that maximizes delivery efficiency and aims to reduce time and costs.

[0090] A "terminal" is an information device used by a driver to display delivery routes and route information and to communicate with the server.

[0091] "Recalculation in real time" is a process that instantly calculates and provides new delivery routes based on current conditions.

[0092] "Feedback" is information provided by the driver after delivery, including reasons for delivery delays and special notes from the delivery destination.

[0093] "Means for providing route information to the driver" refers to a method for communicating the calculated delivery route and detailed instructions to the driver via the terminal.

[0094] The "means for receiving and processing a route change request from a terminal" is a method in which a server receives a route change request sent by a driver from a terminal, calculates and provides a new route based on the request.

[0095] This invention is a system for improving logistics efficiency, which uses generative artificial intelligence (generative AI model) to calculate optimal delivery routes and provide them to drivers. This system consists of a server, terminals, and users, and handles the collection, preprocessing, analysis, display, and feedback of delivery-related data.

[0096] Data collection and preprocessing

[0097] server

[0098] The server collects data using the following means:

[0099] 1. Delivery-related data: Obtain past delivery records from the database. For example, obtain data such as "2023-10-01, Address A, Delivery successful."

[0100] 2. Delivery locker usage status: Obtain real-time data using the API of an external delivery locker management system. For example, obtain data such as "Delivery locker B, available."

[0101] 3. Traffic and weather information: Get data from traffic APIs and weather APIs. For example, get data such as "Road C, traffic jam" or "Area D, rain."

[0102] The server then preprocesses the data, calculating the percentage of people at home during specific time periods from delivery-related data and cleaning up incomplete and duplicate data.

[0103] Calculating the best route

[0104] server

[0105] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route. This generative AI model takes into account delivery history, the current state of the parcel lockers, traffic congestion, and weather data to generate an efficient delivery route.

[0106] Specifically, the generative AI model calculates the following route:

[0107] 1. Prioritize deliveries to areas where people are more likely to be at home in the morning.

[0108] 2. Select a route that avoids traffic jams based on real-time traffic information.

[0109] 3. If the weather is bad, choose a route that uses more delivery boxes.

[0110] Route distribution and display

[0111] Servers and Terminals

[0112] The server sends the calculated delivery route to each driver's device. The device that receives the route information displays it to the user (driver). The display format is a map or list, and includes delivery destinations, routes, and important points. Specific examples of instructions include detailed instructions such as "Next delivery destination is XXX Building" and "Turn right -> go straight -> turn left."

[0113] Real-time updates and feedback

[0114] Users and Servers

[0115] The user (driver) can input new information during the delivery. For example, if the user encounters a traffic jam, the user can report the situation to the server from the terminal: "Road G, traffic jam starting."

[0116] The server recalculates the route in real time based on the new data and resends it to the device. After the delivery is completed, the user can provide feedback such as the reason for the delivery delay or any special notes at the delivery destination, and the server will reflect this in the next route generation.

[0117] Examples and prompts

[0118] Specific examples

[0119] One day, a driver is scheduled to make five deliveries. Based on past performance data, the server prioritizes addresses with a high rate of people at home around 10:00 a.m. in the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most vacant boxes. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C." The device that receives the route displays detailed route information to the driver along with a map. If the driver encounters traffic congestion along the way, new information is sent from the device to the server. The server recalculates the route in real time and presents a new route. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to generate future routes.

[0120] Prompt example

[0121] "Generate the optimal delivery route to efficiently complete five deliveries based on past delivery data, drop-box availability, traffic information, and weather data."

[0122] In this way, the system of the present invention enables efficient delivery routes and real-time responses, and by reflecting driver feedback in the next delivery, it achieves continuous improvement in delivery efficiency.

[0123] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0124] Step 1: Collect data

[0125] The server collects delivery-related data, parcel box usage, traffic information, and weather information.

[0126] Input: Requests to external APIs or databases

[0127] Data processing: Receives the response from the API, analyzes its contents, and saves it in the database.

[0128] Output: Raw data awaiting preprocessing

[0129] What it does: Performs database queries and external API calls to gather delivery records, parcel locker status, and real-time traffic and weather information.

[0130] Step 2: Preprocessing the data

[0131] The server pre-processes the collected data.

[0132] Input: Delivery-related data, parcel locker usage, traffic information, weather information

[0133] Data processing: missing data imputation, duplicate data removal, outlier detection and correction

[0134] Output: Cleansed data

[0135] What it does: It runs a data cleansing algorithm to remove incomplete records, merge duplicates, and calculates the percentage of people at home for a specific time period from delivery performance data.

[0136] Step 3: Calculate the optimal route

[0137] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route.

[0138] Input: Cleansed delivery data, parcel locker usage, traffic information, weather information

[0139] Data processing: All data is integrated and input into a generative AI model, which then runs an algorithm to generate the optimal route.

[0140] Output: Optimal delivery route

[0141] How it works: The generative AI model plans the optimal delivery route based on past delivery performance, real-time traffic and weather information, and available delivery lockers.

[0142] Step 4: Distributing the Route

[0143] The server sends the calculated delivery route to each driver's device.

[0144] Input: Optimal delivery route data

[0145] Data processing: Converting route information into a format that can be processed by the device

[0146] Output: Route information sent to the device

[0147] Specific operation: Encode the optimal route in JSON format or similar and send it to the device using a communication protocol (e.g., HTTP).

[0148] Step 5: View Routes

[0149] The terminal displays the received route information to the user (driver).

[0150] Input: Route information sent

[0151] Data processing: Converting route information into a visually understandable format

[0152] Output: Route information displayed in map and list format

[0153] What it does: Renders route information on the interface of a map or delivery app, showing instructions such as "Next delivery stop is XXX building" or "Turn right -> go straight -> turn left."

[0154] Step 6: Real-time updates

[0155] The server receives route change requests from users (drivers) and recalculates the route in real time.

[0156] Input: Route change request (e.g., "Route G, traffic jam starting")

[0157] Data processing: Integrate new situational data and recalculate the optimal route

[0158] Output: Updated optimal route

[0159] Specific operation: Re-run the generative AI model based on the new traffic information, generate a new route, and send it to the device.

[0160] Step 7: Use the feedback

[0161] The server receives feedback from the user (driver) after the delivery is completed and reflects it in the next route generation.

[0162] Input: Feedback data (e.g. "Delivery destination X, address unknown")

[0163] Data processing: Analyze feedback data and adjust parameters of route generation algorithm

[0164] Output: Improved route generation algorithm

[0165] Specific behavior: Save the feedback in a database and use it the next time you generate a route, or take other measures to improve it.

[0166] (Application example 1)

[0167] 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."

[0168] In logistics, conventional delivery route calculation systems have not been able to sufficiently improve delivery efficiency. In particular, it is difficult to provide optimal routes that take into account the rate at which recipients are at home, real-time traffic conditions, and weather conditions. This also makes it inefficient for drivers to check information en route. Additionally, the cost and time wasted by redelivery have also become an issue.

[0169] 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.

[0170] In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery boxes, means for externally acquiring traffic congestion and weather data, means for pre-processing the acquired data, means for calculating an optimal delivery route using a generation AI, means for transmitting the calculated delivery route to a terminal and displaying it on the smart glasses, means for recalculating the route in real time and transmitting it again to the terminal if a change in the route is necessary, and means for receiving feedback after delivery is completed and reflecting it in the next route generation, thereby enabling improved delivery efficiency and providing information in real time.

[0171] "Delivery performance data" refers to data that includes information such as the date and time of past deliveries, addresses, and delivery success / failure information.

[0172] "Delivery box availability" is data that shows the location and usage status of each box in real time.

[0173] "Traffic congestion status" is data indicating the state of traffic congestion on a road, and is acquired from an external source.

[0174] "Weather data" is data that indicates the weather forecast and current weather conditions for the area.

[0175] "Preprocessing" is a process of cleansing acquired data and eliminating incomplete data and duplicate data.

[0176] "Generative AI" is an artificial intelligence model that generates or predicts something based on given data.

[0177] The "optimal delivery route" is an efficient delivery route that minimizes time and costs, taking into account factors such as the rate at which recipients are at home, traffic congestion, and weather conditions.

[0178] "Device" means a device used by a driver, including a smartphone, tablet, smart glasses, etc.

[0179] "Smart glasses" are glasses-type devices that can transparently display visual information, allowing drivers to check the information hands-free.

[0180] "Recalculation in real time" refers to the process of instantly calculating a new route in response to changes in the situation and sending it back to the terminal.

[0181] "Feedback" refers to the reason for the delivery delay and other information provided by the driver after completing the delivery, and is data used to plan the next route.

[0182] To implement this invention, a server, smart glasses, other terminals, a generative AI model, an external API, etc. are used. Specific embodiments are described below.

[0183] Data collection and preprocessing

[0184] The server collects delivery performance data, delivery locker availability data, traffic congestion information, and weather data from external APIs. Delivery performance data includes the date and time, address, and success / failure information of past deliveries. Delivery locker availability information is obtained in real time and reflects the location and usage status of each locker. Traffic congestion and weather data are collected using external APIs.

[0185] The server then pre-processes this data, which includes cleaning out incomplete and duplicate data, and calculating the percentage of customers at home during specific time periods from the delivery performance data.

[0186] Calculating the best route

[0187] The server inputs the preprocessed data into a generative AI model to calculate the optimal delivery route. The generative AI model generates an efficient route by taking into account past delivery performance, the current state of delivery boxes, traffic congestion, and weather data. For example, it can prioritize deliveries to areas with a high rate of people at home in the morning based on past delivery data, and calculate a route that avoids traffic congestion using real-time traffic information. In addition, when the weather is bad, it can select a route that uses many delivery boxes.

[0188] Route distribution and display

[0189] The server sends the calculated route to the driver's device, which is a pair of smart glasses. The smart glasses visually display the calculated route information to the driver, allowing them to check the information hands-free. The display format is a map or list, and includes the next delivery destination, route, and important points. For example, if the next delivery destination is "AAA Building," specific instructions such as "turn right -> go straight -> turn left" are displayed.

[0190] Real-time updates and feedback

[0191] If the driver needs to change the route during a delivery, the information is sent to the server in real time via the smart glasses. The server recalculates the route in real time based on the new data and displays the new route on the smart glasses. After completing the delivery, the driver provides feedback, which is sent to the server via the smart glasses. The server analyzes the feedback and uses it to generate the next route.

[0192] Specific examples

[0193] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses where the number of people at home around 10:00 AM is high when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is displayed on the driver's smart glasses in the form of "Point A -> Point B -> Point C."

[0194] If the driver encounters a traffic jam along the way, the new information is sent to the server via the smart glasses. The server recalculates the route in real time and displays the new route visually. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to plan future routes.

[0195] Prompt Sentence Examples

[0196] "The driver's current location is in Shibuya Ward, Tokyo, and the specified delivery addresses are in Chuo Ward, Minato Ward, and Shinjuku Ward, Tokyo. The delivery must be completed between 9:00 AM and 5:00 PM. Please calculate the optimal route taking into account traffic congestion and weather conditions. Also, please take into account the usage status of each delivery box."

[0197] This system will improve delivery efficiency and enable drivers to carry out their delivery duties more efficiently.

[0198] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0199] Step 1: Data collection

[0200] The server collects delivery performance data, delivery locker availability data, traffic congestion and weather data from external APIs. Specifically, the server sends requests to each API, and obtains delivery performance data such as delivery date and time, address, and delivery success / failure information. Delivery locker availability is obtained in real time, and the location and usage status of the locker are returned as an API response. Traffic congestion and weather data are also collected from the API, allowing current traffic and weather information to be obtained. This data is temporarily stored on the server.

[0201] Step 2: Data Preprocessing

[0202] The server cleanses the collected data and eliminates incomplete and duplicate data. The delivery performance data also calculates the percentage of people at home during specific time periods. As a result of data preprocessing, a clean dataset suitable for calculating delivery routes is prepared. For example, in the delivery performance data, areas with a high percentage of people at home between 9:00 and 11:00 a.m. are extracted. The cleansed dataset is generated as an output.

[0203] Step 3: Calculate the optimal route

[0204] The server inputs the preprocessed data into a generative AI model to calculate the optimal delivery route. The generative AI model generates an efficient route by comprehensively taking into account delivery history, the current state of the delivery box, traffic congestion, and weather data. For example, based on past data, the model prioritizes addresses with a high probability of the driver being at home in the morning, and calculates a route that avoids traffic congestion using real-time traffic congestion information. The optimal route information is generated as the output.

[0205] Step 4: Distribute and display your route

[0206] The server sends the calculated optimal route to the driver's device. Specifically, smart glasses are used. The smart glasses display a map and route guidance in the driver's field of vision based on the route information received from the server. For example, if the next delivery destination is "AAA Building," specific instructions such as "turn right -> go straight -> turn left" are visually displayed. The output is a visually verifiable route guidance.

[0207] Step 5: Real-time updates

[0208] If the user (driver) needs to change the route during a delivery, the smart glasses send new information to the server in real time. The server then recalculates the route using the generative AI model based on the newly received traffic and weather information. For example, if the driver encounters a traffic jam, a new route is calculated and resent to the device. The updated route guidance is then displayed on the smart glasses as an output.

[0209] Step 6: Collect and use feedback

[0210] After completing a delivery, the user (driver) provides the server with feedback, including the reason for the delivery delay, through the smart glasses. The server stores and analyzes this feedback data, which is then used to generate the optimal route for the next delivery. For example, if delivery to a specific address was difficult, this information is reflected in the next route generation. This allows for more accurate route generation as an output.

[0211] 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.

[0212] This invention combines an emotion engine that recognizes user emotions with the calculation of optimal delivery routes using generative AI to thoroughly improve logistics efficiency. Detailed embodiments are described below.

[0213] Data collection and preprocessing

[0214] First, the server collects delivery history data, delivery box availability data, traffic congestion and weather data. Delivery history data includes information on past deliveries, including the date and time, address, and success or failure of deliveries. Delivery box availability data is obtained in real time, and reflects the location and usage status of each box in real time. Traffic congestion and weather data are collected using an external API.

[0215] The server then preprocesses this data. It calculates the percentage of people at home for a specific time period from the delivery performance data. The percentage of people at home is estimated probabilistically from past data. Next, it cleanses all data, removing incomplete and duplicate data and standardizing the format.

[0216] Calculating the best route

[0217] The server inputs the preprocessed data into the generation AI, which calculates the optimal delivery route. The generation AI considers past performance, the current state of the delivery box, traffic congestion, and weather data to generate an efficient route. This makes it possible to select a route that takes into account times when delivery recipients are most likely to be at home.

[0218] For example, based on past delivery data, the system can prioritize deliveries to areas with a high rate of people at home in the morning, and calculate routes that avoid traffic jams using real-time traffic information.In addition, when the weather is bad, it can select routes that use many delivery boxes.

[0219] Route distribution and display

[0220] The server sends the calculated route to each driver's device. The device then displays the received route information to the user. The display format is a map or list, and includes delivery destinations, routes, and important points. Specific instructions are displayed, such as "Turn right -> Go straight -> Turn left," indicating that the next delivery destination is "XXX Building."

[0221] Emotion recognition by emotion engine

[0222] The emotion engine installed in the device analyzes the driver's voice and input data to recognize their emotions at that time. The emotion engine determines emotions based on voice tone, input speed, selected vocabulary, etc. For example, if the driver is feeling stressed, information such as "The user is feeling stressed" is sent to the server.

[0223] Real-time updates and feedback

[0224] If the user (driver) needs to change the route during a delivery, they send that information from their device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the driver provides feedback on the delivery on their device. The emotion engine analyzes this feedback information and sends emotional information, such as stress or dissatisfaction felt by the driver, to the server.

[0225] Utilizing emotional feedback

[0226] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the system will avoid that route or consider alternative options. Furthermore, the system aims to improve driver satisfaction by adjusting driver rest stops and delivery pace based on the emotional data.

[0227] Specific examples

[0228] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses with a high rate of people at home around 10:00 AM when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C."

[0229] If the driver encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and submits a new route. After completing the delivery, the driver enters the reason for the delivery delay, and the emotion engine analyzes the driver's emotions at that time and sends the result as feedback to the server.

[0230] Through these processes, the system of the present invention not only improves delivery efficiency and reduces redelivery, but also comprehensively supports the resolution of logistics issues by utilizing driver emotional data.

[0231] The processing flow will be explained below.

[0232] Step 1:

[0233] When the server logs in, it retrieves delivery performance data for the delivery area over the past year from the database. This data includes the address of each delivery destination, delivery time, and delivery success / failure information.

[0234] Step 2:

[0235] The server retrieves current availability data from the delivery locker API in real time. The dataset includes each delivery locker's ID, location, and usage status (available / in use).

[0236] Step 3:

[0237] The server retrieves traffic and weather data for the day from an external API (e.g., Google Maps API or Japan Meteorological Agency API), including the degree of traffic congestion on major roads and weather conditions (sunny, rainy, snowy, etc.).

[0238] Step 4:

[0239] The server cleanses the data it receives, removes incomplete and duplicate data, completes it, and standardizes the format of all data.

[0240] Step 5:

[0241] The server calculates the percentage of delivery destinations at home during a specific time period based on past delivery performance data. For example, it identifies delivery destinations with a high percentage of delivery destinations at home between 9:00 AM and 11:00 AM.

[0242] Step 6:

[0243] The server receives a list of packages to be delivered and delivery area information, including the delivery address and desired delivery time for each package.

[0244] Step 7:

[0245] The server uses the generated AI to calculate the optimal delivery route based on the pre-processed data. The AI ​​model is input with collected past data, the current availability of delivery boxes, traffic congestion information, weather conditions, etc.

[0246] Step 8:

[0247] The server sends the optimal route calculated as a result to each driver's device.

[0248] Step 9:

[0249] The device displays detailed route information received to the user (driver) in map or list format, showing the next delivery destination, specific route, and important points to note.

[0250] Step 10:

[0251] The emotion engine installed in the device analyzes the user's (driver's) voice and input data to recognize their emotion at that time. For example, it can determine their emotion based on their voice tone and input speed.

[0252] Step 11:

[0253] If the user (driver) needs to change the route during a delivery, the device sends that information to the server, which recalculates the route in real time based on the new data and sends the new route back to the device.

[0254] Step 12:

[0255] After the user (driver) completes a delivery, they input the reason for the delay or any problems they encountered into the terminal. At the same time, the terminal's emotion engine analyzes the emotion data at that time.

[0256] Step 13:

[0257] The device collects feedback and emotion data and sends it to a server, which stores it and uses it for future route planning and driver care.

[0258] Step 14:

[0259] The server uses the collected feedback and emotional data to consider how to improve the next delivery route and experience, such as avoiding routes that frequently cause stress and suggesting appropriate rest stops.

[0260] This will not only improve delivery efficiency and reduce redelivery, but also realize comprehensive logistics optimization by utilizing driver emotional data.

[0261] Example 2

[0262] 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."

[0263] In conventional logistics systems, optimizing delivery routes is important for improving delivery efficiency, but human factors such as driver emotions and stress are not taken into consideration. Furthermore, response to real-time changes in the situation during delivery is insufficient, leaving a need for recalculation of optimal routes and improvement of delivery efficiency. The present invention aims to solve these problems and improve delivery efficiency while reducing driver stress.

[0264] 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.

[0265] In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery boxes, means for externally acquiring traffic congestion and weather data, means for pre-processing the acquired data, means for calculating the optimal delivery route using a generation AI, means for sending the calculated delivery route to the terminal, means for analyzing emotion data and recognizing the user's emotion, means for recalculating the route in real time and sending it again to the terminal if a change in the route is necessary, and means for receiving feedback and emotion data after delivery is completed and reflecting them in the next route generation. This not only improves delivery efficiency but also reduces driver stress and enables route improvements based on emotion data.

[0266] "Delivery performance data" refers to data that includes information on past deliveries, such as the date and time, address, and whether delivery was successful or unsuccessful.

[0267] "Availability of delivery box" is information indicating the location and real-time usage status of each delivery box.

[0268] "Traffic congestion status" is information related to traffic volume, and is data indicating the flow of traffic on roads and the degree of congestion.

[0269] "Weather data" is meteorological information obtained from an external source, and is data indicating weather conditions such as temperature, precipitation, and wind speed.

[0270] "Preprocessing" is the process of cleansing the collected data, removing duplicate data, standardizing formats, etc.

[0271] "Generative AI" is an artificial intelligence model that calculates optimal delivery routes based on collected and pre-processed data.

[0272] A "terminal" is a portable computing device used by a delivery driver, and is a device that displays route information and inputs emotional data.

[0273] "Emotional data" is information that indicates the driver's emotional state, and is data analyzed from voice tone, input speed, selected vocabulary, etc.

[0274] "Feedback" refers to information entered by the driver after completing a delivery about the reason for the delivery delay or incidents that occurred during the journey.

[0275] "Recalculation" refers to the process of recalculating the optimal route based on the latest information when a route change is necessary during delivery.

[0276] "Analysis" is the process of analyzing the acquired data in detail to find specific patterns and trends.

[0277] "Reflection" is the process of applying the analysis results to the next route generation or system improvement.

[0278] This invention is a system that improves efficiency in delivery work and reduces driver stress. It not only calculates the optimal delivery route using a generative AI model, but also combines it with an emotion engine that recognizes the user's emotions. Detailed embodiments of this invention are described below.

[0279] Data collection and preprocessing

[0280] Collection of delivery performance data

[0281] The server retrieves delivery performance data. It uses a high-performance server (e.g., a cloud server) and a database management system (e.g., MySQL) to extract data from a database containing information on past delivery dates and times, addresses, and delivery success and failures.

[0282] Collection of delivery box availability data

[0283] The server obtains the availability data of the delivery lockers in real time. The location and real-time usage status of each delivery locker are obtained from the delivery locker management system using a RESTful API.

[0284] Traffic and weather data collection

[0285] The server uses external APIs (e.g., map information API, weather information API) to obtain traffic congestion and weather data, thereby providing real-time traffic and weather information.

[0286] Next, the server preprocesses the collected data. Specifically, it performs the following operations:

[0287] Calculating the rate of at-home delivery: Using delivery performance data, we probabilistically estimate the rate of at-home delivery during a specific time period.

[0288] Data cleansing: Use the Pandas library to remove incomplete and duplicate data and standardize the data format.

[0289] Calculating the best route

[0290] The server inputs the preprocessed data into a generative AI model (e.g., GPT-3) to calculate the optimal delivery route. The generative AI model generates an efficient route by taking into account delivery history, the current state of the delivery box, traffic congestion data, and weather data.

[0291] For example, by giving the following prompt sentence to the generative AI model, it can calculate the optimal route.

[0292] Based on past data, prioritize areas with a high delivery success rate between 10:00 a.m. and noon and calculate routes that avoid traffic congestion.

[0293] The generative AI model outputs the optimal route based on this prompt.

[0294] Route distribution and display

[0295] The server sends the calculated route information to the driver's device via a RESTful API. The device then displays the received route information to the user in map or list format and also provides voice guidance.

[0296] Emotion Recognition and Data Transmission

[0297] The device is equipped with an emotion engine that recognizes emotions by analyzing the user's voice and input data. For example, if the user says, "I'm a little tired," the device analyzes the tiredness from the tone of the voice and sends the information that "the user is feeling tired" to the server.

[0298] Real-time updates and feedback

[0299] If a route change is necessary during a delivery, the user sends that information from their device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the user enters feedback into the device, which the emotion engine analyzes and sends emotional information to the server.

[0300] Utilizing emotional feedback

[0301] The server analyzes the collected feedback and emotion data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the server will avoid that route or consider alternatives, such as providing rest stops for drivers.

[0302] Specific examples

[0303] If a driver is scheduled to make five deliveries on a given day, the server will prioritize addresses with a high rate of people at home around 10:00 a.m. based on past performance data. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data and sends it to the driver's device in the form of "Point A -> Point B -> Point C."

[0304] If the user encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and sends a new route. After the delivery is completed, the user enters the reason for the delivery delay, and the emotion engine analyzes the user's emotions at that time and sends the result as feedback to the server.

[0305] As described above, the present invention provides a specific system configuration and procedure for improving efficiency in delivery work and reducing stress on drivers.

[0306] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0307] Step 1: Data collection

[0308] The server collects delivery performance data, delivery box availability data, traffic congestion data, and weather data.

[0309] Input: Delivery date and time, address, delivery success / failure information, location and usage status of delivery lockers, traffic conditions and weather data from external APIs.

[0310] Data processing: The server collects data using SQL queries and RESTful APIs and stores it in the appropriate data format.

[0311] Output: A consolidated dataset for use in preprocessing.

[0312] Step 2: Data Preprocessing

[0313] The server pre-processes the collected data.

[0314] Input: Unified dataset.

[0315] Data processing: The server cleanses the data, removes incomplete and duplicate data, and standardizes the format. It also calculates the percentage of customers at home during specific time periods based on delivery performance data.

[0316] Output: A preprocessed and clean dataset.

[0317] Step 3: Calculate the optimal route

[0318] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route.

[0319] Input: Preprocessed and clean dataset.

[0320] Data processing: The server provides a prompt to a generative AI model (e.g., GPT-3) and requests it to calculate a route. Specifically, it calculates an efficient delivery route taking into account the probability of delivery destinations being at home, real-time traffic congestion, and weather conditions.

[0321] Output: Optimal delivery route information.

[0322] Step 4: Distributing the Route

[0323] The server sends the calculated route to each driver's device.

[0324] Input: Optimal delivery route information.

[0325] Data processing: The server converts the route information into JSON or other appropriate format and sends it to the device via a RESTful API.

[0326] Output: Data containing route instructions is sent to the driver's device.

[0327] Step 5: View Routes

[0328] The terminal displays the received route information to the user.

[0329] Input: Route information received from the server.

[0330] Data processing: The device displays route information in map and list format. Voice guidance can also be enabled for hands-free operation.

[0331] Output: A route display that the user can see and hear.

[0332] Step 6: Send real-time updates during delivery

[0333] If the user needs to change the route during delivery, the user sends that information from the terminal to the server.

[0334] Input: Real-time information such as traffic jams encountered during delivery.

[0335] Data processing: The device sends the information from the user to the server, which recalculates the route in real time based on the new data.

[0336] Output: The newly calculated route information is sent to the device.

[0337] Step 7: Emotion Recognition

[0338] The emotion engine installed in the device recognizes the user's emotions.

[0339] Input: User speech and input data.

[0340] Data processing: The emotion engine analyzes voice tone, typing speed, and selected vocabulary to determine the user's emotions.

[0341] Output: The recognized emotion data is sent to the server.

[0342] Step 8: Feedback after delivery

[0343] After completing the delivery, the user provides feedback, which is then sent to the server by the device and analyzed by the emotion engine.

[0344] Input: Feedback information after delivery completion, user sentiment information.

[0345] Data processing: The server analyzes the feedback information and combines it with the accumulated emotion data.

[0346] Output: Analysis results that will be reflected in the next route generation and improvements to the operation interface.

[0347] Step 9: Use emotional feedback

[0348] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface.

[0349] Input: Aggregated feedback and sentiment data.

[0350] Data processing: The server uses data analysis software (e.g., Python data analysis libraries) to find patterns and trends.

[0351] Output: Improved route generation algorithm and updated user interface.

[0352] (Application example 2)

[0353] 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."

[0354] While maximizing delivery efficiency at logistics centers, there is a need to reduce drivers' working environments and psychological stress. However, conventional systems focus only on optimizing delivery routes without considering the driver's emotional state. This can lead to driver stress and dissatisfaction, which can lead to a decline in delivery quality and efficiency. Therefore, in addition to optimizing delivery routes, it is important to utilize driver emotional data to improve the working environment.

[0355] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery lockers, means for acquiring traffic congestion and weather data from outside, means for pre-processing the acquired data, means for calculating the optimal delivery route using a generation AI, means for transmitting the calculated delivery route to the terminal, means for recalculating the route in real time and transmitting it again to the terminal if a change in the route is necessary, means for receiving feedback after delivery is completed and reflecting this in the generation of the next route, means for recognizing and analyzing the emotions of the delivery person from voice and input data using an emotion engine, and means for improving the interface and route based on the driver's emotion data. This maximizes delivery efficiency while reducing the driver's working environment and psychological stress.

[0356] "Delivery performance data" refers to data that contains information about deliveries that have been made in the past, including the date and time, address, and whether the delivery was successful or unsuccessful.

[0357] "Delivery box availability" is data indicating whether each delivery box is in use or not, and is acquired in real time.

[0358] "Traffic congestion status" is data showing the state of traffic congestion on roads and is collected through an external API.

[0359] "Weather Data" means data containing information about current weather and forecasts, collected through external APIs.

[0360] "Preprocessing" refers to the process of cleansing the collected raw data, removing missing values ​​and duplicate data, and converting it into a format suitable for analysis and interpretation.

[0361] "Generative AI" is a system that uses artificial intelligence technology to generate optimal delivery routes, making predictions and optimizations based on past delivery performance and real-time data.

[0362] "Terminal" refers to the mobile device or smartphone carried by the delivery person, which is used to input delivery route information and feedback from the server.

[0363] "Real-time recalculation" refers to the process of instantly recalculating a new delivery route based on new information obtained during the delivery route.

[0364] The "emotion engine" is a system that recognizes and analyzes user emotions from voice and text input, and is used to understand the driver's stress and satisfaction.

[0365] "Feedback" refers to information provided by the driver after completing a delivery, including information about any troubles or emotional stress experienced during the delivery.

[0366] "Interface" refers to the parts that the user directly operates, including the screen and operations of the delivery application.

[0367] The "at-home rate of delivery destination" is data indicating the probability that the delivery destination will be at home during a specific time period, and is calculated from past delivery performance data.

[0368] The present invention is a system for realizing efficient delivery in a logistics center, and includes the following steps.

[0369] Data collection and preprocessing

[0370] The server uses external APIs to collect traffic and weather data, and retrieves delivery performance data and delivery locker availability data from local or cloud databases. The retrieved data is cleansed using the Python library "pandas," removing incomplete data and unifying duplicate data. The rate at home during specific time periods is also calculated and used as the basis for analysis.

[0371] Calculating the best route

[0372] Based on the preprocessed data, the server uses the "ortools" library to convert it into a format suitable for the generative AI model and calculates the optimal delivery route. An efficient route is generated taking into account past delivery data, real-time traffic congestion information, and weather data. This allows the system to select routes that avoid traffic congestion and times when the delivery destination is most likely to be at home. For example, it can prioritize areas where the number of people at home is high in the morning.

[0373] Route distribution and display

[0374] The calculated optimal route information is sent from the server to the driver's device. The device displays this information in map or list format, making it intuitive for the driver. For example, if the next delivery destination is "XXX Building," specific instructions such as "Turn right -> Go straight -> Turn left" are displayed.

[0375] Emotion recognition by emotion engine

[0376] The emotion engine on the device analyzes the driver's voice and manual input data to recognize emotions. This uses emotion analysis models from the "transformers" library. For example, if a driver says "I'm stressed today" via voice input, the emotion engine will analyze it and determine the emotion as "stress."

[0377] Real-time updates and feedback

[0378] If the user (driver) requests a route change during a delivery, that information is sent from the device to the server. The server immediately recalculates the route based on the new data and resends it to the device. Also, when the driver provides feedback on the delivery on the device after completing the delivery, the emotion engine analyzes this feedback information and sends the stress or dissatisfaction the driver felt to the server.

[0379] Utilizing emotional feedback

[0380] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the system will avoid that route or consider alternative options. Furthermore, by adjusting driver rest points and delivery pace based on the emotional data, the system also contributes to improving driver satisfaction.

[0381] Specific examples

[0382] If a driver has five deliveries scheduled for a given day, the server will use past performance data to include addresses with a high rate of people at home around 10:00 a.m. in the route, and will also obtain information on delivery box usage and include boxes with many empty spaces in the route. A route that avoids traffic jams is generated based on traffic congestion data, and this is sent to the driver's device in the form of "Point A -> Point B -> Point C." If the driver encounters traffic jams along the way, a new route is instantly recalculated and distributed. After completing a delivery, the driver enters the reason for the delay, and at the same time, the emotion analysis engine recognizes "stress." This will lead to improvements to the next delivery route and interface.

[0383] Example prompt: "Please identify the emotion from the following speech input: 'I'm feeling stressed today.'"

[0384] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0385] Step 1: Data collection

[0386] The server uses an external API to collect traffic and weather data. It also retrieves delivery history data and delivery box availability data from a local or cloud database. Specifically, it uses the requests library to access the API and retrieve data in JSON format. An API key and endpoint are required as input, and the retrieved data is passed to the server in JSON format. This makes it possible to collect various required data in real time.

[0387] Step 2: Data Preprocessing

[0388] The server cleanses the acquired data using the pandas library, removing incomplete and duplicate data. It also extracts important information such as date, time, and address, and calculates the at-home rate for a specific time period. The collected raw data is passed as input, and cleaned data and at-home rate data are generated as output. This process formats the data in a format suitable for analysis and prediction.

[0389] Step 3: Calculate the optimal route

[0390] The server uses the ortools library to calculate the optimal delivery route based on the preprocessed data. Specifically, collected delivery performance data, real-time traffic information, weather data, etc. are input into a generative AI model to generate an efficient route. Clean data and at-home rate data are used as input, and optimal route information is generated as output. This makes it possible to select routes that take into account time periods when delivery destinations are most likely to be at home and traffic conditions.

[0391] Step 4: Distribute and display your route

[0392] The server sends the calculated optimal route information to each driver's device. The device displays this information in map or list format, allowing the driver to understand it intuitively. For example, delivery destinations and routes are displayed in detail. The optimal route information is passed to the device as input, and is displayed as output in a visually easy-to-understand format. This allows the driver to easily check the next delivery destination and route.

[0393] Step 5: Emotion Recognition

[0394] The emotion engine installed on the device analyzes the driver's voice input and manual input data to recognize emotions. Specifically, it uses the emotion analysis model from the Transformers library. The driver's voice and text are passed as input, and analyzed emotional information is generated as output. This allows the driver's current emotional state to be understood.

[0395] Step 6: Real-time updates

[0396] If the user (driver) requests a route change during a delivery, that information is sent from the device to the server. The server recalculates the route based on the new data and resends it. Real-time data from the driver is used as input, and new optimal route information is generated as output. This allows for flexible route changes to adapt to real-world conditions.

[0397] Step 7: Gather feedback

[0398] After completing a delivery, the user (driver) enters feedback about the delivery on the terminal. The emotion engine simultaneously analyzes the driver's emotions and sends the data to the server. The driver's feedback text and emotional data are passed as input, and the analyzed feedback information is saved as output. This accumulates data that will be useful for generating the next delivery route and improving the system.

[0399] Step 8: Use emotional feedback

[0400] The server analyzes the collected feedback and emotion data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, it will avoid that route or consider alternative options. The analyzed feedback data is used as input, and improvement suggestions are generated as output. This improves driver satisfaction and maximizes delivery efficiency.

[0401] 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.

[0402] 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.

[0403] 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.

[0404] [Second embodiment]

[0405] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0406] 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.

[0407] 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).

[0408] 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.

[0409] 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.

[0410] 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).

[0411] 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.

[0412] 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.

[0413] 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.

[0414] 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.

[0415] 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.

[0416] 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."

[0417] This invention is a system for improving the efficiency of logistics, which uses generative AI to calculate optimal delivery routes and provide them to truck drivers. This system is based mainly on interactions between a server, terminals, and users, and detailed embodiments are described below.

[0418] Data collection and preprocessing

[0419] First, the server collects delivery history data, delivery box availability data, traffic congestion data, and weather data. Delivery history data includes the date, time, address, and delivery success / failure information for past deliveries. Delivery box availability data is obtained in real time and reflects the location and usage status of each box. Traffic congestion and weather data are collected using an external API.

[0420] The server then pre-processes this data. From the delivery performance data, the rate of at-home delivery during a specific time period is calculated. Data pre-processing cleans incomplete and duplicate data.

[0421] Calculating the best route

[0422] The server inputs the preprocessed data into the generation AI to calculate the optimal delivery route. The generation AI takes into account past performance, the current state of the delivery lockers, traffic congestion, and weather data to generate an efficient route.

[0423] For example, based on past delivery data, the system can prioritize deliveries to areas with a high rate of people at home in the morning, and calculate routes that avoid traffic jams using real-time traffic information.In addition, if the weather is bad, it can select a route that uses many delivery boxes.

[0424] Route distribution and display

[0425] The server sends the calculated route to each driver's device. The device then displays the received route information to the user. The display format is a map or list, and includes delivery destinations, routes, and important points. For example, if the next delivery destination is "XXX Building," specific instructions such as "Turn right -> Go straight -> Turn left" are displayed.

[0426] Real-time updates and feedback

[0427] If the user (driver) needs to change the route during a delivery, the information is sent from the device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the user provides feedback, which is sent from the device to the server. The server analyzes the feedback and uses it to generate the next route.

[0428] Specific examples

[0429] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses with a high rate of people being at home around 10:00 AM when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C." The device that receives the information displays detailed route information to the driver along with a map.

[0430] If the driver encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and submits a new route. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to generate future routes.

[0431] In this way, the system of the present invention improves delivery efficiency and reduces redelivery, thereby contributing to solving logistics issues.

[0432] The processing flow will be explained below.

[0433] Step 1:

[0434] When the server logs in, it retrieves delivery performance data for the delivery area over the past year from the database. This data includes the address of each delivery destination, delivery time, and delivery success / failure information.

[0435] Step 2:

[0436] The server retrieves current availability data from the delivery locker API in real time. The dataset includes the ID, location, and usage status (available / in use) of each delivery locker.

[0437] Step 3:

[0438] The server retrieves traffic and weather data for the day from an external API (e.g., Google Maps API or Japan Meteorological Agency API), including the degree of traffic congestion on major roads and weather conditions (sunny, rainy, snowy, etc.).

[0439] Step 4:

[0440] The server cleanses the data it receives, removes incomplete and duplicate data, completes it, and standardizes the format of all data.

[0441] Step 5:

[0442] The server calculates the percentage of delivery destinations at home during a specific time period based on past delivery performance data. For example, it identifies delivery destinations with a high percentage of delivery destinations at home between 9:00 AM and 11:00 AM.

[0443] Step 6:

[0444] The server receives a list of packages to be delivered and delivery area information, including the delivery address and desired delivery time for each package.

[0445] Step 7:

[0446] The server uses the generated AI to calculate the optimal delivery route based on the pre-processed data. The AI ​​model is input with collected past data, the current availability of delivery boxes, traffic congestion information, weather conditions, etc.

[0447] Step 8:

[0448] The server sends the optimal route calculated as a result to each driver's device.

[0449] Step 9:

[0450] The device displays detailed route information received to the user (driver) in map or list format, showing the next delivery destination, specific route, and important points to note.

[0451] Step 10:

[0452] If the user (driver) needs to change the route during a delivery, the device sends that information to the server, which recalculates the route in real time based on the new data and sends the new route back to the device.

[0453] Step 11:

[0454] After completing a delivery, the user (driver) enters the reason for the delivery delay or any problems that occurred as feedback into the terminal.

[0455] Step 12:

[0456] The device sends the received feedback data to the server, which analyzes the collected feedback and reflects it in the next route generation.

[0457] These processing steps create a system that improves delivery efficiency and reduces redelivery.

[0458] Example 1

[0459] 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."

[0460] Conventional logistics systems have problems with insufficient optimization of delivery routes, making it difficult to achieve efficient deliveries. It is also difficult to respond in real time to traffic congestion and weather changes, leading to redeliveries and delivery delays. Furthermore, there is an insufficient system for utilizing feedback from drivers in the next delivery, making it difficult to improve delivery efficiency.

[0461] 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.

[0462] In this invention, the server includes means for acquiring delivery-related data, means for acquiring the usage status of delivery boxes, means for externally acquiring traffic and weather information, means for preprocessing the acquired data, means for calculating an optimal delivery route using generative artificial intelligence, means for transmitting the calculated delivery route to the terminal, means for recalculating the route in real time and transmitting it again to the terminal if a route change is necessary, means for receiving feedback after delivery is completed and reflecting it in the generation of the next route, means for providing the driver with route information displayed on the terminal, and means for receiving and processing route change requests from the terminal during delivery. This enables efficient delivery route optimization and real-time response. Furthermore, by reflecting driver feedback in the next delivery, continuous improvement in delivery efficiency can be achieved.

[0463] "Delivery-related data" means data containing information about past and current deliveries, including delivery dates and times, addresses, and delivery success and failure information.

[0464] "Delivery box usage status" is data that indicates the current usage status of each delivery box, and includes the location of the box, availability status, usage history, etc.

[0465] "Traffic information" is data that indicates the current traffic situation, including road congestion and accident information.

[0466] "Weather information" refers to data that indicates current and forecast weather conditions, including precipitation, temperature, wind speed, and the like.

[0467] "Preprocessing" refers to the process of preparing acquired data so that it can be analyzed, and includes the deletion of incomplete data, the integration of duplicate data, and the extraction of necessary data.

[0468] "Generative artificial intelligence" is a system that uses artificial intelligence techniques to generate new information based on data, which is then used to calculate optimal delivery routes.

[0469] An "optimal delivery route" is a delivery sequence and route that maximizes delivery efficiency and aims to reduce time and costs.

[0470] A "terminal" is an information device used by a driver to display delivery routes and route information and to communicate with the server.

[0471] "Recalculation in real time" is a process that instantly calculates and provides new delivery routes based on current conditions.

[0472] "Feedback" is information provided by the driver after delivery, including reasons for delivery delays and special notes from the delivery destination.

[0473] "Means for providing route information to the driver" refers to a method for communicating the calculated delivery route and detailed instructions to the driver via the terminal.

[0474] The "means for receiving and processing a route change request from a terminal" is a method in which a server receives a route change request sent by a driver from a terminal, calculates and provides a new route based on the request.

[0475] This invention is a system for improving logistics efficiency, which uses generative artificial intelligence (generative AI model) to calculate optimal delivery routes and provide them to drivers. This system consists of a server, terminals, and users, and handles the collection, preprocessing, analysis, display, and feedback of delivery-related data.

[0476] Data collection and preprocessing

[0477] server

[0478] The server collects data using the following means:

[0479] 1. Delivery-related data: Obtain past delivery records from the database. For example, obtain data such as "2023-10-01, Address A, Delivery successful."

[0480] 2. Delivery locker usage status: Obtain real-time data using the API of an external delivery locker management system. For example, obtain data such as "Delivery locker B, available."

[0481] 3. Traffic and weather information: Get data from traffic APIs and weather APIs. For example, get data such as "Road C, traffic jam" or "Area D, rain."

[0482] The server then preprocesses the data, calculating the percentage of people at home during specific time periods from delivery-related data and cleaning up incomplete and duplicate data.

[0483] Calculating the best route

[0484] server

[0485] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route. This generative AI model takes into account delivery history, the current state of the parcel lockers, traffic congestion, and weather data to generate an efficient delivery route.

[0486] Specifically, the generative AI model calculates the following route:

[0487] 1. Prioritize deliveries to areas where people are more likely to be at home in the morning.

[0488] 2. Select a route that avoids traffic jams based on real-time traffic information.

[0489] 3. If the weather is bad, choose a route that uses more delivery boxes.

[0490] Route distribution and display

[0491] Servers and Terminals

[0492] The server sends the calculated delivery route to each driver's device. The device that receives the route information displays it to the user (driver). The display format is a map or list, and includes delivery destinations, routes, and important points. Specific examples of instructions include detailed instructions such as "Next delivery destination is XXX Building" and "Turn right -> go straight -> turn left."

[0493] Real-time updates and feedback

[0494] Users and Servers

[0495] The user (driver) can input new information during the delivery. For example, if the user encounters a traffic jam, the user can report the situation to the server from the terminal: "Road G, traffic jam starting."

[0496] The server recalculates the route in real time based on the new data and resends it to the device. After the delivery is completed, the user can provide feedback such as the reason for the delivery delay or any special notes at the delivery destination, and the server will reflect this in the next route generation.

[0497] Examples and prompts

[0498] Specific examples

[0499] One day, a driver is scheduled to make five deliveries. Based on past performance data, the server prioritizes addresses with a high rate of people at home around 10:00 a.m. in the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most vacant boxes. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C." The device that receives the route displays detailed route information to the driver along with a map. If the driver encounters traffic congestion along the way, new information is sent from the device to the server. The server recalculates the route in real time and presents a new route. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to generate future routes.

[0500] Prompt example

[0501] "Generate the optimal delivery route to efficiently complete five deliveries based on past delivery data, drop-box availability, traffic information, and weather data."

[0502] In this way, the system of the present invention enables efficient delivery routes and real-time responses, and by reflecting driver feedback in the next delivery, it achieves continuous improvement in delivery efficiency.

[0503] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0504] Step 1: Collect data

[0505] The server collects delivery-related data, parcel box usage, traffic information, and weather information.

[0506] Input: Requests to external APIs or databases

[0507] Data processing: Receives the response from the API, analyzes its contents, and saves it in the database.

[0508] Output: Raw data awaiting preprocessing

[0509] What it does: Performs database queries and external API calls to gather delivery records, parcel locker status, and real-time traffic and weather information.

[0510] Step 2: Preprocessing the data

[0511] The server pre-processes the collected data.

[0512] Input: Delivery-related data, parcel locker usage, traffic information, weather information

[0513] Data processing: missing data imputation, duplicate data removal, outlier detection and correction

[0514] Output: Cleansed data

[0515] What it does: It runs a data cleansing algorithm to remove incomplete records, merge duplicates, and calculates the percentage of people at home for a specific time period from delivery performance data.

[0516] Step 3: Calculate the optimal route

[0517] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route.

[0518] Input: Cleansed delivery data, parcel locker usage, traffic information, weather information

[0519] Data processing: All data is integrated and input into a generative AI model, which then runs an algorithm to generate the optimal route.

[0520] Output: Optimal delivery route

[0521] How it works: The generative AI model plans the optimal delivery route based on past delivery performance, real-time traffic and weather information, and available delivery lockers.

[0522] Step 4: Distributing the Route

[0523] The server sends the calculated delivery route to each driver's device.

[0524] Input: Optimal delivery route data

[0525] Data processing: Converting route information into a format that can be processed by the device

[0526] Output: Route information sent to the device

[0527] Specific operation: Encode the optimal route in JSON format or similar and send it to the device using a communication protocol (e.g., HTTP).

[0528] Step 5: View Routes

[0529] The terminal displays the received route information to the user (driver).

[0530] Input: Route information sent

[0531] Data processing: Converting route information into a visually understandable format

[0532] Output: Route information displayed in map and list format

[0533] What it does: Renders route information on the interface of a map or delivery app, showing instructions such as "Next delivery stop is XXX building" or "Turn right -> go straight -> turn left."

[0534] Step 6: Real-time updates

[0535] The server receives route change requests from users (drivers) and recalculates the route in real time.

[0536] Input: Route change request (e.g., "Route G, traffic jam starting")

[0537] Data processing: Integrate new situational data and recalculate the optimal route

[0538] Output: Updated optimal route

[0539] Specific operation: Re-run the generative AI model based on the new traffic information, generate a new route, and send it to the device.

[0540] Step 7: Use the feedback

[0541] The server receives feedback from the user (driver) after the delivery is completed and reflects it in the next route generation.

[0542] Input: Feedback data (e.g. "Delivery destination X, address unknown")

[0543] Data processing: Analyze feedback data and adjust parameters of route generation algorithm

[0544] Output: Improved route generation algorithm

[0545] Specific behavior: Save the feedback in a database and use it the next time you generate a route, or take other measures to improve it.

[0546] (Application example 1)

[0547] 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."

[0548] In logistics, conventional delivery route calculation systems have not been able to sufficiently improve delivery efficiency. In particular, it is difficult to provide optimal routes that take into account the rate at which recipients are at home, real-time traffic conditions, and weather conditions. This also makes it inefficient for drivers to check information en route. Additionally, the cost and time wasted by redelivery have also become an issue.

[0549] 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.

[0550] In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery boxes, means for externally acquiring traffic congestion and weather data, means for pre-processing the acquired data, means for calculating an optimal delivery route using a generation AI, means for transmitting the calculated delivery route to a terminal and displaying it on the smart glasses, means for recalculating the route in real time and transmitting it again to the terminal if a change in the route is necessary, and means for receiving feedback after delivery is completed and reflecting it in the next route generation, thereby enabling improved delivery efficiency and providing information in real time.

[0551] "Delivery performance data" refers to data that includes information such as the date and time of past deliveries, addresses, and delivery success / failure information.

[0552] "Delivery box availability" is data that shows the location and usage status of each box in real time.

[0553] "Traffic congestion status" is data indicating the state of traffic congestion on a road, and is acquired from an external source.

[0554] "Weather data" is data that indicates the weather forecast and current weather conditions for the area.

[0555] "Preprocessing" is a process of cleansing acquired data and eliminating incomplete data and duplicate data.

[0556] "Generative AI" is an artificial intelligence model that generates or predicts something based on given data.

[0557] The "optimal delivery route" is an efficient delivery route that minimizes time and costs, taking into account factors such as the rate at which recipients are at home, traffic congestion, and weather conditions.

[0558] "Device" means a device used by a driver, including a smartphone, tablet, smart glasses, etc.

[0559] "Smart glasses" are glasses-type devices that can transparently display visual information, allowing drivers to check the information hands-free.

[0560] "Recalculation in real time" refers to the process of instantly calculating a new route in response to changes in the situation and sending it back to the terminal.

[0561] "Feedback" refers to the reason for the delivery delay and other information provided by the driver after completing the delivery, and is data used to plan the next route.

[0562] To implement this invention, a server, smart glasses, other terminals, a generative AI model, an external API, etc. are used. Specific embodiments are described below.

[0563] Data collection and preprocessing

[0564] The server collects delivery performance data, delivery locker availability data, traffic congestion information, and weather data from external APIs. Delivery performance data includes the date and time, address, and success / failure information of past deliveries. Delivery locker availability information is obtained in real time and reflects the location and usage status of each locker. Traffic congestion and weather data are collected using external APIs.

[0565] The server then pre-processes this data, which includes cleaning out incomplete and duplicate data, and calculating the percentage of customers at home during specific time periods from the delivery performance data.

[0566] Calculating the best route

[0567] The server inputs the preprocessed data into a generative AI model to calculate the optimal delivery route. The generative AI model generates an efficient route by taking into account past delivery performance, the current state of delivery boxes, traffic congestion, and weather data. For example, it can prioritize deliveries to areas with a high rate of people at home in the morning based on past delivery data, and calculate a route that avoids traffic congestion using real-time traffic information. In addition, when the weather is bad, it can select a route that uses many delivery boxes.

[0568] Route distribution and display

[0569] The server sends the calculated route to the driver's device, which is a pair of smart glasses. The smart glasses visually display the calculated route information to the driver, allowing them to check the information hands-free. The display format is a map or list, and includes the next delivery destination, route, and important points. For example, if the next delivery destination is "AAA Building," specific instructions such as "turn right -> go straight -> turn left" are displayed.

[0570] Real-time updates and feedback

[0571] If the driver needs to change the route during a delivery, the information is sent to the server in real time via the smart glasses. The server recalculates the route in real time based on the new data and displays the new route on the smart glasses. After completing the delivery, the driver provides feedback, which is sent to the server via the smart glasses. The server analyzes the feedback and uses it to generate the next route.

[0572] Specific examples

[0573] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses where the number of people at home around 10:00 AM is high when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is displayed on the driver's smart glasses in the form of "Point A -> Point B -> Point C."

[0574] If the driver encounters a traffic jam along the way, the new information is sent to the server via the smart glasses. The server recalculates the route in real time and displays the new route visually. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to plan future routes.

[0575] Prompt Sentence Examples

[0576] "The driver's current location is in Shibuya Ward, Tokyo, and the specified delivery addresses are in Chuo Ward, Minato Ward, and Shinjuku Ward, Tokyo. The delivery must be completed between 9:00 AM and 5:00 PM. Please calculate the optimal route taking into account traffic congestion and weather conditions. Also, please take into account the usage status of each delivery box."

[0577] This system will improve delivery efficiency and enable drivers to carry out their delivery duties more efficiently.

[0578] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0579] Step 1: Data collection

[0580] The server collects delivery performance data, delivery locker availability data, traffic congestion and weather data from external APIs. Specifically, the server sends requests to each API, and obtains delivery performance data such as delivery date and time, address, and delivery success / failure information. Delivery locker availability is obtained in real time, and the location and usage status of the locker are returned as an API response. Traffic congestion and weather data are also collected from the API, allowing current traffic and weather information to be obtained. This data is temporarily stored on the server.

[0581] Step 2: Data Preprocessing

[0582] The server cleanses the collected data and eliminates incomplete and duplicate data. The delivery performance data also calculates the percentage of people at home during specific time periods. As a result of data preprocessing, a clean dataset suitable for calculating delivery routes is prepared. For example, in the delivery performance data, areas with a high percentage of people at home between 9:00 and 11:00 a.m. are extracted. The cleansed dataset is generated as an output.

[0583] Step 3: Calculate the optimal route

[0584] The server inputs the preprocessed data into a generative AI model to calculate the optimal delivery route. The generative AI model generates an efficient route by comprehensively taking into account delivery history, the current state of the delivery box, traffic congestion, and weather data. For example, based on past data, the model prioritizes addresses with a high probability of the driver being at home in the morning, and calculates a route that avoids traffic congestion using real-time traffic congestion information. The optimal route information is generated as the output.

[0585] Step 4: Distribute and display your route

[0586] The server sends the calculated optimal route to the driver's device. Specifically, smart glasses are used. The smart glasses display a map and route guidance in the driver's field of vision based on the route information received from the server. For example, if the next delivery destination is "AAA Building," specific instructions such as "turn right -> go straight -> turn left" are visually displayed. The output is a visually verifiable route guidance.

[0587] Step 5: Real-time updates

[0588] If the user (driver) needs to change the route during a delivery, the smart glasses send new information to the server in real time. The server then recalculates the route using the generative AI model based on the newly received traffic and weather information. For example, if the driver encounters a traffic jam, a new route is calculated and resent to the device. The updated route guidance is then displayed on the smart glasses as an output.

[0589] Step 6: Collect and use feedback

[0590] After completing a delivery, the user (driver) provides the server with feedback, including the reason for the delivery delay, through the smart glasses. The server stores and analyzes this feedback data, which is then used to generate the optimal route for the next delivery. For example, if delivery to a specific address was difficult, this information is reflected in the next route generation. This allows for more accurate route generation as an output.

[0591] 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.

[0592] This invention combines an emotion engine that recognizes user emotions with the calculation of optimal delivery routes using generative AI to thoroughly improve logistics efficiency. Detailed embodiments are described below.

[0593] Data collection and preprocessing

[0594] First, the server collects delivery history data, delivery box availability data, traffic congestion and weather data. Delivery history data includes information on past deliveries, including the date and time, address, and success or failure of deliveries. Delivery box availability data is obtained in real time, and reflects the location and usage status of each box in real time. Traffic congestion and weather data are collected using an external API.

[0595] The server then preprocesses this data. It calculates the percentage of people at home for a specific time period from the delivery performance data. The percentage of people at home is estimated probabilistically from past data. Next, it cleanses all data, removing incomplete and duplicate data and standardizing the format.

[0596] Calculating the best route

[0597] The server inputs the preprocessed data into the generation AI, which calculates the optimal delivery route. The generation AI considers past performance, the current state of the delivery box, traffic congestion, and weather data to generate an efficient route. This makes it possible to select a route that takes into account times when delivery recipients are most likely to be at home.

[0598] For example, based on past delivery data, the system can prioritize deliveries to areas with a high rate of people at home in the morning, and calculate routes that avoid traffic jams using real-time traffic information.In addition, when the weather is bad, it can select routes that use many delivery boxes.

[0599] Route distribution and display

[0600] The server sends the calculated route to each driver's device. The device then displays the received route information to the user. The display format is a map or list, and includes delivery destinations, routes, and important points. Specific instructions are displayed, such as "Turn right -> Go straight -> Turn left," indicating that the next delivery destination is "XXX Building."

[0601] Emotion recognition by emotion engine

[0602] The emotion engine installed in the device analyzes the driver's voice and input data to recognize their emotions at that time. The emotion engine determines emotions based on voice tone, input speed, selected vocabulary, etc. For example, if the driver is feeling stressed, information such as "The user is feeling stressed" is sent to the server.

[0603] Real-time updates and feedback

[0604] If the user (driver) needs to change the route during a delivery, they send that information from their device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the driver provides feedback on the delivery on their device. The emotion engine analyzes this feedback information and sends emotional information, such as stress or dissatisfaction felt by the driver, to the server.

[0605] Utilizing emotional feedback

[0606] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the system will avoid that route or consider alternative options. Furthermore, the system aims to improve driver satisfaction by adjusting driver rest stops and delivery pace based on the emotional data.

[0607] Specific examples

[0608] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses with a high rate of people at home around 10:00 AM when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C."

[0609] If the driver encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and submits a new route. After completing the delivery, the driver enters the reason for the delivery delay, and the emotion engine analyzes the driver's emotions at that time and sends the result as feedback to the server.

[0610] Through these processes, the system of the present invention not only improves delivery efficiency and reduces redelivery, but also comprehensively supports the resolution of logistics issues by utilizing driver emotional data.

[0611] The processing flow will be explained below.

[0612] Step 1:

[0613] When the server logs in, it retrieves delivery performance data for the delivery area over the past year from the database. This data includes the address of each delivery destination, delivery time, and delivery success / failure information.

[0614] Step 2:

[0615] The server retrieves current availability data from the delivery locker API in real time. The dataset includes the ID, location, and usage status (available / in use) of each delivery locker.

[0616] Step 3:

[0617] The server retrieves traffic and weather data for the day from an external API (e.g., Google Maps API or Japan Meteorological Agency API), including the degree of traffic congestion on major roads and weather conditions (sunny, rainy, snowy, etc.).

[0618] Step 4:

[0619] The server cleanses the data it receives, removes incomplete and duplicate data, completes it, and standardizes the format of all data.

[0620] Step 5:

[0621] The server calculates the percentage of delivery destinations at home during a specific time period based on past delivery performance data. For example, it identifies delivery destinations with a high percentage of delivery destinations at home between 9:00 AM and 11:00 AM.

[0622] Step 6:

[0623] The server receives a list of packages to be delivered and delivery area information, including the delivery address and desired delivery time for each package.

[0624] Step 7:

[0625] The server uses the generated AI to calculate the optimal delivery route based on the pre-processed data. The AI ​​model is input with collected past data, the current availability of delivery boxes, traffic congestion information, weather conditions, etc.

[0626] Step 8:

[0627] The server sends the optimal route calculated as a result to each driver's device.

[0628] Step 9:

[0629] The device displays detailed route information received to the user (driver) in map or list format, showing the next delivery destination, specific route, and important points to note.

[0630] Step 10:

[0631] The emotion engine installed in the device analyzes the user's (driver's) voice and input data to recognize their emotion at that time. For example, it can determine their emotion based on their voice tone and input speed.

[0632] Step 11:

[0633] If the user (driver) needs to change the route during a delivery, the device sends that information to the server, which recalculates the route in real time based on the new data and sends the new route back to the device.

[0634] Step 12:

[0635] After the user (driver) completes a delivery, they input the reason for the delay or any problems they encountered into the terminal. At the same time, the terminal's emotion engine analyzes the emotion data at that time.

[0636] Step 13:

[0637] The device collects feedback and emotion data and sends it to a server, which stores it and uses it for future route planning and driver care.

[0638] Step 14:

[0639] The server uses the collected feedback and emotional data to consider how to improve the next delivery route and experience, such as avoiding routes that frequently cause stress and suggesting appropriate rest stops.

[0640] This will not only improve delivery efficiency and reduce redelivery, but also realize comprehensive logistics optimization by utilizing driver emotional data.

[0641] Example 2

[0642] 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."

[0643] In conventional logistics systems, optimizing delivery routes is important for improving delivery efficiency, but human factors such as driver emotions and stress are not taken into consideration. Furthermore, response to real-time changes in the situation during delivery is insufficient, leaving a need for recalculation of optimal routes and improvement of delivery efficiency. The present invention aims to solve these problems and improve delivery efficiency while reducing driver stress.

[0644] 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.

[0645] In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery boxes, means for externally acquiring traffic congestion and weather data, means for pre-processing the acquired data, means for calculating the optimal delivery route using a generation AI, means for sending the calculated delivery route to the terminal, means for analyzing emotion data and recognizing the user's emotion, means for recalculating the route in real time and sending it again to the terminal if a change in the route is necessary, and means for receiving feedback and emotion data after delivery is completed and reflecting them in the next route generation. This not only improves delivery efficiency but also reduces driver stress and enables route improvements based on emotion data.

[0646] "Delivery performance data" refers to data that includes information on past deliveries, such as the date and time, address, and whether delivery was successful or unsuccessful.

[0647] "Availability of delivery box" is information indicating the location and real-time usage status of each delivery box.

[0648] "Traffic congestion status" is information related to traffic volume, and is data indicating the flow of traffic on roads and the degree of congestion.

[0649] "Weather data" is meteorological information obtained from an external source, and is data indicating weather conditions such as temperature, precipitation, and wind speed.

[0650] "Preprocessing" is the process of cleansing the collected data, removing duplicate data, standardizing formats, etc.

[0651] "Generative AI" is an artificial intelligence model that calculates optimal delivery routes based on collected and pre-processed data.

[0652] A "terminal" is a portable computing device used by a delivery driver, and is a device that displays route information and inputs emotional data.

[0653] "Emotional data" is information that indicates the driver's emotional state, and is data analyzed from voice tone, input speed, selected vocabulary, etc.

[0654] "Feedback" refers to information entered by the driver after completing a delivery about the reason for the delivery delay or incidents that occurred during the journey.

[0655] "Recalculation" refers to the process of recalculating the optimal route based on the latest information when a route change is necessary during delivery.

[0656] "Analysis" is the process of analyzing the acquired data in detail to find specific patterns and trends.

[0657] "Reflection" is the process of applying the analysis results to the next route generation or system improvement.

[0658] This invention is a system that improves efficiency in delivery work and reduces driver stress. It not only calculates the optimal delivery route using a generative AI model, but also combines it with an emotion engine that recognizes the user's emotions. Detailed embodiments of this invention are described below.

[0659] Data collection and preprocessing

[0660] Collection of delivery performance data

[0661] The server retrieves delivery performance data. It uses a high-performance server (e.g., a cloud server) and a database management system (e.g., MySQL) to extract data from a database containing information on past delivery dates and times, addresses, and delivery success and failures.

[0662] Collection of delivery box availability data

[0663] The server obtains the availability data of the delivery lockers in real time. The location and real-time usage status of each delivery locker are obtained from the delivery locker management system using a RESTful API.

[0664] Traffic and weather data collection

[0665] The server uses external APIs (e.g., map information API, weather information API) to obtain traffic congestion and weather data, thereby providing real-time traffic and weather information.

[0666] Next, the server preprocesses the collected data. Specifically, it performs the following operations:

[0667] Calculating the rate of at-home delivery: Using delivery performance data, we probabilistically estimate the rate of at-home delivery during a specific time period.

[0668] Data cleansing: Use the Pandas library to remove incomplete and duplicate data and standardize the data format.

[0669] Calculating the best route

[0670] The server inputs the preprocessed data into a generative AI model (e.g., GPT-3) to calculate the optimal delivery route. The generative AI model generates an efficient route by taking into account delivery history, the current state of the delivery box, traffic congestion data, and weather data.

[0671] For example, by giving the following prompt sentence to the generative AI model, it can calculate the optimal route.

[0672] Based on past data, prioritize areas with a high delivery success rate between 10:00 a.m. and noon and calculate routes that avoid traffic congestion.

[0673] The generative AI model outputs the optimal route based on this prompt.

[0674] Route distribution and display

[0675] The server sends the calculated route information to the driver's device via a RESTful API. The device then displays the received route information to the user in map or list format and also provides voice guidance.

[0676] Emotion Recognition and Data Transmission

[0677] The device is equipped with an emotion engine that recognizes emotions by analyzing the user's voice and input data. For example, if the user says "I'm a little tired," the device analyzes the tiredness from the tone of the voice and sends the information that "the user is feeling tired" to the server.

[0678] Real-time updates and feedback

[0679] If a route change is necessary during a delivery, the user sends that information from their device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the user enters feedback into the device, which the emotion engine analyzes and sends emotional information to the server.

[0680] Utilizing emotional feedback

[0681] The server analyzes the collected feedback and emotion data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the server will avoid that route or consider alternatives, such as providing rest stops for drivers.

[0682] Specific examples

[0683] If a driver is scheduled to make five deliveries on a given day, the server will prioritize addresses with a high rate of people at home around 10:00 a.m. based on past performance data. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data and sends it to the driver's device in the form of "Point A -> Point B -> Point C."

[0684] If the user encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and sends a new route. After the delivery is completed, the user enters the reason for the delivery delay, and the emotion engine analyzes the user's emotions at that time and sends the result as feedback to the server.

[0685] As described above, the present invention provides a specific system configuration and procedure for improving efficiency in delivery work and reducing stress on drivers.

[0686] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0687] Step 1: Data collection

[0688] The server collects delivery performance data, delivery box availability data, traffic congestion data, and weather data.

[0689] Input: Delivery date and time, address, delivery success / failure information, location and usage status of delivery lockers, traffic conditions and weather data from external APIs.

[0690] Data processing: The server collects data using SQL queries and RESTful APIs and stores it in the appropriate data format.

[0691] Output: A consolidated dataset for use in preprocessing.

[0692] Step 2: Data Preprocessing

[0693] The server pre-processes the collected data.

[0694] Input: Unified dataset.

[0695] Data processing: The server cleanses the data, removes incomplete and duplicate data, and standardizes the format. It also calculates the percentage of customers at home during specific time periods based on delivery performance data.

[0696] Output: A preprocessed and clean dataset.

[0697] Step 3: Calculate the optimal route

[0698] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route.

[0699] Input: Preprocessed and clean dataset.

[0700] Data processing: The server provides a prompt to a generative AI model (e.g., GPT-3) and requests it to calculate a route. Specifically, it calculates an efficient delivery route taking into account the probability of delivery destinations being at home, real-time traffic congestion, and weather conditions.

[0701] Output: Optimal delivery route information.

[0702] Step 4: Distributing the Route

[0703] The server sends the calculated route to each driver's device.

[0704] Input: Optimal delivery route information.

[0705] Data processing: The server converts the route information into JSON or other appropriate format and sends it to the device via a RESTful API.

[0706] Output: Data containing route instructions is sent to the driver's device.

[0707] Step 5: View Routes

[0708] The terminal displays the received route information to the user.

[0709] Input: Route information received from the server.

[0710] Data processing: The device displays route information in map and list format. Voice guidance can also be enabled for hands-free operation.

[0711] Output: A route display that the user can see and hear.

[0712] Step 6: Send real-time updates during delivery

[0713] If the user needs to change the route during delivery, the user sends that information from the terminal to the server.

[0714] Input: Real-time information such as traffic jams encountered during delivery.

[0715] Data processing: The device sends the information from the user to the server, which recalculates the route in real time based on the new data.

[0716] Output: The newly calculated route information is sent to the device.

[0717] Step 7: Emotion Recognition

[0718] The emotion engine installed in the device recognizes the user's emotions.

[0719] Input: User speech and input data.

[0720] Data processing: The emotion engine analyzes voice tone, typing speed, and selected vocabulary to determine the user's emotions.

[0721] Output: The recognized emotion data is sent to the server.

[0722] Step 8: Feedback after delivery

[0723] After completing the delivery, the user provides feedback, which is then sent to the server by the device and analyzed by the emotion engine.

[0724] Input: Feedback information after delivery completion, user sentiment information.

[0725] Data processing: The server analyzes the feedback information and combines it with the accumulated emotion data.

[0726] Output: Analysis results that will be reflected in the next route generation and improvements to the operation interface.

[0727] Step 9: Use emotional feedback

[0728] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface.

[0729] Input: Aggregated feedback and sentiment data.

[0730] Data processing: The server uses data analysis software (e.g., Python data analysis libraries) to find patterns and trends.

[0731] Output: Improved route generation algorithm and updated user interface.

[0732] (Application example 2)

[0733] 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."

[0734] While maximizing delivery efficiency at logistics centers, there is a need to reduce drivers' working environments and psychological stress. However, conventional systems focus only on optimizing delivery routes without considering the driver's emotional state. This can lead to driver stress and dissatisfaction, which can lead to a decline in delivery quality and efficiency. Therefore, in addition to optimizing delivery routes, it is important to utilize driver emotional data to improve the working environment.

[0735] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery lockers, means for acquiring traffic congestion and weather data from outside, means for pre-processing the acquired data, means for calculating the optimal delivery route using a generation AI, means for transmitting the calculated delivery route to the terminal, means for recalculating the route in real time and transmitting it again to the terminal if a change in the route is necessary, means for receiving feedback after delivery is completed and reflecting this in the generation of the next route, means for recognizing and analyzing the emotions of the delivery person from voice and input data using an emotion engine, and means for improving the interface and route based on the driver's emotion data. This maximizes delivery efficiency while reducing the driver's working environment and psychological stress.

[0736] "Delivery performance data" refers to data that contains information about deliveries that have been made in the past, including the date and time, address, and whether the delivery was successful or unsuccessful.

[0737] "Delivery box availability" is data indicating whether each delivery box is in use or not, and is acquired in real time.

[0738] "Traffic congestion status" is data showing the state of traffic congestion on roads and is collected through an external API.

[0739] "Weather Data" means data containing information about current weather and forecasts, collected through external APIs.

[0740] "Preprocessing" refers to the process of cleansing the collected raw data, removing missing values ​​and duplicate data, and converting it into a format suitable for analysis and interpretation.

[0741] "Generative AI" is a system that uses artificial intelligence technology to generate optimal delivery routes, making predictions and optimizations based on past delivery performance and real-time data.

[0742] "Terminal" refers to the mobile device or smartphone carried by the delivery person, which is used to input delivery route information and feedback from the server.

[0743] "Real-time recalculation" refers to the process of instantly recalculating a new delivery route based on new information obtained during the delivery route.

[0744] The "emotion engine" is a system that recognizes and analyzes user emotions from voice and text input, and is used to understand the driver's stress and satisfaction.

[0745] "Feedback" refers to information provided by the driver after completing a delivery, including information about any troubles or emotional stress experienced during the delivery.

[0746] "Interface" refers to the parts that the user directly operates, including the screen and operations of the delivery application.

[0747] The "at-home rate of delivery destination" is data indicating the probability that the delivery destination will be at home during a specific time period, and is calculated from past delivery performance data.

[0748] The present invention is a system for realizing efficient delivery in a logistics center, and includes the following steps.

[0749] Data collection and preprocessing

[0750] The server uses external APIs to collect traffic and weather data, and retrieves delivery performance data and delivery locker availability data from local or cloud databases. The retrieved data is cleansed using the Python library "pandas," removing incomplete data and unifying duplicate data. The rate at home during specific time periods is also calculated and used as the basis for analysis.

[0751] Calculating the best route

[0752] Based on the preprocessed data, the server uses the "ortools" library to convert it into a format suitable for the generative AI model and calculates the optimal delivery route. An efficient route is generated taking into account past delivery data, real-time traffic congestion information, and weather data. This allows the system to select routes that avoid traffic congestion and times when the delivery destination is most likely to be at home. For example, it can prioritize areas where the number of people at home is high in the morning.

[0753] Route distribution and display

[0754] The calculated optimal route information is sent from the server to the driver's device. The device displays this information in map or list format, making it intuitive for the driver. For example, if the next delivery destination is "XXX Building," specific instructions such as "Turn right -> Go straight -> Turn left" are displayed.

[0755] Emotion recognition by emotion engine

[0756] The emotion engine on the device analyzes the driver's voice and manual input data to recognize emotions. This uses emotion analysis models from the "transformers" library. For example, if a driver says "I'm stressed today" via voice input, the emotion engine will analyze it and determine the emotion as "stress."

[0757] Real-time updates and feedback

[0758] If the user (driver) requests a route change during a delivery, that information is sent from the device to the server. The server immediately recalculates the route based on the new data and resends it to the device. Also, when the driver provides feedback on the delivery on the device after completing the delivery, the emotion engine analyzes this feedback information and sends the stress or dissatisfaction the driver felt to the server.

[0759] Utilizing emotional feedback

[0760] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the system will avoid that route or consider alternative options. Furthermore, by adjusting driver rest points and delivery pace based on the emotional data, the system also contributes to improving driver satisfaction.

[0761] Specific examples

[0762] If a driver has five deliveries scheduled for a given day, the server will use past performance data to include addresses with a high rate of people at home around 10:00 a.m. in the route, and will also obtain information on delivery box usage and include boxes with many empty spaces in the route. A route that avoids traffic jams is generated based on traffic congestion data, and this is sent to the driver's device in the form of "Point A -> Point B -> Point C." If the driver encounters traffic jams along the way, a new route is instantly recalculated and distributed. After completing a delivery, the driver enters the reason for the delay, and at the same time, the emotion analysis engine recognizes "stress." This will lead to improvements to the next delivery route and interface.

[0763] Example prompt: "Please identify the emotion from the following speech input: 'I'm feeling stressed today.'"

[0764] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0765] Step 1: Data collection

[0766] The server uses an external API to collect traffic and weather data. It also retrieves delivery history data and delivery box availability data from a local or cloud database. Specifically, it uses the requests library to access the API and retrieve data in JSON format. An API key and endpoint are required as input, and the retrieved data is passed to the server in JSON format. This makes it possible to collect various required data in real time.

[0767] Step 2: Data Preprocessing

[0768] The server cleanses the acquired data using the pandas library, removing incomplete and duplicate data. It also extracts important information such as date, time, and address, and calculates the at-home rate for a specific time period. The collected raw data is passed as input, and cleaned data and at-home rate data are generated as output. This process formats the data in a format suitable for analysis and prediction.

[0769] Step 3: Calculate the optimal route

[0770] The server uses the ortools library to calculate the optimal delivery route based on the preprocessed data. Specifically, collected delivery performance data, real-time traffic information, weather data, etc. are input into a generative AI model to generate an efficient route. Clean data and at-home rate data are used as input, and optimal route information is generated as output. This makes it possible to select routes that take into account time periods when delivery destinations are most likely to be at home and traffic conditions.

[0771] Step 4: Distribute and display your route

[0772] The server sends the calculated optimal route information to each driver's device. The device displays this information in map or list format, allowing the driver to understand it intuitively. For example, delivery destinations and routes are displayed in detail. The optimal route information is passed to the device as input, and is displayed as output in a visually easy-to-understand format. This allows the driver to easily check the next delivery destination and route.

[0773] Step 5: Emotion Recognition

[0774] The emotion engine installed on the device analyzes the driver's voice input and manual input data to recognize emotions. Specifically, it uses the emotion analysis model from the Transformers library. The driver's voice and text are passed as input, and analyzed emotional information is generated as output. This allows the driver's current emotional state to be understood.

[0775] Step 6: Real-time updates

[0776] If the user (driver) requests a route change during a delivery, that information is sent from the device to the server. The server recalculates the route based on the new data and resends it. Real-time data from the driver is used as input, and new optimal route information is generated as output. This allows for flexible route changes to adapt to real-world conditions.

[0777] Step 7: Gather feedback

[0778] After completing a delivery, the user (driver) enters feedback about the delivery on the terminal. The emotion engine simultaneously analyzes the driver's emotions and sends the data to the server. The driver's feedback text and emotional data are passed as input, and the analyzed feedback information is saved as output. This accumulates data that will be useful for generating the next delivery route and improving the system.

[0779] Step 8: Use emotional feedback

[0780] The server analyzes the collected feedback and emotion data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, it will avoid that route or consider alternative options. The analyzed feedback data is used as input, and improvement suggestions are generated as output. This improves driver satisfaction and maximizes delivery efficiency.

[0781] 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.

[0782] 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.

[0783] 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.

[0784] [Third embodiment]

[0785] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0786] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0787] 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).

[0788] 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.

[0789] 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.

[0790] 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).

[0791] 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.

[0792] 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.

[0793] 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.

[0794] 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.

[0795] 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.

[0796] 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."

[0797] This invention is a system for improving the efficiency of logistics, which uses generative AI to calculate optimal delivery routes and provide them to truck drivers. This system is based mainly on interactions between a server, terminals, and users, and detailed embodiments are described below.

[0798] Data collection and preprocessing

[0799] First, the server collects delivery history data, delivery box availability data, traffic congestion data, and weather data. Delivery history data includes the date, time, address, and delivery success / failure information for past deliveries. Delivery box availability data is obtained in real time and reflects the location and usage status of each box. Traffic congestion and weather data are collected using an external API.

[0800] The server then pre-processes this data. From the delivery performance data, the rate of at-home delivery during a specific time period is calculated. Data pre-processing cleans incomplete and duplicate data.

[0801] Calculating the best route

[0802] The server inputs the preprocessed data into the generation AI to calculate the optimal delivery route. The generation AI takes into account past performance, the current state of the delivery lockers, traffic congestion, and weather data to generate an efficient route.

[0803] For example, based on past delivery data, the system can prioritize deliveries to areas with a high rate of people at home in the morning, and calculate routes that avoid traffic jams using real-time traffic information.In addition, if the weather is bad, it can select a route that uses many delivery boxes.

[0804] Route distribution and display

[0805] The server sends the calculated route to each driver's device. The device then displays the received route information to the user. The display format is a map or list, and includes delivery destinations, routes, and important points. For example, if the next delivery destination is "XXX Building," specific instructions such as "Turn right -> Go straight -> Turn left" are displayed.

[0806] Real-time updates and feedback

[0807] If the user (driver) needs to change the route during a delivery, the information is sent from the device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the user provides feedback, which is sent from the device to the server. The server analyzes the feedback and uses it to generate the next route.

[0808] Specific examples

[0809] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses with a high rate of people being at home around 10:00 AM when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C." The device that receives the information displays detailed route information to the driver along with a map.

[0810] If the driver encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and submits a new route. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to generate future routes.

[0811] In this way, the system of the present invention improves delivery efficiency and reduces redelivery, thereby contributing to solving logistics issues.

[0812] The processing flow will be explained below.

[0813] Step 1:

[0814] When the server logs in, it retrieves delivery performance data for the delivery area over the past year from the database. This data includes the address of each delivery destination, delivery time, and delivery success / failure information.

[0815] Step 2:

[0816] The server retrieves current availability data from the delivery locker API in real time. The dataset includes the ID, location, and usage status (available / in use) of each delivery locker.

[0817] Step 3:

[0818] The server retrieves traffic and weather data for the day from an external API (e.g., Google Maps API or Japan Meteorological Agency API), including the degree of traffic congestion on major roads and weather conditions (sunny, rainy, snowy, etc.).

[0819] Step 4:

[0820] The server cleanses the data it receives, removes incomplete and duplicate data, completes it, and standardizes the format of all data.

[0821] Step 5:

[0822] The server calculates the percentage of delivery destinations at home during a specific time period based on past delivery performance data. For example, it identifies delivery destinations with a high percentage of delivery destinations at home between 9:00 AM and 11:00 AM.

[0823] Step 6:

[0824] The server receives a list of packages to be delivered and delivery area information, including the delivery address and desired delivery time for each package.

[0825] Step 7:

[0826] The server uses the generated AI to calculate the optimal delivery route based on the pre-processed data. The AI ​​model is input with collected past data, the current availability of delivery boxes, traffic congestion information, weather conditions, etc.

[0827] Step 8:

[0828] The server sends the optimal route calculated as a result to each driver's device.

[0829] Step 9:

[0830] The device displays detailed route information received to the user (driver) in map or list format, showing the next delivery destination, specific route, and important points to note.

[0831] Step 10:

[0832] If the user (driver) needs to change the route during a delivery, the device sends that information to the server, which recalculates the route in real time based on the new data and sends the new route back to the device.

[0833] Step 11:

[0834] After completing a delivery, the user (driver) enters the reason for the delivery delay or any problems that occurred as feedback into the terminal.

[0835] Step 12:

[0836] The device sends the received feedback data to the server, which analyzes the collected feedback and reflects it in the next route generation.

[0837] These processing steps create a system that improves delivery efficiency and reduces redelivery.

[0838] Example 1

[0839] 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."

[0840] Conventional logistics systems have problems with insufficient optimization of delivery routes, making it difficult to achieve efficient deliveries. It is also difficult to respond in real time to traffic congestion and weather changes, leading to redeliveries and delivery delays. Furthermore, there is an insufficient system for utilizing feedback from drivers in the next delivery, making it difficult to improve delivery efficiency.

[0841] 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.

[0842] In this invention, the server includes means for acquiring delivery-related data, means for acquiring the usage status of delivery boxes, means for externally acquiring traffic and weather information, means for preprocessing the acquired data, means for calculating an optimal delivery route using generative artificial intelligence, means for transmitting the calculated delivery route to the terminal, means for recalculating the route in real time and transmitting it again to the terminal if a route change is necessary, means for receiving feedback after delivery is completed and reflecting it in the generation of the next route, means for providing the driver with route information displayed on the terminal, and means for receiving and processing route change requests from the terminal during delivery. This enables efficient delivery route optimization and real-time response. Furthermore, by reflecting driver feedback in the next delivery, continuous improvement in delivery efficiency can be achieved.

[0843] "Delivery-related data" means data containing information about past and current deliveries, including delivery dates and times, addresses, and delivery success and failure information.

[0844] "Delivery box usage status" is data that indicates the current usage status of each delivery box, and includes the location of the box, availability status, usage history, etc.

[0845] "Traffic information" is data that indicates the current traffic situation, including road congestion and accident information.

[0846] "Weather information" refers to data that indicates current and forecast weather conditions, including precipitation, temperature, wind speed, and the like.

[0847] "Preprocessing" refers to the process of preparing acquired data so that it can be analyzed, and includes the deletion of incomplete data, the integration of duplicate data, and the extraction of necessary data.

[0848] "Generative artificial intelligence" is a system that uses artificial intelligence techniques to generate new information based on data, which is then used to calculate optimal delivery routes.

[0849] An "optimal delivery route" is a delivery sequence and route that maximizes delivery efficiency and aims to reduce time and costs.

[0850] A "terminal" is an information device used by a driver to display delivery routes and route information and to communicate with the server.

[0851] "Recalculation in real time" is a process that instantly calculates and provides new delivery routes based on current conditions.

[0852] "Feedback" is information provided by the driver after delivery, including reasons for delivery delays and special notes from the delivery destination.

[0853] "Means for providing route information to the driver" refers to a method for communicating the calculated delivery route and detailed instructions to the driver via the terminal.

[0854] The "means for receiving and processing a route change request from a terminal" is a method in which a server receives a route change request sent by a driver from a terminal, calculates and provides a new route based on the request.

[0855] This invention is a system for improving logistics efficiency, which uses generative artificial intelligence (generative AI model) to calculate optimal delivery routes and provide them to drivers. This system consists of a server, terminals, and users, and handles the collection, preprocessing, analysis, display, and feedback of delivery-related data.

[0856] Data collection and preprocessing

[0857] server

[0858] The server collects data using the following means:

[0859] 1. Delivery-related data: Obtain past delivery records from the database. For example, obtain data such as "2023-10-01, Address A, Delivery successful."

[0860] 2. Delivery locker usage status: Obtain real-time data using the API of an external delivery locker management system. For example, obtain data such as "Delivery locker B, available."

[0861] 3. Traffic and weather information: Get data from traffic APIs and weather APIs. For example, get data such as "Road C, traffic jam" or "Area D, rain."

[0862] The server then preprocesses the data, calculating the percentage of people at home during specific time periods from delivery-related data and cleaning up incomplete and duplicate data.

[0863] Calculating the best route

[0864] server

[0865] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route. This generative AI model takes into account delivery history, the current state of the parcel lockers, traffic congestion, and weather data to generate an efficient delivery route.

[0866] Specifically, the generative AI model calculates the following route:

[0867] 1. Prioritize deliveries to areas where people are more likely to be at home in the morning.

[0868] 2. Select a route that avoids traffic jams based on real-time traffic information.

[0869] 3. If the weather is bad, choose a route that uses more delivery boxes.

[0870] Route distribution and display

[0871] Servers and Terminals

[0872] The server sends the calculated delivery route to each driver's device. The device that receives the route information displays it to the user (driver). The display format is a map or list, and includes delivery destinations, routes, and important points. Specific examples of instructions include detailed instructions such as "Next delivery destination is XXX Building" and "Turn right -> go straight -> turn left."

[0873] Real-time updates and feedback

[0874] Users and Servers

[0875] The user (driver) can input new information during the delivery. For example, if the user encounters a traffic jam, the user can report the situation to the server from the terminal: "Road G, traffic jam starting."

[0876] The server recalculates the route in real time based on the new data and resends it to the device. After the delivery is completed, the user can provide feedback such as the reason for the delivery delay or any special notes at the delivery destination, and the server will reflect this in the next route generation.

[0877] Examples and prompts

[0878] Specific examples

[0879] One day, a driver is scheduled to make five deliveries. Based on past performance data, the server prioritizes addresses with a high rate of people at home around 10:00 a.m. in the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most vacant boxes. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C." The device that receives the route displays detailed route information to the driver along with a map. If the driver encounters traffic congestion along the way, new information is sent from the device to the server. The server recalculates the route in real time and presents a new route. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to generate future routes.

[0880] Prompt example

[0881] "Generate the optimal delivery route to efficiently complete five deliveries based on past delivery data, drop-box availability, traffic information, and weather data."

[0882] In this way, the system of the present invention enables efficient delivery routes and real-time responses, and by reflecting driver feedback in the next delivery, it achieves continuous improvement in delivery efficiency.

[0883] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0884] Step 1: Collect data

[0885] The server collects delivery-related data, parcel box usage, traffic information, and weather information.

[0886] Input: Requests to external APIs or databases

[0887] Data processing: Receives the response from the API, analyzes its contents, and saves it in the database.

[0888] Output: Raw data awaiting preprocessing

[0889] What it does: Performs database queries and external API calls to gather delivery records, parcel locker status, and real-time traffic and weather information.

[0890] Step 2: Preprocessing the data

[0891] The server pre-processes the collected data.

[0892] Input: Delivery-related data, parcel locker usage, traffic information, weather information

[0893] Data processing: missing data imputation, duplicate data removal, outlier detection and correction

[0894] Output: Cleansed data

[0895] What it does: It runs a data cleansing algorithm to remove incomplete records, merge duplicates, and calculates the percentage of people at home for a specific time period from delivery performance data.

[0896] Step 3: Calculate the optimal route

[0897] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route.

[0898] Input: Cleansed delivery data, parcel locker usage, traffic information, weather information

[0899] Data processing: All data is integrated and input into a generative AI model, which then runs an algorithm to generate the optimal route.

[0900] Output: Optimal delivery route

[0901] How it works: The generative AI model plans the optimal delivery route based on past delivery performance, real-time traffic and weather information, and available delivery lockers.

[0902] Step 4: Distributing the Route

[0903] The server sends the calculated delivery route to each driver's device.

[0904] Input: Optimal delivery route data

[0905] Data processing: Converting route information into a format that can be processed by the device

[0906] Output: Route information sent to the device

[0907] Specific operation: Encode the optimal route in JSON format or similar and send it to the device using a communication protocol (e.g., HTTP).

[0908] Step 5: View Routes

[0909] The terminal displays the received route information to the user (driver).

[0910] Input: Route information sent

[0911] Data processing: Converting route information into a visually understandable format

[0912] Output: Route information displayed in map and list format

[0913] What it does: Renders route information on the interface of a map or delivery app, showing instructions such as "Next delivery stop is XXX building" or "Turn right -> go straight -> turn left."

[0914] Step 6: Real-time updates

[0915] The server receives route change requests from users (drivers) and recalculates the route in real time.

[0916] Input: Route change request (e.g., "Route G, traffic jam starting")

[0917] Data processing: Integrate new situational data and recalculate the optimal route

[0918] Output: Updated optimal route

[0919] Specific operation: Re-run the generative AI model based on the new traffic information, generate a new route, and send it to the device.

[0920] Step 7: Use the feedback

[0921] The server receives feedback from the user (driver) after the delivery is completed and reflects it in the next route generation.

[0922] Input: Feedback data (e.g. "Delivery destination X, address unknown")

[0923] Data processing: Analyze feedback data and adjust parameters of route generation algorithm

[0924] Output: Improved route generation algorithm

[0925] Specific behavior: Save the feedback in a database and use it the next time you generate a route, or take other measures to improve it.

[0926] (Application example 1)

[0927] 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."

[0928] In logistics, conventional delivery route calculation systems have not been able to sufficiently improve delivery efficiency. In particular, it is difficult to provide optimal routes that take into account the rate at which recipients are at home, real-time traffic conditions, and weather conditions. This also makes it inefficient for drivers to check information en route. Additionally, the cost and time wasted by redelivery have also become an issue.

[0929] 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.

[0930] In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery boxes, means for externally acquiring traffic congestion and weather data, means for pre-processing the acquired data, means for calculating an optimal delivery route using a generation AI, means for transmitting the calculated delivery route to a terminal and displaying it on the smart glasses, means for recalculating the route in real time and transmitting it again to the terminal if a change in the route is necessary, and means for receiving feedback after delivery is completed and reflecting it in the next route generation, thereby enabling improved delivery efficiency and providing information in real time.

[0931] "Delivery performance data" refers to data that includes information such as the date and time of past deliveries, addresses, and delivery success / failure information.

[0932] "Delivery box availability" is data that shows the location and usage status of each box in real time.

[0933] "Traffic congestion status" is data indicating the state of traffic congestion on a road, and is acquired from an external source.

[0934] "Weather data" is data that indicates the weather forecast and current weather conditions for the area.

[0935] "Preprocessing" is a process of cleansing acquired data and eliminating incomplete data and duplicate data.

[0936] "Generative AI" is an artificial intelligence model that generates or predicts something based on given data.

[0937] The "optimal delivery route" is an efficient delivery route that minimizes time and costs, taking into account factors such as the rate at which recipients are at home, traffic congestion, and weather conditions.

[0938] "Device" means a device used by a driver, including a smartphone, tablet, smart glasses, etc.

[0939] "Smart glasses" are glasses-type devices that can transparently display visual information, allowing drivers to check the information hands-free.

[0940] "Recalculation in real time" refers to the process of instantly calculating a new route in response to changes in the situation and sending it back to the terminal.

[0941] "Feedback" refers to the reason for the delivery delay and other information provided by the driver after completing the delivery, and is data used to plan the next route.

[0942] To implement this invention, a server, smart glasses, other terminals, a generative AI model, an external API, etc. are used. Specific embodiments are described below.

[0943] Data collection and preprocessing

[0944] The server collects delivery performance data, delivery locker availability data, traffic congestion information, and weather data from external APIs. Delivery performance data includes the date and time, address, and success / failure information of past deliveries. Delivery locker availability information is obtained in real time and reflects the location and usage status of each locker. Traffic congestion and weather data are collected using external APIs.

[0945] The server then pre-processes this data, which includes cleaning out incomplete and duplicate data, and calculating the percentage of customers at home during specific time periods from the delivery performance data.

[0946] Calculating the best route

[0947] The server inputs the preprocessed data into a generative AI model to calculate the optimal delivery route. The generative AI model generates an efficient route by taking into account past delivery performance, the current state of delivery boxes, traffic congestion, and weather data. For example, it can prioritize deliveries to areas with a high rate of people at home in the morning based on past delivery data, and calculate a route that avoids traffic congestion using real-time traffic information. In addition, when the weather is bad, it can select a route that uses many delivery boxes.

[0948] Route distribution and display

[0949] The server sends the calculated route to the driver's device, which is a pair of smart glasses. The smart glasses visually display the calculated route information to the driver, allowing them to check the information hands-free. The display format is a map or list, and includes the next delivery destination, route, and important points. For example, if the next delivery destination is "AAA Building," specific instructions such as "turn right -> go straight -> turn left" are displayed.

[0950] Real-time updates and feedback

[0951] If the driver needs to change the route during a delivery, the information is sent to the server in real time via the smart glasses. The server recalculates the route in real time based on the new data and displays the new route on the smart glasses. After completing the delivery, the driver provides feedback, which is sent to the server via the smart glasses. The server analyzes the feedback and uses it to generate the next route.

[0952] Specific examples

[0953] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses where the number of people at home around 10:00 AM is high when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is displayed on the driver's smart glasses in the form of "Point A -> Point B -> Point C."

[0954] If the driver encounters a traffic jam along the way, the new information is sent to the server via the smart glasses. The server recalculates the route in real time and displays the new route visually. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to plan future routes.

[0955] Prompt Sentence Examples

[0956] "The driver's current location is in Shibuya Ward, Tokyo, and the specified delivery addresses are in Chuo Ward, Minato Ward, and Shinjuku Ward, Tokyo. The delivery must be completed between 9:00 AM and 5:00 PM. Please calculate the optimal route taking into account traffic congestion and weather conditions. Also, please take into account the usage status of each delivery box."

[0957] This system will improve delivery efficiency and enable drivers to carry out their delivery duties more efficiently.

[0958] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0959] Step 1: Data collection

[0960] The server collects delivery performance data, delivery locker availability data, traffic congestion and weather data from external APIs. Specifically, the server sends requests to each API, and obtains delivery performance data such as delivery date and time, address, and delivery success / failure information. Delivery locker availability is obtained in real time, and the location and usage status of the locker are returned as an API response. Traffic congestion and weather data are also collected from the API, allowing current traffic and weather information to be obtained. This data is temporarily stored on the server.

[0961] Step 2: Data Preprocessing

[0962] The server cleanses the collected data and eliminates incomplete and duplicate data. The delivery performance data also calculates the percentage of people at home during specific time periods. As a result of data preprocessing, a clean dataset suitable for calculating delivery routes is prepared. For example, in the delivery performance data, areas with a high percentage of people at home between 9:00 and 11:00 a.m. are extracted. The cleansed dataset is generated as an output.

[0963] Step 3: Calculate the optimal route

[0964] The server inputs the preprocessed data into a generative AI model to calculate the optimal delivery route. The generative AI model generates an efficient route by comprehensively taking into account delivery history, the current state of the delivery box, traffic congestion, and weather data. For example, based on past data, the model prioritizes addresses with a high probability of the driver being at home in the morning, and calculates a route that avoids traffic congestion using real-time traffic congestion information. The optimal route information is generated as the output.

[0965] Step 4: Distribute and display your route

[0966] The server sends the calculated optimal route to the driver's device. Specifically, smart glasses are used. The smart glasses display a map and route guidance in the driver's field of vision based on the route information received from the server. For example, if the next delivery destination is "AAA Building," specific instructions such as "turn right -> go straight -> turn left" are visually displayed. The output is a visually verifiable route guidance.

[0967] Step 5: Real-time updates

[0968] If the user (driver) needs to change the route during a delivery, the smart glasses send new information to the server in real time. The server then recalculates the route using the generative AI model based on the newly received traffic and weather information. For example, if the driver encounters a traffic jam, a new route is calculated and resent to the device. The updated route guidance is then displayed on the smart glasses as an output.

[0969] Step 6: Collect and use feedback

[0970] After completing a delivery, the user (driver) provides the server with feedback, including the reason for the delivery delay, through the smart glasses. The server stores and analyzes this feedback data, which is then used to generate the optimal route for the next delivery. For example, if delivery to a specific address was difficult, this information is reflected in the next route generation. This allows for more accurate route generation as an output.

[0971] 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.

[0972] This invention combines an emotion engine that recognizes user emotions with the calculation of optimal delivery routes using generative AI to thoroughly improve logistics efficiency. Detailed embodiments are described below.

[0973] Data collection and preprocessing

[0974] First, the server collects delivery history data, delivery box availability data, traffic congestion and weather data. Delivery history data includes information on past deliveries, including the date and time, address, and success or failure of deliveries. Delivery box availability data is obtained in real time, and reflects the location and usage status of each box in real time. Traffic congestion and weather data are collected using an external API.

[0975] The server then preprocesses this data. It calculates the percentage of people at home for a specific time period from the delivery performance data. The percentage of people at home is estimated probabilistically from past data. Next, it cleanses all data, removing incomplete and duplicate data and standardizing the format.

[0976] Calculating the best route

[0977] The server inputs the preprocessed data into the generation AI, which calculates the optimal delivery route. The generation AI considers past performance, the current state of the delivery box, traffic congestion, and weather data to generate an efficient route. This makes it possible to select a route that takes into account times when delivery recipients are most likely to be at home.

[0978] For example, based on past delivery data, the system can prioritize deliveries to areas with a high rate of people at home in the morning, and calculate routes that avoid traffic jams using real-time traffic information.In addition, when the weather is bad, it can select routes that use many delivery boxes.

[0979] Route distribution and display

[0980] The server sends the calculated route to each driver's device. The device then displays the received route information to the user. The display format is a map or list, and includes delivery destinations, routes, and important points. Specific instructions are displayed, such as "Turn right -> Go straight -> Turn left," indicating that the next delivery destination is "XXX Building."

[0981] Emotion recognition by emotion engine

[0982] The emotion engine installed in the device analyzes the driver's voice and input data to recognize their emotions at that time. The emotion engine determines emotions based on voice tone, input speed, selected vocabulary, etc. For example, if the driver is feeling stressed, information such as "The user is feeling stressed" is sent to the server.

[0983] Real-time updates and feedback

[0984] If the user (driver) needs to change the route during a delivery, they send that information from their device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the driver provides feedback on the delivery on their device. The emotion engine analyzes this feedback information and sends emotional information, such as stress or dissatisfaction felt by the driver, to the server.

[0985] Utilizing emotional feedback

[0986] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the system will avoid that route or consider alternative options. Furthermore, the system aims to improve driver satisfaction by adjusting driver rest stops and delivery pace based on the emotional data.

[0987] Specific examples

[0988] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses with a high rate of people at home around 10:00 AM when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C."

[0989] If the driver encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and submits a new route. After completing the delivery, the driver enters the reason for the delivery delay, and the emotion engine analyzes the driver's emotions at that time and sends the result as feedback to the server.

[0990] Through these processes, the system of the present invention not only improves delivery efficiency and reduces redelivery, but also comprehensively supports the resolution of logistics issues by utilizing driver emotional data.

[0991] The processing flow will be explained below.

[0992] Step 1:

[0993] When the server logs in, it retrieves delivery performance data for the delivery area over the past year from the database. This data includes the address of each delivery destination, delivery time, and delivery success / failure information.

[0994] Step 2:

[0995] The server retrieves current availability data from the delivery locker API in real time. The dataset includes the ID, location, and usage status (available / in use) of each delivery locker.

[0996] Step 3:

[0997] The server retrieves traffic and weather data for the day from an external API (e.g., Google Maps API or Japan Meteorological Agency API), including the degree of traffic congestion on major roads and weather conditions (sunny, rainy, snowy, etc.).

[0998] Step 4:

[0999] The server cleanses the data it receives, removes incomplete and duplicate data, completes it, and standardizes the format of all data.

[1000] Step 5:

[1001] The server calculates the percentage of delivery destinations at home during a specific time period based on past delivery performance data. For example, it identifies delivery destinations with a high percentage of delivery destinations at home between 9:00 AM and 11:00 AM.

[1002] Step 6:

[1003] The server receives a list of packages to be delivered and delivery area information, including the delivery address and desired delivery time for each package.

[1004] Step 7:

[1005] The server uses the generated AI to calculate the optimal delivery route based on the pre-processed data. The AI ​​model is input with collected past data, the current availability of delivery boxes, traffic congestion information, weather conditions, etc.

[1006] Step 8:

[1007] The server sends the optimal route calculated as a result to each driver's device.

[1008] Step 9:

[1009] The device displays detailed route information received to the user (driver) in map or list format, showing the next delivery destination, specific route, and important points to note.

[1010] Step 10:

[1011] The emotion engine installed in the device analyzes the user's (driver's) voice and input data to recognize their emotion at that time. For example, it can determine their emotion based on their voice tone and input speed.

[1012] Step 11:

[1013] If the user (driver) needs to change the route during a delivery, the device sends that information to the server, which recalculates the route in real time based on the new data and sends the new route back to the device.

[1014] Step 12:

[1015] After the user (driver) completes a delivery, they input the reason for the delay or any problems they encountered into the terminal. At the same time, the terminal's emotion engine analyzes the emotion data at that time.

[1016] Step 13:

[1017] The device collects feedback and emotion data and sends it to a server, which stores it and uses it for future route planning and driver care.

[1018] Step 14:

[1019] The server uses the collected feedback and emotional data to consider how to improve the next delivery route and experience, such as avoiding routes that frequently cause stress and suggesting appropriate rest stops.

[1020] This will not only improve delivery efficiency and reduce redelivery, but also realize comprehensive logistics optimization by utilizing driver emotional data.

[1021] Example 2

[1022] 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."

[1023] In conventional logistics systems, optimizing delivery routes is important for improving delivery efficiency, but human factors such as driver emotions and stress are not taken into consideration. Furthermore, response to real-time changes in the situation during delivery is insufficient, leaving a need for recalculation of optimal routes and improvement of delivery efficiency. The present invention aims to solve these problems and improve delivery efficiency while reducing driver stress.

[1024] 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.

[1025] In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery boxes, means for externally acquiring traffic congestion and weather data, means for pre-processing the acquired data, means for calculating the optimal delivery route using a generation AI, means for sending the calculated delivery route to the terminal, means for analyzing emotion data and recognizing the user's emotion, means for recalculating the route in real time and sending it again to the terminal if a change in the route is necessary, and means for receiving feedback and emotion data after delivery is completed and reflecting them in the next route generation. This not only improves delivery efficiency but also reduces driver stress and enables route improvements based on emotion data.

[1026] "Delivery performance data" refers to data that includes information on past deliveries, such as the date and time, address, and whether delivery was successful or unsuccessful.

[1027] "Availability of delivery box" is information indicating the location and real-time usage status of each delivery box.

[1028] "Traffic congestion status" is information related to traffic volume, and is data indicating the flow of traffic on roads and the degree of congestion.

[1029] "Weather data" is meteorological information obtained from an external source, and is data indicating weather conditions such as temperature, precipitation, and wind speed.

[1030] "Preprocessing" is the process of cleansing the collected data, removing duplicate data, standardizing formats, etc.

[1031] "Generative AI" is an artificial intelligence model that calculates optimal delivery routes based on collected and pre-processed data.

[1032] A "terminal" is a portable computing device used by a delivery driver, and is a device that displays route information and inputs emotional data.

[1033] "Emotional data" is information that indicates the driver's emotional state, and is data analyzed from voice tone, input speed, selected vocabulary, etc.

[1034] "Feedback" refers to information entered by the driver after completing a delivery about the reason for the delivery delay or incidents that occurred during the journey.

[1035] "Recalculation" refers to the process of recalculating the optimal route based on the latest information when a route change is necessary during delivery.

[1036] "Analysis" is the process of analyzing the acquired data in detail to find specific patterns and trends.

[1037] "Reflection" is the process of applying the analysis results to the next route generation or system improvement.

[1038] This invention is a system that improves efficiency in delivery work and reduces driver stress. It not only calculates the optimal delivery route using a generative AI model, but also combines it with an emotion engine that recognizes the user's emotions. Detailed embodiments of this invention are described below.

[1039] Data collection and preprocessing

[1040] Collection of delivery performance data

[1041] The server retrieves delivery performance data. It uses a high-performance server (e.g., a cloud server) and a database management system (e.g., MySQL) to extract data from a database containing information on past delivery dates and times, addresses, and delivery success and failures.

[1042] Collection of delivery box availability data

[1043] The server obtains the availability data of the delivery lockers in real time. The location and real-time usage status of each delivery locker are obtained from the delivery locker management system using a RESTful API.

[1044] Traffic and weather data collection

[1045] The server uses external APIs (e.g., map information API, weather information API) to obtain traffic congestion and weather data, thereby providing real-time traffic and weather information.

[1046] Next, the server preprocesses the collected data. Specifically, it performs the following operations:

[1047] Calculating the rate of at-home delivery: Using delivery performance data, we probabilistically estimate the rate of at-home delivery during a specific time period.

[1048] Data cleansing: Use the Pandas library to remove incomplete and duplicate data and standardize the data format.

[1049] Calculating the best route

[1050] The server inputs the preprocessed data into a generative AI model (e.g., GPT-3) to calculate the optimal delivery route. The generative AI model generates an efficient route by taking into account delivery history, the current state of the delivery box, traffic congestion data, and weather data.

[1051] For example, by giving the following prompt sentence to the generative AI model, it can calculate the optimal route.

[1052] Based on past data, prioritize areas with a high delivery success rate between 10:00 a.m. and noon and calculate routes that avoid traffic congestion.

[1053] The generative AI model outputs the optimal route based on this prompt.

[1054] Route distribution and display

[1055] The server sends the calculated route information to the driver's device via a RESTful API. The device then displays the received route information to the user in map or list format and also provides voice guidance.

[1056] Emotion Recognition and Data Transmission

[1057] The device is equipped with an emotion engine that recognizes emotions by analyzing the user's voice and input data. For example, if the user says "I'm a little tired," the device analyzes the tiredness from the tone of the voice and sends the information that "the user is feeling tired" to the server.

[1058] Real-time updates and feedback

[1059] If a route change is necessary during a delivery, the user sends that information from their device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the user enters feedback into the device, which the emotion engine analyzes and sends emotional information to the server.

[1060] Utilizing emotional feedback

[1061] The server analyzes the collected feedback and emotion data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the server will avoid that route or consider alternatives, such as providing rest stops for drivers.

[1062] Specific examples

[1063] If a driver is scheduled to make five deliveries on a given day, the server will prioritize addresses with a high rate of people at home around 10:00 a.m. based on past performance data. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data and sends it to the driver's device in the form of "Point A -> Point B -> Point C."

[1064] If the user encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and sends a new route. After the delivery is completed, the user enters the reason for the delivery delay, and the emotion engine analyzes the user's emotions at that time and sends the result as feedback to the server.

[1065] As described above, the present invention provides a specific system configuration and procedure for improving efficiency in delivery work and reducing stress on drivers.

[1066] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1067] Step 1: Data collection

[1068] The server collects delivery performance data, delivery box availability data, traffic congestion data, and weather data.

[1069] Input: Delivery date and time, address, delivery success / failure information, location and usage status of delivery lockers, traffic conditions and weather data from external APIs.

[1070] Data processing: The server collects data using SQL queries and RESTful APIs and stores it in the appropriate data format.

[1071] Output: A consolidated dataset for use in preprocessing.

[1072] Step 2: Data Preprocessing

[1073] The server pre-processes the collected data.

[1074] Input: Unified dataset.

[1075] Data processing: The server cleanses the data, removes incomplete and duplicate data, and standardizes the format. It also calculates the percentage of customers at home during specific time periods based on delivery performance data.

[1076] Output: A preprocessed and clean dataset.

[1077] Step 3: Calculate the optimal route

[1078] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route.

[1079] Input: Preprocessed and clean dataset.

[1080] Data processing: The server provides a prompt to a generative AI model (e.g., GPT-3) and requests it to calculate a route. Specifically, it calculates an efficient delivery route taking into account the probability of delivery destinations being at home, real-time traffic congestion, and weather conditions.

[1081] Output: Optimal delivery route information.

[1082] Step 4: Distributing the Route

[1083] The server sends the calculated route to each driver's device.

[1084] Input: Optimal delivery route information.

[1085] Data processing: The server converts the route information into JSON or other appropriate format and sends it to the device via a RESTful API.

[1086] Output: Data containing route instructions is sent to the driver's device.

[1087] Step 5: View Routes

[1088] The terminal displays the received route information to the user.

[1089] Input: Route information received from the server.

[1090] Data processing: The device displays route information in map and list format. Voice guidance can also be enabled for hands-free operation.

[1091] Output: A route display that the user can see and hear.

[1092] Step 6: Send real-time updates during delivery

[1093] If the user needs to change the route during delivery, the user sends that information from the terminal to the server.

[1094] Input: Real-time information such as traffic jams encountered during delivery.

[1095] Data processing: The device sends the information from the user to the server, which recalculates the route in real time based on the new data.

[1096] Output: The newly calculated route information is sent to the device.

[1097] Step 7: Emotion Recognition

[1098] The emotion engine installed in the device recognizes the user's emotions.

[1099] Input: User speech and input data.

[1100] Data processing: The emotion engine analyzes voice tone, typing speed, and selected vocabulary to determine the user's emotions.

[1101] Output: The recognized emotion data is sent to the server.

[1102] Step 8: Feedback after delivery

[1103] After completing the delivery, the user provides feedback, which is then sent to the server by the device and analyzed by the emotion engine.

[1104] Input: Feedback information after delivery completion, user sentiment information.

[1105] Data processing: The server analyzes the feedback information and combines it with the accumulated emotion data.

[1106] Output: Analysis results that will be reflected in the next route generation and improvements to the operation interface.

[1107] Step 9: Use emotional feedback

[1108] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface.

[1109] Input: Aggregated feedback and sentiment data.

[1110] Data processing: The server uses data analysis software (e.g., Python data analysis libraries) to find patterns and trends.

[1111] Output: Improved route generation algorithm and updated user interface.

[1112] (Application example 2)

[1113] 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."

[1114] While maximizing delivery efficiency at logistics centers, there is a need to reduce drivers' working environments and psychological stress. However, conventional systems focus only on optimizing delivery routes without considering the driver's emotional state. This can lead to driver stress and dissatisfaction, which can lead to a decline in delivery quality and efficiency. Therefore, in addition to optimizing delivery routes, it is important to utilize driver emotional data to improve the working environment.

[1115] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery lockers, means for acquiring traffic congestion and weather data from outside, means for pre-processing the acquired data, means for calculating the optimal delivery route using a generation AI, means for transmitting the calculated delivery route to the terminal, means for recalculating the route in real time and transmitting it again to the terminal if a change in the route is necessary, means for receiving feedback after delivery is completed and reflecting this in the generation of the next route, means for recognizing and analyzing the emotions of the delivery person from voice and input data using an emotion engine, and means for improving the interface and route based on the driver's emotion data. This maximizes delivery efficiency while reducing the driver's working environment and psychological stress.

[1116] "Delivery performance data" refers to data that contains information about deliveries that have been made in the past, including the date and time, address, and whether the delivery was successful or unsuccessful.

[1117] "Delivery box availability" is data indicating whether each delivery box is in use or not, and is acquired in real time.

[1118] "Traffic congestion status" is data showing the state of traffic congestion on roads and is collected through an external API.

[1119] "Weather Data" means data containing information about current weather and forecasts, collected through external APIs.

[1120] "Preprocessing" refers to the process of cleansing the collected raw data, removing missing values ​​and duplicate data, and converting it into a format suitable for analysis and interpretation.

[1121] "Generative AI" is a system that uses artificial intelligence technology to generate optimal delivery routes, making predictions and optimizations based on past delivery performance and real-time data.

[1122] "Terminal" refers to the mobile device or smartphone carried by the delivery person, which is used to input delivery route information and feedback from the server.

[1123] "Real-time recalculation" refers to the process of instantly recalculating a new delivery route based on new information obtained during the delivery route.

[1124] The "emotion engine" is a system that recognizes and analyzes user emotions from voice and text input, and is used to understand the driver's stress and satisfaction.

[1125] "Feedback" refers to information provided by the driver after completing a delivery, including information about any troubles or emotional stress experienced during the delivery.

[1126] "Interface" refers to the parts that the user directly operates, including the screen and operations of the delivery application.

[1127] The "at-home rate of delivery destination" is data indicating the probability that the delivery destination will be at home during a specific time period, and is calculated from past delivery performance data.

[1128] The present invention is a system for realizing efficient delivery in a logistics center, and includes the following steps.

[1129] Data collection and preprocessing

[1130] The server uses external APIs to collect traffic and weather data, and retrieves delivery performance data and delivery locker availability data from local or cloud databases. The retrieved data is cleansed using the Python library "pandas," removing incomplete data and unifying duplicate data. The rate at home during specific time periods is also calculated and used as the basis for analysis.

[1131] Calculating the best route

[1132] Based on the preprocessed data, the server uses the "ortools" library to convert it into a format suitable for the generative AI model and calculates the optimal delivery route. An efficient route is generated taking into account past delivery data, real-time traffic congestion information, and weather data. This allows the system to select routes that avoid traffic congestion and times when the delivery destination is most likely to be at home. For example, it can prioritize areas where the number of people at home is high in the morning.

[1133] Route distribution and display

[1134] The calculated optimal route information is sent from the server to the driver's device. The device displays this information in map or list format, making it intuitive for the driver. For example, if the next delivery destination is "XXX Building," specific instructions such as "Turn right -> Go straight -> Turn left" are displayed.

[1135] Emotion recognition by emotion engine

[1136] The emotion engine on the device analyzes the driver's voice and manual input data to recognize emotions. This uses emotion analysis models from the "transformers" library. For example, if a driver says "I'm stressed today" via voice input, the emotion engine will analyze it and determine the emotion as "stress."

[1137] Real-time updates and feedback

[1138] If the user (driver) requests a route change during a delivery, that information is sent from the device to the server. The server immediately recalculates the route based on the new data and resends it to the device. Also, when the driver provides feedback on the delivery on the device after completing the delivery, the emotion engine analyzes this feedback information and sends the stress or dissatisfaction the driver felt to the server.

[1139] Utilizing emotional feedback

[1140] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the system will avoid that route or consider alternative options. Furthermore, by adjusting driver rest points and delivery pace based on the emotional data, the system also contributes to improving driver satisfaction.

[1141] Specific examples

[1142] If a driver has five deliveries scheduled for a given day, the server will use past performance data to include addresses with a high rate of people at home around 10:00 a.m. in the route, and will also obtain information on delivery box usage and include boxes with many empty spaces in the route. A route that avoids traffic jams is generated based on traffic congestion data, and this is sent to the driver's device in the form of "Point A -> Point B -> Point C." If the driver encounters traffic jams along the way, a new route is instantly recalculated and distributed. After completing a delivery, the driver enters the reason for the delay, and at the same time, the emotion analysis engine recognizes "stress." This will lead to improvements to the next delivery route and interface.

[1143] Example prompt: "Please identify the emotion from the following speech input: 'I'm feeling stressed today.'"

[1144] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1145] Step 1: Data collection

[1146] The server uses an external API to collect traffic and weather data. It also retrieves delivery history data and delivery box availability data from a local or cloud database. Specifically, it uses the requests library to access the API and retrieve data in JSON format. An API key and endpoint are required as input, and the retrieved data is passed to the server in JSON format. This makes it possible to collect various required data in real time.

[1147] Step 2: Data Preprocessing

[1148] The server cleanses the acquired data using the pandas library, removing incomplete and duplicate data. It also extracts important information such as date, time, and address, and calculates the at-home rate for a specific time period. The collected raw data is passed as input, and cleaned data and at-home rate data are generated as output. This process formats the data in a format suitable for analysis and prediction.

[1149] Step 3: Calculate the optimal route

[1150] The server uses the ortools library to calculate the optimal delivery route based on the preprocessed data. Specifically, collected delivery performance data, real-time traffic information, weather data, etc. are input into a generative AI model to generate an efficient route. Clean data and at-home rate data are used as input, and optimal route information is generated as output. This makes it possible to select routes that take into account time periods when delivery destinations are most likely to be at home and traffic conditions.

[1151] Step 4: Distribute and display your route

[1152] The server sends the calculated optimal route information to each driver's device. The device displays this information in map or list format, allowing the driver to understand it intuitively. For example, delivery destinations and routes are displayed in detail. The optimal route information is passed to the device as input, and is displayed as output in a visually easy-to-understand format. This allows the driver to easily check the next delivery destination and route.

[1153] Step 5: Emotion Recognition

[1154] The emotion engine installed on the device analyzes the driver's voice input and manual input data to recognize emotions. Specifically, it uses the emotion analysis model from the Transformers library. The driver's voice and text are passed as input, and analyzed emotional information is generated as output. This allows the driver's current emotional state to be understood.

[1155] Step 6: Real-time updates

[1156] If the user (driver) requests a route change during a delivery, that information is sent from the device to the server. The server recalculates the route based on the new data and resends it. Real-time data from the driver is used as input, and new optimal route information is generated as output. This allows for flexible route changes to adapt to real-world conditions.

[1157] Step 7: Gather feedback

[1158] After completing a delivery, the user (driver) enters feedback about the delivery on the terminal. The emotion engine simultaneously analyzes the driver's emotions and sends the data to the server. The driver's feedback text and emotional data are passed as input, and the analyzed feedback information is saved as output. This accumulates data that will be useful for generating the next delivery route and improving the system.

[1159] Step 8: Use emotional feedback

[1160] The server analyzes the collected feedback and emotion data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, it will avoid that route or consider alternative options. The analyzed feedback data is used as input, and improvement suggestions are generated as output. This improves driver satisfaction and maximizes delivery efficiency.

[1161] 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.

[1162] 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.

[1163] 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.

[1164] [Fourth embodiment]

[1165] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1166] 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.

[1167] 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).

[1168] 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.

[1169] 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.

[1170] 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).

[1171] 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.

[1172] 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.

[1173] 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.

[1174] 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.

[1175] 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.

[1176] 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.

[1177] 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."

[1178] This invention is a system for improving the efficiency of logistics, which uses generative AI to calculate optimal delivery routes and provide them to truck drivers. This system is based mainly on interactions between a server, terminals, and users, and detailed embodiments are described below.

[1179] Data collection and preprocessing

[1180] First, the server collects delivery history data, delivery box availability data, traffic congestion data, and weather data. Delivery history data includes the date, time, address, and delivery success / failure information for past deliveries. Delivery box availability data is obtained in real time and reflects the location and usage status of each box. Traffic congestion and weather data are collected using an external API.

[1181] The server then pre-processes this data. From the delivery performance data, the rate of at-home delivery during a specific time period is calculated. Data pre-processing cleans incomplete and duplicate data.

[1182] Calculating the best route

[1183] The server inputs the preprocessed data into the generation AI to calculate the optimal delivery route. The generation AI takes into account past performance, the current state of the delivery lockers, traffic congestion, and weather data to generate an efficient route.

[1184] For example, based on past delivery data, the system can prioritize deliveries to areas with a high rate of people at home in the morning, and calculate routes that avoid traffic jams using real-time traffic information.In addition, if the weather is bad, it can select a route that uses many delivery boxes.

[1185] Route distribution and display

[1186] The server sends the calculated route to each driver's device. The device then displays the received route information to the user. The display format is a map or list, and includes delivery destinations, routes, and important points. For example, if the next delivery destination is "XXX Building," specific instructions such as "Turn right -> Go straight -> Turn left" are displayed.

[1187] Real-time updates and feedback

[1188] If the user (driver) needs to change the route during a delivery, the information is sent from the device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the user provides feedback, which is sent from the device to the server. The server analyzes the feedback and uses it to generate the next route.

[1189] Specific examples

[1190] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses with a high rate of people being at home around 10:00 AM when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C." The device that receives the information displays detailed route information to the driver along with a map.

[1191] If the driver encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and submits a new route. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to generate future routes.

[1192] In this way, the system of the present invention improves delivery efficiency and reduces redelivery, thereby contributing to solving logistics issues.

[1193] The processing flow will be explained below.

[1194] Step 1:

[1195] When the server logs in, it retrieves delivery performance data for the delivery area over the past year from the database. This data includes the address of each delivery destination, delivery time, and delivery success / failure information.

[1196] Step 2:

[1197] The server retrieves current availability data from the delivery locker API in real time. The dataset includes the ID, location, and usage status (available / in use) of each delivery locker.

[1198] Step 3:

[1199] The server retrieves traffic and weather data for the day from an external API (e.g., Google Maps API or Japan Meteorological Agency API), including the degree of traffic congestion on major roads and weather conditions (sunny, rainy, snowy, etc.).

[1200] Step 4:

[1201] The server cleanses the data it receives, removes incomplete and duplicate data, completes it, and standardizes the format of all data.

[1202] Step 5:

[1203] The server calculates the percentage of delivery destinations at home during a specific time period based on past delivery performance data. For example, it identifies delivery destinations with a high percentage of delivery destinations at home between 9:00 AM and 11:00 AM.

[1204] Step 6:

[1205] The server receives a list of packages to be delivered and delivery area information, including the delivery address and desired delivery time for each package.

[1206] Step 7:

[1207] The server uses the generated AI to calculate the optimal delivery route based on the pre-processed data. The AI ​​model is input with collected past data, the current availability of delivery boxes, traffic congestion information, weather conditions, etc.

[1208] Step 8:

[1209] The server sends the optimal route calculated as a result to each driver's device.

[1210] Step 9:

[1211] The device displays detailed route information received to the user (driver) in map or list format, showing the next delivery destination, specific route, and important points to note.

[1212] Step 10:

[1213] If the user (driver) needs to change the route during a delivery, the device sends that information to the server, which recalculates the route in real time based on the new data and sends the new route back to the device.

[1214] Step 11:

[1215] After completing a delivery, the user (driver) enters the reason for the delivery delay or any problems that occurred as feedback into the terminal.

[1216] Step 12:

[1217] The device sends the received feedback data to the server, which analyzes the collected feedback and reflects it in the next route generation.

[1218] These processing steps create a system that improves delivery efficiency and reduces redelivery.

[1219] Example 1

[1220] 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."

[1221] Conventional logistics systems have problems with insufficient optimization of delivery routes, making it difficult to achieve efficient deliveries. It is also difficult to respond in real time to traffic congestion and weather changes, leading to redeliveries and delivery delays. Furthermore, there is an insufficient system for utilizing feedback from drivers in the next delivery, making it difficult to improve delivery efficiency.

[1222] 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.

[1223] In this invention, the server includes means for acquiring delivery-related data, means for acquiring the usage status of delivery boxes, means for externally acquiring traffic and weather information, means for preprocessing the acquired data, means for calculating an optimal delivery route using generative artificial intelligence, means for transmitting the calculated delivery route to the terminal, means for recalculating the route in real time and transmitting it again to the terminal if a route change is necessary, means for receiving feedback after delivery is completed and reflecting it in the generation of the next route, means for providing the driver with route information displayed on the terminal, and means for receiving and processing route change requests from the terminal during delivery. This enables efficient delivery route optimization and real-time response. Furthermore, by reflecting driver feedback in the next delivery, continuous improvement in delivery efficiency can be achieved.

[1224] "Delivery-related data" means data containing information about past and current deliveries, including delivery dates and times, addresses, and delivery success and failure information.

[1225] "Delivery box usage status" is data that indicates the current usage status of each delivery box, and includes the location of the box, availability status, usage history, etc.

[1226] "Traffic information" is data that indicates the current traffic situation, including road congestion and accident information.

[1227] "Weather information" refers to data that indicates current and forecast weather conditions, including precipitation, temperature, wind speed, and the like.

[1228] "Preprocessing" refers to the process of preparing acquired data so that it can be analyzed, and includes the deletion of incomplete data, the integration of duplicate data, and the extraction of necessary data.

[1229] "Generative artificial intelligence" is a system that uses artificial intelligence techniques to generate new information based on data, which is then used to calculate optimal delivery routes.

[1230] An "optimal delivery route" is a delivery sequence and route that maximizes delivery efficiency and aims to reduce time and costs.

[1231] A "terminal" is an information device used by a driver to display delivery routes and route information and to communicate with the server.

[1232] "Recalculation in real time" is a process that instantly calculates and provides new delivery routes based on current conditions.

[1233] "Feedback" is information provided by the driver after delivery, including reasons for delivery delays and special notes from the delivery destination.

[1234] "Means for providing route information to the driver" refers to a method for communicating the calculated delivery route and detailed instructions to the driver via the terminal.

[1235] The "means for receiving and processing a route change request from a terminal" is a method in which a server receives a route change request sent by a driver from a terminal, calculates and provides a new route based on the request.

[1236] This invention is a system for improving logistics efficiency, which uses generative artificial intelligence (generative AI model) to calculate optimal delivery routes and provide them to drivers. This system consists of a server, terminals, and users, and handles the collection, preprocessing, analysis, display, and feedback of delivery-related data.

[1237] Data collection and preprocessing

[1238] server

[1239] The server collects data using the following means:

[1240] 1. Delivery-related data: Obtain past delivery records from the database. For example, obtain data such as "2023-10-01, Address A, Delivery successful."

[1241] 2. Delivery locker usage status: Obtain real-time data using the API of an external delivery locker management system. For example, obtain data such as "Delivery locker B, available."

[1242] 3. Traffic and weather information: Get data from traffic APIs and weather APIs. For example, get data such as "Road C, traffic jam" or "Area D, rain."

[1243] The server then preprocesses the data, calculating the percentage of people at home during specific time periods from delivery-related data and cleaning up incomplete and duplicate data.

[1244] Calculating the best route

[1245] server

[1246] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route. This generative AI model takes into account delivery history, the current state of the parcel lockers, traffic congestion, and weather data to generate an efficient delivery route.

[1247] Specifically, the generative AI model calculates the following route:

[1248] 1. Prioritize deliveries to areas where people are more likely to be at home in the morning.

[1249] 2. Select a route that avoids traffic jams based on real-time traffic information.

[1250] 3. If the weather is bad, choose a route that uses more delivery boxes.

[1251] Route distribution and display

[1252] Servers and Terminals

[1253] The server sends the calculated delivery route to each driver's device. The device that receives the route information displays it to the user (driver). The display format is a map or list, and includes delivery destinations, routes, and important points. Specific examples of instructions include detailed instructions such as "Next delivery destination is XXX Building" and "Turn right -> go straight -> turn left."

[1254] Real-time updates and feedback

[1255] Users and Servers

[1256] The user (driver) can input new information during the delivery. For example, if the user encounters a traffic jam, the user can report the situation to the server from the terminal: "Road G, traffic jam starting."

[1257] The server recalculates the route in real time based on the new data and resends it to the device. After the delivery is completed, the user can provide feedback such as the reason for the delivery delay or any special notes at the delivery destination, and the server will reflect this in the next route generation.

[1258] Examples and prompts

[1259] Specific examples

[1260] One day, a driver is scheduled to make five deliveries. Based on past performance data, the server prioritizes addresses with a high rate of people at home around 10:00 a.m. in the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most vacant boxes. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C." The device that receives the route displays detailed route information to the driver along with a map. If the driver encounters traffic congestion along the way, new information is sent from the device to the server. The server recalculates the route in real time and presents a new route. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to generate future routes.

[1261] Prompt example

[1262] "Generate the optimal delivery route to efficiently complete five deliveries based on past delivery data, drop-box availability, traffic information, and weather data."

[1263] In this way, the system of the present invention enables efficient delivery routes and real-time responses, and by reflecting driver feedback in the next delivery, it achieves continuous improvement in delivery efficiency.

[1264] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1265] Step 1: Collect data

[1266] The server collects delivery-related data, parcel box usage, traffic information, and weather information.

[1267] Input: Requests to external APIs or databases

[1268] Data processing: Receives the response from the API, analyzes its contents, and saves it in the database.

[1269] Output: Raw data awaiting preprocessing

[1270] What it does: Performs database queries and external API calls to gather delivery records, parcel locker status, and real-time traffic and weather information.

[1271] Step 2: Preprocessing the data

[1272] The server pre-processes the collected data.

[1273] Input: Delivery-related data, parcel locker usage, traffic information, weather information

[1274] Data processing: missing data imputation, duplicate data removal, outlier detection and correction

[1275] Output: Cleansed data

[1276] What it does: It runs a data cleansing algorithm to remove incomplete records, merge duplicates, and calculates the percentage of people at home for a specific time period from delivery performance data.

[1277] Step 3: Calculate the optimal route

[1278] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route.

[1279] Input: Cleansed delivery data, parcel locker usage, traffic information, weather information

[1280] Data processing: All data is integrated and input into a generative AI model, which then runs an algorithm to generate the optimal route.

[1281] Output: Optimal delivery route

[1282] How it works: The generative AI model plans the optimal delivery route based on past delivery performance, real-time traffic and weather information, and available delivery lockers.

[1283] Step 4: Distributing the Route

[1284] The server sends the calculated delivery route to each driver's device.

[1285] Input: Optimal delivery route data

[1286] Data processing: Converting route information into a format that can be processed by the device

[1287] Output: Route information sent to the device

[1288] Specific operation: Encode the optimal route in JSON format or similar and send it to the device using a communication protocol (e.g., HTTP).

[1289] Step 5: View Routes

[1290] The terminal displays the received route information to the user (driver).

[1291] Input: Route information sent

[1292] Data processing: Converting route information into a visually understandable format

[1293] Output: Route information displayed in map and list format

[1294] What it does: Renders route information on the interface of a map or delivery app, showing instructions such as "Next delivery stop is XXX building" or "Turn right -> go straight -> turn left."

[1295] Step 6: Real-time updates

[1296] The server receives route change requests from users (drivers) and recalculates the route in real time.

[1297] Input: Route change request (e.g., "Route G, traffic jam starting")

[1298] Data processing: Integrate new situational data and recalculate the optimal route

[1299] Output: Updated optimal route

[1300] Specific operation: Re-run the generative AI model based on the new traffic information, generate a new route, and send it to the device.

[1301] Step 7: Use the feedback

[1302] The server receives feedback from the user (driver) after the delivery is completed and reflects it in the next route generation.

[1303] Input: Feedback data (e.g. "Delivery destination X, address unknown")

[1304] Data processing: Analyze feedback data and adjust parameters of route generation algorithm

[1305] Output: Improved route generation algorithm

[1306] Specific behavior: Save the feedback in a database and use it the next time you generate a route, or take other measures to improve it.

[1307] (Application example 1)

[1308] 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."

[1309] In logistics, conventional delivery route calculation systems have not been able to sufficiently improve delivery efficiency. In particular, it is difficult to provide optimal routes that take into account the rate at which recipients are at home, real-time traffic conditions, and weather conditions. This also makes it inefficient for drivers to check information en route. Additionally, the cost and time wasted by redelivery have also become an issue.

[1310] 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.

[1311] In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery boxes, means for externally acquiring traffic congestion and weather data, means for pre-processing the acquired data, means for calculating an optimal delivery route using a generation AI, means for transmitting the calculated delivery route to a terminal and displaying it on the smart glasses, means for recalculating the route in real time and transmitting it again to the terminal if a change in the route is necessary, and means for receiving feedback after delivery is completed and reflecting it in the next route generation, thereby enabling improved delivery efficiency and providing information in real time.

[1312] "Delivery performance data" refers to data that includes information such as the date and time of past deliveries, addresses, and delivery success / failure information.

[1313] "Delivery box availability" is data that shows the location and usage status of each box in real time.

[1314] "Traffic congestion status" is data indicating the state of traffic congestion on a road, and is acquired from an external source.

[1315] "Weather data" is data that indicates the weather forecast and current weather conditions for the area.

[1316] "Preprocessing" is a process of cleansing acquired data and eliminating incomplete data and duplicate data.

[1317] "Generative AI" is an artificial intelligence model that generates or predicts something based on given data.

[1318] The "optimal delivery route" is an efficient delivery route that minimizes time and costs, taking into account factors such as the rate at which recipients are at home, traffic congestion, and weather conditions.

[1319] "Device" means a device used by a driver, including a smartphone, tablet, smart glasses, etc.

[1320] "Smart glasses" are glasses-type devices that can transparently display visual information, allowing drivers to check the information hands-free.

[1321] "Recalculation in real time" refers to the process of instantly calculating a new route in response to changes in the situation and sending it back to the terminal.

[1322] "Feedback" refers to the reason for the delivery delay and other information provided by the driver after completing the delivery, and is data used to plan the next route.

[1323] To implement this invention, a server, smart glasses, other terminals, a generative AI model, an external API, etc. are used. Specific embodiments are described below.

[1324] Data collection and preprocessing

[1325] The server collects delivery performance data, delivery locker availability data, traffic congestion information, and weather data from external APIs. Delivery performance data includes the date and time, address, and success / failure information of past deliveries. Delivery locker availability information is obtained in real time and reflects the location and usage status of each locker. Traffic congestion and weather data are collected using external APIs.

[1326] The server then pre-processes this data, which includes cleaning out incomplete and duplicate data, and calculating the percentage of customers at home during specific time periods from the delivery performance data.

[1327] Calculating the best route

[1328] The server inputs the preprocessed data into a generative AI model to calculate the optimal delivery route. The generative AI model generates an efficient route by taking into account past delivery performance, the current state of delivery boxes, traffic congestion, and weather data. For example, it can prioritize deliveries to areas with a high rate of people at home in the morning based on past delivery data, and calculate a route that avoids traffic congestion using real-time traffic information. In addition, when the weather is bad, it can select a route that uses many delivery boxes.

[1329] Route distribution and display

[1330] The server sends the calculated route to the driver's device, which is a pair of smart glasses. The smart glasses visually display the calculated route information to the driver, allowing them to check the information hands-free. The display format is a map or list, and includes the next delivery destination, route, and important points. For example, if the next delivery destination is "AAA Building," specific instructions such as "turn right -> go straight -> turn left" are displayed.

[1331] Real-time updates and feedback

[1332] If the driver needs to change the route during a delivery, the information is sent to the server in real time via the smart glasses. The server recalculates the route in real time based on the new data and displays the new route on the smart glasses. After completing the delivery, the driver provides feedback, which is sent to the server via the smart glasses. The server analyzes the feedback and uses it to generate the next route.

[1333] Specific examples

[1334] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses where the number of people at home around 10:00 AM is high when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is displayed on the driver's smart glasses in the form of "Point A -> Point B -> Point C."

[1335] If the driver encounters a traffic jam along the way, the new information is sent to the server via the smart glasses. The server recalculates the route in real time and displays the new route visually. After completing the delivery, the driver provides feedback, such as the reason for the delivery delay, which the server uses to plan future routes.

[1336] Prompt Sentence Examples

[1337] "The driver's current location is in Shibuya Ward, Tokyo, and the specified delivery addresses are in Chuo Ward, Minato Ward, and Shinjuku Ward, Tokyo. The delivery must be completed between 9:00 AM and 5:00 PM. Please calculate the optimal route taking into account traffic congestion and weather conditions. Also, please take into account the usage status of each delivery box."

[1338] This system will improve delivery efficiency and enable drivers to carry out their delivery duties more efficiently.

[1339] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1340] Step 1: Data collection

[1341] The server collects delivery performance data, delivery locker availability data, traffic congestion and weather data from external APIs. Specifically, the server sends requests to each API, and obtains delivery performance data such as delivery date and time, address, and delivery success / failure information. Delivery locker availability is obtained in real time, and the location and usage status of the locker are returned as an API response. Traffic congestion and weather data are also collected from the API, allowing current traffic and weather information to be obtained. This data is temporarily stored on the server.

[1342] Step 2: Data Preprocessing

[1343] The server cleanses the collected data and eliminates incomplete and duplicate data. The delivery performance data also calculates the percentage of people at home during specific time periods. As a result of data preprocessing, a clean dataset suitable for calculating delivery routes is prepared. For example, in the delivery performance data, areas with a high percentage of people at home between 9:00 and 11:00 a.m. are extracted. The cleansed dataset is generated as an output.

[1344] Step 3: Calculate the optimal route

[1345] The server inputs the preprocessed data into a generative AI model to calculate the optimal delivery route. The generative AI model generates an efficient route by comprehensively taking into account delivery history, the current state of the delivery box, traffic congestion, and weather data. For example, based on past data, the model prioritizes addresses with a high probability of the driver being at home in the morning, and calculates a route that avoids traffic congestion using real-time traffic congestion information. The optimal route information is generated as the output.

[1346] Step 4: Distribute and display your route

[1347] The server sends the calculated optimal route to the driver's device. Specifically, smart glasses are used. The smart glasses display a map and route guidance in the driver's field of vision based on the route information received from the server. For example, if the next delivery destination is "AAA Building," specific instructions such as "turn right -> go straight -> turn left" are visually displayed. The output is a visually verifiable route guidance.

[1348] Step 5: Real-time updates

[1349] If the user (driver) needs to change the route during a delivery, the smart glasses send new information to the server in real time. The server then recalculates the route using the generative AI model based on the newly received traffic and weather information. For example, if the driver encounters a traffic jam, a new route is calculated and resent to the device. The updated route guidance is then displayed on the smart glasses as an output.

[1350] Step 6: Collect and use feedback

[1351] After completing a delivery, the user (driver) provides the server with feedback, including the reason for the delivery delay, through the smart glasses. The server stores and analyzes this feedback data, which is then used to generate the optimal route for the next delivery. For example, if delivery to a specific address was difficult, this information is reflected in the next route generation. This allows for more accurate route generation as an output.

[1352] 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.

[1353] This invention combines an emotion engine that recognizes user emotions with the calculation of optimal delivery routes using generative AI to thoroughly improve logistics efficiency. Detailed embodiments are described below.

[1354] Data collection and preprocessing

[1355] First, the server collects delivery history data, delivery box availability data, traffic congestion and weather data. Delivery history data includes information on past deliveries, including the date and time, address, and success or failure of deliveries. Delivery box availability data is obtained in real time, and reflects the location and usage status of each box in real time. Traffic congestion and weather data are collected using an external API.

[1356] The server then preprocesses this data. It calculates the percentage of people at home for a specific time period from the delivery performance data. The percentage of people at home is estimated probabilistically from past data. Next, it cleanses all data, removing incomplete and duplicate data and standardizing the format.

[1357] Calculating the best route

[1358] The server inputs the preprocessed data into the generation AI, which calculates the optimal delivery route. The generation AI considers past performance, the current state of the delivery box, traffic congestion, and weather data to generate an efficient route. This makes it possible to select a route that takes into account times when delivery recipients are most likely to be at home.

[1359] For example, based on past delivery data, the system can prioritize deliveries to areas with a high rate of people at home in the morning, and calculate routes that avoid traffic jams using real-time traffic information.In addition, when the weather is bad, it can select routes that use many delivery boxes.

[1360] Route distribution and display

[1361] The server sends the calculated route to each driver's device. The device then displays the received route information to the user. The display format is a map or list, and includes delivery destinations, routes, and important points. Specific instructions are displayed, such as "Turn right -> Go straight -> Turn left," indicating that the next delivery destination is "XXX Building."

[1362] Emotion recognition by emotion engine

[1363] The emotion engine installed in the device analyzes the driver's voice and input data to recognize their emotions at that time. The emotion engine determines emotions based on voice tone, input speed, selected vocabulary, etc. For example, if the driver is feeling stressed, information such as "The user is feeling stressed" is sent to the server.

[1364] Real-time updates and feedback

[1365] If the user (driver) needs to change the route during a delivery, they send that information from their device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the driver provides feedback on the delivery on their device. The emotion engine analyzes this feedback information and sends emotional information, such as stress or dissatisfaction felt by the driver, to the server.

[1366] Utilizing emotional feedback

[1367] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the system will avoid that route or consider alternative options. Furthermore, the system aims to improve driver satisfaction by adjusting driver rest stops and delivery pace based on the emotional data.

[1368] Specific examples

[1369] Let's say a driver has five deliveries scheduled for one day. Based on past performance data, the server prioritizes addresses with a high rate of people at home around 10:00 AM when incorporating them into the route. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data. The calculated route is sent to the driver's device in the form of "Point A -> Point B -> Point C."

[1370] If the driver encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and submits a new route. After completing the delivery, the driver enters the reason for the delivery delay, and the emotion engine analyzes the driver's emotions at that time and sends the result as feedback to the server.

[1371] Through these processes, the system of the present invention not only improves delivery efficiency and reduces redelivery, but also comprehensively supports the resolution of logistics issues by utilizing driver emotional data.

[1372] The processing flow will be explained below.

[1373] Step 1:

[1374] When the server logs in, it retrieves delivery performance data for the delivery area over the past year from the database. This data includes the address of each delivery destination, delivery time, and delivery success / failure information.

[1375] Step 2:

[1376] The server retrieves current availability data from the delivery locker API in real time. The dataset includes the ID, location, and usage status (available / in use) of each delivery locker.

[1377] Step 3:

[1378] The server retrieves traffic and weather data for the day from an external API (e.g., Google Maps API or Japan Meteorological Agency API), including the degree of traffic congestion on major roads and weather conditions (sunny, rainy, snowy, etc.).

[1379] Step 4:

[1380] The server cleanses the data it receives, removes incomplete and duplicate data, completes it, and standardizes the format of all data.

[1381] Step 5:

[1382] The server calculates the percentage of delivery destinations at home during a specific time period based on past delivery performance data. For example, it identifies delivery destinations with a high percentage of delivery destinations at home between 9:00 AM and 11:00 AM.

[1383] Step 6:

[1384] The server receives a list of packages to be delivered and delivery area information, including the delivery address and desired delivery time for each package.

[1385] Step 7:

[1386] The server uses the generated AI to calculate the optimal delivery route based on the pre-processed data. The AI ​​model is input with collected past data, the current availability of delivery boxes, traffic congestion information, weather conditions, etc.

[1387] Step 8:

[1388] The server sends the optimal route calculated as a result to each driver's device.

[1389] Step 9:

[1390] The device displays detailed route information received to the user (driver) in map or list format, showing the next delivery destination, specific route, and important points to note.

[1391] Step 10:

[1392] The emotion engine installed in the device analyzes the user's (driver's) voice and input data to recognize their emotion at that time. For example, it can determine their emotion based on their voice tone and input speed.

[1393] Step 11:

[1394] If the user (driver) needs to change the route during a delivery, the device sends that information to the server, which recalculates the route in real time based on the new data and sends the new route back to the device.

[1395] Step 12:

[1396] After the user (driver) completes a delivery, they input the reason for the delay or any problems they encountered into the terminal. At the same time, the terminal's emotion engine analyzes the emotion data at that time.

[1397] Step 13:

[1398] The device collects feedback and emotion data and sends it to a server, which stores it and uses it for future route planning and driver care.

[1399] Step 14:

[1400] The server uses the collected feedback and emotional data to consider how to improve the next delivery route and experience, such as avoiding routes that frequently cause stress and suggesting appropriate rest stops.

[1401] This will not only improve delivery efficiency and reduce redelivery, but also realize comprehensive logistics optimization by utilizing driver emotional data.

[1402] Example 2

[1403] 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."

[1404] In conventional logistics systems, optimizing delivery routes is important for improving delivery efficiency, but human factors such as driver emotions and stress are not taken into consideration. Furthermore, response to real-time changes in the situation during delivery is insufficient, leaving a need for recalculation of optimal routes and improvement of delivery efficiency. The present invention aims to solve these problems and improve delivery efficiency while reducing driver stress.

[1405] 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.

[1406] In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery boxes, means for externally acquiring traffic congestion and weather data, means for pre-processing the acquired data, means for calculating the optimal delivery route using a generation AI, means for sending the calculated delivery route to the terminal, means for analyzing emotion data and recognizing the user's emotion, means for recalculating the route in real time and sending it again to the terminal if a change in the route is necessary, and means for receiving feedback and emotion data after delivery is completed and reflecting them in the next route generation. This not only improves delivery efficiency but also reduces driver stress and enables route improvements based on emotion data.

[1407] "Delivery performance data" refers to data that includes information on past deliveries, such as the date and time, address, and whether delivery was successful or unsuccessful.

[1408] "Availability of delivery box" is information indicating the location and real-time usage status of each delivery box.

[1409] "Traffic congestion status" is information related to traffic volume, and is data indicating the flow of traffic on roads and the degree of congestion.

[1410] "Weather data" is meteorological information obtained from an external source, and is data indicating weather conditions such as temperature, precipitation, and wind speed.

[1411] "Preprocessing" is the process of cleansing the collected data, removing duplicate data, standardizing formats, etc.

[1412] "Generative AI" is an artificial intelligence model that calculates optimal delivery routes based on collected and pre-processed data.

[1413] A "terminal" is a portable computing device used by a delivery driver, and is a device that displays route information and inputs emotional data.

[1414] "Emotional data" is information that indicates the driver's emotional state, and is data analyzed from voice tone, input speed, selected vocabulary, etc.

[1415] "Feedback" refers to information entered by the driver after completing a delivery about the reason for the delivery delay or incidents that occurred during the journey.

[1416] "Recalculation" refers to the process of recalculating the optimal route based on the latest information when a route change is necessary during delivery.

[1417] "Analysis" is the process of analyzing the acquired data in detail to find specific patterns and trends.

[1418] "Reflection" is the process of applying the analysis results to the next route generation or system improvement.

[1419] This invention is a system that improves efficiency in delivery work and reduces driver stress. It not only calculates the optimal delivery route using a generative AI model, but also combines it with an emotion engine that recognizes the user's emotions. Detailed embodiments of this invention are described below.

[1420] Data collection and preprocessing

[1421] Collection of delivery performance data

[1422] The server retrieves delivery performance data. It uses a high-performance server (e.g., a cloud server) and a database management system (e.g., MySQL) to extract data from a database containing information on past delivery dates and times, addresses, and delivery success and failures.

[1423] Collection of delivery box availability data

[1424] The server obtains the availability data of the delivery lockers in real time. The location and real-time usage status of each delivery locker are obtained from the delivery locker management system using a RESTful API.

[1425] Traffic and weather data collection

[1426] The server uses external APIs (e.g., map information API, weather information API) to obtain traffic congestion and weather data, thereby providing real-time traffic and weather information.

[1427] Next, the server preprocesses the collected data. Specifically, it performs the following operations:

[1428] Calculating the rate of at-home delivery: Using delivery performance data, we probabilistically estimate the rate of at-home delivery during a specific time period.

[1429] Data cleansing: Use the Pandas library to remove incomplete and duplicate data and standardize the data format.

[1430] Calculating the best route

[1431] The server inputs the preprocessed data into a generative AI model (e.g., GPT-3) to calculate the optimal delivery route. The generative AI model generates an efficient route by taking into account delivery history, the current state of the delivery box, traffic congestion data, and weather data.

[1432] For example, by giving the following prompt sentence to the generative AI model, it can calculate the optimal route.

[1433] Based on past data, prioritize areas with a high delivery success rate between 10:00 a.m. and noon and calculate routes that avoid traffic congestion.

[1434] The generative AI model outputs the optimal route based on this prompt.

[1435] Route distribution and display

[1436] The server sends the calculated route information to the driver's device via a RESTful API. The device then displays the received route information to the user in map or list format and also provides voice guidance.

[1437] Emotion Recognition and Data Transmission

[1438] The device is equipped with an emotion engine that recognizes emotions by analyzing the user's voice and input data. For example, if the user says "I'm a little tired," the device analyzes the tiredness from the tone of the voice and sends the information that "the user is feeling tired" to the server.

[1439] Real-time updates and feedback

[1440] If a route change is necessary during a delivery, the user sends that information from their device to the server. The server recalculates the route in real time based on the new data and sends it back to the device. After completing the delivery, the user enters feedback into the device, which the emotion engine analyzes and sends emotional information to the server.

[1441] Utilizing emotional feedback

[1442] The server analyzes the collected feedback and emotion data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the server will avoid that route or consider alternatives, such as providing rest stops for drivers.

[1443] Specific examples

[1444] If a driver is scheduled to make five deliveries on a given day, the server will prioritize addresses with a high rate of people at home around 10:00 a.m. based on past performance data. It also obtains the usage status of delivery boxes in the specified area and selects a route that uses boxes with the most free space. It also generates a route that avoids traffic congestion based on the obtained traffic congestion data and sends it to the driver's device in the form of "Point A -> Point B -> Point C."

[1445] If the user encounters a traffic jam along the way, the device sends new information to the server. The server recalculates the route in real time and sends a new route. After the delivery is completed, the user enters the reason for the delivery delay, and the emotion engine analyzes the user's emotions at that time and sends the result as feedback to the server.

[1446] As described above, the present invention provides a specific system configuration and procedure for improving efficiency in delivery work and reducing stress on drivers.

[1447] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1448] Step 1: Data collection

[1449] The server collects delivery performance data, delivery box availability data, traffic congestion data, and weather data.

[1450] Input: Delivery date and time, address, delivery success / failure information, location and usage status of delivery lockers, traffic conditions and weather data from external APIs.

[1451] Data processing: The server collects data using SQL queries and RESTful APIs and stores it in the appropriate data format.

[1452] Output: A consolidated dataset for use in preprocessing.

[1453] Step 2: Data Preprocessing

[1454] The server pre-processes the collected data.

[1455] Input: Unified dataset.

[1456] Data processing: The server cleanses the data, removes incomplete and duplicate data, and standardizes the format. It also calculates the percentage of customers at home during specific time periods based on delivery performance data.

[1457] Output: A preprocessed and clean dataset.

[1458] Step 3: Calculate the optimal route

[1459] The server inputs the pre-processed data into a generative AI model to calculate the optimal delivery route.

[1460] Input: Preprocessed and clean dataset.

[1461] Data processing: The server provides a prompt to a generative AI model (e.g., GPT-3) and requests it to calculate a route. Specifically, it calculates an efficient delivery route taking into account the probability of delivery destinations being at home, real-time traffic congestion, and weather conditions.

[1462] Output: Optimal delivery route information.

[1463] Step 4: Distributing the Route

[1464] The server sends the calculated route to each driver's device.

[1465] Input: Optimal delivery route information.

[1466] Data processing: The server converts the route information into JSON or other appropriate format and sends it to the device via a RESTful API.

[1467] Output: Data containing route instructions is sent to the driver's device.

[1468] Step 5: View Routes

[1469] The terminal displays the received route information to the user.

[1470] Input: Route information received from the server.

[1471] Data processing: The device displays route information in map and list format. Voice guidance can also be enabled for hands-free operation.

[1472] Output: A route display that the user can see and hear.

[1473] Step 6: Send real-time updates during delivery

[1474] If the user needs to change the route during delivery, the user sends that information from the terminal to the server.

[1475] Input: Real-time information such as traffic jams encountered during delivery.

[1476] Data processing: The device sends the information from the user to the server, which recalculates the route in real time based on the new data.

[1477] Output: The newly calculated route information is sent to the device.

[1478] Step 7: Emotion Recognition

[1479] The emotion engine installed in the device recognizes the user's emotions.

[1480] Input: User speech and input data.

[1481] Data processing: The emotion engine analyzes voice tone, typing speed, and selected vocabulary to determine the user's emotions.

[1482] Output: The recognized emotion data is sent to the server.

[1483] Step 8: Feedback after delivery

[1484] After completing the delivery, the user provides feedback, which is then sent to the server by the device and analyzed by the emotion engine.

[1485] Input: Feedback information after delivery completion, user sentiment information.

[1486] Data processing: The server analyzes the feedback information and combines it with the accumulated emotion data.

[1487] Output: Analysis results that will be reflected in the next route generation and improvements to the operation interface.

[1488] Step 9: Use emotional feedback

[1489] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface.

[1490] Input: Aggregated feedback and sentiment data.

[1491] Data processing: The server uses data analysis software (e.g., Python data analysis libraries) to find patterns and trends.

[1492] Output: Improved route generation algorithm and updated user interface.

[1493] (Application example 2)

[1494] 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."

[1495] While maximizing delivery efficiency at logistics centers, there is a need to reduce drivers' working environments and psychological stress. However, conventional systems focus only on optimizing delivery routes without considering the driver's emotional state. This can lead to driver stress and dissatisfaction, which can lead to a decline in delivery quality and efficiency. Therefore, in addition to optimizing delivery routes, it is important to utilize driver emotional data to improve the working environment.

[1496] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring delivery performance data, means for acquiring the availability of delivery lockers, means for acquiring traffic congestion and weather data from outside, means for pre-processing the acquired data, means for calculating the optimal delivery route using a generation AI, means for transmitting the calculated delivery route to the terminal, means for recalculating the route in real time and transmitting it again to the terminal if a change in the route is necessary, means for receiving feedback after delivery is completed and reflecting this in the generation of the next route, means for recognizing and analyzing the emotions of the delivery person from voice and input data using an emotion engine, and means for improving the interface and route based on the driver's emotion data. This maximizes delivery efficiency while reducing the driver's working environment and psychological stress.

[1497] "Delivery performance data" refers to data that contains information about deliveries that have been made in the past, including the date and time, address, and whether the delivery was successful or unsuccessful.

[1498] "Delivery box availability" is data indicating whether each delivery box is in use or not, and is acquired in real time.

[1499] "Traffic congestion status" is data showing the state of traffic congestion on roads and is collected through an external API.

[1500] "Weather Data" means data containing information about current weather and forecasts, collected through external APIs.

[1501] "Preprocessing" refers to the process of cleansing the collected raw data, removing missing values ​​and duplicate data, and converting it into a format suitable for analysis and interpretation.

[1502] "Generative AI" is a system that uses artificial intelligence technology to generate optimal delivery routes, making predictions and optimizations based on past delivery performance and real-time data.

[1503] "Terminal" refers to the mobile device or smartphone carried by the delivery person, which is used to input delivery route information and feedback from the server.

[1504] "Real-time recalculation" refers to the process of instantly recalculating a new delivery route based on new information obtained during the delivery route.

[1505] The "emotion engine" is a system that recognizes and analyzes user emotions from voice and text input, and is used to understand the driver's stress and satisfaction.

[1506] "Feedback" refers to information provided by the driver after completing a delivery, including information about any troubles or emotional stress experienced during the delivery.

[1507] "Interface" refers to the parts that the user directly operates, including the screen and operations of the delivery application.

[1508] The "at-home rate of delivery destination" is data indicating the probability that the delivery destination will be at home during a specific time period, and is calculated from past delivery performance data.

[1509] The present invention is a system for realizing efficient delivery in a logistics center, and includes the following steps.

[1510] Data collection and preprocessing

[1511] The server uses external APIs to collect traffic and weather data, and retrieves delivery performance data and delivery locker availability data from local or cloud databases. The retrieved data is cleansed using the Python library "pandas," removing incomplete data and unifying duplicate data. The rate at home during specific time periods is also calculated and used as the basis for analysis.

[1512] Calculating the best route

[1513] Based on the preprocessed data, the server uses the "ortools" library to convert it into a format suitable for the generative AI model and calculates the optimal delivery route. An efficient route is generated taking into account past delivery data, real-time traffic congestion information, and weather data. This allows the system to select routes that avoid traffic congestion and times when the delivery destination is most likely to be at home. For example, it can prioritize areas where the number of people at home is high in the morning.

[1514] Route distribution and display

[1515] The calculated optimal route information is sent from the server to the driver's device. The device displays this information in map or list format, making it intuitive for the driver. For example, if the next delivery destination is "XXX Building," specific instructions such as "Turn right -> Go straight -> Turn left" are displayed.

[1516] Emotion recognition by emotion engine

[1517] The emotion engine on the device analyzes the driver's voice and manual input data to recognize emotions. This uses emotion analysis models from the "transformers" library. For example, if a driver says "I'm stressed today" via voice input, the emotion engine will analyze it and determine the emotion as "stress."

[1518] Real-time updates and feedback

[1519] If the user (driver) requests a route change during a delivery, that information is sent from the device to the server. The server immediately recalculates the route based on the new data and resends it to the device. Also, when the driver provides feedback on the delivery on the device after completing the delivery, the emotion engine analyzes this feedback information and sends the stress or dissatisfaction the driver felt to the server.

[1520] Utilizing emotional feedback

[1521] The server analyzes the collected feedback and emotional data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, the system will avoid that route or consider alternative options. Furthermore, by adjusting driver rest points and delivery pace based on the emotional data, the system also contributes to improving driver satisfaction.

[1522] Specific examples

[1523] If a driver has five deliveries scheduled for a given day, the server will use past performance data to include addresses with a high rate of people at home around 10:00 a.m. in the route, and will also obtain information on delivery box usage and include boxes with many empty spaces in the route. A route that avoids traffic jams is generated based on traffic congestion data, and this is sent to the driver's device in the form of "Point A -> Point B -> Point C." If the driver encounters traffic jams along the way, a new route is instantly recalculated and distributed. After completing a delivery, the driver enters the reason for the delay, and at the same time, the emotion analysis engine recognizes "stress." This will lead to improvements to the next delivery route and interface.

[1524] Example prompt: "Please identify the emotion from the following speech input: 'I'm feeling stressed today.'"

[1525] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1526] Step 1: Data collection

[1527] The server uses an external API to collect traffic and weather data. It also retrieves delivery history data and delivery box availability data from a local or cloud database. Specifically, it uses the requests library to access the API and retrieve data in JSON format. An API key and endpoint are required as input, and the retrieved data is passed to the server in JSON format. This makes it possible to collect various required data in real time.

[1528] Step 2: Data Preprocessing

[1529] The server cleanses the acquired data using the pandas library, removing incomplete and duplicate data. It also extracts important information such as date, time, and address, and calculates the at-home rate for a specific time period. The collected raw data is passed as input, and cleaned data and at-home rate data are generated as output. This process formats the data in a format suitable for analysis and prediction.

[1530] Step 3: Calculate the optimal route

[1531] The server uses the ortools library to calculate the optimal delivery route based on the preprocessed data. Specifically, collected delivery performance data, real-time traffic information, weather data, etc. are input into a generative AI model to generate an efficient route. Clean data and at-home rate data are used as input, and optimal route information is generated as output. This makes it possible to select routes that take into account time periods when delivery destinations are most likely to be at home and traffic conditions.

[1532] Step 4: Distribute and display your route

[1533] The server sends the calculated optimal route information to each driver's device. The device displays this information in map or list format, allowing the driver to understand it intuitively. For example, delivery destinations and routes are displayed in detail. The optimal route information is passed to the device as input, and is displayed as output in a visually easy-to-understand format. This allows the driver to easily check the next delivery destination and route.

[1534] Step 5: Emotion Recognition

[1535] The emotion engine installed on the device analyzes the driver's voice input and manual input data to recognize emotions. Specifically, it uses the emotion analysis model from the Transformers library. The driver's voice and text are passed as input, and analyzed emotional information is generated as output. This allows the driver's current emotional state to be understood.

[1536] Step 6: Real-time updates

[1537] If the user (driver) requests a route change during a delivery, that information is sent from the device to the server. The server recalculates the route based on the new data and resends it. Real-time data from the driver is used as input, and new optimal route information is generated as output. This allows for flexible route changes to adapt to real-world conditions.

[1538] Step 7: Gather feedback

[1539] After completing a delivery, the user (driver) enters feedback about the delivery on the terminal. The emotion engine simultaneously analyzes the driver's emotions and sends the data to the server. The driver's feedback text and emotional data are passed as input, and the analyzed feedback information is saved as output. This accumulates data that will be useful for generating the next delivery route and improving the system.

[1540] Step 8: Use emotional feedback

[1541] The server analyzes the collected feedback and emotion data and reflects it in the next route generation and improvements to the operation interface. For example, if many drivers feel stressed on a particular route, it will avoid that route or consider alternative options. The analyzed feedback data is used as input, and improvement suggestions are generated as output. This improves driver satisfaction and maximizes delivery efficiency.

[1542] 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.

[1543] 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.

[1544] 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 robot 414.

[1545] 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.

[1546] FIG. 9 is a diagram illustrating 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 actions 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.

[1547] 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.

[1548] 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).

[1549] 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.

[1550] 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."

[1551] 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.

[1552] 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).

[1553] 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.

[1554] 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.

[1555] 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.

[1556] 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.

[1557] 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 FP...

Claims

1. A means for acquiring delivery performance data; A means to check the availability of delivery boxes, A means for externally obtaining traffic congestion and weather data; means for pre-processing the acquired data; A means of calculating the optimal delivery route using generative AI; means for transmitting the calculated delivery route to the terminal; If a route change is required, the route will be recalculated in real time and sent to the device again. A means to receive feedback after delivery is completed and reflect it in the next route generation, A system including:

2. 2. The system according to claim 1, further comprising means for calculating the rate at home of delivery destinations from past delivery performance data.

3. The system according to claim 1, further comprising means for acquiring the availability of delivery boxes in real time using an API.

Images