system

Abstract

Provide a system. [Solution] a means of forecasting resource needs during disasters; means for allocating resources based on the predicted demand; A means of collecting information on the status of the affected areas in real time, a means for providing information to users and accepting donations; A means to manage and track the delivery of donated resources; A system including:

Description

[Technical Field]

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

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

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

[0004] In modern times, when a disaster occurs, large amounts of resources are needed rapidly, and they need to be distributed quickly to the appropriate places. However, it is extremely difficult to accurately predict how much resources will be needed in each area and to distribute them effectively. Furthermore, when collecting donations and other support, it is difficult to fully grasp the specific needs of the affected areas, which can lead to uneven distribution of support. This can lead to the problem of some areas receiving excess supplies while others lack the necessary supplies. [Means for solving the problem]

[0005] To solve this problem, the present invention provides a system that combines the following elements. First, it includes a means for predicting resource demand during a disaster. This means uses a machine learning algorithm to train a predictive model based on past disaster data and the current state of the disaster-stricken area, and predicts demand in real time. Next, it includes a means for allocating resources based on the predicted demand. This means formulates a specific resource allocation plan based on the prediction results and manages and tracks deliveries in cooperation with logistics companies. Furthermore, it includes a means for collecting information on the state of the disaster-stricken area in real time, acquiring the latest data using sensors and drones. In addition, it includes a means for providing information to users and accepting donations. Users can check the specific needs of the disaster-stricken area through their terminals and provide appropriate support.

[0006] "Means for predicting resource demand during disasters" refers to a function that uses data from past disasters and current conditions in affected areas to use machine learning algorithms to predict the specific amount of supplies and funds needed for each affected area.

[0007] "Means for allocating resources based on forecasted demand" refers to the function of formulating specific resource allocation plans based on forecast results and quickly and effectively allocating necessary supplies and funds to appropriate areas.

[0008] "Means of collecting information on the status of disaster-stricken areas in real time" refers to the function of instantly acquiring and updating data on the status of disaster-stricken areas (such as the number of evacuees, shortages of supplies, and the extent of damage) using sensors, drones, etc.

[0009] "Means of providing information to users and accepting donations" refers to the function of allowing users to check the specific needs of disaster-stricken areas through their devices and accepting donations of funds, supplies, etc.

[0010] "Means for managing and tracking the delivery of donated resources" refers to the function of managing the process of resources donated by users until they are actually delivered to the disaster area and tracking the progress of delivery. [Brief explanation of the drawings]

[0011] [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

[0012] 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.

[0013] First, the terms used in the following description will be explained.

[0014] 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).

[0015] 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.

[0016] 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.

[0017] 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.

[0018] 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."

[0019] [First embodiment]

[0020] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0021] 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.

[0022] 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).

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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.

[0027] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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."

[0032] The present invention is a system for optimizing resource demand prediction and allocation in the event of a disaster, and specific embodiments thereof will be described below.

[0033] First, the system is centered around a server, which predicts the demand for supplies and funds in each region based on past disaster data and real-time information on current disaster-stricken areas.

[0034] Program operation description:

[0035] 1. Data Collection:

[0036] The server collects data on past disasters from government and NGO databases, analyzing, for example, how much supplies were needed at each evacuation shelter during a typhoon the previous year.

[0037] The server also uses sensors and drones to collect real-time data from the affected areas, including the current capacity of evacuation centers, a list of needed supplies, and the extent of the damage.

[0038] 2. Data preprocessing and model training:

[0039] The server cleanses the collected data, corrects outliers, and fills in missing values.

[0040] The server will then use the cleansed data to train machine learning models that can predict the specific supplies and funding needs of each shelter.

[0041] 3. Demand forecasting:

[0042] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies will be needed at each evacuation center.

[0043] 4. Resource Allocation:

[0044] The server then formulates a specific resource allocation plan based on the predicted demand. For example, it determines that shelter A needs 1,000 bottles of water and 500 packs of food.

[0045] The server will follow this plan and work with logistics companies to deliver the supplies.

[0046] 5. User Notification and Donation Acceptance:

[0047] The user can check the information provided by the system through the terminal. For example, they can learn that shelter A is experiencing a water shortage.

[0048] Users use their devices to donate specific items or funds, and once the donation is complete, the device sends this information to the server.

[0049] 6. Donation Management and Delivery Tracking:

[0050] The server records the donation information in a database and updates the donation status in real time.

[0051] The server coordinates with logistics companies to ensure the proper delivery of donated supplies and funds, and tracks the progress of deliveries. For example, it checks whether 50 bottles of water have arrived safely at Evacuation Center A.

[0052] Users can check the results of their donations through their devices. For example, they can confirm that the supplies they donated have arrived at shelter A.

[0053] Examples:

[0054] As a specific example, consider the case where a typhoon hits a part of Japan. Shelter A can accommodate 1,000 people, and shelter B has 800 people. From the collected data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles. Based on this prediction, the server submits a specific resource allocation plan to the logistics company and begins delivery. At the same time, the user checks the water shortage information for shelter A on their device and donates 50 bottles of water. The device sends this information to the server, which updates the donation status in real time and manages the donated water until it arrives safely at shelter A.

[0055] By using the above-described method and system, the present invention can quickly and effectively allocate necessary resources in the event of a disaster, thereby preventing uneven distribution of support.

[0056] The processing flow will be explained below.

[0057] Step 1: Data collection

[0058] The server collects past disaster data from a database of past disasters, including the amount of supplies and funds needed at each evacuation shelter, the number of evacuees, and the extent of damage caused by past disasters.

[0059] At the same time, the server collects real-time information on the current situation in the disaster area, using sensors and drones to obtain data such as the number of people in evacuation shelters, the current stock of supplies, and the extent of the damage.

[0060] Step 2: Data Preprocessing

[0061] The server cleanses the collected data on past disasters and current situations, correcting outliers and filling in missing values ​​to improve data quality.

[0062] The server converts the pre-processed data into a suitable format and prepares it for the machine learning algorithms.

[0063] Step 3: Predictive model training

[0064] The server uses machine learning algorithms to train a demand forecasting model based on the preprocessed dataset, creating a model capable of predicting each shelter's future supply and financial needs.

[0065] The server stores the trained model and prepares it for real-time predictions.

[0066] Step 4: Demand forecast

[0067] When a new disaster occurs, the server inputs current disaster area data collected in real time into the predictive model.

[0068] Based on this data, the server predicts the amount of supplies and funds that will be needed at each evacuation shelter and in the affected area.

[0069] Step 5: Resource Allocation Planning

[0070] The server creates a resource allocation plan based on predicted demand, specifically determining how much supplies should be sent from which distribution center to which shelter.

[0071] The server notifies the distribution plan to the logistics company and instructs them to prepare for delivery.

[0072] Step 6: Notify users and accept donations

[0073] The user checks the information provided by the system through their own terminal. For example, they learn that shelter A is experiencing a water shortage.

[0074] Users use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[0075] Step 7: Donation Management and Delivery Tracking

[0076] The server records donation information in a database, updates donation status in real time, and, if necessary, reallocates donated supplies and funds to areas in need.

[0077] The server tracks the delivery status of donated goods, receives delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination.

[0078] Users can check how their donations have been used through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[0079] The above are the specific processing steps of the system of the present invention. This system is expected to enable rapid and effective forecasting of demand for resources and appropriate allocation in the event of a disaster.

[0080] Example 1

[0081] 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."

[0082] In times of disaster, rapid and accurate resource allocation is required, but conventional systems have had difficulty achieving this. In particular, it was difficult to use past disaster data and create immediate demand forecasts and distribution plans based on the latest information on affected areas. Another issue was the lack of functionality to reflect user donations in real time and manage and track their delivery status.

[0083] 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.

[0084] In this invention, the server includes means for collecting past disaster data from a database, means for collecting information on disaster-stricken areas in real time using sensors and remote control devices, means for cleansing the collected data, correcting outliers and filling in missing values, means for training a machine learning model based on the cleansed data and predicting demand for supplies and funds, means for formulating a specific resource allocation plan based on the predicted demand and delivering supplies in cooperation with delivery agencies, means for providing shelter information and accepting donations via a user terminal, means for recording donation information in a database and tracking the delivery of donated resources, and means for confirming donation results via a user terminal. This makes it possible to quickly and accurately allocate necessary resources in the event of a disaster and prevent uneven distribution of support.

[0085] A "database" is an electronic system designed to efficiently store, manage, and retrieve information.

[0086] "Disaster data" refers to data containing information about natural and man-made disasters, including the date and time of the disaster, location, type and extent of damage, and necessary supplies.

[0087] A "sensor" is a device that detects a physical phenomenon and converts it into an electrical signal.

[0088] A "remote control device" is a device used to operate and control equipment or systems from a remote location.

[0089] "Real-time" refers to a state in which data and information are acquired and processed immediately.

[0090] "Data cleansing" is the process of correcting errors and missing values ​​and maintaining consistency in order to improve the quality of data.

[0091] A "machine learning model" is a mathematical model that learns patterns and rules based on large amounts of data and makes predictions and classifications for new data.

[0092] "Prediction" is the prediction of future events or conditions based on past and present data.

[0093] A "resource allocation plan" is a detailed plan for optimally distributing available resources and providing them where needed.

[0094] A "delivery organization" is an organization or infrastructure that transports goods or services to designated locations.

[0095] A "user terminal" is a device used to access the system and exchange information.

[0096] "Shelter information" refers to data about shelters, including the number of people they can accommodate, the supplies they need, and current issues.

[0097] A "donation" is when an individual or organization provides goods or funds for disaster relief.

[0098] "Recording in a database" means storing information in a database in a manner that allows it to be accessed later.

[0099] "Tracking" is the process of monitoring and confirming how goods or information are moving and where they are currently located.

[0100] "Donation results" is information showing how donated goods and funds were used and where they ended up.

[0101] MODE FOR CARRYING OUT THE INVENTION

[0102] The present invention relates to a system for optimizing resource demand forecasting and allocation in the event of a disaster, and specific embodiments thereof are described below. The core of this system is a server that makes full use of advanced data collection and machine learning.

[0103] Hardware and software used

[0104] 1. Server:

[0105] Hardware: General-purpose high-performance computer (e.g., AWS EC2 instance)

[0106] Software: Apache Kafka, Python, TENSORFLOW, Flask

[0107] 2. Terminal:

[0108] The device from which the user accesses the site (e.g., smartphone, tablet, PC)

[0109] Software: React

[0110] 3. Data Collection Equipment:

[0111] Sensors, drones, etc. (collecting on-site conditions)

[0112] Program processing overview

[0113] 1. Data Collection:

[0114] The server collects historical disaster data from government and NGO databases, specifically using Apache Kafka to retrieve time-stamped disaster data (location, type of damage, type and quantity of needed supplies).

[0115] The server uses sensors and drones to collect real-time data from the affected area, including the capacity of evacuation shelters, lists of needed supplies, and the extent of damage.

[0116] 2. Data preprocessing and model training:

[0117] The server cleanses the collected data using Python scripts, correcting outliers and completing missing values.

[0118] Next, the machine learning library TensorFlow is used to train a demand forecasting model based on disaster data, resulting in a model capable of predicting how much supplies will be needed at each evacuation center.

[0119] 3. Demand forecasting and resource allocation:

[0120] When a new disaster occurs, the server inputs real-time data into a predictive model to forecast the specific supply and funding needs of each evacuation shelter.

[0121] The server then formulates a resource allocation plan based on the predicted data and coordinates with delivery agencies to deliver the supplies to the required locations. Delivery arrangements are made through specific logs and APIs.

[0122] 4. User Notification and Donation Acceptance:

[0123] Users can check information about evacuation shelters through their devices. For example, information such as "evacuation shelter A is short of water" or "evacuation shelter B is short of food" will be displayed.

[0124] Users can donate specific items or funds using their own devices, and the information is quickly transmitted to the server.

[0125] 5. Donation Management and Delivery Tracking:

[0126] The server records the donation information in a database and updates the status in real time.

[0127] The server tracks the progress of donated goods to ensure they are delivered appropriately, and users can view on their devices how their donations are being used.

[0128] Specific examples

[0129] For example, consider a situation where a typhoon hits a part of Japan, with 1,000 people housed in shelter A and 800 people evacuated to shelter B. Based on real-time data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles of water. Based on this prediction, the server will contact a logistics company with a specific delivery schedule and arrange for the necessary supplies to be delivered to each shelter. At the same time, a user checks the water shortage information for shelter A on their device and donates 50 bottles of water. This information is updated in real time on the server, where tracking and management are performed, and the user's device is notified of the donation results.

[0130] Example prompts for generative AI models

[0131] "If a typhoon hits a certain area of ​​Japan, 1,000 people will be evacuated to shelter A and 800 people will be evacuated to shelter B. Please create an appropriate supply distribution plan for this situation and tell us which supplies and how much should be delivered to shelters A and B."

[0132] This system will enable the rapid and effective allocation of necessary resources in the event of a disaster, preventing uneven distribution of aid.

[0133] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0134] Program steps and their specific explanations:

[0135] Step 1:

[0136] Data collection

[0137] The server collects past disaster data from government and NGO databases. Specifically, it uses Apache Kafka to obtain data such as the "date and time of the disaster," "location of damage," and "type and quantity of needed supplies." The input data here is past event information related to disasters, and the output is raw data for cleansing. For example, it collects data such as "how much supplies were needed at which evacuation shelters during the previous year's typhoon."

[0138] The server also collects information on the affected areas in real time, using sensors and drones to obtain data such as the capacity of evacuation shelters, a list of necessary supplies, and the extent of damage. This input data is obtained directly from the field, and the output is real-time evacuation shelter information.

[0139] Step 2:

[0140] Data preprocessing and model training

[0141] The server cleanses the collected data using a Python script. It corrects outliers and fills in missing values ​​to maintain data integrity. It applies data cleansing processes to the raw data as input and outputs clean data. For example, if the demand for a material is incorrectly negative, it is corrected to 0 or an appropriate positive value.

[0142] The server uses the cleansed data to train a machine learning model in TensorFlow. The clean data is fed as input to the machine learning algorithm, which then outputs a demand forecasting model. For example, a model can be built to forecast future demand using past supply demand data for each evacuation shelter.

[0143] Step 3:

[0144] Demand forecasting

[0145] When a new disaster occurs, the server uses a machine learning model to predict demand based on data collected in real time. The input is real-time evacuation shelter data and a trained demand prediction model, and the output is the specific supply and funding needs of each evacuation shelter. For example, it inputs the current number of evacuees and weather information and predicts demand for water and temporary toilets when rising temperatures and strong winds are predicted.

[0146] Step 4:

[0147] Resource Allocation

[0148] The server formulates a specific resource allocation plan based on the predicted demand. The input data is the output of the demand forecasting model and the resources of logistics agencies, and the output is specific delivery instructions. For example, a specific plan may be created that "Deliver 1,000 bottles of 2-liter water to shelter A and 800 bottles of food to shelter B."

[0149] The server sends delivery instructions to logistics organizations via API or email according to the plan. For example, it might instruct a logistics company to "arrange three trucks and deliver supplies to the designated evacuation center."

[0150] Step 5:

[0151] User Notification and Donation Acceptance

[0152] The user checks the evacuation shelter information provided by the system through their terminal. The input data is the real-time evacuation shelter situation, and the output is the information displayed on the user's terminal. For example, the information displayed is "Evacuation shelter A is experiencing a water shortage."

[0153] Users use their devices to donate specific items or funds. The input for the donation is the item or amount selected by the user, and the output is the donation data sent to the server. For example, the operation is "User donates 50 bottles of water to shelter A."

[0154] Step 6:

[0155] Donation management and delivery tracking

[0156] The server records donation information in a database. The input is information about donated goods or funds, and the output is an updated database record. For example, 50 bottles of water are donated.

[0157] The server coordinates with logistics agencies to ensure the proper delivery of donated supplies and tracks the progress of deliveries. The inputs are delivery instructions and feedback from logistics agencies, and the output is delivery status updates. For example, it confirms that a truck has arrived at shelter A and that 50 bottles of water have been delivered safely.

[0158] The user checks the results of their donation on their device. The input is feedback data from the system, and the output is confirmation of the donation results displayed on the user's device. For example, a notification may be displayed stating, "The 50 bottles of donated water have arrived safely at Shelter A."

[0159] These steps will ensure that necessary resources are allocated quickly and effectively in the event of a disaster and prevent uneven distribution of aid.

[0160] (Application example 1)

[0161] 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."

[0162] During disasters, there is a sudden surge in demand for supplies and funds at evacuation centers and other facilities, making it important to distribute them quickly and appropriately. Current systems make it difficult to accurately grasp the real-time situation in disaster-stricken areas, often resulting in uneven resource distribution and delays. Furthermore, effective management and tracking of donated resources is lacking, calling for greater transparency and simplified collaboration with donors.

[0163] 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.

[0164] In this invention, the server includes a means for predicting resource demand during a disaster, a means for allocating resources based on the predicted demand, a means for collecting information on the status of disaster-stricken areas in real time, a means for providing information to users and accepting donations, a means for managing and tracking the delivery of donated resources, and a means for real-time management and optimization of distribution of supplies at a logistics center. This enables appropriate and prompt allocation of resources during a disaster, and effective support for disaster-stricken areas. It also improves information provision and transparency for donors, leading to more efficient donation management.

[0165] A "means for predicting resource demand during a disaster" is a device or program that predicts the supplies and funds required in each region and evacuation shelter based on data from past disasters and data collected in real time.

[0166] A "resource allocation tool" is a device or program that optimally allocates necessary supplies and funds to each evacuation center or region based on a predictive model.

[0167] "Means for collecting information on the status of disaster-stricken areas in real time" refers to hardware and software such as sensors, drones, and communication devices used to collect information from disaster-stricken areas in real time.

[0168] The "means for providing information to users and accepting donations" refers to a device or program that notifies users of the current situation at the evacuation shelter and the supplies they need, and accepts donations.

[0169] "Means for managing and tracking the delivery of donated resources" refers to a device or program that manages and tracks the delivery status of materials and funds donated by users and monitors their progress until they finally reach the recipient.

[0170] "Means for real-time management and distribution optimization of materials in logistics centers" refers to a device or program that monitors materials stored and managed in logistics centers in real time and distributes them optimally.

[0171] A "smartphone" is a mobile terminal that combines the functions of a mobile phone and a computer, and is a device on which applications can be installed and run.

[0172] "Smart glasses" are glasses-type wearable devices that have the function of displaying visual information and providing various types of information to the wearer.

[0173] A "logistics robot" is a robot that automatically transports and sorts materials at logistics centers and other locations.

[0174] A "predictive model" is a statistical or machine learning model built to forecast future demand based on historical data.

[0175] "Efficient donation management" means streamlining the management process from receipt of donated goods and funds to final delivery.

[0176] The present invention is a system for optimizing resource demand prediction and allocation in the event of a disaster, and specific embodiments thereof will be described below.

[0177] First, the server predicts the demand for supplies and funds in each region based on past disaster data and real-time information on current disaster areas. Past disaster data is collected from databases provided by government agencies and relief organizations. For example, it analyzes how much supplies were needed at each evacuation center during the previous year's typhoon.

[0178] The server then uses sensors and drones to obtain real-time data from the affected area, including the current capacity of evacuation centers, a list of needed supplies, and the extent of the damage.

[0179] The server cleanses the collected data, correcting outliers and filling in missing values. The server then trains a machine learning model on the cleansed data to build the ability to predict the specific supply and funding needs of each shelter. The server uses Python and the Scikit-learn library to train the model and make predictions.

[0180] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies each shelter needs. The server then creates a specific resource allocation plan based on the predicted demand. For example, it determines that shelter A needs 1,000 bottles of water and 500 packs of food.

[0181] The server then coordinates with logistics companies to deliver supplies according to this plan. Furthermore, users can check information provided by the system through their devices. For example, they may learn that shelter A is experiencing a water shortage. They can then use their devices to donate specific supplies or funds. Once the donation is complete, the device sends this information to the server.

[0182] The server records donation information in a database and updates the donation status in real time. The server coordinates with logistics companies to ensure the proper delivery of donated supplies and funds, and tracks the progress of deliveries. For example, the server can confirm whether 50 bottles of water have safely arrived at Evacuation Center A. Users can check the results of their donations on their devices.

[0183] Smartphones, smart glasses, and logistics robots are also used in logistics centers to manage and optimize the distribution of materials in real time. Materials stored and managed within the logistics center can be monitored in real time and optimally distributed. This will improve the efficiency of material distribution within the logistics center.

[0184] As a specific example, consider the case where a typhoon hits a part of Japan. Shelter A has a capacity of 1,000 people, and shelter B has 800 people. From the collected data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles. Based on this prediction, the server submits a specific resource allocation plan to the logistics company and begins delivery. At the same time, the user checks the water shortage information for shelter A on their device and donates 50 bottles of water. The device sends this information to the server, which updates the donation status in real time and manages the donated water until it reaches shelter A.

[0185] Examples of prompts to input to a generative AI model include:

[0186] "Please tell me the procedure for logistics centers to use smartphones to check demand forecasts in disaster-stricken areas in real time and allocate supplies optimally."

[0187] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0188] Step 1: Data collection

[0189] The server collects past disaster data from government and relief organization databases, and also obtains real-time data from disaster-stricken areas using sensors and drones. The server collects information such as the number of people in the affected area, a list of necessary supplies, and the extent of damage.

[0190] Input: Past disaster data, real-time data

[0191] Output: Collected data

[0192] Specific operation: The server uses APIs to retrieve data from the database, collect real-time data from each sensor and drone, and store the data.

[0193] Step 2: Data Preprocessing

[0194] The server cleanses the collected data, corrects outliers, and fills in missing values, thereby ensuring data quality.

[0195] Input: Collected data

[0196] Output: Cleansed data

[0197] Specific operation: The server uses the Pandas library to cleanse the data, correcting outliers and imputing missing values ​​automatically.

[0198] Step 3: Model training

[0199] The server will then use the cleansed data to train machine learning models that can predict the specific supply and funding needs of each shelter.

[0200] Input: Cleansed data

[0201] Output: A trained predictive model

[0202] How it works: The server uses the Scikit-learn library to train a machine learning model and generate a predictive model based on past data.

[0203] Step 4: Demand forecast

[0204] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies will be needed at each evacuation center.

[0205] Input: Trained predictive model, real-time data

[0206] Output: Forecasted demand

[0207] Specific operation: The server inputs real-time data into a predictive model to estimate the demand for supplies and funds for each evacuation shelter.

[0208] Step 5: Resource allocation

[0209] The server formulates a specific resource allocation plan based on the predicted demand and works with logistics companies to deliver supplies.

[0210] Input: Forecasted demand

[0211] Output: Resource allocation plan

[0212] Specific operation: The server creates an optimal resource allocation plan based on the prediction results and notifies the logistics company of the plan via an API.

[0213] Step 6: Notify users and accept donations

[0214] Users check the information provided by the system on their terminals and donate the necessary supplies. The donation information is then sent to the server.

[0215] Input: Required information for each evacuation shelter

[0216] Output: Donation information

[0217] Specific operation: The application on the device notifies the user of supply shortages at evacuation centers, and when the user completes the donation procedure, the information is sent to the server.

[0218] Step 7: Donation Management and Delivery Tracking

[0219] The server records donation information in a database, updates donation status in real time, and manages and tracks donated goods and funds to ensure their proper delivery.

[0220] Input: Donation information

[0221] Output: Updated donation status, delivery status of donated goods

[0222] Specific operation: The server records donation information in a database, works with logistics companies to track the delivery status of supplies, and provides feedback to users.

[0223] Step 8: Optimized material management

[0224] The logistics center will use smartphones, smart glasses, and logistics robots to manage supplies in real time and optimize their distribution.

[0225] Input: Resource allocation plan, real-time management data

[0226] Output: Optimized resource allocation

[0227] Specific operation: Each device in the logistics center moves and sorts materials in real time based on a resource allocation plan, maintaining optimal conditions.

[0228] 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.

[0229] The present invention relates to a system for realizing more effective relief activities by combining a system for predicting demand for and optimizing allocation of resources during disasters with an emotion engine that recognizes user emotions. The system according to the present invention can be implemented as follows.

[0230] Program operation description:

[0231] 1. Data Collection:

[0232] The server collects past disaster data from a database, including the amount of supplies needed, the number of evacuees, and the extent of damage caused by past disasters.

[0233] The server also collects real-time information on the current state of the affected area, using sensors and drones to obtain data such as the capacity of evacuation centers, current stock of supplies, and the extent of the damage.

[0234] 2. Data preprocessing and model training:

[0235] The server preprocesses the collected data, correcting outliers and filling in missing values, and then feeds the preprocessed data into machine learning algorithms.

[0236] The server uses the preprocessed data to train machine learning models that predict each shelter's future supply and financial needs.

[0237] 3. Demand forecasting:

[0238] When a new disaster occurs, the server inputs disaster area data collected in real time into the predictive model.

[0239] Based on this data, the server predicts the amount of supplies and funds that will be needed at each evacuation shelter and in the affected area.

[0240] 4. Resource Allocation Planning:

[0241] The server creates a resource allocation plan based on the predicted demand, specifically determining how much supplies should be sent from which distribution center to which evacuation shelter.

[0242] The server notifies the distribution plan to the logistics company and instructs them to prepare for delivery.

[0243] 5. User Notification and Donation Acceptance:

[0244] The server notifies the user's device of information about the affected area, and the user can use their device to obtain information about supplies needed to get to an evacuation shelter on foot.

[0245] Users can use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[0246] 6. Emotion recognition:

[0247] The server is equipped with an emotion engine that recognizes the user's emotions, and collects and analyzes the user's emotion data.

[0248] The terminal identifies the user's emotions in real time and presents the user with assistance options according to the emotions.

[0249] 7. Emotion-based donation suggestions:

[0250] The server proposes appropriate donation options to the user based on the user's emotional data provided by the emotion engine. For example, if the server determines that the user is willing to donate a large amount, it will present the user with larger donation options.

[0251] 8. Donation Management and Delivery Tracking:

[0252] The server records donation information in a database, updates donation status in real time, and, if necessary, reallocates donated supplies and funds to areas in need.

[0253] The server tracks the delivery status of donated goods, receives delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination.

[0254] Users can check the results of their donations through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[0255] Examples:

[0256] For example, suppose a large earthquake occurs and devastates parts of Japan. Shelter A can accommodate 1,000 people, while shelter B has 800. Based on past disaster data and real-time collected information, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800. Based on this prediction, the server notifies the logistics company of a specific resource allocation plan and instructs delivery. At the same time, the user can check the water shortage information for shelter A through their device and be presented with appropriate donation options based on their emotional data. Once the user completes their donation, this information is sent from the device to the server, which updates the donation status in real time. The server tracks the delivery status of the donated supplies and confirms that the water has safely arrived at shelter A. The user can then check how their donation has been used through their device.

[0257] By using the above method and system, the present invention is expected to make disaster resource demand forecasts and appropriate allocations quick and effective, thereby improving the efficiency of relief activities. By combining it with an emotion engine, it becomes possible to propose relief based on the user's emotions, realizing the provision of more appropriate relief.

[0258] The processing flow will be explained below.

[0259] Step 1: Data collection

[0260] The server collects disaster data from a database of past disasters, including the amount of supplies and funds needed at each evacuation shelter, the number of evacuees, and the extent of damage caused by past disasters.

[0261] The server also uses sensors and drones to collect real-time data on the current disaster area, providing information such as shelter capacity, current supply inventory, and the extent of the damage.

[0262] Step 2: Data Preprocessing

[0263] The server cleanses the collected data, corrects outliers, and fills in missing values, allowing machine learning algorithms to learn accurately.

[0264] Step 3: Predictive model training

[0265] The server uses machine learning algorithms to train a demand forecasting model based on the pre-processed data, creating a model capable of predicting each shelter's future supply and financial needs.

[0266] Step 4: Demand forecast

[0267] When a new disaster occurs, the server inputs current data from the affected area, collected in real time, into the predictive model.

[0268] Based on this data, the server predicts the specific amount of supplies and funds needed at each evacuation center and disaster area.

[0269] Step 5: Resource Allocation Planning

[0270] The server creates a resource allocation plan based on the predicted demand, specifically determining how much supplies should be sent from which distribution center to which evacuation shelter.

[0271] The server issues delivery instructions to the logistics company based on this plan.

[0272] Step 6: Notify users and accept donations

[0273] The server notifies the user's device of information about the affected area, allowing the user to check on their device what supplies are in short supply at evacuation shelters.

[0274] Users use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[0275] Step 7: Emotion Recognition

[0276] The server uses an emotion engine that recognizes the user's emotions to collect and analyze the user's emotion data.

[0277] The device analyzes the user's emotions in real time and sends the results to the server.

[0278] Step 8: Emotionally Based Donation Proposals

[0279] The server presents appropriate donation options to the user based on the user's emotional data obtained from the emotion engine, for example, presenting an urgent donation option if the user is emotionally moved.

[0280] Step 9: Donation Management and Delivery Tracking

[0281] The server records user donation information in a database, updates the donation status in real time, and verifies that donated goods and funds are distributed appropriately, as needed.

[0282] The server tracks the delivery of donated goods, obtaining delivery status from logistics providers and managing the process until the goods arrive safely at their destination.

[0283] Users can check how their donations have been used through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[0284] The above are the specific processing steps of the system of the present invention. This system is expected to make relief activities more efficient by quickly and effectively predicting demand for resources during disasters and allocating them appropriately. By combining it with an emotion engine, it becomes possible to propose relief based on the user's emotions, realizing the provision of more appropriate relief.

[0285] Example 2

[0286] 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."

[0287] During a disaster, the condition of the affected area and the demand for supplies change from moment to moment, so it is necessary to allocate resources quickly and accurately.In addition, it is necessary to realize effective relief activities by making appropriate relief proposals based on the user's emotions and motivation, but current systems are not able to adequately address these issues.

[0288] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting past disaster data and correcting outliers and imputing missing values ​​based on the data; means for training a machine learning model based on the preprocessed data and predicting future supply and financial needs at each evacuation shelter; means for collecting disaster area data in real time when a new disaster occurs and inputting it into the machine learning model; means for formulating a resource allocation plan based on the predicted demand and notifying logistics companies; means for notifying a user's terminal of disaster area information and accepting donations; means for proposing appropriate support options using an emotion engine that recognizes the user's emotions; means for recording donation information in a database, updating the donation status in real time, and tracking the delivery status of donated supplies; and means for the user to check the results of the donation via the terminal. This allows for rapid and accurate resource demand prediction and appropriate allocation during a disaster, enabling effective support activities based on the user's emotions.

[0289] "Past disaster data" refers to information about disasters that have occurred in the past, including data on the demand for supplies, the number of evacuees, and the extent of damage.

[0290] "Outlier correction" refers to the process of detecting and correcting statistically abnormal values ​​in acquired data.

[0291] "Missing value imputation" refers to the operation of filling in missing values ​​in a dataset, and generally uses statistical methods such as the mean or median.

[0292] "Preprocessed data" refers to data that has been converted into a form suitable for machine learning models by correcting outliers and filling in missing values.

[0293] A "machine learning model" refers to an algorithm or statistical model that can make predictions or classifications based on past data.

[0294] "Collecting disaster area data in real time" refers to the operation of instantly collecting data on ongoing disasters and sending it to a server.

[0295] A "resource allocation plan" refers to a plan that determines how much supplies to send from which logistics center to which shelter based on predicted demand.

[0296] "Logistics company" refers to a company or organization responsible for transporting goods.

[0297] "User terminal" refers to a computer device such as a smartphone or tablet held by a user, and is a means for communicating with a server.

[0298] "Emotion engine" refers to a system for recognizing and analyzing a user's emotional state.

[0299] "Support options" refers to the support methods and types of donations suggested to users.

[0300] "Donation Information" refers to the details of a user's donation (donation recipient, donated items, amount, date and time, etc.).

[0301] A "database" refers to a system for systematically storing and managing data.

[0302] "Donation results" refers to information about how the donation made by the user was used and what results it produced.

[0303] The present invention relates to a system for optimizing resource demand forecasting and allocation in the event of a disaster. This system is mainly composed of a server, terminals, and users.

[0304] First, the server collects past disaster data from a database. The database stores data on past disasters, such as the demand for supplies, the number of evacuees, and the extent of damage. Next, the server uses sensors and drones to collect real-time information on the current situation in the disaster-stricken area. Specifically, it obtains information such as the capacity of evacuation shelters, the current stock of supplies, and the extent of damage.

[0305] The collected data is preprocessed by the server. During the preprocessing process, outliers are corrected, missing values ​​are imputed, and the data is normalized to make it suitable for the machine learning model. The server then trains the machine learning model based on the preprocessed data. This creates a model that predicts the future supply and funding needs of each shelter.

[0306] When a new disaster occurs, the server inputs real-time data on the affected area into the trained prediction model. Based on this data, it predicts the amount of supplies and funds needed at each evacuation shelter and in the affected area. Based on the predicted demand, the server formulates a resource allocation plan, specifically determining how much supplies to send from which logistics center to which evacuation shelter. The server then notifies the distribution plan to logistics companies and issues delivery instructions.

[0307] Users can receive information about disaster areas through their devices and make donations to specific shelters or supplies. Once a donation is completed, this information is sent from the device to a server, which updates the donation status in real time.

[0308] Furthermore, the server is equipped with an emotion engine that recognizes the user's emotions. The device identifies the user's emotions in real time from their facial expressions and voice data, and presents the user with support options that correspond to those emotions. The emotion engine analyzes the user's emotional data and suggests appropriate donation options. For example, if it is inferred that the user is willing to make a large donation, the user will be presented with larger support options.

[0309] Donation information is recorded in a database, and the server updates the donation status in real time. If necessary, donated goods or funds are redistributed to areas where they are in short supply. The server tracks the delivery status of donated goods, obtains delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination. Users can check the results of their donations on their own devices.

[0310] Specific examples

[0311] For example, let's assume that a large-scale earthquake occurs and damages parts of Japan. Shelter A has a capacity of 1,000 people, and shelter B has 800 people. Based on past disaster data and information collected in real time, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800. Based on this prediction, the server notifies the logistics company of a resource allocation plan and instructs delivery. At the same time, the user checks water shortage information for shelter A through their device and is presented with the optimal donation option based on their emotional data. Once the user completes their donation, this information is sent from the device to the server, which updates the donation status in real time. The server tracks the delivery status of the donated supplies and confirms whether the water has arrived safely at shelter A. The user can then check how their donation has been used through their device.

[0312] Prompt Sentence Examples

[0313] "In an area affected by a large earthquake, 1,000 people are evacuated to shelter A and 800 people are evacuated to shelter B. Please predict the supplies needed at each shelter based on past data and real-time information, and issue instructions to optimize relief activities."

[0314] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0315] Step 1: Data collection

[0316] The server collects past disaster data from a database. Specifically, it uses SQL queries to obtain data such as the demand for supplies, the number of evacuees, and the extent of damage during past disasters. It also uses sensors and drones to collect real-time information on the current situation in disaster-stricken areas. Data from the sensors, such as the number of people accommodated, stock of supplies, and the extent of damage, is sent to the server.

[0317] Input: Past disaster database, sensors, drones

[0318] Output: Collected disaster data

[0319] Step 2: Data Preprocessing

[0320] The server preprocesses the collected data, correcting outliers and imputing missing values, and normalizing the data. Missing values ​​are imputed with the mean or median, and outliers are corrected using statistical methods. The data is then scaled and converted into a format suitable for machine learning models.

[0321] Input: Collected disaster data

[0322] Output: Preprocessed data

[0323] Step 3: Model training

[0324] The server trains a machine learning model using the preprocessed data. It splits the dataset into a training set and a test set, trains a regression model or deep learning model using the training set, evaluates the model's performance on the test set, and tunes hyperparameters as needed.

[0325] Input: Preprocessed data

[0326] Output: A trained machine learning model

[0327] Step 4: Demand forecast

[0328] When a new disaster occurs, the server inputs real-time data on the affected area into the trained prediction model, which then uses the model to predict the amount of supplies and funds needed at each evacuation center and in the affected area.

[0329] Input: Real-time disaster area data, trained machine learning model

[0330] Output: Forecasted demand for supplies and funds for each shelter

[0331] Step 5: Resource Allocation Planning

[0332] The server creates a resource allocation plan based on predicted demand. It uses an optimization algorithm to determine how much supplies should be sent from which distribution center to which shelter. It then notifies the distribution company of this plan via an API, instructing them to prepare for delivery.

[0333] Input: Forecasted demand data

[0334] Output: Resource allocation plan, notification to logistics companies

[0335] Step 6: Notify users and accept donations

[0336] The server notifies users of information about the affected areas via their devices. Users can then donate to specific shelters or supplies via their devices. Once a donation is complete, the information is sent from the device to the server, and the donation status is updated in real time.

[0337] Input: Disaster area information, user donation data

[0338] Output: User notification, donation acceptance information

[0339] Step 7: Emotion Recognition

[0340] The server uses an emotion engine to recognize the user's emotions. The device collects the user's facial expressions and voice data and analyzes them using the emotion engine. As a result, assistance options based on the user's emotions are presented.

[0341] Input: User's facial expression and voice data

[0342] Output: Analyzed emotion data, support options

[0343] Step 8: Emotionally Based Donation Proposals

[0344] The server proposes appropriate donation options to the user based on the analysis results of the emotion engine. If the server determines that the user is willing to donate a large amount, it presents the option of large-scale support.

[0345] Input: Parsed emotion data

[0346] Output: Donation proposal

[0347] Step 9: Donation Management and Delivery Tracking

[0348] The server records users' donation information in a database and updates the donation status in real time. It tracks the delivery status of donated goods and funds and obtains delivery status from logistics companies, managing the donated goods until they arrive safely at their destination. Users can check the results of their donations through their devices.

[0349] Input: Donation information, delivery status from logistics company

[0350] Output: Donation management data, delivery tracking information, user notifications

[0351] (Application example 2)

[0352] 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."

[0353] Conventional resource demand forecasting and allocation systems during disasters do not consider the psychological and emotional state of disaster victims when proposing assistance, making it difficult to provide effective assistance. Furthermore, assistance proposals and effective resource allocations are carried out in stages, making it difficult to respond quickly. The present invention aims to solve these problems by combining assistance proposals that consider users' emotions with real-time demand forecasting.

[0354] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0355] In this invention, the server includes means for predicting resource demand during a disaster, means for allocating resources based on the predicted demand, and means for collecting information on the status of the disaster area in real time, thereby enabling means for providing information to users and accepting donations, means for managing and tracking the delivery of donated resources, and means for identifying user emotions in real time and generating support proposals based on those emotions.

[0356] A "means for predicting resource demand during a disaster" is a device or program that predicts the amount of supplies and resources that will be needed when a disaster occurs.

[0357] The "means for allocating resources based on the predicted demand" is a device or program that efficiently allocates materials and resources to each location based on the predicted resource demand.

[0358] "Means for collecting information on the status of disaster-affected areas in real time" refers to devices or systems that use sensors, drones, APIs, etc. to instantly obtain the current status of areas affected by disasters.

[0359] The "means for providing information and accepting donations" refers to an interface or application that provides users with the necessary information and accepts donations from users.

[0360] "Means for managing and tracking the delivery of donated resources" refers to a system that manages the delivery of donated goods and funds to ensure they are delivered to the designated locations and tracks the delivery status in real time.

[0361] The "means for identifying a user's emotions in real time and generating assistance suggestions based on those emotions" refers to a device or system that analyzes a user's emotions in real time and suggests appropriate assistance options based on the results.

[0362] "Means of collecting data on past disasters and training a predictive model based on that data" refers to a system that collects data on past disasters and uses that data to train a demand prediction model for future disasters using a machine learning algorithm.

[0363] "Means for using a predictive model to predict the amount of supplies and funds required for each evacuation shelter" refers to a device or system that uses a trained predictive model to specifically calculate the amount of supplies and funds required for each evacuation shelter.

[0364] The present invention provides a system that combines emotion analysis with a resource demand forecasting and allocation system in the event of a disaster, and proposes assistance based on the user's emotions. Specific embodiments of the present invention are described below.

[0365] System Configuration

[0366] The system consists of the following major components:

[0367] 1. Server

[0368] 2. Terminal

[0369] 3. Users

[0370] The role and specific functions of each component

[0371] server

[0372] The server has several options:

[0373] Data collection method: The server collects past disaster data and real-time local data using sensors, drones, and APIs. For example, it obtains information about the disaster area using Google Maps API.

[0374] Prediction model training method: The server trains a prediction model using a machine learning algorithm (e.g., Scikit-learn's LinearRegression) based on the collected past disaster data.

[0375] Demand forecasting tools: Using trained predictive models, we forecast the amount of supplies and funding needed at each shelter.

[0376] Sentiment analysis method: Emotions are identified in real time using TensorFlow and OpenCV based on data (images and audio) sent by the user.

[0377] Support proposal generation means: Based on the results of sentiment analysis, appropriate donation options and support plans are proposed to the user.

[0378] Donation Management and Tracking: Manage donations and track delivery status in real time using a MySQL® database.

[0379] Terminal

[0380] The terminal has the following functions:

[0381] Information provision function: Provides users with information on the status of the disaster area and the necessary support.

[0382] Donation acceptance feature: Users can make donations to specific shelters or supplies.

[0383] Emotional data transmission function: Collects user emotional data using the smartphone's camera and microphone and transmits it to the server.

[0384] User

[0385] The user does the following:

[0386] Information reception: Receive information about the situation in the disaster area and the need for assistance.

[0387] Donate: Make a donation through your device.

[0388] Emotion data provision: Send your own emotional data to the server via your device.

[0389] Example

[0390] For example, suppose a large earthquake occurs and some areas are affected. Based on past earthquake data and real-time data, the server predicts that shelter A will need 1,000 bottles of water and shelter B will need 800 bottles of water. A user checks the water shortage information for shelter A on their device, senses the urgency, and makes a donation. Once the donation is complete, the server records the donation information in a database and tracks the delivery status.

[0391] Prompt Sentence Examples

[0392] "Designing a disaster food delivery application that optimizes donations through emotion recognition. Features include real-time demand forecasting, user emotion analysis, and donation suggestions. For example, it includes a function to estimate the food needs of shelter A and provide donation options based on the user's emotion."

[0393] It is expected that this system will solve the problems faced by conventional disaster response systems and enable efficient and effective support activities.

[0394] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0395] Step 1: Data collection

[0396] The server collects past disaster data from a database and uses sensors, drones, and APIs (e.g., Google Maps API) to obtain real-time local data (such as shelter capacity, supply inventory, and damage extent).

[0397] Input: Past disaster data, local real-time data

[0398] Data processing: Correction of outliers, completion of missing values

[0399] Output: Preprocessed dataset

[0400] Step 2: Model training

[0401] The server trains a predictive model using a machine learning algorithm (Scikit-learn's LinearRegression) based on the preprocessed data.

[0402] Input: Preprocessed dataset

[0403] Data Computing: Model Training with Machine Learning

[0404] Output: A trained predictive model

[0405] Step 3: Demand forecast

[0406] When a new disaster occurs, the server inputs the collected real-time data into a trained predictive model to predict the amount of supplies and funds needed at each evacuation center.

[0407] Input: Real-time data, trained predictive model

[0408] Data calculation: Running demand forecasting algorithms

[0409] Output: The amount of supplies and funds needed for each shelter

[0410] Step 4: Provide information

[0411] The terminal notifies the user of information about the disaster area provided by the server, and the user can check the accommodation situation at evacuation centers and the shortage of supplies.

[0412] Input: Disaster area information from the server

[0413] Data processing: Convert to notification format

[0414] Output: Information to the user

[0415] Step 5: Accept donations

[0416] Users make donations to specific shelters and supplies through their devices, which then send the donation information to the server.

[0417] Input: User donation information

[0418] Data processing: recording and transmitting donation information

[0419] Output: Send donation information to the server

[0420] Step 6: Collect emotional data

[0421] The device uses the smartphone's camera and microphone to collect the user's emotional data and transmits it to a server.

[0422] Input: User voice and image data

[0423] Data Computing: Identifying emotions through emotion analysis models

[0424] Output: Sending the identified emotion data to the server

[0425] Step 7: Generate support proposals

[0426] The server generates appropriate donation options and support plans for the user based on the results of the sentiment analysis.

[0427] Input: User emotion data

[0428] Data Computing: Implementing an Emotion-Based Donation Suggestion Algorithm

[0429] Output: Generates donation options and support plans for the user

[0430] Step 8: Donation Management and Tracking

[0431] The server manages and tracks in real time how much donated supplies and funds have been delivered to which shelters, using a MySQL database to record donations and get status updates from the distribution center.

[0432] Input: Delivery status from the distribution center, donation information

[0433] Data processing: Merging donation information and delivery status

[0434] Output: Real-time donation management status

[0435] Through these processing steps, the system for predicting and allocating resource demand during a disaster can propose assistance based on the user's emotions, enabling efficient and effective assistance activities.

[0436] 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.

[0437] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0438] 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.

[0439] [Second embodiment]

[0440] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0441] 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.

[0442] 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).

[0443] 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.

[0444] 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.

[0445] 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).

[0446] 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.

[0447] 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.

[0448] 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.

[0449] 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.

[0450] In the smart glasses 214, 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.

[0451] 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."

[0452] The present invention is a system for optimizing resource demand prediction and allocation in the event of a disaster, and specific embodiments thereof will be described below.

[0453] First, the system is centered around a server, which predicts the demand for supplies and funds in each region based on past disaster data and real-time information on current disaster-stricken areas.

[0454] Program operation description:

[0455] 1. Data Collection:

[0456] The server collects data on past disasters from government and NGO databases, analyzing, for example, how much supplies were needed at each evacuation shelter during a typhoon the previous year.

[0457] The server also uses sensors and drones to collect real-time data from the affected areas, including the current capacity of evacuation centers, a list of needed supplies, and the extent of the damage.

[0458] 2. Data preprocessing and model training:

[0459] The server cleanses the collected data, corrects outliers, and fills in missing values.

[0460] The server will then use the cleansed data to train machine learning models that can predict the specific supplies and funding needs of each shelter.

[0461] 3. Demand forecasting:

[0462] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies will be needed at each evacuation center.

[0463] 4. Resource Allocation:

[0464] The server then formulates a specific resource allocation plan based on the predicted demand. For example, it determines that shelter A needs 1,000 bottles of water and 500 packs of food.

[0465] The server will follow this plan and work with logistics companies to deliver the supplies.

[0466] 5. User Notification and Donation Acceptance:

[0467] The user can check the information provided by the system through the terminal. For example, they can learn that shelter A is experiencing a water shortage.

[0468] Users use their devices to donate specific items or funds, and once the donation is complete, the device sends this information to the server.

[0469] 6. Donation Management and Delivery Tracking:

[0470] The server records the donation information in a database and updates the donation status in real time.

[0471] The server coordinates with logistics companies to ensure the proper delivery of donated supplies and funds, and tracks the progress of deliveries. For example, it checks whether 50 bottles of water have arrived safely at Evacuation Center A.

[0472] Users can check the results of their donations through their devices. For example, they can confirm that the supplies they donated have arrived at shelter A.

[0473] Examples:

[0474] As a specific example, consider the case where a typhoon hits a part of Japan. Shelter A can accommodate 1,000 people, and shelter B has 800 people. From the collected data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles. Based on this prediction, the server submits a specific resource allocation plan to the logistics company and begins delivery. At the same time, the user checks the water shortage information for shelter A on their device and donates 50 bottles of water. The device sends this information to the server, which updates the donation status in real time and manages the donated water until it arrives safely at shelter A.

[0475] By using the above-described method and system, the present invention can quickly and effectively allocate necessary resources in the event of a disaster, thereby preventing uneven distribution of support.

[0476] The processing flow will be explained below.

[0477] Step 1: Data collection

[0478] The server collects past disaster data from a database of past disasters, including the amount of supplies and funds needed at each evacuation shelter, the number of evacuees, and the extent of damage caused by past disasters.

[0479] At the same time, the server collects real-time information on the current situation in the disaster area, using sensors and drones to obtain data such as the number of people in evacuation shelters, the current stock of supplies, and the extent of the damage.

[0480] Step 2: Data Preprocessing

[0481] The server cleanses the collected data on past disasters and current situations, correcting outliers and filling in missing values ​​to improve data quality.

[0482] The server converts the pre-processed data into a suitable format and prepares it for the machine learning algorithms.

[0483] Step 3: Predictive model training

[0484] The server uses machine learning algorithms to train a demand forecasting model based on the preprocessed dataset, creating a model capable of predicting each shelter's future supply and financial needs.

[0485] The server stores the trained model and prepares it for real-time predictions.

[0486] Step 4: Demand forecast

[0487] When a new disaster occurs, the server inputs current disaster area data collected in real time into the predictive model.

[0488] Based on this data, the server predicts the amount of supplies and funds that will be needed at each evacuation shelter and in the affected area.

[0489] Step 5: Resource Allocation Planning

[0490] The server creates a resource allocation plan based on predicted demand, specifically determining how much supplies should be sent from which distribution center to which shelter.

[0491] The server notifies the distribution plan to the logistics company and instructs them to prepare for delivery.

[0492] Step 6: Notify users and accept donations

[0493] The user checks the information provided by the system through their own terminal. For example, they learn that shelter A is experiencing a water shortage.

[0494] Users use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[0495] Step 7: Donation Management and Delivery Tracking

[0496] The server records donation information in a database, updates donation status in real time, and, if necessary, reallocates donated supplies and funds to areas in need.

[0497] The server tracks the delivery status of donated goods, receives delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination.

[0498] Users can check how their donations have been used through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[0499] The above are the specific processing steps of the system of the present invention. This system is expected to enable rapid and effective forecasting of demand for resources and appropriate allocation in the event of a disaster.

[0500] Example 1

[0501] 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."

[0502] In times of disaster, rapid and accurate resource allocation is required, but conventional systems have had difficulty achieving this. In particular, it was difficult to use past disaster data and create immediate demand forecasts and distribution plans based on the latest information on affected areas. Another issue was the lack of functionality to reflect user donations in real time and manage and track their delivery status.

[0503] 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.

[0504] In this invention, the server includes means for collecting past disaster data from a database, means for collecting information on disaster-stricken areas in real time using sensors and remote control devices, means for cleansing the collected data, correcting outliers and filling in missing values, means for training a machine learning model based on the cleansed data and predicting demand for supplies and funds, means for formulating a specific resource allocation plan based on the predicted demand and delivering supplies in cooperation with delivery agencies, means for providing shelter information and accepting donations via a user terminal, means for recording donation information in a database and tracking the delivery of donated resources, and means for confirming donation results via a user terminal. This makes it possible to quickly and accurately allocate necessary resources in the event of a disaster and prevent uneven distribution of support.

[0505] A "database" is an electronic system designed to efficiently store, manage, and retrieve information.

[0506] "Disaster data" refers to data containing information about natural and man-made disasters, including the date and time of the disaster, location, type and extent of damage, and necessary supplies.

[0507] A "sensor" is a device that detects a physical phenomenon and converts it into an electrical signal.

[0508] A "remote control device" is a device used to operate and control equipment or systems from a remote location.

[0509] "Real-time" refers to a state in which data and information are acquired and processed immediately.

[0510] "Data cleansing" is the process of correcting errors and missing values ​​and maintaining consistency in order to improve the quality of data.

[0511] A "machine learning model" is a mathematical model that learns patterns and rules based on large amounts of data and makes predictions and classifications for new data.

[0512] "Prediction" is the prediction of future events or conditions based on past and present data.

[0513] A "resource allocation plan" is a detailed plan for optimally distributing available resources and providing them where needed.

[0514] A "delivery organization" is an organization or infrastructure that transports goods or services to designated locations.

[0515] A "user terminal" is a device used to access the system and exchange information.

[0516] "Shelter information" refers to data about shelters, including the number of people they can accommodate, the supplies they need, and current issues.

[0517] A "donation" is when an individual or organization provides goods or funds for disaster relief.

[0518] "Recording in a database" means storing information in a database in a manner that allows it to be accessed later.

[0519] "Tracking" is the process of monitoring and confirming how goods or information are moving and where they are currently located.

[0520] "Donation results" is information showing how donated goods and funds were used and where they ended up.

[0521] MODE FOR CARRYING OUT THE INVENTION

[0522] The present invention relates to a system for optimizing resource demand forecasting and allocation in the event of a disaster, and specific embodiments thereof are described below. The core of this system is a server that makes full use of advanced data collection and machine learning.

[0523] Hardware and software used

[0524] 1. Server:

[0525] Hardware: General-purpose high-performance computer (e.g., AWS EC2 instance)

[0526] Software: Apache Kafka, Python, TensorFlow, Flask

[0527] 2. Terminal:

[0528] The device from which the user accesses the site (e.g., smartphone, tablet, PC)

[0529] Software: React

[0530] 3. Data Collection Equipment:

[0531] Sensors, drones, etc. (collecting on-site conditions)

[0532] Program processing overview

[0533] 1. Data Collection:

[0534] The server collects historical disaster data from government and NGO databases, specifically using Apache Kafka to retrieve time-stamped disaster data (location, type of damage, type and quantity of needed supplies).

[0535] The server uses sensors and drones to collect real-time data from the affected area, including the capacity of evacuation shelters, lists of needed supplies, and the extent of damage.

[0536] 2. Data preprocessing and model training:

[0537] The server cleanses the collected data using Python scripts, correcting outliers and completing missing values.

[0538] Next, the machine learning library TensorFlow is used to train a demand forecasting model based on disaster data, resulting in a model capable of predicting how much supplies will be needed at each evacuation center.

[0539] 3. Demand forecasting and resource allocation:

[0540] When a new disaster occurs, the server inputs real-time data into a predictive model to forecast the specific supply and funding needs of each evacuation shelter.

[0541] The server then formulates a resource allocation plan based on the predicted data and coordinates with delivery agencies to deliver the supplies to the required locations. Delivery arrangements are made through specific logs and APIs.

[0542] 4. User Notification and Donation Acceptance:

[0543] Users can check information about evacuation shelters through their devices. For example, information such as "evacuation shelter A is short of water" or "evacuation shelter B is short of food" will be displayed.

[0544] Users can donate specific items or funds using their own devices, and the information is quickly transmitted to the server.

[0545] 5. Donation Management and Delivery Tracking:

[0546] The server records the donation information in a database and updates the status in real time.

[0547] The server tracks the progress of donated goods to ensure they are delivered appropriately, and users can view on their devices how their donations are being used.

[0548] Specific examples

[0549] For example, consider a situation where a typhoon hits a part of Japan, with 1,000 people housed in shelter A and 800 people evacuated to shelter B. Based on real-time data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles of water. Based on this prediction, the server will contact a logistics company with a specific delivery schedule and arrange for the necessary supplies to be delivered to each shelter. At the same time, a user checks the water shortage information for shelter A on their device and donates 50 bottles of water. This information is updated in real time on the server, where tracking and management are performed, and the user's device is notified of the donation results.

[0550] Example prompts for generative AI models

[0551] "If a typhoon hits a certain area of ​​Japan, 1,000 people will be evacuated to shelter A and 800 people will be evacuated to shelter B. Please create an appropriate supply distribution plan for this situation and tell us which supplies and how much should be delivered to shelters A and B."

[0552] This system will enable the rapid and effective allocation of necessary resources in the event of a disaster, preventing uneven distribution of aid.

[0553] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0554] Program steps and their specific explanations:

[0555] Step 1:

[0556] Data collection

[0557] The server collects past disaster data from government and NGO databases. Specifically, it uses Apache Kafka to obtain data such as the "date and time of the disaster," "location of damage," and "type and quantity of needed supplies." The input data here is past event information related to disasters, and the output is raw data for cleansing. For example, it collects data such as "how much supplies were needed at which evacuation shelters during the previous year's typhoon."

[0558] The server also collects information on the affected areas in real time, using sensors and drones to obtain data such as the capacity of evacuation shelters, a list of necessary supplies, and the extent of damage. This input data is obtained directly from the field, and the output is real-time evacuation shelter information.

[0559] Step 2:

[0560] Data preprocessing and model training

[0561] The server cleanses the collected data using a Python script. It corrects outliers and fills in missing values ​​to maintain data integrity. It applies data cleansing processes to the raw data as input and outputs clean data. For example, if the demand for a material is incorrectly negative, it is corrected to 0 or an appropriate positive value.

[0562] The server uses the cleansed data to train a machine learning model in TensorFlow. The clean data is fed as input to the machine learning algorithm, which then outputs a demand forecasting model. For example, a model can be built to forecast future demand using past supply demand data for each evacuation shelter.

[0563] Step 3:

[0564] Demand forecasting

[0565] When a new disaster occurs, the server uses a machine learning model to predict demand based on data collected in real time. The input is real-time evacuation shelter data and a trained demand prediction model, and the output is the specific supply and funding needs of each evacuation shelter. For example, it inputs the current number of evacuees and weather information and predicts demand for water and temporary toilets when rising temperatures and strong winds are predicted.

[0566] Step 4:

[0567] Resource Allocation

[0568] The server formulates a specific resource allocation plan based on the predicted demand. The input data is the output of the demand forecasting model and the resources of logistics agencies, and the output is specific delivery instructions. For example, a specific plan may be created that "Deliver 1,000 bottles of 2-liter water to shelter A and 800 bottles of food to shelter B."

[0569] The server sends delivery instructions to logistics organizations via API or email according to the plan. For example, it might instruct a logistics company to "arrange three trucks and deliver supplies to the designated evacuation center."

[0570] Step 5:

[0571] User Notification and Donation Acceptance

[0572] The user checks the evacuation shelter information provided by the system through their terminal. The input data is the real-time evacuation shelter situation, and the output is the information displayed on the user's terminal. For example, the information displayed is "Evacuation shelter A is experiencing a water shortage."

[0573] Users use their devices to donate specific items or funds. The input for the donation is the item or amount selected by the user, and the output is the donation data sent to the server. For example, the operation is "User donates 50 bottles of water to shelter A."

[0574] Step 6:

[0575] Donation management and delivery tracking

[0576] The server records donation information in a database. The input is information about donated goods or funds, and the output is an updated database record. For example, 50 bottles of water are donated.

[0577] The server coordinates with logistics agencies to ensure the proper delivery of donated supplies and tracks the progress of deliveries. The inputs are delivery instructions and feedback from logistics agencies, and the output is delivery status updates. For example, it confirms that a truck has arrived at shelter A and that 50 bottles of water have been delivered safely.

[0578] The user checks the results of their donation on their device. The input is feedback data from the system, and the output is confirmation of the donation results displayed on the user's device. For example, a notification may be displayed stating, "The 50 bottles of donated water have arrived safely at Shelter A."

[0579] These steps will ensure that necessary resources are allocated quickly and effectively in the event of a disaster and prevent uneven distribution of aid.

[0580] (Application example 1)

[0581] 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."

[0582] During disasters, there is a sudden surge in demand for supplies and funds at evacuation centers and other facilities, making it important to distribute them quickly and appropriately. Current systems make it difficult to accurately grasp the real-time situation in disaster-stricken areas, often resulting in uneven resource distribution and delays. Furthermore, effective management and tracking of donated resources is lacking, calling for greater transparency and simplified collaboration with donors.

[0583] 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.

[0584] In this invention, the server includes a means for predicting resource demand during a disaster, a means for allocating resources based on the predicted demand, a means for collecting information on the status of disaster-stricken areas in real time, a means for providing information to users and accepting donations, a means for managing and tracking the delivery of donated resources, and a means for real-time management and optimization of distribution of supplies at a logistics center. This enables appropriate and prompt allocation of resources during a disaster, and effective support for disaster-stricken areas. It also improves information provision and transparency for donors, leading to more efficient donation management.

[0585] A "means for predicting resource demand during a disaster" is a device or program that predicts the supplies and funds required in each region and evacuation shelter based on data from past disasters and data collected in real time.

[0586] A "resource allocation tool" is a device or program that optimally allocates necessary supplies and funds to each evacuation center or region based on a predictive model.

[0587] "Means for collecting information on the status of disaster-stricken areas in real time" refers to hardware and software such as sensors, drones, and communication devices used to collect information from disaster-stricken areas in real time.

[0588] The "means for providing information to users and accepting donations" refers to a device or program that notifies users of the current situation at the evacuation shelter and the supplies they need, and accepts donations.

[0589] "Means for managing and tracking the delivery of donated resources" refers to a device or program that manages and tracks the delivery status of materials and funds donated by users and monitors their progress until they finally reach the recipient.

[0590] "Means for real-time management and distribution optimization of materials in logistics centers" refers to a device or program that monitors materials stored and managed in logistics centers in real time and distributes them optimally.

[0591] A "smartphone" is a mobile terminal that combines the functions of a mobile phone and a computer, and is a device on which applications can be installed and run.

[0592] "Smart glasses" are glasses-type wearable devices that have the function of displaying visual information and providing various types of information to the wearer.

[0593] A "logistics robot" is a robot that automatically transports and sorts materials at logistics centers and other locations.

[0594] A "predictive model" is a statistical or machine learning model built to forecast future demand based on historical data.

[0595] "Efficient donation management" means streamlining the management process from receipt of donated goods and funds to final delivery.

[0596] The present invention is a system for optimizing resource demand prediction and allocation in the event of a disaster, and specific embodiments thereof will be described below.

[0597] First, the server predicts the demand for supplies and funds in each region based on past disaster data and real-time information on current disaster areas. Past disaster data is collected from databases provided by government agencies and relief organizations. For example, it analyzes how much supplies were needed at each evacuation center during the previous year's typhoon.

[0598] The server then uses sensors and drones to obtain real-time data from the affected area, including the current capacity of evacuation centers, a list of needed supplies, and the extent of the damage.

[0599] The server cleanses the collected data, correcting outliers and filling in missing values. The server then trains a machine learning model on the cleansed data to build the ability to predict the specific supply and funding needs of each shelter. The server uses Python and the Scikit-learn library to train the model and make predictions.

[0600] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies each shelter needs. The server then creates a specific resource allocation plan based on the predicted demand. For example, it determines that shelter A needs 1,000 bottles of water and 500 packs of food.

[0601] The server then coordinates with logistics companies to deliver supplies according to this plan. Furthermore, users can check information provided by the system through their devices. For example, they may learn that shelter A is experiencing a water shortage. They can then use their devices to donate specific supplies or funds. Once the donation is complete, the device sends this information to the server.

[0602] The server records donation information in a database and updates the donation status in real time. The server coordinates with logistics companies to ensure the proper delivery of donated supplies and funds, and tracks the progress of deliveries. For example, the server can confirm whether 50 bottles of water have safely arrived at Evacuation Center A. Users can check the results of their donations on their devices.

[0603] Smartphones, smart glasses, and logistics robots are also used in logistics centers to manage and optimize the distribution of materials in real time. Materials stored and managed within the logistics center can be monitored in real time and optimally distributed. This will improve the efficiency of material distribution within the logistics center.

[0604] As a specific example, consider the case where a typhoon hits a part of Japan. Shelter A has a capacity of 1,000 people, and shelter B has 800 people. From the collected data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles. Based on this prediction, the server submits a specific resource allocation plan to the logistics company and begins delivery. At the same time, the user checks the water shortage information for shelter A on their device and donates 50 bottles of water. The device sends this information to the server, which updates the donation status in real time and manages the donated water until it reaches shelter A.

[0605] Examples of prompts to input to a generative AI model include:

[0606] "Please tell me the procedure for logistics centers to use smartphones to check demand forecasts in disaster-stricken areas in real time and allocate supplies optimally."

[0607] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0608] Step 1: Data collection

[0609] The server collects past disaster data from government and relief organization databases, and also obtains real-time data from disaster-stricken areas using sensors and drones. The server collects information such as the number of people in the affected area, a list of necessary supplies, and the extent of damage.

[0610] Input: Past disaster data, real-time data

[0611] Output: Collected data

[0612] Specific operation: The server uses APIs to retrieve data from the database, collect real-time data from each sensor and drone, and store the data.

[0613] Step 2: Data Preprocessing

[0614] The server cleanses the collected data, corrects outliers, and fills in missing values, thereby ensuring data quality.

[0615] Input: Collected data

[0616] Output: Cleansed data

[0617] Specific operation: The server uses the Pandas library to cleanse the data, correcting outliers and imputing missing values ​​automatically.

[0618] Step 3: Model training

[0619] The server will then use the cleansed data to train machine learning models that can predict the specific supply and funding needs of each shelter.

[0620] Input: Cleansed data

[0621] Output: A trained predictive model

[0622] How it works: The server uses the Scikit-learn library to train a machine learning model and generate a predictive model based on past data.

[0623] Step 4: Demand forecast

[0624] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies will be needed at each evacuation center.

[0625] Input: Trained predictive model, real-time data

[0626] Output: Forecasted demand

[0627] Specific operation: The server inputs real-time data into a predictive model to estimate the demand for supplies and funds for each evacuation shelter.

[0628] Step 5: Resource allocation

[0629] The server formulates a specific resource allocation plan based on the predicted demand and works with logistics companies to deliver supplies.

[0630] Input: Forecasted demand

[0631] Output: Resource allocation plan

[0632] Specific operation: The server creates an optimal resource allocation plan based on the prediction results and notifies the logistics company of the plan via an API.

[0633] Step 6: Notify users and accept donations

[0634] Users check the information provided by the system on their terminals and donate the necessary supplies. The donation information is then sent to the server.

[0635] Input: Required information for each evacuation shelter

[0636] Output: Donation information

[0637] Specific operation: The application on the device notifies the user of supply shortages at evacuation centers, and when the user completes the donation procedure, the information is sent to the server.

[0638] Step 7: Donation Management and Delivery Tracking

[0639] The server records donation information in a database, updates donation status in real time, and manages and tracks donated goods and funds to ensure their proper delivery.

[0640] Input: Donation information

[0641] Output: Updated donation status, delivery status of donated goods

[0642] Specific operation: The server records donation information in a database, works with logistics companies to track the delivery status of supplies, and provides feedback to users.

[0643] Step 8: Optimized material management

[0644] The logistics center will use smartphones, smart glasses, and logistics robots to manage supplies in real time and optimize their distribution.

[0645] Input: Resource allocation plan, real-time management data

[0646] Output: Optimized resource allocation

[0647] Specific operation: Each device in the logistics center moves and sorts materials in real time based on a resource allocation plan, maintaining optimal conditions.

[0648] 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.

[0649] The present invention relates to a system for realizing more effective relief activities by combining a system for predicting demand for and optimizing allocation of resources during disasters with an emotion engine that recognizes user emotions. The system according to the present invention can be implemented as follows.

[0650] Program operation description:

[0651] 1. Data Collection:

[0652] The server collects past disaster data from a database, including the amount of supplies needed, the number of evacuees, and the extent of damage caused by past disasters.

[0653] The server also collects real-time information on the current state of the affected area, using sensors and drones to obtain data such as the capacity of evacuation centers, current stock of supplies, and the extent of the damage.

[0654] 2. Data preprocessing and model training:

[0655] The server preprocesses the collected data, correcting outliers and filling in missing values, and then feeds the preprocessed data into machine learning algorithms.

[0656] The server uses the preprocessed data to train machine learning models that predict each shelter's future supply and financial needs.

[0657] 3. Demand forecasting:

[0658] When a new disaster occurs, the server inputs disaster area data collected in real time into the predictive model.

[0659] Based on this data, the server predicts the amount of supplies and funds that will be needed at each evacuation shelter and in the affected area.

[0660] 4. Resource Allocation Planning:

[0661] The server creates a resource allocation plan based on the predicted demand, specifically determining how much supplies should be sent from which distribution center to which evacuation shelter.

[0662] The server notifies the distribution plan to the logistics company and instructs them to prepare for delivery.

[0663] 5. User Notification and Donation Acceptance:

[0664] The server notifies the user's device of information about the affected area, and the user can use their device to obtain information about supplies needed to get to an evacuation shelter on foot.

[0665] Users can use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[0666] 6. Emotion recognition:

[0667] The server is equipped with an emotion engine that recognizes the user's emotions, and collects and analyzes the user's emotion data.

[0668] The terminal identifies the user's emotions in real time and presents the user with assistance options according to the emotions.

[0669] 7. Emotion-based donation suggestions:

[0670] The server proposes appropriate donation options to the user based on the user's emotional data provided by the emotion engine. For example, if the server determines that the user is willing to donate a large amount, it will present the user with larger donation options.

[0671] 8. Donation Management and Delivery Tracking:

[0672] The server records donation information in a database, updates donation status in real time, and, if necessary, reallocates donated supplies and funds to areas in need.

[0673] The server tracks the delivery status of donated goods, receives delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination.

[0674] Users can check the results of their donations through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[0675] Examples:

[0676] For example, suppose a large earthquake occurs and devastates parts of Japan. Shelter A can accommodate 1,000 people, while shelter B has 800. Based on past disaster data and real-time collected information, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800. Based on this prediction, the server notifies the logistics company of a specific resource allocation plan and instructs delivery. At the same time, the user can check the water shortage information for shelter A through their device and be presented with appropriate donation options based on their emotional data. Once the user completes their donation, this information is sent from the device to the server, which updates the donation status in real time. The server tracks the delivery status of the donated supplies and confirms that the water has safely arrived at shelter A. The user can then check how their donation has been used through their device.

[0677] By using the above method and system, the present invention is expected to make disaster resource demand forecasts and appropriate allocations quick and effective, thereby improving the efficiency of relief activities. By combining it with an emotion engine, it becomes possible to propose relief based on the user's emotions, realizing the provision of more appropriate relief.

[0678] The processing flow will be explained below.

[0679] Step 1: Data collection

[0680] The server collects disaster data from a database of past disasters, including the amount of supplies and funds needed at each evacuation shelter, the number of evacuees, and the extent of damage caused by past disasters.

[0681] The server also uses sensors and drones to collect real-time data on the current disaster area, providing information such as shelter capacity, current supply inventory, and the extent of the damage.

[0682] Step 2: Data Preprocessing

[0683] The server cleanses the collected data, corrects outliers, and fills in missing values, allowing machine learning algorithms to learn accurately.

[0684] Step 3: Predictive model training

[0685] The server uses machine learning algorithms to train a demand forecasting model based on the pre-processed data, creating a model capable of predicting each shelter's future supply and financial needs.

[0686] Step 4: Demand forecast

[0687] When a new disaster occurs, the server inputs current data from the affected area, collected in real time, into the predictive model.

[0688] Based on this data, the server predicts the specific amount of supplies and funds needed at each evacuation center and disaster area.

[0689] Step 5: Resource Allocation Planning

[0690] The server creates a resource allocation plan based on the predicted demand, specifically determining how much supplies should be sent from which distribution center to which evacuation shelter.

[0691] The server issues delivery instructions to the logistics company based on this plan.

[0692] Step 6: Notify users and accept donations

[0693] The server notifies the user's device of information about the affected area, allowing the user to check on their device what supplies are in short supply at evacuation shelters.

[0694] Users use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[0695] Step 7: Emotion Recognition

[0696] The server uses an emotion engine that recognizes the user's emotions to collect and analyze the user's emotion data.

[0697] The device analyzes the user's emotions in real time and sends the results to the server.

[0698] Step 8: Emotionally Based Donation Proposals

[0699] The server presents appropriate donation options to the user based on the user's emotional data obtained from the emotion engine, for example, presenting an urgent donation option if the user is emotionally moved.

[0700] Step 9: Donation Management and Delivery Tracking

[0701] The server records user donation information in a database, updates the donation status in real time, and verifies that donated goods and funds are distributed appropriately, as needed.

[0702] The server tracks the delivery of donated goods, obtaining delivery status from logistics providers and managing the process until the goods arrive safely at their destination.

[0703] Users can check how their donations have been used through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[0704] The above are the specific processing steps of the system of the present invention. This system is expected to make relief activities more efficient by quickly and effectively predicting demand for resources during disasters and allocating them appropriately. By combining it with an emotion engine, it becomes possible to propose relief based on the user's emotions, realizing the provision of more appropriate relief.

[0705] Example 2

[0706] 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."

[0707] During a disaster, the condition of the affected area and the demand for supplies change from moment to moment, so it is necessary to allocate resources quickly and accurately.In addition, it is necessary to realize effective relief activities by making appropriate relief proposals based on the user's emotions and motivation, but current systems are not able to adequately address these issues.

[0708] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting past disaster data and correcting outliers and imputing missing values ​​based on the data; means for training a machine learning model based on the preprocessed data and predicting future supply and financial needs at each evacuation shelter; means for collecting disaster area data in real time when a new disaster occurs and inputting it into the machine learning model; means for formulating a resource allocation plan based on the predicted demand and notifying logistics companies; means for notifying a user's terminal of disaster area information and accepting donations; means for proposing appropriate support options using an emotion engine that recognizes the user's emotions; means for recording donation information in a database, updating the donation status in real time, and tracking the delivery status of donated supplies; and means for the user to check the results of the donation via the terminal. This allows for rapid and accurate resource demand prediction and appropriate allocation during a disaster, enabling effective support activities based on the user's emotions.

[0709] "Past disaster data" refers to information about disasters that have occurred in the past, including data on the demand for supplies, the number of evacuees, and the extent of damage.

[0710] "Outlier correction" refers to the process of detecting and correcting statistically abnormal values ​​in acquired data.

[0711] "Missing value imputation" refers to the operation of filling in missing values ​​in a dataset, and generally uses statistical methods such as the mean or median.

[0712] "Preprocessed data" refers to data that has been converted into a form suitable for machine learning models by correcting outliers and filling in missing values.

[0713] A "machine learning model" refers to an algorithm or statistical model that can make predictions or classifications based on past data.

[0714] "Collecting disaster area data in real time" refers to the operation of instantly collecting data on ongoing disasters and sending it to a server.

[0715] A "resource allocation plan" refers to a plan that determines how much supplies to send from which logistics center to which shelter based on predicted demand.

[0716] "Logistics company" refers to a company or organization responsible for transporting goods.

[0717] "User terminal" refers to a computer device such as a smartphone or tablet held by a user, and is a means for communicating with a server.

[0718] "Emotion engine" refers to a system for recognizing and analyzing a user's emotional state.

[0719] "Support options" refers to the support methods and types of donations suggested to users.

[0720] "Donation Information" refers to the details of a user's donation (donation recipient, donated items, amount, date and time, etc.).

[0721] A "database" refers to a system for systematically storing and managing data.

[0722] "Donation results" refers to information about how the donation made by the user was used and what results it produced.

[0723] The present invention relates to a system for optimizing resource demand forecasting and allocation in the event of a disaster. This system is mainly composed of a server, terminals, and users.

[0724] First, the server collects past disaster data from a database. The database stores data on past disasters, such as the demand for supplies, the number of evacuees, and the extent of damage. Next, the server uses sensors and drones to collect real-time information on the current situation in the disaster-stricken area. Specifically, it obtains information such as the capacity of evacuation shelters, the current stock of supplies, and the extent of damage.

[0725] The collected data is preprocessed by the server. During the preprocessing process, outliers are corrected, missing values ​​are imputed, and the data is normalized to make it suitable for the machine learning model. The server then trains the machine learning model based on the preprocessed data. This creates a model that predicts the future supply and funding needs of each shelter.

[0726] When a new disaster occurs, the server inputs real-time data on the affected area into the trained prediction model. Based on this data, it predicts the amount of supplies and funds needed at each evacuation shelter and in the affected area. Based on the predicted demand, the server formulates a resource allocation plan, specifically determining how much supplies to send from which logistics center to which evacuation shelter. The server then notifies the distribution plan to logistics companies and issues delivery instructions.

[0727] Users can receive information about disaster areas through their devices and make donations to specific shelters or supplies. Once a donation is completed, this information is sent from the device to a server, which updates the donation status in real time.

[0728] Furthermore, the server is equipped with an emotion engine that recognizes the user's emotions. The device identifies the user's emotions in real time from their facial expressions and voice data, and presents the user with support options that correspond to those emotions. The emotion engine analyzes the user's emotional data and suggests appropriate donation options. For example, if it is inferred that the user is willing to make a large donation, the user will be presented with larger support options.

[0729] Donation information is recorded in a database, and the server updates the donation status in real time. If necessary, donated goods or funds are redistributed to areas where they are in short supply. The server tracks the delivery status of donated goods, obtains delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination. Users can check the results of their donations on their own devices.

[0730] Specific examples

[0731] For example, let's assume that a large-scale earthquake occurs and damages parts of Japan. Shelter A has a capacity of 1,000 people, and shelter B has 800 people. Based on past disaster data and information collected in real time, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800. Based on this prediction, the server notifies the logistics company of a resource allocation plan and instructs delivery. At the same time, the user checks water shortage information for shelter A through their device and is presented with the optimal donation option based on their emotional data. Once the user completes their donation, this information is sent from the device to the server, which updates the donation status in real time. The server tracks the delivery status of the donated supplies and confirms whether the water has arrived safely at shelter A. The user can then check how their donation has been used through their device.

[0732] Prompt Sentence Examples

[0733] "In an area affected by a large earthquake, 1,000 people are evacuated to shelter A and 800 people are evacuated to shelter B. Please predict the supplies needed at each shelter based on past data and real-time information, and issue instructions to optimize relief activities."

[0734] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0735] Step 1: Data collection

[0736] The server collects past disaster data from a database. Specifically, it uses SQL queries to obtain data such as the demand for supplies, the number of evacuees, and the extent of damage during past disasters. It also uses sensors and drones to collect real-time information on the current situation in disaster-stricken areas. Data from the sensors, such as the number of people accommodated, stock of supplies, and the extent of damage, is sent to the server.

[0737] Input: Past disaster database, sensors, drones

[0738] Output: Collected disaster data

[0739] Step 2: Data Preprocessing

[0740] The server preprocesses the collected data, correcting outliers and imputing missing values, and normalizing the data. Missing values ​​are imputed with the mean or median, and outliers are corrected using statistical methods. The data is then scaled and converted into a format suitable for machine learning models.

[0741] Input: Collected disaster data

[0742] Output: Preprocessed data

[0743] Step 3: Model training

[0744] The server trains a machine learning model using the preprocessed data. It splits the dataset into a training set and a test set, trains a regression model or deep learning model using the training set, evaluates the model's performance on the test set, and tunes hyperparameters as needed.

[0745] Input: Preprocessed data

[0746] Output: A trained machine learning model

[0747] Step 4: Demand forecast

[0748] When a new disaster occurs, the server inputs real-time data on the affected area into the trained prediction model, which then uses the model to predict the amount of supplies and funds needed at each evacuation center and in the affected area.

[0749] Input: Real-time disaster area data, trained machine learning model

[0750] Output: Forecasted demand for supplies and funds for each shelter

[0751] Step 5: Resource Allocation Planning

[0752] The server creates a resource allocation plan based on predicted demand. It uses an optimization algorithm to determine how much supplies should be sent from which distribution center to which shelter. It then notifies the distribution company of this plan via an API, instructing them to prepare for delivery.

[0753] Input: Forecasted demand data

[0754] Output: Resource allocation plan, notification to logistics companies

[0755] Step 6: Notify users and accept donations

[0756] The server notifies users of information about the affected areas via their devices. Users can then donate to specific shelters or supplies via their devices. Once a donation is complete, the information is sent from the device to the server, and the donation status is updated in real time.

[0757] Input: Disaster area information, user donation data

[0758] Output: User notification, donation acceptance information

[0759] Step 7: Emotion Recognition

[0760] The server uses an emotion engine to recognize the user's emotions. The device collects the user's facial expressions and voice data and analyzes them using the emotion engine. As a result, assistance options based on the user's emotions are presented.

[0761] Input: User's facial expression and voice data

[0762] Output: Analyzed emotion data, support options

[0763] Step 8: Emotionally Based Donation Proposals

[0764] The server proposes appropriate donation options to the user based on the analysis results of the emotion engine. If the server determines that the user is willing to donate a large amount, it presents the option of large-scale support.

[0765] Input: Parsed emotion data

[0766] Output: Donation proposal

[0767] Step 9: Donation Management and Delivery Tracking

[0768] The server records users' donation information in a database and updates the donation status in real time. It tracks the delivery status of donated goods and funds and obtains delivery status from logistics companies, managing the donated goods until they arrive safely at their destination. Users can check the results of their donations through their devices.

[0769] Input: Donation information, delivery status from logistics company

[0770] Output: Donation management data, delivery tracking information, user notifications

[0771] (Application example 2)

[0772] 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."

[0773] Conventional resource demand forecasting and allocation systems during disasters do not consider the psychological and emotional state of disaster victims when proposing assistance, making it difficult to provide effective assistance. Furthermore, assistance proposals and effective resource allocations are carried out in stages, making it difficult to respond quickly. The present invention aims to solve these problems by combining assistance proposals that consider users' emotions with real-time demand forecasting.

[0774] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0775] In this invention, the server includes means for predicting resource demand during a disaster, means for allocating resources based on the predicted demand, and means for collecting information on the status of the disaster area in real time, thereby enabling means for providing information to users and accepting donations, means for managing and tracking the delivery of donated resources, and means for identifying user emotions in real time and generating support proposals based on those emotions.

[0776] A "means for predicting resource demand during a disaster" is a device or program that predicts the amount of supplies and resources that will be needed when a disaster occurs.

[0777] The "means for allocating resources based on the predicted demand" is a device or program that efficiently allocates materials and resources to each location based on the predicted resource demand.

[0778] "Means for collecting information on the status of disaster-affected areas in real time" refers to devices or systems that use sensors, drones, APIs, etc. to instantly obtain the current status of areas affected by disasters.

[0779] The "means for providing information and accepting donations" refers to an interface or application that provides users with the necessary information and accepts donations from users.

[0780] "Means for managing and tracking the delivery of donated resources" refers to a system that manages the delivery of donated goods and funds to ensure they are delivered to the designated locations and tracks the delivery status in real time.

[0781] The "means for identifying a user's emotions in real time and generating assistance suggestions based on those emotions" refers to a device or system that analyzes a user's emotions in real time and suggests appropriate assistance options based on the results.

[0782] "Means of collecting data on past disasters and training a predictive model based on that data" refers to a system that collects data on past disasters and uses that data to train a demand prediction model for future disasters using a machine learning algorithm.

[0783] "Means for using a predictive model to predict the amount of supplies and funds required for each evacuation shelter" refers to a device or system that uses a trained predictive model to specifically calculate the amount of supplies and funds required for each evacuation shelter.

[0784] The present invention provides a system that combines emotion analysis with a resource demand forecasting and allocation system in the event of a disaster, and proposes assistance based on the user's emotions. Specific embodiments of the present invention are described below.

[0785] System Configuration

[0786] The system consists of the following major components:

[0787] 1. Server

[0788] 2. Terminal

[0789] 3. Users

[0790] The role and specific functions of each component

[0791] server

[0792] The server has several options:

[0793] Data collection method: The server collects past disaster data and real-time local data using sensors, drones, and APIs. For example, it uses Google Maps API to obtain information on disaster areas.

[0794] Prediction model training method: The server trains a prediction model using a machine learning algorithm (e.g., Scikit-learn's LinearRegression) based on the collected past disaster data.

[0795] Demand forecasting tools: Using trained predictive models, we forecast the amount of supplies and funding needed at each shelter.

[0796] Sentiment analysis method: Emotions are identified in real time using TensorFlow and OpenCV based on data (images and audio) sent by the user.

[0797] Support proposal generation means: Based on the results of sentiment analysis, appropriate donation options and support plans are proposed to the user.

[0798] Donation Management and Tracking: Manage donations and track delivery status in real time using a MySQL database.

[0799] Terminal

[0800] The terminal has the following functions:

[0801] Information provision function: Provides users with information on the status of the disaster area and the necessary support.

[0802] Donation acceptance feature: Users can make donations to specific shelters or supplies.

[0803] Emotional data transmission function: Collects user emotional data using the smartphone's camera and microphone and transmits it to the server.

[0804] User

[0805] The user does the following:

[0806] Information reception: Receive information about the situation in the disaster area and the need for assistance.

[0807] Donate: Make a donation through your device.

[0808] Emotion data provision: Send your own emotional data to the server via your device.

[0809] Example

[0810] For example, suppose a large earthquake occurs and some areas are affected. Based on past earthquake data and real-time data, the server predicts that shelter A will need 1,000 bottles of water and shelter B will need 800 bottles of water. A user checks the water shortage information for shelter A on their device, senses the urgency, and makes a donation. Once the donation is complete, the server records the donation information in a database and tracks the delivery status.

[0811] Prompt Sentence Examples

[0812] "Designing a disaster food delivery application that optimizes donations through emotion recognition. Features include real-time demand forecasting, user emotion analysis, and donation suggestions. For example, it includes a function to estimate the food needs of shelter A and provide donation options based on the user's emotion."

[0813] It is expected that this system will solve the problems faced by conventional disaster response systems and enable efficient and effective support activities.

[0814] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0815] Step 1: Data collection

[0816] The server collects past disaster data from a database and uses sensors, drones, and APIs (e.g., Google Maps API) to obtain real-time local data (such as shelter capacity, supply inventory, and damage extent).

[0817] Input: Past disaster data, local real-time data

[0818] Data processing: Correction of outliers, completion of missing values

[0819] Output: Preprocessed dataset

[0820] Step 2: Model training

[0821] The server trains a predictive model using a machine learning algorithm (Scikit-learn's LinearRegression) based on the preprocessed data.

[0822] Input: Preprocessed dataset

[0823] Data Computing: Model Training with Machine Learning

[0824] Output: A trained predictive model

[0825] Step 3: Demand forecast

[0826] When a new disaster occurs, the server inputs the collected real-time data into a trained predictive model to predict the amount of supplies and funds needed at each evacuation center.

[0827] Input: Real-time data, trained predictive model

[0828] Data calculation: Running demand forecasting algorithms

[0829] Output: The amount of supplies and funds needed for each shelter

[0830] Step 4: Provide information

[0831] The terminal notifies the user of information about the disaster area provided by the server, and the user can check the accommodation situation at evacuation centers and the shortage of supplies.

[0832] Input: Disaster area information from the server

[0833] Data processing: Convert to notification format

[0834] Output: Information to the user

[0835] Step 5: Accept donations

[0836] Users make donations to specific shelters and supplies through their devices, which then send the donation information to the server.

[0837] Input: User donation information

[0838] Data processing: recording and transmitting donation information

[0839] Output: Send donation information to the server

[0840] Step 6: Collect emotional data

[0841] The device uses the smartphone's camera and microphone to collect the user's emotional data and transmits it to a server.

[0842] Input: User voice and image data

[0843] Data Computing: Identifying emotions through emotion analysis models

[0844] Output: Sending the identified emotion data to the server

[0845] Step 7: Generate support proposals

[0846] The server generates appropriate donation options and support plans for the user based on the results of the sentiment analysis.

[0847] Input: User emotion data

[0848] Data Computing: Implementing an Emotion-Based Donation Suggestion Algorithm

[0849] Output: Generates donation options and support plans for the user

[0850] Step 8: Donation Management and Tracking

[0851] The server manages and tracks in real time how much donated supplies and funds have been delivered to which shelters, using a MySQL database to record donations and get status updates from the distribution center.

[0852] Input: Delivery status from the distribution center, donation information

[0853] Data processing: Merging donation information and delivery status

[0854] Output: Real-time donation management status

[0855] Through these processing steps, the system for predicting and allocating resource demand during a disaster can propose assistance based on the user's emotions, enabling efficient and effective assistance activities.

[0856] 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.

[0857] 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.

[0858] 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.

[0859] [Third embodiment]

[0860] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0861] 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.

[0862] 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).

[0863] 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.

[0864] 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.

[0865] 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).

[0866] 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.

[0867] 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.

[0868] 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.

[0869] 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.

[0870] 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.

[0871] 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."

[0872] The present invention is a system for optimizing resource demand prediction and allocation in the event of a disaster, and specific embodiments thereof will be described below.

[0873] First, the system is centered around a server, which predicts the demand for supplies and funds in each region based on past disaster data and real-time information on current disaster-stricken areas.

[0874] Program operation description:

[0875] 1. Data Collection:

[0876] The server collects data on past disasters from government and NGO databases, analyzing, for example, how much supplies were needed at each evacuation shelter during a typhoon the previous year.

[0877] The server also uses sensors and drones to collect real-time data from the affected areas, including the current capacity of evacuation centers, a list of needed supplies, and the extent of the damage.

[0878] 2. Data preprocessing and model training:

[0879] The server cleanses the collected data, corrects outliers, and fills in missing values.

[0880] The server will then use the cleansed data to train machine learning models that can predict the specific supplies and funding needs of each shelter.

[0881] 3. Demand forecasting:

[0882] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies will be needed at each evacuation center.

[0883] 4. Resource Allocation:

[0884] The server then formulates a specific resource allocation plan based on the predicted demand. For example, it determines that shelter A needs 1,000 bottles of water and 500 packs of food.

[0885] The server will follow this plan and work with logistics companies to deliver the supplies.

[0886] 5. User Notification and Donation Acceptance:

[0887] The user can check the information provided by the system through the terminal. For example, they can learn that shelter A is experiencing a water shortage.

[0888] Users use their devices to donate specific items or funds, and once the donation is complete, the device sends this information to the server.

[0889] 6. Donation Management and Delivery Tracking:

[0890] The server records the donation information in a database and updates the donation status in real time.

[0891] The server coordinates with logistics companies to ensure the proper delivery of donated supplies and funds, and tracks the progress of deliveries. For example, it checks whether 50 bottles of water have arrived safely at Evacuation Center A.

[0892] Users can check the results of their donations through their devices. For example, they can confirm that the supplies they donated have arrived at shelter A.

[0893] Examples:

[0894] As a specific example, consider the case where a typhoon hits a part of Japan. Shelter A can accommodate 1,000 people, and shelter B has 800 people. From the collected data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles. Based on this prediction, the server submits a specific resource allocation plan to the logistics company and begins delivery. At the same time, the user checks the water shortage information for shelter A on their device and donates 50 bottles of water. The device sends this information to the server, which updates the donation status in real time and manages the donated water until it arrives safely at shelter A.

[0895] By using the above-described method and system, the present invention can quickly and effectively allocate necessary resources in the event of a disaster, thereby preventing uneven distribution of support.

[0896] The processing flow will be explained below.

[0897] Step 1: Data collection

[0898] The server collects past disaster data from a database of past disasters, including the amount of supplies and funds needed at each evacuation shelter, the number of evacuees, and the extent of damage caused by past disasters.

[0899] At the same time, the server collects real-time information on the current situation in the disaster area, using sensors and drones to obtain data such as the number of people in evacuation shelters, the current stock of supplies, and the extent of the damage.

[0900] Step 2: Data Preprocessing

[0901] The server cleanses the collected data on past disasters and current situations, correcting outliers and filling in missing values ​​to improve data quality.

[0902] The server converts the pre-processed data into a suitable format and prepares it for the machine learning algorithms.

[0903] Step 3: Predictive model training

[0904] The server uses machine learning algorithms to train a demand forecasting model based on the preprocessed dataset, creating a model capable of predicting each shelter's future supply and financial needs.

[0905] The server stores the trained model and prepares it for real-time predictions.

[0906] Step 4: Demand forecast

[0907] When a new disaster occurs, the server inputs current disaster area data collected in real time into the predictive model.

[0908] Based on this data, the server predicts the amount of supplies and funds that will be needed at each evacuation shelter and in the affected area.

[0909] Step 5: Resource Allocation Planning

[0910] The server creates a resource allocation plan based on predicted demand, specifically determining how much supplies should be sent from which distribution center to which shelter.

[0911] The server notifies the distribution plan to the logistics company and instructs them to prepare for delivery.

[0912] Step 6: Notify users and accept donations

[0913] The user checks the information provided by the system through their own terminal. For example, they learn that shelter A is experiencing a water shortage.

[0914] Users use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[0915] Step 7: Donation Management and Delivery Tracking

[0916] The server records donation information in a database, updates donation status in real time, and, if necessary, reallocates donated supplies and funds to areas in need.

[0917] The server tracks the delivery status of donated goods, receives delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination.

[0918] Users can check how their donations have been used through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[0919] The above are the specific processing steps of the system of the present invention. This system is expected to enable rapid and effective forecasting of demand for resources and appropriate allocation in the event of a disaster.

[0920] Example 1

[0921] 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."

[0922] In times of disaster, rapid and accurate resource allocation is required, but conventional systems have had difficulty achieving this. In particular, it was difficult to use past disaster data and create immediate demand forecasts and distribution plans based on the latest information on affected areas. Another issue was the lack of functionality to reflect user donations in real time and manage and track their delivery status.

[0923] 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.

[0924] In this invention, the server includes means for collecting past disaster data from a database, means for collecting information on disaster-stricken areas in real time using sensors and remote control devices, means for cleansing the collected data, correcting outliers and filling in missing values, means for training a machine learning model based on the cleansed data and predicting demand for supplies and funds, means for formulating a specific resource allocation plan based on the predicted demand and delivering supplies in cooperation with delivery agencies, means for providing shelter information and accepting donations via a user terminal, means for recording donation information in a database and tracking the delivery of donated resources, and means for confirming donation results via a user terminal. This makes it possible to quickly and accurately allocate necessary resources in the event of a disaster and prevent uneven distribution of support.

[0925] A "database" is an electronic system designed to efficiently store, manage, and retrieve information.

[0926] "Disaster data" refers to data containing information about natural and man-made disasters, including the date and time of the disaster, location, type and extent of damage, and necessary supplies.

[0927] A "sensor" is a device that detects a physical phenomenon and converts it into an electrical signal.

[0928] A "remote control device" is a device used to operate and control equipment or systems from a remote location.

[0929] "Real-time" refers to a state in which data and information are acquired and processed immediately.

[0930] "Data cleansing" is the process of correcting errors and missing values ​​and maintaining consistency in order to improve the quality of data.

[0931] A "machine learning model" is a mathematical model that learns patterns and rules based on large amounts of data and makes predictions and classifications for new data.

[0932] "Prediction" is the prediction of future events or conditions based on past and present data.

[0933] A "resource allocation plan" is a detailed plan for optimally distributing available resources and providing them where needed.

[0934] A "delivery organization" is an organization or infrastructure that transports goods or services to designated locations.

[0935] A "user terminal" is a device used to access the system and exchange information.

[0936] "Shelter information" refers to data about shelters, including the number of people they can accommodate, the supplies they need, and current issues.

[0937] A "donation" is when an individual or organization provides goods or funds for disaster relief.

[0938] "Recording in a database" means storing information in a database in a manner that allows it to be accessed later.

[0939] "Tracking" is the process of monitoring and confirming how goods or information are moving and where they are currently located.

[0940] "Donation results" is information showing how donated goods and funds were used and where they ended up.

[0941] MODE FOR CARRYING OUT THE INVENTION

[0942] The present invention relates to a system for optimizing resource demand forecasting and allocation in the event of a disaster, and specific embodiments thereof are described below. The core of this system is a server that makes full use of advanced data collection and machine learning.

[0943] Hardware and software used

[0944] 1. Server:

[0945] Hardware: General-purpose high-performance computer (e.g., AWS EC2 instance)

[0946] Software: Apache Kafka, Python, TensorFlow, Flask

[0947] 2. Terminal:

[0948] The device from which the user accesses the site (e.g., smartphone, tablet, PC)

[0949] Software: React

[0950] 3. Data Collection Equipment:

[0951] Sensors, drones, etc. (collecting on-site conditions)

[0952] Program processing overview

[0953] 1. Data Collection:

[0954] The server collects historical disaster data from government and NGO databases, specifically using Apache Kafka to retrieve time-stamped disaster data (location, type of damage, type and quantity of needed supplies).

[0955] The server uses sensors and drones to collect real-time data from the affected area, including the capacity of evacuation shelters, lists of needed supplies, and the extent of damage.

[0956] 2. Data preprocessing and model training:

[0957] The server cleanses the collected data using Python scripts, correcting outliers and completing missing values.

[0958] Next, the machine learning library TensorFlow is used to train a demand forecasting model based on disaster data, resulting in a model capable of predicting how much supplies will be needed at each evacuation center.

[0959] 3. Demand forecasting and resource allocation:

[0960] When a new disaster occurs, the server inputs real-time data into a predictive model to forecast the specific supply and funding needs of each evacuation shelter.

[0961] The server then formulates a resource allocation plan based on the predicted data and coordinates with delivery agencies to deliver the supplies to the required locations. Delivery arrangements are made through specific logs and APIs.

[0962] 4. User Notification and Donation Acceptance:

[0963] Users can check information about evacuation shelters through their devices. For example, information such as "evacuation shelter A is short of water" or "evacuation shelter B is short of food" will be displayed.

[0964] Users can donate specific items or funds using their own devices, and the information is quickly transmitted to the server.

[0965] 5. Donation Management and Delivery Tracking:

[0966] The server records the donation information in a database and updates the status in real time.

[0967] The server tracks the progress of donated goods to ensure they are delivered appropriately, and users can view on their devices how their donations are being used.

[0968] Specific examples

[0969] For example, consider a situation where a typhoon hits a part of Japan, with 1,000 people housed in shelter A and 800 people evacuated to shelter B. Based on real-time data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles of water. Based on this prediction, the server will contact a logistics company with a specific delivery schedule and arrange for the necessary supplies to be delivered to each shelter. At the same time, a user checks the water shortage information for shelter A on their device and donates 50 bottles of water. This information is updated in real time on the server, where tracking and management are performed, and the user's device is notified of the donation results.

[0970] Example prompts for generative AI models

[0971] "If a typhoon hits a certain area of ​​Japan, 1,000 people will be evacuated to shelter A and 800 people will be evacuated to shelter B. Please create an appropriate supply distribution plan for this situation and tell us which supplies and how much should be delivered to shelters A and B."

[0972] This system will enable the rapid and effective allocation of necessary resources in the event of a disaster, preventing uneven distribution of aid.

[0973] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0974] Program steps and their specific explanations:

[0975] Step 1:

[0976] Data collection

[0977] The server collects past disaster data from government and NGO databases. Specifically, it uses Apache Kafka to obtain data such as the "date and time of the disaster," "location of damage," and "type and quantity of needed supplies." The input data here is past event information related to disasters, and the output is raw data for cleansing. For example, it collects data such as "how much supplies were needed at which evacuation shelters during the previous year's typhoon."

[0978] The server also collects information on the affected areas in real time, using sensors and drones to obtain data such as the capacity of evacuation shelters, a list of necessary supplies, and the extent of damage. This input data is obtained directly from the field, and the output is real-time evacuation shelter information.

[0979] Step 2:

[0980] Data preprocessing and model training

[0981] The server cleanses the collected data using a Python script. It corrects outliers and fills in missing values ​​to maintain data integrity. It applies data cleansing processes to the raw data as input and outputs clean data. For example, if the demand for a material is incorrectly negative, it is corrected to 0 or an appropriate positive value.

[0982] The server uses the cleansed data to train a machine learning model in TensorFlow. The clean data is fed as input to the machine learning algorithm, which then outputs a demand forecasting model. For example, a model can be built to forecast future demand using past supply demand data for each evacuation shelter.

[0983] Step 3:

[0984] Demand forecasting

[0985] When a new disaster occurs, the server uses a machine learning model to predict demand based on data collected in real time. The input is real-time evacuation shelter data and a trained demand prediction model, and the output is the specific supply and funding needs of each evacuation shelter. For example, it inputs the current number of evacuees and weather information and predicts demand for water and temporary toilets when rising temperatures and strong winds are predicted.

[0986] Step 4:

[0987] Resource Allocation

[0988] The server formulates a specific resource allocation plan based on the predicted demand. The input data is the output of the demand forecasting model and the resources of logistics agencies, and the output is specific delivery instructions. For example, a specific plan may be created that "Deliver 1,000 bottles of 2-liter water to shelter A and 800 bottles of food to shelter B."

[0989] The server sends delivery instructions to logistics organizations via API or email according to the plan. For example, it might instruct a logistics company to "arrange three trucks and deliver supplies to the designated evacuation center."

[0990] Step 5:

[0991] User Notification and Donation Acceptance

[0992] The user checks the evacuation shelter information provided by the system through their terminal. The input data is the real-time evacuation shelter situation, and the output is the information displayed on the user's terminal. For example, the information displayed is "Evacuation shelter A is experiencing a water shortage."

[0993] Users use their devices to donate specific items or funds. The input for the donation is the item or amount selected by the user, and the output is the donation data sent to the server. For example, the operation is "User donates 50 bottles of water to shelter A."

[0994] Step 6:

[0995] Donation management and delivery tracking

[0996] The server records donation information in a database. The input is information about donated goods or funds, and the output is an updated database record. For example, 50 bottles of water are donated.

[0997] The server coordinates with logistics agencies to ensure the proper delivery of donated supplies and tracks the progress of deliveries. The inputs are delivery instructions and feedback from logistics agencies, and the output is delivery status updates. For example, it confirms that a truck has arrived at shelter A and that 50 bottles of water have been delivered safely.

[0998] The user checks the results of their donation on their device. The input is feedback data from the system, and the output is confirmation of the donation results displayed on the user's device. For example, a notification may be displayed stating, "The 50 bottles of donated water have arrived safely at Shelter A."

[0999] These steps will ensure that necessary resources are allocated quickly and effectively in the event of a disaster and prevent uneven distribution of aid.

[1000] (Application example 1)

[1001] 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."

[1002] During disasters, there is a sudden surge in demand for supplies and funds at evacuation centers and other facilities, making it important to distribute them quickly and appropriately. Current systems make it difficult to accurately grasp the real-time situation in disaster-stricken areas, often resulting in uneven resource distribution and delays. Furthermore, effective management and tracking of donated resources is lacking, calling for greater transparency and simplified collaboration with donors.

[1003] 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.

[1004] In this invention, the server includes a means for predicting resource demand during a disaster, a means for allocating resources based on the predicted demand, a means for collecting information on the status of disaster-stricken areas in real time, a means for providing information to users and accepting donations, a means for managing and tracking the delivery of donated resources, and a means for real-time management and optimization of distribution of supplies at a logistics center. This enables appropriate and prompt allocation of resources during a disaster, and effective support for disaster-stricken areas. It also improves information provision and transparency for donors, leading to more efficient donation management.

[1005] A "means for predicting resource demand during a disaster" is a device or program that predicts the supplies and funds required in each region and evacuation shelter based on data from past disasters and data collected in real time.

[1006] A "resource allocation tool" is a device or program that optimally allocates necessary supplies and funds to each evacuation center or region based on a predictive model.

[1007] "Means for collecting information on the status of disaster-stricken areas in real time" refers to hardware and software such as sensors, drones, and communication devices used to collect information from disaster-stricken areas in real time.

[1008] The "means for providing information to users and accepting donations" refers to a device or program that notifies users of the current situation at the evacuation shelter and the supplies they need, and accepts donations.

[1009] "Means for managing and tracking the delivery of donated resources" refers to a device or program that manages and tracks the delivery status of materials and funds donated by users and monitors their progress until they finally reach the recipient.

[1010] "Means for real-time management and distribution optimization of materials in logistics centers" refers to a device or program that monitors materials stored and managed in logistics centers in real time and distributes them optimally.

[1011] A "smartphone" is a mobile terminal that combines the functions of a mobile phone and a computer, and is a device on which applications can be installed and run.

[1012] "Smart glasses" are glasses-type wearable devices that have the function of displaying visual information and providing various types of information to the wearer.

[1013] A "logistics robot" is a robot that automatically transports and sorts materials at logistics centers and other locations.

[1014] A "predictive model" is a statistical or machine learning model built to forecast future demand based on historical data.

[1015] "Efficient donation management" means streamlining the management process from receipt of donated goods and funds to final delivery.

[1016] The present invention is a system for optimizing resource demand prediction and allocation in the event of a disaster, and specific embodiments thereof will be described below.

[1017] First, the server predicts the demand for supplies and funds in each region based on past disaster data and real-time information on current disaster areas. Past disaster data is collected from databases provided by government agencies and relief organizations. For example, it analyzes how much supplies were needed at each evacuation center during the previous year's typhoon.

[1018] The server then uses sensors and drones to obtain real-time data from the affected area, including the current capacity of evacuation centers, a list of needed supplies, and the extent of the damage.

[1019] The server cleanses the collected data, correcting outliers and filling in missing values. The server then trains a machine learning model on the cleansed data to build the ability to predict the specific supply and funding needs of each shelter. The server uses Python and the Scikit-learn library to train the model and make predictions.

[1020] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies each shelter needs. The server then creates a specific resource allocation plan based on the predicted demand. For example, it determines that shelter A needs 1,000 bottles of water and 500 packs of food.

[1021] The server then coordinates with logistics companies to deliver supplies according to this plan. Furthermore, users can check information provided by the system through their devices. For example, they may learn that shelter A is experiencing a water shortage. They can then use their devices to donate specific supplies or funds. Once the donation is complete, the device sends this information to the server.

[1022] The server records donation information in a database and updates the donation status in real time. The server coordinates with logistics companies to ensure the proper delivery of donated supplies and funds, and tracks the progress of deliveries. For example, the server can confirm whether 50 bottles of water have safely arrived at Evacuation Center A. Users can check the results of their donations on their devices.

[1023] Smartphones, smart glasses, and logistics robots are also used in logistics centers to manage and optimize the distribution of materials in real time. Materials stored and managed within the logistics center can be monitored in real time and optimally distributed. This will improve the efficiency of material distribution within the logistics center.

[1024] As a specific example, consider the case where a typhoon hits a part of Japan. Shelter A has a capacity of 1,000 people, and shelter B has 800 people. From the collected data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles. Based on this prediction, the server submits a specific resource allocation plan to the logistics company and begins delivery. At the same time, the user checks the water shortage information for shelter A on their device and donates 50 bottles of water. The device sends this information to the server, which updates the donation status in real time and manages the donated water until it reaches shelter A.

[1025] Examples of prompts to input to a generative AI model include:

[1026] "Please tell me the procedure for logistics centers to use smartphones to check demand forecasts in disaster-stricken areas in real time and allocate supplies optimally."

[1027] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1028] Step 1: Data collection

[1029] The server collects past disaster data from government and relief organization databases, and also obtains real-time data from disaster-stricken areas using sensors and drones. The server collects information such as the number of people in the affected area, a list of necessary supplies, and the extent of damage.

[1030] Input: Past disaster data, real-time data

[1031] Output: Collected data

[1032] Specific operation: The server uses APIs to retrieve data from the database, collect real-time data from each sensor and drone, and store the data.

[1033] Step 2: Data Preprocessing

[1034] The server cleanses the collected data, corrects outliers, and fills in missing values, thereby ensuring data quality.

[1035] Input: Collected data

[1036] Output: Cleansed data

[1037] Specific operation: The server uses the Pandas library to cleanse the data, correcting outliers and imputing missing values ​​automatically.

[1038] Step 3: Model training

[1039] The server will then use the cleansed data to train machine learning models that can predict the specific supply and funding needs of each shelter.

[1040] Input: Cleansed data

[1041] Output: A trained predictive model

[1042] How it works: The server uses the Scikit-learn library to train a machine learning model and generate a predictive model based on past data.

[1043] Step 4: Demand forecast

[1044] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies will be needed at each evacuation center.

[1045] Input: Trained predictive model, real-time data

[1046] Output: Forecasted demand

[1047] Specific operation: The server inputs real-time data into a predictive model to estimate the demand for supplies and funds for each evacuation shelter.

[1048] Step 5: Resource allocation

[1049] The server formulates a specific resource allocation plan based on the predicted demand and works with logistics companies to deliver supplies.

[1050] Input: Forecasted demand

[1051] Output: Resource allocation plan

[1052] Specific operation: The server creates an optimal resource allocation plan based on the prediction results and notifies the logistics company of the plan via an API.

[1053] Step 6: Notify users and accept donations

[1054] Users check the information provided by the system on their terminals and donate the necessary supplies. The donation information is then sent to the server.

[1055] Input: Required information for each evacuation shelter

[1056] Output: Donation information

[1057] Specific operation: The application on the device notifies the user of supply shortages at evacuation centers, and when the user completes the donation procedure, the information is sent to the server.

[1058] Step 7: Donation Management and Delivery Tracking

[1059] The server records donation information in a database, updates donation status in real time, and manages and tracks donated goods and funds to ensure their proper delivery.

[1060] Input: Donation information

[1061] Output: Updated donation status, delivery status of donated goods

[1062] Specific operation: The server records donation information in a database, works with logistics companies to track the delivery status of supplies, and provides feedback to users.

[1063] Step 8: Optimized material management

[1064] The logistics center will use smartphones, smart glasses, and logistics robots to manage supplies in real time and optimize their distribution.

[1065] Input: Resource allocation plan, real-time management data

[1066] Output: Optimized resource allocation

[1067] Specific operation: Each device in the logistics center moves and sorts materials in real time based on a resource allocation plan, maintaining optimal conditions.

[1068] 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.

[1069] The present invention relates to a system for realizing more effective relief activities by combining a system for predicting demand for and optimizing allocation of resources during disasters with an emotion engine that recognizes user emotions. The system according to the present invention can be implemented as follows.

[1070] Program operation description:

[1071] 1. Data Collection:

[1072] The server collects past disaster data from a database, including the amount of supplies needed, the number of evacuees, and the extent of damage caused by past disasters.

[1073] The server also collects real-time information on the current state of the affected area, using sensors and drones to obtain data such as the capacity of evacuation centers, current stock of supplies, and the extent of the damage.

[1074] 2. Data preprocessing and model training:

[1075] The server preprocesses the collected data, correcting outliers and filling in missing values, and then feeds the preprocessed data into machine learning algorithms.

[1076] The server uses the preprocessed data to train machine learning models that predict each shelter's future supply and financial needs.

[1077] 3. Demand forecasting:

[1078] When a new disaster occurs, the server inputs disaster area data collected in real time into the predictive model.

[1079] Based on this data, the server predicts the amount of supplies and funds that will be needed at each evacuation shelter and in the affected area.

[1080] 4. Resource Allocation Planning:

[1081] The server creates a resource allocation plan based on the predicted demand, specifically determining how much supplies should be sent from which distribution center to which evacuation shelter.

[1082] The server notifies the distribution plan to the logistics company and instructs them to prepare for delivery.

[1083] 5. User Notification and Donation Acceptance:

[1084] The server notifies the user's device of information about the affected area, and the user can use their device to obtain information about supplies needed to get to an evacuation shelter on foot.

[1085] Users can use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[1086] 6. Emotion recognition:

[1087] The server is equipped with an emotion engine that recognizes the user's emotions, and collects and analyzes the user's emotion data.

[1088] The terminal identifies the user's emotions in real time and presents the user with assistance options according to the emotions.

[1089] 7. Emotion-based donation suggestions:

[1090] The server proposes appropriate donation options to the user based on the user's emotional data provided by the emotion engine. For example, if the server determines that the user is willing to donate a large amount, it will present the user with larger donation options.

[1091] 8. Donation Management and Delivery Tracking:

[1092] The server records donation information in a database, updates donation status in real time, and, if necessary, reallocates donated supplies and funds to areas in need.

[1093] The server tracks the delivery status of donated goods, receives delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination.

[1094] Users can check the results of their donations through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[1095] Examples:

[1096] For example, suppose a large earthquake occurs and devastates parts of Japan. Shelter A can accommodate 1,000 people, while shelter B has 800. Based on past disaster data and real-time collected information, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800. Based on this prediction, the server notifies the logistics company of a specific resource allocation plan and instructs delivery. At the same time, the user can check the water shortage information for shelter A through their device and be presented with appropriate donation options based on their emotional data. Once the user completes their donation, this information is sent from the device to the server, which updates the donation status in real time. The server tracks the delivery status of the donated supplies and confirms that the water has safely arrived at shelter A. The user can then check how their donation has been used through their device.

[1097] By using the above method and system, the present invention is expected to make disaster resource demand forecasts and appropriate allocations quick and effective, thereby improving the efficiency of relief activities. By combining it with an emotion engine, it becomes possible to propose relief based on the user's emotions, realizing the provision of more appropriate relief.

[1098] The processing flow will be explained below.

[1099] Step 1: Data collection

[1100] The server collects disaster data from a database of past disasters, including the amount of supplies and funds needed at each evacuation shelter, the number of evacuees, and the extent of damage caused by past disasters.

[1101] The server also uses sensors and drones to collect real-time data on the current disaster area, providing information such as shelter capacity, current supply inventory, and the extent of the damage.

[1102] Step 2: Data Preprocessing

[1103] The server cleanses the collected data, corrects outliers, and fills in missing values, allowing machine learning algorithms to learn accurately.

[1104] Step 3: Predictive model training

[1105] The server uses machine learning algorithms to train a demand forecasting model based on the pre-processed data, creating a model capable of predicting each shelter's future supply and financial needs.

[1106] Step 4: Demand forecast

[1107] When a new disaster occurs, the server inputs current data from the affected area, collected in real time, into the predictive model.

[1108] Based on this data, the server predicts the specific amount of supplies and funds needed at each evacuation center and disaster area.

[1109] Step 5: Resource Allocation Planning

[1110] The server creates a resource allocation plan based on the predicted demand, specifically determining how much supplies should be sent from which distribution center to which evacuation shelter.

[1111] The server issues delivery instructions to the logistics company based on this plan.

[1112] Step 6: Notify users and accept donations

[1113] The server notifies the user's device of information about the affected area, allowing the user to check on their device what supplies are in short supply at evacuation shelters.

[1114] Users use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[1115] Step 7: Emotion Recognition

[1116] The server uses an emotion engine that recognizes the user's emotions to collect and analyze the user's emotion data.

[1117] The device analyzes the user's emotions in real time and sends the results to the server.

[1118] Step 8: Emotionally Based Donation Proposals

[1119] The server presents appropriate donation options to the user based on the user's emotional data obtained from the emotion engine, for example, presenting an urgent donation option if the user is emotionally moved.

[1120] Step 9: Donation Management and Delivery Tracking

[1121] The server records user donation information in a database, updates the donation status in real time, and verifies that donated goods and funds are distributed appropriately, as needed.

[1122] The server tracks the delivery of donated goods, obtaining delivery status from logistics providers and managing the process until the goods arrive safely at their destination.

[1123] Users can check how their donations have been used through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[1124] The above are the specific processing steps of the system of the present invention. This system is expected to make relief activities more efficient by quickly and effectively predicting demand for resources during disasters and allocating them appropriately. By combining it with an emotion engine, it becomes possible to propose relief based on the user's emotions, realizing the provision of more appropriate relief.

[1125] Example 2

[1126] 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."

[1127] During a disaster, the condition of the affected area and the demand for supplies change from moment to moment, so it is necessary to allocate resources quickly and accurately.In addition, it is necessary to realize effective relief activities by making appropriate relief proposals based on the user's emotions and motivation, but current systems are not able to adequately address these issues.

[1128] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting past disaster data and correcting outliers and imputing missing values ​​based on the data; means for training a machine learning model based on the preprocessed data and predicting future supply and financial needs at each evacuation shelter; means for collecting disaster area data in real time when a new disaster occurs and inputting it into the machine learning model; means for formulating a resource allocation plan based on the predicted demand and notifying logistics companies; means for notifying a user's terminal of disaster area information and accepting donations; means for proposing appropriate support options using an emotion engine that recognizes the user's emotions; means for recording donation information in a database, updating the donation status in real time, and tracking the delivery status of donated supplies; and means for the user to check the results of the donation via the terminal. This allows for rapid and accurate resource demand prediction and appropriate allocation during a disaster, enabling effective support activities based on the user's emotions.

[1129] "Past disaster data" refers to information about disasters that have occurred in the past, including data on the demand for supplies, the number of evacuees, and the extent of damage.

[1130] "Outlier correction" refers to the process of detecting and correcting statistically abnormal values ​​in acquired data.

[1131] "Missing value imputation" refers to the operation of filling in missing values ​​in a dataset, and generally uses statistical methods such as the mean or median.

[1132] "Preprocessed data" refers to data that has been converted into a form suitable for machine learning models by correcting outliers and filling in missing values.

[1133] A "machine learning model" refers to an algorithm or statistical model that can make predictions or classifications based on past data.

[1134] "Collecting disaster area data in real time" refers to the operation of instantly collecting data on ongoing disasters and sending it to a server.

[1135] A "resource allocation plan" refers to a plan that determines how much supplies to send from which logistics center to which shelter based on predicted demand.

[1136] "Logistics company" refers to a company or organization responsible for transporting goods.

[1137] "User terminal" refers to a computer device such as a smartphone or tablet held by a user, and is a means for communicating with a server.

[1138] "Emotion engine" refers to a system for recognizing and analyzing a user's emotional state.

[1139] "Support options" refers to the support methods and types of donations suggested to users.

[1140] "Donation Information" refers to the details of a user's donation (donation recipient, donated items, amount, date and time, etc.).

[1141] A "database" refers to a system for systematically storing and managing data.

[1142] "Donation results" refers to information about how the donation made by the user was used and what results it produced.

[1143] The present invention relates to a system for optimizing resource demand forecasting and allocation in the event of a disaster. This system is mainly composed of a server, terminals, and users.

[1144] First, the server collects past disaster data from a database. The database stores data on past disasters, such as the demand for supplies, the number of evacuees, and the extent of damage. Next, the server uses sensors and drones to collect real-time information on the current situation in the disaster-stricken area. Specifically, it obtains information such as the capacity of evacuation shelters, the current stock of supplies, and the extent of damage.

[1145] The collected data is preprocessed by the server. During the preprocessing process, outliers are corrected, missing values ​​are imputed, and the data is normalized to make it suitable for the machine learning model. The server then trains the machine learning model based on the preprocessed data. This creates a model that predicts the future supply and funding needs of each shelter.

[1146] When a new disaster occurs, the server inputs real-time data on the affected area into the trained prediction model. Based on this data, it predicts the amount of supplies and funds needed at each evacuation shelter and in the affected area. Based on the predicted demand, the server formulates a resource allocation plan, specifically determining how much supplies to send from which logistics center to which evacuation shelter. The server then notifies the distribution plan to logistics companies and issues delivery instructions.

[1147] Users can receive information about disaster areas through their devices and make donations to specific shelters or supplies. Once a donation is completed, this information is sent from the device to a server, which updates the donation status in real time.

[1148] Furthermore, the server is equipped with an emotion engine that recognizes the user's emotions. The device identifies the user's emotions in real time from their facial expressions and voice data, and presents the user with support options that correspond to those emotions. The emotion engine analyzes the user's emotional data and suggests appropriate donation options. For example, if it is inferred that the user is willing to make a large donation, the user will be presented with larger support options.

[1149] Donation information is recorded in a database, and the server updates the donation status in real time. If necessary, donated goods or funds are redistributed to areas where they are in short supply. The server tracks the delivery status of donated goods, obtains delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination. Users can check the results of their donations on their own devices.

[1150] Specific examples

[1151] For example, let's assume that a large-scale earthquake occurs and damages parts of Japan. Shelter A has a capacity of 1,000 people, and shelter B has 800 people. Based on past disaster data and information collected in real time, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800. Based on this prediction, the server notifies the logistics company of a resource allocation plan and instructs delivery. At the same time, the user checks water shortage information for shelter A through their device and is presented with the optimal donation option based on their emotional data. Once the user completes their donation, this information is sent from the device to the server, which updates the donation status in real time. The server tracks the delivery status of the donated supplies and confirms whether the water has arrived safely at shelter A. The user can then check how their donation has been used through their device.

[1152] Prompt Sentence Examples

[1153] "In an area affected by a large earthquake, 1,000 people are evacuated to shelter A and 800 people are evacuated to shelter B. Please predict the supplies needed at each shelter based on past data and real-time information, and issue instructions to optimize relief activities."

[1154] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1155] Step 1: Data collection

[1156] The server collects past disaster data from a database. Specifically, it uses SQL queries to obtain data such as the demand for supplies, the number of evacuees, and the extent of damage during past disasters. It also uses sensors and drones to collect real-time information on the current situation in disaster-stricken areas. Data from the sensors, such as the number of people accommodated, stock of supplies, and the extent of damage, is sent to the server.

[1157] Input: Past disaster database, sensors, drones

[1158] Output: Collected disaster data

[1159] Step 2: Data Preprocessing

[1160] The server preprocesses the collected data, correcting outliers and imputing missing values, and normalizing the data. Missing values ​​are imputed with the mean or median, and outliers are corrected using statistical methods. The data is then scaled and converted into a format suitable for machine learning models.

[1161] Input: Collected disaster data

[1162] Output: Preprocessed data

[1163] Step 3: Model training

[1164] The server trains a machine learning model using the preprocessed data. It splits the dataset into a training set and a test set, trains a regression model or deep learning model using the training set, evaluates the model's performance on the test set, and tunes hyperparameters as needed.

[1165] Input: Preprocessed data

[1166] Output: A trained machine learning model

[1167] Step 4: Demand forecast

[1168] When a new disaster occurs, the server inputs real-time data on the affected area into the trained prediction model, which then uses the model to predict the amount of supplies and funds needed at each evacuation center and in the affected area.

[1169] Input: Real-time disaster area data, trained machine learning model

[1170] Output: Forecasted demand for supplies and funds for each shelter

[1171] Step 5: Resource Allocation Planning

[1172] The server creates a resource allocation plan based on predicted demand. It uses an optimization algorithm to determine how much supplies should be sent from which distribution center to which shelter. It then notifies the distribution company of this plan via an API, instructing them to prepare for delivery.

[1173] Input: Forecasted demand data

[1174] Output: Resource allocation plan, notification to logistics companies

[1175] Step 6: Notify users and accept donations

[1176] The server notifies users of information about the affected areas via their devices. Users can then donate to specific shelters or supplies via their devices. Once a donation is complete, the information is sent from the device to the server, and the donation status is updated in real time.

[1177] Input: Disaster area information, user donation data

[1178] Output: User notification, donation acceptance information

[1179] Step 7: Emotion Recognition

[1180] The server uses an emotion engine to recognize the user's emotions. The device collects the user's facial expressions and voice data and analyzes them using the emotion engine. As a result, assistance options based on the user's emotions are presented.

[1181] Input: User's facial expression and voice data

[1182] Output: Analyzed emotion data, support options

[1183] Step 8: Emotionally Based Donation Proposals

[1184] The server proposes appropriate donation options to the user based on the analysis results of the emotion engine. If the server determines that the user is willing to donate a large amount, it presents the option of large-scale support.

[1185] Input: Parsed emotion data

[1186] Output: Donation proposal

[1187] Step 9: Donation Management and Delivery Tracking

[1188] The server records users' donation information in a database and updates the donation status in real time. It tracks the delivery status of donated goods and funds and obtains delivery status from logistics companies, managing the donated goods until they arrive safely at their destination. Users can check the results of their donations through their devices.

[1189] Input: Donation information, delivery status from logistics company

[1190] Output: Donation management data, delivery tracking information, user notifications

[1191] (Application example 2)

[1192] 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."

[1193] Conventional resource demand forecasting and allocation systems during disasters do not consider the psychological and emotional state of disaster victims when proposing assistance, making it difficult to provide effective assistance. Furthermore, assistance proposals and effective resource allocations are carried out in stages, making it difficult to respond quickly. The present invention aims to solve these problems by combining assistance proposals that consider users' emotions with real-time demand forecasting.

[1194] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1195] In this invention, the server includes means for predicting resource demand during a disaster, means for allocating resources based on the predicted demand, and means for collecting information on the status of the disaster area in real time, thereby enabling means for providing information to users and accepting donations, means for managing and tracking the delivery of donated resources, and means for identifying user emotions in real time and generating support proposals based on those emotions.

[1196] A "means for predicting resource demand during a disaster" is a device or program that predicts the amount of supplies and resources that will be needed when a disaster occurs.

[1197] The "means for allocating resources based on the predicted demand" is a device or program that efficiently allocates materials and resources to each location based on the predicted resource demand.

[1198] "Means for collecting information on the status of disaster-affected areas in real time" refers to devices or systems that use sensors, drones, APIs, etc. to instantly obtain the current status of areas affected by disasters.

[1199] The "means for providing information and accepting donations" refers to an interface or application that provides users with the necessary information and accepts donations from users.

[1200] "Means for managing and tracking the delivery of donated resources" refers to a system that manages the delivery of donated goods and funds to ensure they are delivered to the designated locations and tracks the delivery status in real time.

[1201] The "means for identifying a user's emotions in real time and generating assistance suggestions based on those emotions" refers to a device or system that analyzes a user's emotions in real time and suggests appropriate assistance options based on the results.

[1202] "Means of collecting data on past disasters and training a predictive model based on that data" refers to a system that collects data on past disasters and uses that data to train a demand prediction model for future disasters using a machine learning algorithm.

[1203] "Means for using a predictive model to predict the amount of supplies and funds required for each evacuation shelter" refers to a device or system that uses a trained predictive model to specifically calculate the amount of supplies and funds required for each evacuation shelter.

[1204] The present invention provides a system that combines emotion analysis with a resource demand forecasting and allocation system in the event of a disaster, and proposes assistance based on the user's emotions. Specific embodiments of the present invention are described below.

[1205] System Configuration

[1206] The system consists of the following major components:

[1207] 1. Server

[1208] 2. Terminal

[1209] 3. Users

[1210] The role and specific functions of each component

[1211] server

[1212] The server has several options:

[1213] Data collection method: The server collects past disaster data and real-time local data using sensors, drones, and APIs. For example, it uses Google Maps API to obtain information on disaster areas.

[1214] Prediction model training method: The server trains a prediction model using a machine learning algorithm (e.g., Scikit-learn's LinearRegression) based on the collected past disaster data.

[1215] Demand forecasting tools: Using trained predictive models, we forecast the amount of supplies and funding needed at each shelter.

[1216] Sentiment analysis method: Emotions are identified in real time using TensorFlow and OpenCV based on data (images and audio) sent by the user.

[1217] Support proposal generation means: Based on the results of sentiment analysis, appropriate donation options and support plans are proposed to the user.

[1218] Donation Management and Tracking: Manage donations and track delivery status in real time using a MySQL database.

[1219] Terminal

[1220] The terminal has the following functions:

[1221] Information provision function: Provides users with information on the status of the disaster area and the necessary support.

[1222] Donation acceptance feature: Users can make donations to specific shelters or supplies.

[1223] Emotional data transmission function: Collects user emotional data using the smartphone's camera and microphone and transmits it to the server.

[1224] User

[1225] The user does the following:

[1226] Information reception: Receive information about the situation in the disaster area and the need for assistance.

[1227] Donate: Make a donation through your device.

[1228] Emotion data provision: Send your own emotional data to the server via your device.

[1229] Example

[1230] For example, suppose a large earthquake occurs and some areas are affected. Based on past earthquake data and real-time data, the server predicts that shelter A will need 1,000 bottles of water and shelter B will need 800 bottles of water. A user checks the water shortage information for shelter A on their device, senses the urgency, and makes a donation. Once the donation is complete, the server records the donation information in a database and tracks the delivery status.

[1231] Prompt Sentence Examples

[1232] "Designing a disaster food delivery application that optimizes donations through emotion recognition. Features include real-time demand forecasting, user emotion analysis, and donation suggestions. For example, it includes a function to estimate the food needs of shelter A and provide donation options based on the user's emotion."

[1233] It is expected that this system will solve the problems faced by conventional disaster response systems and enable efficient and effective support activities.

[1234] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1235] Step 1: Data collection

[1236] The server collects past disaster data from a database and uses sensors, drones, and APIs (e.g., Google Maps API) to obtain real-time local data (such as shelter capacity, supply inventory, and damage extent).

[1237] Input: Past disaster data, local real-time data

[1238] Data processing: Correction of outliers, completion of missing values

[1239] Output: Preprocessed dataset

[1240] Step 2: Model training

[1241] The server trains a predictive model using a machine learning algorithm (Scikit-learn's LinearRegression) based on the preprocessed data.

[1242] Input: Preprocessed dataset

[1243] Data Computing: Model Training with Machine Learning

[1244] Output: A trained predictive model

[1245] Step 3: Demand forecast

[1246] When a new disaster occurs, the server inputs the collected real-time data into a trained predictive model to predict the amount of supplies and funds needed at each evacuation center.

[1247] Input: Real-time data, trained predictive model

[1248] Data calculation: Running demand forecasting algorithms

[1249] Output: The amount of supplies and funds needed for each shelter

[1250] Step 4: Provide information

[1251] The terminal notifies the user of information about the disaster area provided by the server, and the user can check the accommodation situation at evacuation centers and the shortage of supplies.

[1252] Input: Disaster area information from the server

[1253] Data processing: Convert to notification format

[1254] Output: Information to the user

[1255] Step 5: Accept donations

[1256] Users make donations to specific shelters and supplies through their devices, which then send the donation information to the server.

[1257] Input: User donation information

[1258] Data processing: recording and transmitting donation information

[1259] Output: Send donation information to the server

[1260] Step 6: Collect emotional data

[1261] The device uses the smartphone's camera and microphone to collect the user's emotional data and transmits it to a server.

[1262] Input: User voice and image data

[1263] Data Computing: Identifying emotions through emotion analysis models

[1264] Output: Sending the identified emotion data to the server

[1265] Step 7: Generate support proposals

[1266] The server generates appropriate donation options and support plans for the user based on the results of the sentiment analysis.

[1267] Input: User emotion data

[1268] Data Computing: Implementing an Emotion-Based Donation Suggestion Algorithm

[1269] Output: Generates donation options and support plans for the user

[1270] Step 8: Donation Management and Tracking

[1271] The server manages and tracks in real time how much donated supplies and funds have been delivered to which shelters, using a MySQL database to record donations and get status updates from the distribution center.

[1272] Input: Delivery status from the distribution center, donation information

[1273] Data processing: Merging donation information and delivery status

[1274] Output: Real-time donation management status

[1275] Through these processing steps, the system for predicting and allocating resource demand during a disaster can propose assistance based on the user's emotions, enabling efficient and effective assistance activities.

[1276] 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.

[1277] 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.

[1278] 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.

[1279] [Fourth embodiment]

[1280] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1281] 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.

[1282] 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).

[1283] 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.

[1284] 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.

[1285] 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).

[1286] 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.

[1287] 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.

[1288] 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.

[1289] 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.

[1290] 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.

[1291] 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.

[1292] 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."

[1293] The present invention is a system for optimizing resource demand prediction and allocation in the event of a disaster, and specific embodiments thereof will be described below.

[1294] First, the system is centered around a server, which predicts the demand for supplies and funds in each region based on past disaster data and real-time information on current disaster-stricken areas.

[1295] Program operation description:

[1296] 1. Data Collection:

[1297] The server collects data on past disasters from government and NGO databases, analyzing, for example, how much supplies were needed at each evacuation shelter during a typhoon the previous year.

[1298] The server also uses sensors and drones to collect real-time data from the affected areas, including the current capacity of evacuation centers, a list of needed supplies, and the extent of the damage.

[1299] 2. Data preprocessing and model training:

[1300] The server cleanses the collected data, corrects outliers, and fills in missing values.

[1301] The server will then use the cleansed data to train machine learning models that can predict the specific supplies and funding needs of each shelter.

[1302] 3. Demand forecasting:

[1303] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies will be needed at each evacuation center.

[1304] 4. Resource Allocation:

[1305] The server then formulates a specific resource allocation plan based on the predicted demand. For example, it determines that shelter A needs 1,000 bottles of water and 500 packs of food.

[1306] The server will follow this plan and work with logistics companies to deliver the supplies.

[1307] 5. User Notification and Donation Acceptance:

[1308] The user can check the information provided by the system through the terminal. For example, they can learn that shelter A is experiencing a water shortage.

[1309] Users use their devices to donate specific items or funds, and once the donation is complete, the device sends this information to the server.

[1310] 6. Donation Management and Delivery Tracking:

[1311] The server records the donation information in a database and updates the donation status in real time.

[1312] The server coordinates with logistics companies to ensure the proper delivery of donated supplies and funds, and tracks the progress of deliveries. For example, it checks whether 50 bottles of water have arrived safely at Evacuation Center A.

[1313] Users can check the results of their donations through their devices. For example, they can confirm that the supplies they donated have arrived at shelter A.

[1314] Examples:

[1315] As a specific example, consider the case where a typhoon hits a part of Japan. Shelter A can accommodate 1,000 people, and shelter B has 800 people. From the collected data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles. Based on this prediction, the server submits a specific resource allocation plan to the logistics company and begins delivery. At the same time, the user checks the water shortage information for shelter A on their device and donates 50 bottles of water. The device sends this information to the server, which updates the donation status in real time and manages the donated water until it arrives safely at shelter A.

[1316] By using the above-described method and system, the present invention can quickly and effectively allocate necessary resources in the event of a disaster, thereby preventing uneven distribution of support.

[1317] The processing flow will be explained below.

[1318] Step 1: Data collection

[1319] The server collects past disaster data from a database of past disasters, including the amount of supplies and funds needed at each evacuation shelter, the number of evacuees, and the extent of damage caused by past disasters.

[1320] At the same time, the server collects real-time information on the current situation in the disaster area, using sensors and drones to obtain data such as the number of people in evacuation shelters, the current stock of supplies, and the extent of the damage.

[1321] Step 2: Data Preprocessing

[1322] The server cleanses the collected data on past disasters and current situations, correcting outliers and filling in missing values ​​to improve data quality.

[1323] The server converts the pre-processed data into a suitable format and prepares it for the machine learning algorithms.

[1324] Step 3: Predictive model training

[1325] The server uses machine learning algorithms to train a demand forecasting model based on the preprocessed dataset, creating a model capable of predicting each shelter's future supply and financial needs.

[1326] The server stores the trained model and prepares it for real-time predictions.

[1327] Step 4: Demand forecast

[1328] When a new disaster occurs, the server inputs current disaster area data collected in real time into the predictive model.

[1329] Based on this data, the server predicts the amount of supplies and funds that will be needed at each evacuation shelter and in the affected area.

[1330] Step 5: Resource Allocation Planning

[1331] The server creates a resource allocation plan based on predicted demand, specifically determining how much supplies should be sent from which distribution center to which shelter.

[1332] The server notifies the distribution plan to the logistics company and instructs them to prepare for delivery.

[1333] Step 6: Notify users and accept donations

[1334] The user checks the information provided by the system through their own terminal. For example, they learn that shelter A is experiencing a water shortage.

[1335] Users use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[1336] Step 7: Donation Management and Delivery Tracking

[1337] The server records donation information in a database, updates donation status in real time, and, if necessary, reallocates donated supplies and funds to areas in need.

[1338] The server tracks the delivery status of donated goods, receives delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination.

[1339] Users can check how their donations have been used through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[1340] The above are the specific processing steps of the system of the present invention. This system is expected to enable rapid and effective forecasting of demand for resources and appropriate allocation in the event of a disaster.

[1341] Example 1

[1342] 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."

[1343] In times of disaster, rapid and accurate resource allocation is required, but conventional systems have had difficulty achieving this. In particular, it was difficult to use past disaster data and create immediate demand forecasts and distribution plans based on the latest information on affected areas. Another issue was the lack of functionality to reflect user donations in real time and manage and track their delivery status.

[1344] 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.

[1345] In this invention, the server includes means for collecting past disaster data from a database, means for collecting information on disaster-stricken areas in real time using sensors and remote control devices, means for cleansing the collected data, correcting outliers and filling in missing values, means for training a machine learning model based on the cleansed data and predicting demand for supplies and funds, means for formulating a specific resource allocation plan based on the predicted demand and delivering supplies in cooperation with delivery agencies, means for providing shelter information and accepting donations via a user terminal, means for recording donation information in a database and tracking the delivery of donated resources, and means for confirming donation results via a user terminal. This makes it possible to quickly and accurately allocate necessary resources in the event of a disaster and prevent uneven distribution of support.

[1346] A "database" is an electronic system designed to efficiently store, manage, and retrieve information.

[1347] "Disaster data" refers to data containing information about natural and man-made disasters, including the date and time of the disaster, location, type and extent of damage, and necessary supplies.

[1348] A "sensor" is a device that detects a physical phenomenon and converts it into an electrical signal.

[1349] A "remote control device" is a device used to operate and control equipment or systems from a remote location.

[1350] "Real-time" refers to a state in which data and information are acquired and processed immediately.

[1351] "Data cleansing" is the process of correcting errors and missing values ​​and maintaining consistency in order to improve the quality of data.

[1352] A "machine learning model" is a mathematical model that learns patterns and rules based on large amounts of data and makes predictions and classifications for new data.

[1353] "Prediction" is the prediction of future events or conditions based on past and present data.

[1354] A "resource allocation plan" is a detailed plan for optimally distributing available resources and providing them where needed.

[1355] A "delivery organization" is an organization or infrastructure that transports goods or services to designated locations.

[1356] A "user terminal" is a device used to access the system and exchange information.

[1357] "Shelter information" refers to data about shelters, including the number of people they can accommodate, the supplies they need, and current issues.

[1358] A "donation" is when an individual or organization provides goods or funds for disaster relief.

[1359] "Recording in a database" means storing information in a database in a manner that allows it to be accessed later.

[1360] "Tracking" is the process of monitoring and confirming how goods or information are moving and where they are currently located.

[1361] "Donation results" is information showing how donated goods and funds were used and where they ended up.

[1362] MODE FOR CARRYING OUT THE INVENTION

[1363] The present invention relates to a system for optimizing resource demand forecasting and allocation in the event of a disaster, and specific embodiments thereof are described below. The core of this system is a server that makes full use of advanced data collection and machine learning.

[1364] Hardware and software used

[1365] 1. Server:

[1366] Hardware: General-purpose high-performance computer (e.g., AWS EC2 instance)

[1367] Software: Apache Kafka, Python, TensorFlow, Flask

[1368] 2. Terminal:

[1369] The device from which the user accesses the site (e.g., smartphone, tablet, PC)

[1370] Software: React

[1371] 3. Data Collection Equipment:

[1372] Sensors, drones, etc. (collecting on-site conditions)

[1373] Program processing overview

[1374] 1. Data Collection:

[1375] The server collects historical disaster data from government and NGO databases, specifically using Apache Kafka to retrieve time-stamped disaster data (location, type of damage, type and quantity of needed supplies).

[1376] The server uses sensors and drones to collect real-time data from the affected area, including the capacity of evacuation shelters, lists of needed supplies, and the extent of damage.

[1377] 2. Data preprocessing and model training:

[1378] The server cleanses the collected data using Python scripts, correcting outliers and completing missing values.

[1379] Next, the machine learning library TensorFlow is used to train a demand forecasting model based on disaster data, resulting in a model capable of predicting how much supplies will be needed at each evacuation center.

[1380] 3. Demand forecasting and resource allocation:

[1381] When a new disaster occurs, the server inputs real-time data into a predictive model to forecast the specific supply and funding needs of each evacuation shelter.

[1382] The server then formulates a resource allocation plan based on the predicted data and coordinates with delivery agencies to deliver the supplies to the required locations. Delivery arrangements are made through specific logs and APIs.

[1383] 4. User Notification and Donation Acceptance:

[1384] Users can check information about evacuation shelters through their devices. For example, information such as "evacuation shelter A is short of water" or "evacuation shelter B is short of food" will be displayed.

[1385] Users can donate specific items or funds using their own devices, and the information is quickly transmitted to the server.

[1386] 5. Donation Management and Delivery Tracking:

[1387] The server records the donation information in a database and updates the status in real time.

[1388] The server tracks the progress of donated goods to ensure they are delivered appropriately, and users can view on their devices how their donations are being used.

[1389] Specific examples

[1390] For example, consider a situation where a typhoon hits a part of Japan, with 1,000 people housed in shelter A and 800 people evacuated to shelter B. Based on real-time data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles of water. Based on this prediction, the server will contact a logistics company with a specific delivery schedule and arrange for the necessary supplies to be delivered to each shelter. At the same time, a user checks the water shortage information for shelter A on their device and donates 50 bottles of water. This information is updated in real time on the server, where tracking and management are performed, and the user's device is notified of the donation results.

[1391] Example prompts for generative AI models

[1392] "If a typhoon hits a certain area of ​​Japan, 1,000 people will be evacuated to shelter A and 800 people will be evacuated to shelter B. Please create an appropriate supply distribution plan for this situation and tell us which supplies and how much should be delivered to shelters A and B."

[1393] This system will enable the rapid and effective allocation of necessary resources in the event of a disaster, preventing uneven distribution of aid.

[1394] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1395] Program steps and their specific explanations:

[1396] Step 1:

[1397] Data collection

[1398] The server collects past disaster data from government and NGO databases. Specifically, it uses Apache Kafka to obtain data such as the "date and time of the disaster," "location of damage," and "type and quantity of needed supplies." The input data here is past event information related to disasters, and the output is raw data for cleansing. For example, it collects data such as "how much supplies were needed at which evacuation shelters during the previous year's typhoon."

[1399] The server also collects information on the affected areas in real time, using sensors and drones to obtain data such as the capacity of evacuation shelters, a list of necessary supplies, and the extent of damage. This input data is obtained directly from the field, and the output is real-time evacuation shelter information.

[1400] Step 2:

[1401] Data preprocessing and model training

[1402] The server cleanses the collected data using a Python script. It corrects outliers and fills in missing values ​​to maintain data integrity. It applies data cleansing processes to the raw data as input and outputs clean data. For example, if the demand for a material is incorrectly negative, it is corrected to 0 or an appropriate positive value.

[1403] The server uses the cleansed data to train a machine learning model in TensorFlow. The clean data is fed as input to the machine learning algorithm, which then outputs a demand forecasting model. For example, a model can be built to forecast future demand using past supply demand data for each evacuation shelter.

[1404] Step 3:

[1405] Demand forecasting

[1406] When a new disaster occurs, the server uses a machine learning model to predict demand based on data collected in real time. The input is real-time evacuation shelter data and a trained demand prediction model, and the output is the specific supply and funding needs of each evacuation shelter. For example, it inputs the current number of evacuees and weather information and predicts demand for water and temporary toilets when rising temperatures and strong winds are predicted.

[1407] Step 4:

[1408] Resource Allocation

[1409] The server formulates a specific resource allocation plan based on the predicted demand. The input data is the output of the demand forecasting model and the resources of logistics agencies, and the output is specific delivery instructions. For example, a specific plan may be created that "Deliver 1,000 bottles of 2-liter water to shelter A and 800 bottles of food to shelter B."

[1410] The server sends delivery instructions to logistics organizations via API or email according to the plan. For example, it might instruct a logistics company to "arrange three trucks and deliver supplies to the designated evacuation center."

[1411] Step 5:

[1412] User Notification and Donation Acceptance

[1413] The user checks the evacuation shelter information provided by the system through their terminal. The input data is the real-time evacuation shelter situation, and the output is the information displayed on the user's terminal. For example, the information displayed is "Evacuation shelter A is experiencing a water shortage."

[1414] Users use their devices to donate specific items or funds. The input for the donation is the item or amount selected by the user, and the output is the donation data sent to the server. For example, the operation is "User donates 50 bottles of water to shelter A."

[1415] Step 6:

[1416] Donation management and delivery tracking

[1417] The server records donation information in a database. The input is information about donated goods or funds, and the output is an updated database record. For example, 50 bottles of water are donated.

[1418] The server coordinates with logistics agencies to ensure the proper delivery of donated supplies and tracks the progress of deliveries. The inputs are delivery instructions and feedback from logistics agencies, and the output is delivery status updates. For example, it confirms that a truck has arrived at shelter A and that 50 bottles of water have been delivered safely.

[1419] The user checks the results of their donation on their device. The input is feedback data from the system, and the output is confirmation of the donation results displayed on the user's device. For example, a notification may be displayed stating, "The 50 bottles of donated water have arrived safely at Shelter A."

[1420] These steps will ensure that necessary resources are allocated quickly and effectively in the event of a disaster and prevent uneven distribution of aid.

[1421] (Application example 1)

[1422] 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."

[1423] During disasters, there is a sudden surge in demand for supplies and funds at evacuation centers and other facilities, making it important to distribute them quickly and appropriately. Current systems make it difficult to accurately grasp the real-time situation in disaster-stricken areas, often resulting in uneven resource distribution and delays. Furthermore, effective management and tracking of donated resources is lacking, calling for greater transparency and simplified collaboration with donors.

[1424] 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.

[1425] In this invention, the server includes a means for predicting resource demand during a disaster, a means for allocating resources based on the predicted demand, a means for collecting information on the status of disaster-stricken areas in real time, a means for providing information to users and accepting donations, a means for managing and tracking the delivery of donated resources, and a means for real-time management and optimization of distribution of supplies at a logistics center. This enables appropriate and prompt allocation of resources during a disaster, and effective support for disaster-stricken areas. It also improves information provision and transparency for donors, leading to more efficient donation management.

[1426] A "means for predicting resource demand during a disaster" is a device or program that predicts the supplies and funds required in each region and evacuation shelter based on data from past disasters and data collected in real time.

[1427] A "resource allocation tool" is a device or program that optimally allocates necessary supplies and funds to each evacuation center or region based on a predictive model.

[1428] "Means for collecting information on the status of disaster-stricken areas in real time" refers to hardware and software such as sensors, drones, and communication devices used to collect information from disaster-stricken areas in real time.

[1429] The "means for providing information to users and accepting donations" refers to a device or program that notifies users of the current situation at the evacuation shelter and the supplies they need, and accepts donations.

[1430] "Means for managing and tracking the delivery of donated resources" refers to a device or program that manages and tracks the delivery status of materials and funds donated by users and monitors their progress until they finally reach the recipient.

[1431] "Means for real-time management and distribution optimization of materials in logistics centers" refers to a device or program that monitors materials stored and managed in logistics centers in real time and distributes them optimally.

[1432] A "smartphone" is a mobile terminal that combines the functions of a mobile phone and a computer, and is a device on which applications can be installed and run.

[1433] "Smart glasses" are glasses-type wearable devices that have the function of displaying visual information and providing various types of information to the wearer.

[1434] A "logistics robot" is a robot that automatically transports and sorts materials at logistics centers and other locations.

[1435] A "predictive model" is a statistical or machine learning model built to forecast future demand based on historical data.

[1436] "Efficient donation management" means streamlining the management process from receipt of donated goods and funds to final delivery.

[1437] The present invention is a system for optimizing resource demand prediction and allocation in the event of a disaster, and specific embodiments thereof will be described below.

[1438] First, the server predicts the demand for supplies and funds in each region based on past disaster data and real-time information on current disaster areas. Past disaster data is collected from databases provided by government agencies and relief organizations. For example, it analyzes how much supplies were needed at each evacuation center during the previous year's typhoon.

[1439] The server then uses sensors and drones to obtain real-time data from the affected area, including the current capacity of evacuation centers, a list of needed supplies, and the extent of the damage.

[1440] The server cleanses the collected data, correcting outliers and filling in missing values. The server then trains a machine learning model on the cleansed data to build the ability to predict the specific supply and funding needs of each shelter. The server uses Python and the Scikit-learn library to train the model and make predictions.

[1441] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies each shelter needs. The server then creates a specific resource allocation plan based on the predicted demand. For example, it determines that shelter A needs 1,000 bottles of water and 500 packs of food.

[1442] The server then coordinates with logistics companies to deliver supplies according to this plan. Furthermore, users can check information provided by the system through their devices. For example, they may learn that shelter A is experiencing a water shortage. They can then use their devices to donate specific supplies or funds. Once the donation is complete, the device sends this information to the server.

[1443] The server records donation information in a database and updates the donation status in real time. The server coordinates with logistics companies to ensure the proper delivery of donated supplies and funds, and tracks the progress of deliveries. For example, the server can confirm whether 50 bottles of water have safely arrived at Evacuation Center A. Users can check the results of their donations on their devices.

[1444] Smartphones, smart glasses, and logistics robots are also used in logistics centers to manage and optimize the distribution of materials in real time. Materials stored and managed within the logistics center can be monitored in real time and optimally distributed. This will improve the efficiency of material distribution within the logistics center.

[1445] As a specific example, consider the case where a typhoon hits a part of Japan. Shelter A has a capacity of 1,000 people, and shelter B has 800 people. From the collected data, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800 bottles. Based on this prediction, the server submits a specific resource allocation plan to the logistics company and begins delivery. At the same time, the user checks the water shortage information for shelter A on their device and donates 50 bottles of water. The device sends this information to the server, which updates the donation status in real time and manages the donated water until it reaches shelter A.

[1446] Examples of prompts to input to a generative AI model include:

[1447] "Please tell me the procedure for logistics centers to use smartphones to check demand forecasts in disaster-stricken areas in real time and allocate supplies optimally."

[1448] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1449] Step 1: Data collection

[1450] The server collects past disaster data from government and relief organization databases, and also obtains real-time data from disaster-stricken areas using sensors and drones. The server collects information such as the number of people in the affected area, a list of necessary supplies, and the extent of damage.

[1451] Input: Past disaster data, real-time data

[1452] Output: Collected data

[1453] Specific operation: The server uses APIs to retrieve data from the database, collect real-time data from each sensor and drone, and store the data.

[1454] Step 2: Data Preprocessing

[1455] The server cleanses the collected data, corrects outliers, and fills in missing values, thereby ensuring data quality.

[1456] Input: Collected data

[1457] Output: Cleansed data

[1458] Specific operation: The server uses the Pandas library to cleanse the data, correcting outliers and imputing missing values ​​automatically.

[1459] Step 3: Model training

[1460] The server will then use the cleansed data to train machine learning models that can predict the specific supply and funding needs of each shelter.

[1461] Input: Cleansed data

[1462] Output: A trained predictive model

[1463] How it works: The server uses the Scikit-learn library to train a machine learning model and generate a predictive model based on past data.

[1464] Step 4: Demand forecast

[1465] When a new disaster occurs, the server uses a real-time predictive model to estimate how much supplies will be needed at each evacuation center.

[1466] Input: Trained predictive model, real-time data

[1467] Output: Forecasted demand

[1468] Specific operation: The server inputs real-time data into a predictive model to estimate the demand for supplies and funds for each evacuation shelter.

[1469] Step 5: Resource allocation

[1470] The server formulates a specific resource allocation plan based on the predicted demand and works with logistics companies to deliver supplies.

[1471] Input: Forecasted demand

[1472] Output: Resource allocation plan

[1473] Specific operation: The server creates an optimal resource allocation plan based on the prediction results and notifies the logistics company of the plan via an API.

[1474] Step 6: Notify users and accept donations

[1475] Users check the information provided by the system on their terminals and donate the necessary supplies. The donation information is then sent to the server.

[1476] Input: Required information for each evacuation shelter

[1477] Output: Donation information

[1478] Specific operation: The application on the device notifies the user of supply shortages at evacuation centers, and when the user completes the donation procedure, the information is sent to the server.

[1479] Step 7: Donation Management and Delivery Tracking

[1480] The server records donation information in a database, updates donation status in real time, and manages and tracks donated goods and funds to ensure their proper delivery.

[1481] Input: Donation information

[1482] Output: Updated donation status, delivery status of donated goods

[1483] Specific operation: The server records donation information in a database, works with logistics companies to track the delivery status of supplies, and provides feedback to users.

[1484] Step 8: Optimized material management

[1485] The logistics center will use smartphones, smart glasses, and logistics robots to manage supplies in real time and optimize their distribution.

[1486] Input: Resource allocation plan, real-time management data

[1487] Output: Optimized resource allocation

[1488] Specific operation: Each device in the logistics center moves and sorts materials in real time based on a resource allocation plan, maintaining optimal conditions.

[1489] 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.

[1490] The present invention relates to a system for realizing more effective relief activities by combining a system for predicting demand for and optimizing allocation of resources during disasters with an emotion engine that recognizes user emotions. The system according to the present invention can be implemented as follows.

[1491] Program operation description:

[1492] 1. Data Collection:

[1493] The server collects past disaster data from a database, including the amount of supplies needed, the number of evacuees, and the extent of damage caused by past disasters.

[1494] The server also collects real-time information on the current state of the affected area, using sensors and drones to obtain data such as the capacity of evacuation centers, current stock of supplies, and the extent of the damage.

[1495] 2. Data preprocessing and model training:

[1496] The server preprocesses the collected data, correcting outliers and filling in missing values, and then feeds the preprocessed data into machine learning algorithms.

[1497] The server uses the preprocessed data to train machine learning models that predict each shelter's future supply and financial needs.

[1498] 3. Demand forecasting:

[1499] When a new disaster occurs, the server inputs disaster area data collected in real time into the predictive model.

[1500] Based on this data, the server predicts the amount of supplies and funds that will be needed at each evacuation shelter and in the affected area.

[1501] 4. Resource Allocation Planning:

[1502] The server creates a resource allocation plan based on the predicted demand, specifically determining how much supplies should be sent from which distribution center to which evacuation shelter.

[1503] The server notifies the distribution plan to the logistics company and instructs them to prepare for delivery.

[1504] 5. User Notification and Donation Acceptance:

[1505] The server notifies the user's device of information about the affected area, and the user can use their device to obtain information about supplies needed to get to an evacuation shelter on foot.

[1506] Users can use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[1507] 6. Emotion recognition:

[1508] The server is equipped with an emotion engine that recognizes the user's emotions, and collects and analyzes the user's emotion data.

[1509] The terminal identifies the user's emotions in real time and presents the user with assistance options according to the emotions.

[1510] 7. Emotion-based donation suggestions:

[1511] The server proposes appropriate donation options to the user based on the user's emotional data provided by the emotion engine. For example, if the server determines that the user is willing to donate a large amount, it will present the user with larger donation options.

[1512] 8. Donation Management and Delivery Tracking:

[1513] The server records donation information in a database, updates donation status in real time, and, if necessary, reallocates donated supplies and funds to areas in need.

[1514] The server tracks the delivery status of donated goods, receives delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination.

[1515] Users can check the results of their donations through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[1516] Examples:

[1517] For example, suppose a large earthquake occurs and devastates parts of Japan. Shelter A can accommodate 1,000 people, while shelter B has 800. Based on past disaster data and real-time collected information, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800. Based on this prediction, the server notifies the logistics company of a specific resource allocation plan and instructs delivery. At the same time, the user can check the water shortage information for shelter A through their device and be presented with appropriate donation options based on their emotional data. Once the user completes their donation, this information is sent from the device to the server, which updates the donation status in real time. The server tracks the delivery status of the donated supplies and confirms that the water has safely arrived at shelter A. The user can then check how their donation has been used through their device.

[1518] By using the above method and system, the present invention is expected to make disaster resource demand forecasts and appropriate allocations quick and effective, thereby improving the efficiency of relief activities. By combining it with an emotion engine, it becomes possible to propose relief based on the user's emotions, realizing the provision of more appropriate relief.

[1519] The processing flow will be explained below.

[1520] Step 1: Data collection

[1521] The server collects disaster data from a database of past disasters, including the amount of supplies and funds needed at each evacuation shelter, the number of evacuees, and the extent of damage caused by past disasters.

[1522] The server also uses sensors and drones to collect real-time data on the current disaster area, providing information such as shelter capacity, current supply inventory, and the extent of the damage.

[1523] Step 2: Data Preprocessing

[1524] The server cleanses the collected data, corrects outliers, and fills in missing values, allowing machine learning algorithms to learn accurately.

[1525] Step 3: Predictive model training

[1526] The server uses machine learning algorithms to train a demand forecasting model based on the pre-processed data, creating a model capable of predicting each shelter's future supply and financial needs.

[1527] Step 4: Demand forecast

[1528] When a new disaster occurs, the server inputs current data from the affected area, collected in real time, into the predictive model.

[1529] Based on this data, the server predicts the specific amount of supplies and funds needed at each evacuation center and disaster area.

[1530] Step 5: Resource Allocation Planning

[1531] The server creates a resource allocation plan based on the predicted demand, specifically determining how much supplies should be sent from which distribution center to which evacuation shelter.

[1532] The server issues delivery instructions to the logistics company based on this plan.

[1533] Step 6: Notify users and accept donations

[1534] The server notifies the user's device of information about the affected area, allowing the user to check on their device what supplies are in short supply at evacuation shelters.

[1535] Users use their devices to make donations to specific shelters and supplies, and once the donation is complete, the device sends this information to the server.

[1536] Step 7: Emotion Recognition

[1537] The server uses an emotion engine that recognizes the user's emotions to collect and analyze the user's emotion data.

[1538] The device analyzes the user's emotions in real time and sends the results to the server.

[1539] Step 8: Emotionally Based Donation Proposals

[1540] The server presents appropriate donation options to the user based on the user's emotional data obtained from the emotion engine, for example, presenting an urgent donation option if the user is emotionally moved.

[1541] Step 9: Donation Management and Delivery Tracking

[1542] The server records user donation information in a database, updates the donation status in real time, and verifies that donated goods and funds are distributed appropriately, as needed.

[1543] The server tracks the delivery of donated goods, obtaining delivery status from logistics providers and managing the process until the goods arrive safely at their destination.

[1544] Users can check how their donations have been used through their devices. For example, they can see that 50 bottles of water have arrived safely at shelter A.

[1545] The above are the specific processing steps of the system of the present invention. This system is expected to make relief activities more efficient by quickly and effectively predicting demand for resources during disasters and allocating them appropriately. By combining it with an emotion engine, it becomes possible to propose relief based on the user's emotions, realizing the provision of more appropriate relief.

[1546] Example 2

[1547] 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."

[1548] During a disaster, the condition of the affected area and the demand for supplies change from moment to moment, so it is necessary to allocate resources quickly and accurately.In addition, it is necessary to realize effective relief activities by making appropriate relief proposals based on the user's emotions and motivation, but current systems are not able to adequately address these issues.

[1549] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for collecting past disaster data and correcting outliers and imputing missing values ​​based on the data; means for training a machine learning model based on the preprocessed data and predicting future supply and financial needs at each evacuation shelter; means for collecting disaster area data in real time when a new disaster occurs and inputting it into the machine learning model; means for formulating a resource allocation plan based on the predicted demand and notifying logistics companies; means for notifying a user's terminal of disaster area information and accepting donations; means for proposing appropriate support options using an emotion engine that recognizes the user's emotions; means for recording donation information in a database, updating the donation status in real time, and tracking the delivery status of donated supplies; and means for the user to check the results of the donation via the terminal. This allows for rapid and accurate resource demand prediction and appropriate allocation during a disaster, enabling effective support activities based on the user's emotions.

[1550] "Past disaster data" refers to information about disasters that have occurred in the past, including data on the demand for supplies, the number of evacuees, and the extent of damage.

[1551] "Outlier correction" refers to the process of detecting and correcting statistically abnormal values ​​in acquired data.

[1552] "Missing value imputation" refers to the operation of filling in missing values ​​in a dataset, and generally uses statistical methods such as the mean or median.

[1553] "Preprocessed data" refers to data that has been converted into a form suitable for machine learning models by correcting outliers and filling in missing values.

[1554] A "machine learning model" refers to an algorithm or statistical model that can make predictions or classifications based on past data.

[1555] "Collecting disaster area data in real time" refers to the operation of instantly collecting data on ongoing disasters and sending it to a server.

[1556] A "resource allocation plan" refers to a plan that determines how much supplies to send from which logistics center to which shelter based on predicted demand.

[1557] "Logistics company" refers to a company or organization responsible for transporting goods.

[1558] "User terminal" refers to a computer device such as a smartphone or tablet held by a user, and is a means for communicating with a server.

[1559] "Emotion engine" refers to a system for recognizing and analyzing a user's emotional state.

[1560] "Support options" refers to the support methods and types of donations suggested to users.

[1561] "Donation Information" refers to the details of a user's donation (donation recipient, donated items, amount, date and time, etc.).

[1562] A "database" refers to a system for systematically storing and managing data.

[1563] "Donation results" refers to information about how the donation made by the user was used and what results it produced.

[1564] The present invention relates to a system for optimizing resource demand forecasting and allocation in the event of a disaster. This system is mainly composed of a server, terminals, and users.

[1565] First, the server collects past disaster data from a database. The database stores data on past disasters, such as the demand for supplies, the number of evacuees, and the extent of damage. Next, the server uses sensors and drones to collect real-time information on the current situation in the disaster-stricken area. Specifically, it obtains information such as the capacity of evacuation shelters, the current stock of supplies, and the extent of damage.

[1566] The collected data is preprocessed by the server. During the preprocessing process, outliers are corrected, missing values ​​are imputed, and the data is normalized to make it suitable for the machine learning model. The server then trains the machine learning model based on the preprocessed data. This creates a model that predicts the future supply and funding needs of each shelter.

[1567] When a new disaster occurs, the server inputs real-time data on the affected area into the trained prediction model. Based on this data, it predicts the amount of supplies and funds needed at each evacuation shelter and in the affected area. Based on the predicted demand, the server formulates a resource allocation plan, specifically determining how much supplies to send from which logistics center to which evacuation shelter. The server then notifies the distribution plan to logistics companies and issues delivery instructions.

[1568] Users can receive information about disaster areas through their devices and make donations to specific shelters or supplies. Once a donation is completed, this information is sent from the device to a server, which updates the donation status in real time.

[1569] Furthermore, the server is equipped with an emotion engine that recognizes the user's emotions. The device identifies the user's emotions in real time from their facial expressions and voice data, and presents the user with support options that correspond to those emotions. The emotion engine analyzes the user's emotional data and suggests appropriate donation options. For example, if it is inferred that the user is willing to make a large donation, the user will be presented with larger support options.

[1570] Donation information is recorded in a database, and the server updates the donation status in real time. If necessary, donated goods or funds are redistributed to areas where they are in short supply. The server tracks the delivery status of donated goods, obtains delivery status from logistics companies, and manages the process until the donated goods arrive safely at their destination. Users can check the results of their donations on their own devices.

[1571] Specific examples

[1572] For example, let's assume that a large-scale earthquake occurs and damages parts of Japan. Shelter A has a capacity of 1,000 people, and shelter B has 800 people. Based on past disaster data and information collected in real time, the server predicts that shelter A will need 1,000 2-liter bottles of water, and shelter B will need 800. Based on this prediction, the server notifies the logistics company of a resource allocation plan and instructs delivery. At the same time, the user checks water shortage information for shelter A through their device and is presented with the optimal donation option based on their emotional data. Once the user completes their donation, this information is sent from the device to the server, which updates the donation status in real time. The server tracks the delivery status of the donated supplies and confirms whether the water has arrived safely at shelter A. The user can then check how their donation has been used through their device.

[1573] Prompt Sentence Examples

[1574] "In an area affected by a large earthquake, 1,000 people are evacuated to shelter A and 800 people are evacuated to shelter B. Please predict the supplies needed at each shelter based on past data and real-time information, and issue instructions to optimize relief activities."

[1575] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1576] Step 1: Data collection

[1577] The server collects past disaster data from a database. Specifically, it uses SQL queries to obtain data such as the demand for supplies, the number of evacuees, and the extent of damage during past disasters. It also uses sensors and drones to collect real-time information on the current situation in disaster-stricken areas. Data from the sensors, such as the number of people accommodated, stock of supplies, and the extent of damage, is sent to the server.

[1578] Input: Past disaster database, sensors, drones

[1579] Output: Collected disaster data

[1580] Step 2: Data Preprocessing

[1581] The server preprocesses the collected data, correcting outliers and imputing missing values, and normalizing the data. Missing values ​​are imputed with the mean or median, and outliers are corrected using statistical methods. The data is then scaled and converted into a format suitable for machine learning models.

[1582] Input: Collected disaster data

[1583] Output: Preprocessed data

[1584] Step 3: Model training

[1585] The server trains a machine learning model using the preprocessed data. It splits the dataset into a training set and a test set, trains a regression model or deep learning model using the training set, evaluates the model's performance on the test set, and tunes hyperparameters as needed.

[1586] Input: Preprocessed data

[1587] Output: A trained machine learning model

[1588] Step 4: Demand forecast

[1589] When a new disaster occurs, the server inputs real-time data on the affected area into the trained prediction model, which then uses the model to predict the amount of supplies and funds needed at each evacuation center and in the affected area.

[1590] Input: Real-time disaster area data, trained machine learning model

[1591] Output: Forecasted demand for supplies and funds for each shelter

[1592] Step 5: Resource Allocation Planning

[1593] The server creates a resource allocation plan based on predicted demand. It uses an optimization algorithm to determine how much supplies should be sent from which distribution center to which shelter. It then notifies the distribution company of this plan via an API, instructing them to prepare for delivery.

[1594] Input: Forecasted demand data

[1595] Output: Resource allocation plan, notification to logistics companies

[1596] Step 6: Notify users and accept donations

[1597] The server notifies users of information about the affected areas via their devices. Users can then donate to specific shelters or supplies via their devices. Once a donation is complete, the information is sent from the device to the server, and the donation status is updated in real time.

[1598] Input: Disaster area information, user donation data

[1599] Output: User notification, donation acceptance information

[1600] Step 7: Emotion Recognition

[1601] The server uses an emotion engine to recognize the user's emotions. The device collects the user's facial expressions and voice data and analyzes them using the emotion engine. As a result, assistance options based on the user's emotions are presented.

[1602] Input: User's facial expression and voice data

[1603] Output: Analyzed emotion data, support options

[1604] Step 8: Emotionally Based Donation Proposals

[1605] The server proposes appropriate donation options to the user based on the analysis results of the emotion engine. If the server determines that the user is willing to donate a large amount, it presents the option of large-scale support.

[1606] Input: Parsed emotion data

[1607] Output: Donation proposal

[1608] Step 9: Donation Management and Delivery Tracking

[1609] The server records users' donation information in a database and updates the donation status in real time. It tracks the delivery status of donated goods and funds and obtains delivery status from logistics companies, managing the donated goods until they arrive safely at their destination. Users can check the results of their donations through their devices.

[1610] Input: Donation information, delivery status from logistics company

[1611] Output: Donation management data, delivery tracking information, user notifications

[1612] (Application example 2)

[1613] 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."

[1614] Conventional resource demand forecasting and allocation systems during disasters do not consider the psychological and emotional state of disaster victims when proposing assistance, making it difficult to provide effective assistance. Furthermore, assistance proposals and effective resource allocations are carried out in stages, making it difficult to respond quickly. The present invention aims to solve these problems by combining assistance proposals that consider users' emotions with real-time demand forecasting.

[1615] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1616] In this invention, the server includes means for predicting resource demand during a disaster, means for allocating resources based on the predicted demand, and means for collecting information on the status of the disaster area in real time, thereby enabling means for providing information to users and accepting donations, means for managing and tracking the delivery of donated resources, and means for identifying user emotions in real time and generating support proposals based on those emotions.

[1617] A "means for predicting resource demand during a disaster" is a device or program that predicts the amount of supplies and resources that will be needed when a disaster occurs.

[1618] The "means for allocating resources based on the predicted demand" is a device or program that efficiently allocates materials and resources to each location based on the predicted resource demand.

[1619] "Means for collecting information on the status of disaster-affected areas in real time" refers to devices or systems that use sensors, drones, APIs, etc. to instantly obtain the current status of areas affected by disasters.

[1620] The "means for providing information and accepting donations" refers to an interface or application that provides users with the necessary information and accepts donations from users.

[1621] "Means for managing and tracking the delivery of donated resources" refers to a system that manages the delivery of donated goods and funds to ensure they are delivered to the designated locations and tracks the delivery status in real time.

[1622] The "means for identifying a user's emotions in real time and generating assistance suggestions based on those emotions" refers to a device or system that analyzes a user's emotions in real time and suggests appropriate assistance options based on the results.

[1623] "Means of collecting data on past disasters and training a predictive model based on that data" refers to a system that collects data on past disasters and uses that data to train a demand prediction model for future disasters using a machine learning algorithm.

[1624] "Means for using a predictive model to predict the amount of supplies and funds required for each evacuation shelter" refers to a device or system that uses a trained predictive model to specifically calculate the amount of supplies and funds required for each evacuation shelter.

[1625] The present invention provides a system that combines emotion analysis with a resource demand forecasting and allocation system in the event of a disaster, and proposes assistance based on the user's emotions. Specific embodiments of the present invention are described below.

[1626] System Configuration

[1627] The system consists of the following major components:

[1628] 1. Server

[1629] 2. Terminal

[1630] 3. Users

[1631] The role and specific functions of each component

[1632] server

[1633] The server has several options:

[1634] Data collection method: The server collects past disaster data and real-time local data using sensors, drones, and APIs. For example, it uses Google Maps API to obtain information on disaster areas.

[1635] Prediction model training method: The server trains a prediction model using a machine learning algorithm (e.g., Scikit-learn's LinearRegression) based on the collected past disaster data.

[1636] Demand forecasting tools: Using trained predictive models, we forecast the amount of supplies and funding needed at each shelter.

[1637] Sentiment analysis method: Emotions are identified in real time using TensorFlow and OpenCV based on data (images and audio) sent by the user.

[1638] Support proposal generation means: Based on the results of sentiment analysis, appropriate donation options and support plans are proposed to the user.

[1639] Donation Management and Tracking: Manage donations and track delivery status in real time using a MySQL database.

[1640] Terminal

[1641] The terminal has the following functions:

[1642] Information provision function: Provides users with information on the status of the disaster area and the necessary support.

[1643] Donation acceptance feature: Users can make donations to specific shelters or supplies.

[1644] Emotional data transmission function: Collects user emotional data using the smartphone's camera and microphone and transmits it to the server.

[1645] User

[1646] The user does the following:

[1647] Information reception: Receive information about the situation in the disaster area and the need for assistance.

[1648] Donate: Make a donation through your device.

[1649] Emotion data provision: Send your own emotional data to the server via your device.

[1650] Example

[1651] For example, suppose a large earthquake occurs and some areas are affected. Based on past earthquake data and real-time data, the server predicts that shelter A will need 1,000 bottles of water and shelter B will need 800 bottles of water. A user checks the water shortage information for shelter A on their device, senses the urgency, and makes a donation. Once the donation is complete, the server records the donation information in a database and tracks the delivery status.

[1652] Prompt Sentence Examples

[1653] "Designing a disaster food delivery application that optimizes donations through emotion recognition. Features include real-time demand forecasting, user emotion analysis, and donation suggestions. For example, it includes a function to estimate the food needs of shelter A and provide donation options based on the user's emotion."

[1654] It is expected that this system will solve the problems faced by conventional disaster response systems and enable efficient and effective support activities.

[1655] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1656] Step 1: Data collection

[1657] The server collects past disaster data from a database and uses sensors, drones, and APIs (e.g., Google Maps API) to obtain real-time local data (such as shelter capacity, supply inventory, and damage extent).

[1658] Input: Past disaster data, local real-time data

[1659] Data processing: Correction of outliers, completion of missing values

[1660] Output: Preprocessed dataset

[1661] Step 2: Model training

[1662] The server trains a predictive model using a machine learning algorithm (Scikit-learn's LinearRegression) based on the preprocessed data.

[1663] Input: Preprocessed dataset

[1664] Data Computing: Model Training with Machine Learning

[1665] Output: A trained predictive model

[1666] Step 3: Demand forecast

[1667] When a new disaster occurs, the server inputs the collected real-time data into a trained predictive model to predict the amount of supplies and funds needed at each evacuation center.

[1668] Input: Real-time data, trained predictive model

[1669] Data calculation: Running demand forecasting algorithms

[1670] Output: The amount of supplies and funds needed for each shelter

[1671] Step 4: Provide information

[1672] The terminal notifies the user of information about the disaster area provided by the server, and the user can check the accommodation situation at evacuation centers and the shortage of supplies.

[1673] Input: Disaster area information from the server

[1674] Data processing: Convert to notification format

[1675] Output: Information to the user

[1676] Step 5: Accept donations

[1677] Users make donations to specific shelters and supplies through their devices, which then send the donation information to the server.

[1678] Input: User donation information

[1679] Data processing: recording and transmitting donation information

[1680] Output: Send donation information to the server

[1681] Step 6: Collect emotional data

[1682] The device uses the smartphone's camera and microphone to collect the user's emotional data and transmits it to a server.

[1683] Input: User voice and image data

[1684] Data Computing: Identifying emotions through emotion analysis models

[1685] Output: Sending the identified emotion data to the server

[1686] Step 7: Generate support proposals

[1687] The server generates appropriate donation options and support plans for the user based on the results of the sentiment analysis.

[1688] Input: User emotion data

[1689] Data Computing: Implementing an Emotion-Based Donation Suggestion Algorithm

[1690] Output: Generates donation options and support plans for the user

[1691] Step 8: Donation Management and Tracking

[1692] The server manages and tracks in real time how much donated supplies and funds have been delivered to which shelters, using a MySQL database to record donations and get status updates from the distribution center.

[1693] Input: Delivery status from the distribution center, donation information

[1694] Data processing: Merging donation information and delivery status

[1695] Output: Real-time donation management status

[1696] Through these processing steps, the system for predicting and allocating resource demand during a disaster can propose assistance based on the user's emotions, enabling efficient and effective assistance activities.

[1697] 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.

[1698] 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.

[1699] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1700] 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.

[1701] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1702] 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.

[1703] 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).

[1704] 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.

[1705] 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."

[1706] 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.

[1707] 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).

[1708] 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.

[1709] 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.

[1710] 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.

[1711] 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.

[1712] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1713] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1714] As an example of a syst...

Claims

1. a means of forecasting resource needs during disasters; means for allocating resources based on the predicted demand; A means of collecting information on the status of the affected areas in real time, a means for providing information to users and accepting donations; A means to manage and track the delivery of donated resources; A system including:

2. The system according to claim 1 , further comprising means for collecting past disaster data and training the prediction model based on the data.

3. The system of claim 1 further comprising means for forecasting material and financial needs for each evacuation shelter using a forecasting model.

Images