Method and system for providing food safety management services based on artificial intelligence algorithms

Abstract

A system for providing food safety management services based on an artificial intelligence algorithm is disclosed. A system according to one embodiment of the present disclosure includes: a plurality of types of sensors; a first terminal device used by a first worker; a second terminal device used by a manager; and a management server. The management server acquires sensing data related to a first food preparation room from the plurality of types of sensors; and acquires first information regarding the first food preparation room input by the first worker from the first terminal device. Based on the fact that a value representing the difference between the sensing data and the first information is less than a first threshold value, the management server acquires second information regarding a food menu prepared in the first food preparation room input by the first worker from the first terminal device; inputs the first information and the second information into a first artificial intelligence (AI) model to acquire a total evaluation score for the food menu; and, based on the fact that the total evaluation score exceeds a second threshold value, can generate a report related to the food menu.

Description

Technology Field

[0001] The present disclosure relates to food service safety management technology, and more specifically, to a method and system for providing food service safety management services based on an artificial intelligence algorithm. Background Technology

[0003] Recently, food poisoning incidents continue to occur in various food service facilities, and issues such as meals being served in insufficient portions or improperly prepared food are emerging as persistent social concerns. To prevent this, regulations require supervisory inspections of home-based daycare centers; however, simply increasing the frequency of physical visits is far from sufficient to resolve these complex problems.

[0004] With the recent advancement of artificial intelligence technology, there have been increasing attempts to incorporate AI into food service safety management systems; however, an AI-based food service safety management system that meets expectations has not yet been developed. Prior art literature

[0006] Korean Registered Patent No. 10-2189984 (December 7, 2020) The problem to be solved

[0007] The purpose of the present disclosure is to provide a method and system for providing food safety management services based on an artificial intelligence algorithm.

[0008] The problems that this disclosure aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0010] A system for providing a food safety management service based on an artificial intelligence algorithm according to one embodiment of the present disclosure comprises: a plurality of types of sensors; a first terminal device used by a first worker; a second terminal device used by a manager; and a management server, wherein the management server: acquires sensing data related to a first food preparation room from the plurality of types of sensors; and acquires first information regarding the first food preparation room input by the first worker from the first terminal device, and based on the fact that a value indicating the difference between the sensing data and the first information is less than a first threshold value, the management server: acquires second information regarding a food menu prepared in the first food preparation room input by the first worker from the first terminal device; inputs the first information and the second information into a first artificial intelligence (AI) model to acquire a total evaluation score for the food menu; and, based on the fact that the total evaluation score exceeds a second threshold value, can generate a report related to the food menu.

[0011] And, the plurality of types of sensors include an image sensor, a humidity sensor, and a temperature sensor, and the sensing data includes humidity within the first food preparation room measured through the humidity sensor, temperature within the first food preparation room measured through the temperature sensor, and a plurality of images within the first food preparation room through the image sensor, and the first information may include temperature and humidity within the first food preparation room measured by the first operator and a plurality of images within the first food preparation room acquired by the first terminal device.

[0012] And, the first AI model can be trained to: obtain a first intermediate value by applying a first weight to a nutritional balance value related to the meal menu included in the second information; obtain a second intermediate value by applying a second weight to a difference between i) a predefined temperature and humidity for the meal menu and ii) a temperature and humidity within the first meal preparation room included in the first information; and calculate the total evaluation score based on the first intermediate value and the second intermediate value.

[0013] And, based on the difference between the sensing data and the first information being greater than or equal to the first threshold, a message requesting additional inspection to the second terminal device is transmitted by the management server; and based on the reception of the third information regarding the first food preparation room entered by the manager from the second terminal device, the second information and the third information can be input into the first AI model to obtain a total evaluation score for the food menu.

[0014] And, the management server may impose a penalty on the first worker based on the difference between the first information and the third information. Effects of the invention

[0016] By various embodiments of the present disclosure, a method and system for providing food safety management services based on artificial intelligence algorithms may be provided.

[0017] By various embodiments of the present disclosure, as the food safety management system is controlled more efficiently, errors and wasted time resulting from manual entry of food-related record sheets can be prevented.

[0018] The effects of the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below. Brief explanation of the drawing

[0020] FIG. 1 is a drawing for explaining a method and system for providing food safety management services based on an artificial intelligence algorithm according to one embodiment of the present disclosure. FIG. 2 is a diagram illustrating the configuration of a management server that performs a method of providing a food safety management service based on an artificial intelligence algorithm according to one embodiment of the present disclosure. FIG. 3 is a diagram illustrating a method for providing a food safety management service based on an artificial intelligence algorithm according to one embodiment of the present disclosure. Specific details for implementing the invention

[0021] The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure is complete and to fully inform those skilled in the art of the scope of the present disclosure, and the present disclosure is defined only by the scope of the claims.

[0022] The terms used herein are for describing the embodiments and are not intended to limit the disclosure. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used herein, "comprises" and / or "comprising" do not exclude the presence or addition of one or more other components in addition to the components mentioned.

[0023] Throughout the specification, the same reference numerals refer to the same components, and "and / or" includes each of the mentioned components and all combinations of one or more thereof. Although terms such as "first," "second," etc., are used to describe various components, they are not limited by these terms. These terms are used merely to distinguish one component from another. Accordingly, the first component mentioned below may be the second component within the technical scope of this disclosure.

[0024] Unless otherwise defined, all terms used herein (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which this disclosure pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0025] Spatially relative terms such as "below," "beneath," "lower," "above," and "upper" may be used to easily describe the relationship between one component and another, as illustrated in the drawings. Spatially relative terms should be understood as encompassing the different directions of the components during use or operation, in addition to the directions depicted in the drawings.

[0026] For example, if a component depicted in a drawing is inverted, a component described as being "below" or "beneath" another component may be placed "above" the other component. Therefore, the exemplary term "below" may encompass both the downward and upward directions. Components may also be oriented in other directions, and accordingly, spatially relative terms may be interpreted according to the orientation.

[0027] In describing the present disclosure, an artificial intelligence model may be composed of a plurality of neural network layers. Each of the plurality of neural network layers has a plurality of weight values ​​and performs neural network operations through operations between the results of operations of a previous layer and the plurality of weights. The plurality of weights possessed by the plurality of neural network layers may be optimized by the learning results of the artificial intelligence model. For example, the plurality of weights may be updated so that the loss value or cost value obtained from the artificial intelligence model during the learning process is reduced or minimized. The artificial neural network may include a Deep Neural Network (DNN), and examples include, but are not limited to, Convolutional Neural Networks (CNN), Deep Neural Networks (DNN), Recurrent Neural Networks (RNN), Restricted Boltzmann Machines (RBM), Deep Belief Networks (DBN), Bidirectional Recurrent Deep Neural Networks (BRDNN), or Deep Q-Networks.

[0028] Furthermore, blockchain is a general term for data tamper-proof technology based on distributed computing. In other words, blockchain refers to a distributed ledger technology in which transaction data authenticated by the consensus of P2P-connected nodes is stored in a distributed manner in blocks, and each block is linked together in a chain. With the recent advancement of blockchain-related technologies, there are attempts to solve problems in various fields by applying distributed ledger technology to them.

[0029] In describing the present disclosure, a block refers to a bundle of valid information and may include a block hash value serving as a block identifier, a previous block hash value, a Merkle root, and transaction information. Additionally, a hash function may refer to a function that maps data of arbitrary length to data of fixed length. For example, a hash function may be used to generate a hash value of a fixed length regardless of the size of the input data. For example, if the hash function is SHA-256, the hash value may be 256 bits.

[0030] The following is a diagram illustrating a method and system for providing food safety management services based on an artificial intelligence algorithm.

[0031] FIG. 1 is a drawing for explaining a method and system for providing food safety management services based on an artificial intelligence algorithm according to one embodiment of the present disclosure.

[0032] As illustrated in FIG. 1, a system (1000) providing a food safety management service based on an artificial intelligence algorithm may include a management server (100), a first terminal device (200) used by a first worker, a second terminal device (300) used by a manager, and a plurality of sensors (400-1, 400-2, ..., 400-N) (N is a natural number greater than or equal to 2).

[0033] In describing the present disclosure, a terminal device (e.g., a first terminal device (200), a second terminal device (300), etc.) may include at least one of a smartphone, a tablet PC, a desktop, a laptop, or a server device. However, this is merely one embodiment, and the terminal device may be implemented as a device of another type.

[0034] Additionally, the management server (100) can provide food safety management services based on information received from each of the first terminal device (200) used by the first worker, the second terminal device (300) used by the manager, and a plurality of sensors (400-1, 400-2, ..., 400-N). That is, the management server (100) can collectively refer to a server that manages and controls an application that provides food safety management services based on an artificial intelligence algorithm.

[0035] FIG. 1 illustrates a case in which one first worker and one manager are included on the system (1000), but is not limited thereto. The system (1000) may include one or more terminal devices used by each of the first workers and / or managers.

[0036] FIG. 2 is a diagram illustrating the configuration of a management server that performs a method of providing a food safety management service based on an artificial intelligence algorithm according to one embodiment of the present disclosure.

[0037] The management server (100) illustrated in FIG. 2 may include a memory (110), a communication module (120), a display (130), an input module (140), and a processor (150). Here, the management server (100) may be implemented as various types of electronic devices, such as a smartphone, a tablet PC, a wearable device, a laptop, or a desktop.

[0038] However, the configuration illustrated in FIG. 2 is an illustrative diagram for implementing embodiments of the present disclosure, and appropriate hardware and software configurations that are obvious to a person skilled in the art may be additionally included in the management server (100).

[0039] The memory (110) can store one or more instructions for the processor (150) to perform various operations. The memory (110) can store data that supports various functions of the management server (100), programs for the operation of the processor (150), and input / output data.

[0040] The memory (110) may include at least one type of storage medium among flash memory type, hard disk type, SSD type (Solid State Disk type), SSD type (Silicon Disk Drive type), multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), RAM (random access memory; RAM), SRAM (static random access memory), ROM (read-only memory; ROM), EEPROM (electrically erasable programmable read-only memory), PROM (programmable read-only memory), magnetic memory, magnetic disk, and optical disk.

[0041] The communication module (120) may include one or more components including a circuit that enables communication with an external device. For example, the communication module (120) may include at least one of a broadcast receiving module, a wired communication module, a wireless communication module, a short-range communication module, and a location information module.

[0042] The display (130) displays (outputs) information processed by the management server (100). For example, the display (130) can display information on the execution screen of an application running on the management server (100), or UI (User Interface) and GUI (Graphic User Interface) information based on such execution screen information.

[0043] The input module (140) refers to a component for authorizing various input data or / and input interactions (e.g., touch, swipe, etc.) to the management server (100).

[0044] The processor (150) can provide a method for creating and managing digital twin content based on an artificial intelligence algorithm by executing one or more instructions stored in memory (110). That is, the processor (150) can control overall operation and function using each component of the management server (100).

[0045] Specifically, the processor (150) may be implemented as a memory that stores data for an algorithm or a program that reproduces the algorithm for controlling the operation of components within the management server (100), and at least one processor that performs the aforementioned operation using the data stored in the memory. In this case, the memory and the processor may each be implemented as separate chips. Alternatively, the memory and the processor may be implemented as a single chip.

[0046] In addition, the processor (150) may control one or more of the components described above in combination to implement various embodiments according to the present disclosure described in FIG. 3 below on the management server (100). That is, the operation of the management server (100) described with reference to FIG. 3 may include operations performed and controlled by the processor (150).

[0047] FIG. 3 is a diagram illustrating a method for providing a food safety management service based on an artificial intelligence algorithm according to one embodiment of the present disclosure.

[0048] The management server can acquire sensing data related to the first food preparation room from multiple types of sensors (S310).

[0049] Here, as described above, a plurality of types of sensors may include an image sensor, a humidity sensor, and a temperature sensor, and the sensors may be mounted on the first food preparation room.

[0050] In addition, the sensing data may include humidity within the first food preparation room measured through a humidity sensor, temperature within the first food preparation room measured through a temperature sensor, and multiple images within the first food preparation room measured through an image sensor.

[0051] And, the management server can obtain first information about the first food preparation room that the first worker (directly) entered from the first terminal device (S320).

[0052] For example, a first operator may directly input first information regarding a first food preparation room onto a first terminal device, and the first terminal device may transmit the input first information to a management server. Here, the first information may include the temperature and humidity within the first food preparation room and a plurality of images within the first food preparation room, measured by the first operator through individual sensor devices (e.g., a thermometer, a hygrometer, a camera device, or / and a first terminal device equipped with such devices).

[0053] In one example of the present disclosure, based on the fact that the difference between the sensing data and the first information is less than the first threshold value, the management server can obtain second information about the food menu prepared in the first food preparation room that the first worker inputs from the first terminal device (S330).

[0054] Specifically, the management server can primarily verify the reliability of the first worker input information by comparing objective sensing data obtained through multiple types of sensors with the first information manually measured by the first worker.

[0055] Here, the fact that the value representing the difference between objective sensing data acquired through multiple types of sensors and the first information manually measured by the first operator is less than the first threshold means that the first information recorded by the first operator substantially matches the actual manufacturing room environment, thereby ensuring the integrity of the data.

[0056] When the reliability of the first information is confirmed in this way, the management server can obtain second information from the first terminal device, including a meal plan, ingredient composition, cooking method, nutritional balance values, etc., for a meal menu that is scheduled to be manufactured or is being manufactured in the first meal manufacturing room (S340).

[0057] For example, a first operator may input second information onto a first terminal device, which includes a meal plan, ingredient composition, cooking method, and nutritional balance values ​​(e.g., ratios between carbohydrates, protein, and fat) for a meal menu that is scheduled to be manufactured or is currently being manufactured in the corresponding first meal manufacturing room. Accordingly, the first terminal device may transmit the second information to a management server.

[0058] The management server can input the first information and the second information into the first AI model to obtain a total evaluation score for the meal menu (S340).

[0059] According to one embodiment of the present disclosure, the first AI model may be configured as a multi-modal architecture based on a GPU (Graphic Processing Unit) or NPU (Neural Processing Unit) to process input heterogeneous data in parallel.

[0060] That is, the first AI model can be trained to comprehensively analyze not only numerical data such as temperature and humidity included in the first information, but also image feature vectors in latent space extracted from multiple images within the first food preparation room.

[0061] Specifically, the first AI model can calculate a first intermediate value by applying a first weight set for the nutritional balance value related to the meal menu included in the second information. In addition, the first AI model can obtain a second intermediate value by calculating the difference between i) a predefined standard temperature and humidity for optimal cooking of the meal menu and ii) the actual temperature and humidity inside the first meal preparation room included in the first information, and applying a second weight to the environmental difference value.

[0062] Subsequently, the first AI model can be trained to calculate the total evaluation score, which quantifies the overall safety and quality status of the food service menu, based on the calculated first intermediate value (i.e., nutritional aspect) and the second intermediate value (i.e., hygiene and cooking environment aspect).

[0063] Afterwards, the management server determines whether the calculated total evaluation score exceeds a second threshold, and can generate a report related to the meal menu based on whether the total evaluation score exceeds the second threshold (S350).

[0064] Here, the fact that the total evaluation score exceeds the second threshold means that the meal menu was safely prepared in an appropriate hygienic environment with a nutritional balance exceeding the standard.

[0065] The generated report may be automatically recorded in the food safety management ledger within the system's database or transmitted in real time to a second terminal device used by an administrator for additional approval or monitoring. For example, a report related to a food menu may include a total evaluation score, the type of food menu, first information, and second information.

[0066] According to another embodiment of the present disclosure, a management server may calculate the difference between acquired objective sensing data and first information manually recorded by a first operator, and determine that the difference is greater than or equal to a preset first threshold.

[0067] This means that the first information entered by the first worker is significantly different from the actual physical environment (sensing data) of the first food preparation room, and may implicitly indicate the possibility that intentional data manipulation occurred to conceal simple mistakes in the input process or poor hygiene conditions.

[0068] In cases where the reliability of the first information is compromised in this manner, the management server may immediately send a notification message requesting additional inspection to the aforementioned second terminal device to fundamentally prevent the problem of "Garbage In, Garbage Out"—where erroneous data is input into the AI ​​model and leads to incorrect evaluations. This message may be displayed as a warning pop-up on a status board monitored by the administrator or pushed to the administrator's mobile device.

[0069] After receiving the notification message, the manager can directly verify the situation through an on-site visit to the relevant first food preparation room or real-time CCTV remote monitoring, and then input third information regarding the objectively verified environment of the first food preparation room through the second terminal device.

[0070] Subsequently, based on receiving the third information input by the manager from the second terminal device, the management server can exclude the unreliable existing first information and input the previously obtained second information regarding the meal menu and the newly verified third information into the first AI model to obtain a total evaluation score for the meal menu.

[0071] Through this, the system of the present disclosure can continuously maintain the integrity and objectivity of the AI ​​model-based food safety management evaluation despite the omission of information or malicious manipulation by the first worker.

[0072] Furthermore, as a systematic follow-up measure to the aforementioned data discrepancy situation, the management server may automatically impose a penalty on the first worker based on the difference between the first information (i.e., the first worker's initial input value) and the third information (i.e., the manager's verified input value).

[0073] For example, the management server can calculate the error rate between the temperature / humidity values ​​included in the first information and the third information, respectively, or quantify the degree of discrepancy by comparing the cosine similarity between the image feature vectors included in both information.

[0074] In this case, the greater the calculated degree of discrepancy, the higher the weight of the penalty score assigned to the corresponding first worker's identification account or attendance evaluation record. By linking the accumulated penalty score to the condition for retaking hygiene and safety group training for food service employees or utilizing it as basic statistical data for the management agency's differential management system, this can raise awareness among on-site first workers and encourage voluntary compliance with safety management regulations.

[0075] For example, if the accumulated penalty score exceeds a threshold, the management server may send a warning message to the first terminal device that includes the schedule for hygiene / safety group training for the first worker's cafeteria staff.

[0076] According to another embodiment of the present disclosure, the first threshold and the second threshold, which are pre-set by the management server, are not fixed constants but may be configured / set to change dynamically based on environmental factors inside and outside the first food preparation room or / and the history of the first worker, etc.

[0077] First, the first threshold value, which determines the allowable error range between the sensing data and the first information, may be set differentially by the management server based on the accumulated penalty history of the first worker who recorded the first information, the inherent error rate due to the aging of the sensor device in the first food preparation room, or / and seasonal factors (e.g., rapid local temperature and humidity changes during the rainy season).

[0078] For example, if the management server determines that there is a high risk of data manipulation or measurement error when a specific first worker's past penalty score exceeds a certain threshold, it may lower the first threshold by a predefined first value (i.e., strictly strengthen the verification criteria).

[0079] Conversely, if the possibility of mechanical error is detected when the calibration cycle of the temperature sensor or humidity sensor is imminent, the management server can prevent unnecessary administrator intervention (sending notification messages) and system overload by temporarily raising the first threshold value by a predefined second value.

[0080] Meanwhile, the second threshold value, which serves as the pass / fail criterion for the total evaluation score calculated by the first AI model, may be dynamically varied based on the real-time food poisoning index of the region received from an external public data server, weather advisories (e.g., heat wave advisory), the sensitivity to spoilage of ingredients constituting the meal menu, or / and the characteristics of the meal recipients (e.g., the proportion of infants in home daycare centers with weak immune systems).

[0081] Specifically, the management server may increase the second threshold value by a predefined third value above the default setting value in cases where the risk of food poisoning increases sharply, such as during the summer, causing the external food poisoning index to rise to a dangerous level, or when the day's meal menu includes ingredients vulnerable to food poisoning (e.g., seafood).

[0082] This involves applying stricter acceptance standards to the system for hygiene and cooking environments based on external conditions and menu characteristics, aiming to proactively prevent potential food safety accidents and maximize the effectiveness of AI-based evaluations.

[0083] Additionally or alternatively, the database stored on the management server (e.g., DB server) may be built on or configured to be linked to a blockchain network to ensure the integrity and irreversibility of the collected food safety management data.

[0084] Specifically, the management server can obtain first vector data by performing embedding on each of the sensing data of the first food preparation room obtained through the steps described above, the first information entered by the first worker, the third information entered after verification by the manager, the total evaluation score calculated through the first AI model, the report on the first food preparation room generated by the system, and the penalty history assigned to the first worker. The management server obtains a first hash value by applying a predefined hash function to the first vector data and can generate a first block based on the obtained first hash value. That is, the first block is a block related to the first food preparation room.

[0085] In addition, the management server can obtain second vector data by performing embedding on each of the following: sensing data from another food preparation room (e.g., a second food preparation room) (e.g., sensing data obtained from multiple types of sensors installed in the second food preparation room), first information regarding the second food preparation room entered by the second worker, third information regarding the second food preparation room verified and entered by the manager, the total evaluation score calculated through the first AI model, the report regarding the second food preparation room generated by the system, and the penalty history assigned to the second worker. The management server can obtain a second hash value by applying a predefined hash function to the second vector data. The management server can generate a second block based on the second hash value. That is, the second block is a block related to the second food preparation room.

[0086] In addition, the management server can construct a blockchain-based food preparation room database that prevents arbitrary manipulation and falsification of data by cryptographically linking the second block to the first block in a manner that includes the unique (hash) value of the first block in the second block.

[0087] Meanwhile, each generated block may include the encrypted hash value of the corresponding data set, a timestamp indicating the exact time when the data was recorded, and the hash value of the previous block.

[0088] The management server utilizes Distributed Ledger Technology (DLT) in the form of a Private Blockchain or a Consortium Blockchain to share the generated blocks among the central center, the support center, and authorized terminal devices (nodes) equipped at each children's meal service center. The shared blocks are mutually verified according to a pre-configured consensus algorithm and then stored by being continuously connected to an existing chain.

[0089] Additionally or alternatively, the management server can process the information included in the above report related to the meal menu of the first meal preparation room and automatically generate audiovisual video content to be provided to meal consumers (e.g., children, parents, or staff).

[0090] As a specific example, the management server can input multidimensional data, such as the total evaluation score included in the report, the nutrition and ingredient types of the meal menu, the kitchen environment measured by the operator (e.g., first information), and information about the meal menu (e.g., second information), into a pre-trained Large Language Model (LLM) in the form of a prompt.

[0091] The above-mentioned large language model can be trained to generate text data (scripts) that comprehensively analyze input parameters and explain the hygienic safety and nutritional excellence of the corresponding meal menu in natural, colloquial sentences tailored to the user's level. Subsequently, the management server can obtain voice data that reflects the contextual nuances of the text data by applying a Text-to-Speech (TTS) algorithm.

[0092] Furthermore, the management server can generate a 3D avatar based on the actual facial image of the first worker or pre-extracted feature point data, and obtain video content in which the avatar takes motions as if actually speaking in accordance with the voice data.

[0093] Specifically, the management server can implement sophisticated lip-sync by analyzing the audio feature vector of the acquired voice data and dynamically controlling the blendshape weights within the avatar's facial rigging model. Furthermore, rendering quality can be improved by merging natural motion gestures based on the avatar's skeletal hierarchy and skinning weights to match the emphasis of the speech content.

[0094] Finally, the management server can transmit the video content generated in this way in real time to terminal devices (e.g., notification applications) used by multiple users (or their guardians) consuming the meal menu in the first food preparation room, or to displays installed at the dining area. Through this, users receive vivid explanations from an avatar modeled after the actual cooking manager, instead of rigid, text-based reports, thereby maximizing transparency and trust in the food service.

[0095] Meanwhile, the disclosed embodiments may be implemented in the form of a recording medium that stores instructions executable by a computer. The instructions may be stored in the form of program code and, when executed by a processor, may generate a program module to perform the operation of the disclosed embodiments. The recording medium may be implemented as a computer-readable recording medium.

[0096] Computer-readable recording media include all types of recording media that store instructions that can be decoded by a computer. Examples include ROM (Read Only Memory), RAM (Random Access Memory), magnetic tape, magnetic disk, flash memory, optical data storage devices, etc.

[0097] As described above, the disclosed embodiments have been explained with reference to the attached drawings. Those skilled in the art will understand that the present disclosure may be practiced in forms different from the disclosed embodiments without changing the technical spirit or essential features of the present disclosure. The disclosed embodiments are illustrative and should not be interpreted restrictively.

Claims

Claim 1 A system for providing food safety management services based on an artificial intelligence algorithm comprises: a plurality of types of sensors; a first terminal device used by a first worker; a second terminal device used by a manager; and a management server, wherein the management server acquires sensing data related to a first food preparation room from the plurality of types of sensors; and acquires first information regarding the first food preparation room input by the first worker from the first terminal device, and based on the fact that a value representing the difference between the sensing data and the first information is less than a first threshold, the management server acquires second information regarding a food menu prepared in the first food preparation room input by the first worker from the first terminal device; and inputs the first information and the second information into a first artificial intelligence (AI) model to acquire a total evaluation score for the food menu; And based on the total evaluation score exceeding a second threshold, a report related to the meal menu is generated, wherein the plurality of types of sensors include an image sensor, a humidity sensor, and a temperature sensor, and the sensing data includes humidity within the first meal preparation room measured through the humidity sensor, temperature within the first meal preparation room measured through the temperature sensor, and a plurality of images within the first meal preparation room through the image sensor, and the first information includes temperature and humidity within the first meal preparation room measured by the first operator and a plurality of images within the first meal preparation room acquired by the first terminal device, and the first AI model: obtains a first intermediate value by applying a first weight to the nutritional balance value related to the meal menu included in the second information; and obtains a second intermediate value by applying a second weight to the difference between i) a predefined temperature and humidity for the meal menu and ii) the temperature and humidity within the first meal preparation room included in the first information;A system that is trained to calculate the total evaluation score based on the first intermediate value and the second intermediate value, and based on the difference between the sensing data and the first information being greater than or equal to the first threshold, a message requesting additional inspection is transmitted by the management server to the second terminal device; and based on receiving third information regarding the first food preparation room entered by the manager from the second terminal device, the second information and the third information are input into the first AI model to obtain the total evaluation score for the food menu, and the management server imposes a penalty on the first worker based on the difference between the first information and the third information. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 delete

Images