Catering management method and system based on supply chain separation and full-link traceability

Abstract

The invention relates to the technical field of catering management, in particular to a catering management method and system based on supply chain separation and full-link traceability. The method comprises the steps of obtaining and analyzing a user order, and determining an order demand; obtaining real-time food material inventory information, and generating structured recipe data through an AI model according to the order demand and the real-time food material inventory information; a food material purchasing instruction and a processing service instruction are generated; acquiring real-time link monitoring data based on the food material purchasing instruction and the processing service instruction; and determining a traceability display item according to an order demand, analyzing the real-time link monitoring data and the structured recipe data according to the traceability display item, and generating display content. And the recipe data is ensured to adapt to real-time change. Corresponding instructions are generated, the independence and the performability of the instructions are ensured, and the efficiency bottleneck caused by link coupling is avoided. And real-time link monitoring data is acquired, and display content is generated, so that the integrity and the dynamism of traceability display are realized, and the transparency and the credibility of a user are enhanced.

Description

Technical Field

[0001] This application relates to the field of catering management technology, and in particular to a catering management method and system based on supply chain separation and full-chain traceability. Background Technology

[0002] In the current catering industry, supply chain management faces multiple challenges, especially in terms of ingredient procurement, processing optimization, and full-chain traceability.

[0003] Traditional restaurant management systems typically use IoT devices to monitor food inventory and processing in a localized manner. However, because they are often limited to a single link, they cannot effectively trace the source, which makes it impossible to achieve refined management across the entire chain. This results in rigid recipe plans, high food waste rates, delayed instruction generation, high error rates, and an inability to achieve end-to-end automation. This not only increases operating costs but also reduces service response speed and user experience. Summary of the Invention

[0004] This application provides a catering management method and system based on supply chain separation and full-link traceability to solve the above problems.

[0005] Firstly, this application provides a catering management method based on supply chain separation and full-link traceability, the method comprising:

[0006] Obtain user orders; analyze the user orders to determine order requirements;

[0007] Obtain real-time ingredient inventory information, and generate structured recipe data using an AI model based on the order requirements and the real-time ingredient inventory information;

[0008] Based on the structured recipe data, generate ingredient procurement instructions and processing service instructions;

[0009] Based on the aforementioned food procurement instructions and processing service instructions, real-time process monitoring data is obtained;

[0010] Based on the order requirements, determine the traceability display items, and based on the traceability display items, parse the real-time process monitoring data and the structured recipe data to generate display content.

[0011] This solution acquires user orders, preventing process interruptions due to missing data. It analyzes user orders to determine order requirements, eliminating uncertainties in order parsing and ensuring these requirements can be directly used to generate structured recipe data and traceability display projects. Real-time ingredient inventory information is obtained, and based on order requirements and real-time inventory information, structured recipe data is generated using an AI model, improving ingredient utilization and processing accuracy, and ensuring recipe data adapts to real-time changes. Based on the structured recipe data, ingredient procurement instructions and processing service instructions are generated, ensuring the independence and executability of these instructions and avoiding efficiency bottlenecks caused by process coupling. Based on these instructions, real-time process monitoring data is acquired, ensuring full-chain monitoring coverage and supporting operational security verification and process tracking. Based on order requirements, traceability display projects are determined, and based on these projects, real-time process monitoring data and structured recipe data are parsed to generate display content, achieving completeness and dynamism in traceability display, enhancing user transparency and trust.

[0012] Optionally, the step of parsing the real-time process monitoring data and the structured recipe data based on the traceability display project to generate display content includes:

[0013] Analyze the structured recipe data to determine the ingredient processing flow;

[0014] Based on the traceability display items, the food processing flow is selected to determine the target display flow;

[0015] Based on the target display process, the real-time monitoring data of the process is analyzed to generate display content.

[0016] This solution analyzes structured recipe data to determine ingredient processing flows, enabling dynamic generation of these flows. This reduces the risk of fragmented traceability content, improves the accuracy and efficiency of the entire traceability process, and supports end-to-end monitoring. Based on the traceability display project, it filters ingredient processing flows, determines the target display flow, reduces data redundancy, increases traceability transparency, enhances dynamic generation capabilities, and improves consumer trust and compliance. Based on the target display flow, it analyzes real-time monitoring data to generate display content, ensuring that the content includes the latest dynamic details and provides end-to-end transparency, thereby increasing trust and reducing the risk of decision-making delays.

[0017] Optionally, the step of generating structured recipe data using an AI model based on the order requirements and the real-time ingredient inventory information includes:

[0018] Obtain historical cooking datasets;

[0019] Analyze the historical cooking dataset to determine the cooking logic of the dishes;

[0020] The cost of purchasing ingredients is determined based on the order requirements and the real-time ingredient inventory information.

[0021] Based on the cooking logic of the dishes, the order requirements, and the real-time ingredient inventory information, structured recipe data is generated through an AI model.

[0022] This solution acquires historical cooking datasets, allowing them to be directly used to determine the cooking logic for dishes and reducing data query latency. It analyzes these datasets to identify the cooking logic for each dish, ensuring that the logic is based on historical best practices and can be directly used in the recipe generation process. Based on order demand and real-time ingredient inventory information, it determines the cost of purchased ingredients, optimizes ingredient procurement quantities, and avoids oversupply or shortages. Based on the cooking logic, order demand, and real-time ingredient inventory information, an AI model generates structured recipe data, improving the accuracy and adaptability of the recipes.

[0023] Optionally, the step of generating structured recipe data using an AI model based on the dish cooking logic, order requirements, and real-time ingredient inventory information includes:

[0024] Analyze the real-time food inventory information to determine the food's shelf life.

[0025] Obtain data on the food storage environment;

[0026] Based on the food's shelf life data and the storage environment data, a real-time spoilage index is determined through a spoilage rate model analysis.

[0027] The freshness level of the ingredients is determined based on the real-time spoilage index.

[0028] The actual cooking requirements will be determined based on the freshness of the ingredients and the order details.

[0029] Based on the cooking logic of the dish and the actual cooking requirements, the cooking time and cooking temperature of the cooking nodes are determined by an AI model.

[0030] The freshness of the ingredients, the cooking time, and the cooking temperature are determined as structured recipe data.

[0031] This solution analyzes real-time ingredient inventory information to determine ingredient expiration dates, providing a real-time time constraint basis for the entire recipe generation process. Based on expiration date data, ingredient freshness is determined, ensuring consistency in ingredient condition and avoiding data conversion overhead. Based on ingredient freshness and order demand, actual cooking needs are determined, optimizing ingredient allocation and reducing spoilage risk. Based on the dish's cooking logic and actual cooking needs, an AI model determines the cooking time and temperature for each cooking node, improving recipe adaptability and the accuracy of processing instructions. Ingredient freshness, cooking time, and cooking temperature are structured into recipe data, enhancing traceability transparency and consumer trust.

[0032] Optionally, generating ingredient procurement instructions and processing service instructions based on the structured recipe data includes:

[0033] Based on the cooking time and the cooking temperature, a processing service instruction is generated;

[0034] Based on the freshness of the ingredients, determine the available ingredient information;

[0035] Based on the order requirements, determine the required ingredient information;

[0036] Based on the available ingredient information and the ingredient demand information, the procurement information is determined;

[0037] Based on the procurement information, a food procurement instruction is generated.

[0038] This solution generates processing service instructions based on cooking time and temperature, improving recipe adaptability and avoiding processing errors caused by temperature or time variations. It determines available ingredients based on freshness, reducing waste from using spoiled ingredients and optimizing inventory utilization. It identifies ingredient demand based on order requirements, preventing response delays caused by coupling and ensuring demand analysis is independent of procurement logic. Based on available and demand information, it determines procurement information, preventing ingredient shortages or surpluses. Finally, it generates ingredient procurement instructions based on procurement information, improving supply chain responsiveness.

[0039] Optionally, the available ingredient information includes the quantity of available ingredients; the ingredient demand information includes the quantity of ingredients required; and generating an ingredient purchase instruction based on the purchase information includes:

[0040] Analyze the freshness of the ingredients to determine the emergency replenishment threshold;

[0041] If the available amount of ingredients is lower than the required amount of ingredients and the freshness of the ingredients reaches the emergency replenishment threshold, then obtain the supplier's historical fulfillment data;

[0042] Based on the real-time corruption index and the supplier's historical performance data, the optimal supplier is determined through a weighted scoring model.

[0043] Based on the optimal supplier, generate a food procurement instruction.

[0044] This solution analyzes ingredient freshness to determine emergency replenishment thresholds, avoiding the use of unusable ingredients and reducing waste and safety hazards caused by spoilage. If the available ingredient quantity is lower than the required quantity and the ingredient freshness reaches the emergency replenishment threshold, historical supplier performance data is retrieved to ensure efficient resource utilization and avoid processing supplier data in non-emergency scenarios. Based on real-time spoilage index and historical supplier performance data, a weighted scoring model is used to determine the optimal supplier, reducing order execution delays and improving procurement success rates. Based on the optimal supplier, ingredient procurement orders are generated, improving procurement efficiency and ingredient quality, and reducing the risk of supply chain disruptions.

[0045] Optionally, the method further includes:

[0046] Analyze the real-time monitoring data to determine the operator information; the operator information includes the operator's facial data.

[0047] Based on the processing service instruction, analyze the operator information to determine whether the operator is a member of the preset whitelist;

[0048] If the user is not a member of the preset whitelist, the platform's review data will be retrieved based on the operator information.

[0049] The platform review data is matched with the operator information. If the operator information matches any item in the platform review data, the platform review data that matches the operator information is reviewed based on the facial data to determine whether to lock the current processing equipment and suspend instruction execution.

[0050] This solution analyzes real-time monitoring data to identify operator information, including facial data, ensuring accurate identification of operators in the real-time processing environment. Based on processing service instructions, the system analyzes operator information to determine if the operator is on a pre-defined whitelist, enabling rapid screening of authorized personnel and preventing unverified individuals from directly operating processing equipment, thus enhancing operational security. If the operator is not on the whitelist, the system retrieves platform verification data based on the operator information to expand the verification scope, covering potentially authorized but not yet verified operators. This ensures that some operators awaiting verification can still be processed in emergencies, reducing the possibility of operational interruptions. The system matches platform verification data with operator information. If any item matches, the system verifies the matching data based on facial data to determine whether to lock the current processing equipment and suspend instruction execution. This ensures that only authorized or verified personnel can operate the equipment, effectively preventing unauthorized operations and improving the security of the processing process.

[0051] Optionally, the method further includes:

[0052] Based on the processing service instruction, a preset tracking target is determined;

[0053] Analyze the processing service instructions to determine the tracking timestamp;

[0054] Based on the tracking timestamp, analyze real-time process monitoring data to determine the recommended target profile;

[0055] The recommended target contour is matched with the preset tracking target. If they do not match, the command execution is paused and an alarm is issued.

[0056] This solution binds processing service instructions to preset tracking targets (such as the outline of an oil can), enabling real-time monitoring of instruction execution. If the tracking target is not found, it indicates the instruction was not executed correctly, thus improving the transparency and traceability of the production process. Analyzing real-time monitoring data recommends the most likely outline as the target; if the match is below a set threshold, it is considered a violation, automatically pausing subsequent steps and issuing an alarm for timely intervention. This prevents errors from escalating, ensures the production process meets expectations, and ultimately improves overall production efficiency and product quality.

[0057] Optionally, the step of determining the optimal supplier based on the real-time corruption index and the supplier's historical performance data through a weighted scoring model includes:

[0058] Analyze the supplier's historical performance data to determine the supplier's commitment record;

[0059] Obtain the supplier bidding information at the current moment; parse the supplier bidding information to determine the supply logistics characteristics;

[0060] The maximum delay time is determined based on the real-time corruption index.

[0061] The optimal supplier is determined by weighted scoring model analysis based on the supplier's commitment record, supply logistics characteristics, and longest delay time.

[0062] This solution analyzes historical supplier performance data to determine supplier compliance records, ensuring supplier selection is based on objective historical performance. It acquires current supplier bidding information to ensure data timeliness and avoid decision-making biases caused by outdated information. The solution analyzes supplier bidding information to determine supply logistics characteristics, converting logistics details into comparable quantifiable feature values ​​to facilitate analysis of supplier logistics capabilities. Based on a real-time spoilage index, it determines the longest delay time, reflecting the urgency of food spoilage and providing a time constraint for supplier selection, ensuring that replacement ingredients arrive before spoilage. Based on supplier compliance records, supply logistics characteristics, and the longest delay time, a weighted scoring model is used to determine the optimal supplier, enabling rapid and accurate selection of the best supplier and optimizing emergency replenishment decisions.

[0063] Secondly, this application provides a catering management system based on supply chain separation and full-link traceability, the system comprising:

[0064] The order analysis module is used to obtain user orders; analyze the user orders, and determine order requirements;

[0065] The recipe generation module is used to obtain real-time ingredient inventory information and generate structured recipe data through an AI model based on the order requirements and the real-time ingredient inventory information.

[0066] The instruction generation module is used to generate ingredient procurement instructions and processing service instructions based on the structured recipe data;

[0067] The data acquisition module is used to acquire real-time process monitoring data based on the food procurement instructions and processing service instructions.

[0068] The traceability display module is used to determine the traceability display items according to the order requirements, and to parse the real-time process monitoring data and the structured recipe data according to the traceability display items to generate display content. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0070] Figure 1 This is a schematic diagram illustrating an application scenario provided in one embodiment of this application;

[0071] Figure 2 A flowchart illustrating a catering management method based on supply chain separation and full-link traceability, provided as an embodiment of this application;

[0072] Figure 3 This is a schematic diagram of a catering management system based on supply chain separation and full-link traceability, provided as an embodiment of this application. Detailed Implementation

[0073] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0074] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0075] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0076] Traditional restaurant management systems typically use IoT devices to monitor food inventory and processing in a localized manner. However, because they are often limited to a single link, they cannot effectively trace the source, which makes it impossible to achieve refined management across the entire chain. This results in rigid recipe plans, high food waste rates, delayed instruction generation, high error rates, and an inability to achieve end-to-end automation. This not only increases operating costs but also reduces service response speed and user experience.

[0077] Based on this, this application provides a catering management method and system based on supply chain separation and full-link traceability. It acquires user orders, avoiding process interruptions due to data loss. User orders are analyzed to determine order requirements, eliminating uncertainty in order parsing and ensuring that order requirements can be directly used to generate structured recipe data and traceability display items. Real-time ingredient inventory information is acquired, and based on order requirements and real-time ingredient inventory information, structured recipe data is generated through an AI model, improving ingredient utilization and processing accuracy, and ensuring that recipe data adapts to real-time changes. Based on the structured recipe data, ingredient procurement instructions and processing service instructions are generated, ensuring the independence and executability of instructions and avoiding efficiency bottlenecks caused by process coupling. Based on the ingredient procurement instructions and processing service instructions, real-time process monitoring data is acquired, ensuring full-link monitoring coverage and supporting operational safety verification and process tracking. Based on order requirements, traceability display items are determined, and based on the traceability display items, real-time process monitoring data and structured recipe data are parsed to generate display content, achieving completeness and dynamism in traceability display, and enhancing user transparency and trust.

[0078] Figure 1 This is a schematic diagram illustrating an application scenario provided by this application, showing the application of the method provided in this application when performing supply chain separation and full-link traceability in catering management.

[0079] Specifically, the method provided in this application can be applied to any server, where the server interacts with user devices and an inventory management system to obtain user orders. The user orders are analyzed to determine order requirements. Real-time ingredient inventory information is obtained through the inventory management system. Based on order requirements and real-time ingredient inventory information, structured recipe data is generated using an AI model. Based on the structured recipe data, ingredient procurement instructions and processing service instructions are generated, ensuring the independence and executability of the instructions and avoiding efficiency bottlenecks caused by process coupling. Based on the ingredient procurement instructions and processing service instructions, real-time process monitoring data is obtained to ensure full-chain monitoring coverage and support operational security verification and process tracking. Based on order requirements, traceability display items are determined. Based on the traceability display items, real-time process monitoring data and structured recipe data are parsed to generate display content, achieving completeness and dynamism in traceability display and enhancing user transparency and trust.

[0080] For specific implementation details, please refer to the following examples.

[0081] Figure 2 This is a flowchart illustrating a catering management method based on supply chain separation and full-link traceability, provided as an embodiment of this application. The method of this embodiment can be applied to servers in the above scenarios. For example... Figure 2 As shown, the method includes:

[0082] S201. Obtain user orders; analyze user orders to determine order requirements;

[0083] User orders can be data related to a user's food service request, including the name of the dish, quantity, and delivery time.

[0084] Order requirements can be structured requirements parameters extracted from user orders, including the types and quantities of ingredients required, time constraints, and allergen information.

[0085] Specifically, user devices connect to the server via an API interface, retrieving user orders in real time. The system parses user orders, identifying the menu list (the set of dish names requested in the user order) and the quantity field (the number of servings of a single dish requested in the user order); it extracts additional requirements (user's personalized requirements for dishes (taste, allergen exclusion, such as obtaining low-sodium information from the remarks field), and summarizes these to generate order requirements.

[0086] S202. Obtain real-time ingredient inventory information, and generate structured recipe data through an AI model based on order requirements and real-time ingredient inventory information;

[0087] Real-time food inventory information can be dynamically updated data on the status of food inventory, including information such as food type, available quantity, storage location, and freshness indicators.

[0088] AI models can be pre-trained machine learning models used to generate structured recipe data.

[0089] Structured recipe data can be standardized recipe information generated by AI models, including information such as ingredient lists, cooking steps, heat settings, and optimization logic.

[0090] Specifically, real-time ingredient inventory information (such as ingredient ID, available quantity, warehouse location, etc.) is obtained from the inventory management system. An AI model integrates order demand with real-time ingredient inventory information to calculate ingredient shortages (occurring when order demand exceeds real-time ingredient inventory) or surpluses (occurring when real-time ingredient inventory exceeds order demand). Then, combined with historical cooking datasets (datasets storing historical cooking records such as dish names, ingredient quantities, cooking steps, and time parameters), structured recipe data is generated.

[0091] S203. Generate ingredient procurement instructions and processing service instructions based on structured recipe data;

[0092] Food procurement orders can be procurement action orders generated based on structured recipe data.

[0093] Processing service instructions can be operation instructions for processing steps generated based on structured recipe data.

[0094] Specifically, based on the structured recipe data, the ingredient allocation is extracted (the ingredient usage plan for each dish specifies the types and quantities of ingredients); based on the ingredient allocation and real-time ingredient inventory information, the ingredients that need to be replenished (such as the part of insufficient inventory) are calculated and formatted into an ingredient purchase instruction (for example, the ingredient purchase instruction includes ingredient ID, purchase quantity, and priority flag).

[0095] Based on structured recipe data, cooking steps (detailed cooking operation sequences such as chopping and stir-frying, including step order, parameters and time requirements) are extracted and converted into executable processing service instructions.

[0096] S204. Based on food procurement orders and processing service orders, obtain real-time process monitoring data;

[0097] Real-time monitoring data can be dynamic monitoring data during the execution of food procurement orders and processing service orders.

[0098] Specifically, based on the food procurement instructions, the system calls the food supplier's server system to obtain real-time logistics data (such as GPS location and estimated arrival time).

[0099] Kitchen sensors and cameras are deployed at the processing station (to capture processing start / end times), the vegetable cutting area (to monitor the status of food processing), and the cooking stove (to record temperature and time parameters). Then, based on the processing service instructions, operational data (such as processing start time and personnel ID) is captured through kitchen sensors and cameras. Real-time logistics data and operational data are integrated to generate real-time process monitoring data.

[0100] S205. Based on order requirements, determine the traceability display items, and based on the traceability display items, analyze real-time process monitoring data and structured recipe data to generate display content.

[0101] The traceability display item can be a traceability information field defined according to order requirements.

[0102] The displayed content can be the final generated traceability information output.

[0103] Specifically, the traceability display items are determined directly based on order requirements; based on the traceability display items, relevant fields are parsed by combining real-time process monitoring data and structured recipe data (e.g., extracting the supplier name field from real-time process monitoring data and the ingredient usage record field from structured recipe data) to generate structured display content.

[0104] This solution acquires user orders, preventing process interruptions due to missing data. It analyzes user orders to determine order requirements, eliminating uncertainties in order parsing and ensuring these requirements can be directly used to generate structured recipe data and traceability display projects. Real-time ingredient inventory information is obtained, and based on order requirements and real-time inventory information, structured recipe data is generated using an AI model, improving ingredient utilization and processing accuracy, and ensuring recipe data adapts to real-time changes. Based on the structured recipe data, ingredient procurement instructions and processing service instructions are generated, ensuring the independence and executability of these instructions and avoiding efficiency bottlenecks caused by process coupling. Based on these instructions, real-time process monitoring data is acquired, ensuring full-chain monitoring coverage and supporting operational security verification and process tracking. Based on order requirements, traceability display projects are determined, and based on these projects, real-time process monitoring data and structured recipe data are parsed to generate display content, achieving completeness and dynamism in traceability display, enhancing user transparency and trust.

[0105] In some embodiments, structured recipe data is parsed to determine the ingredient processing flow; based on the traceability display project, the ingredient processing flow is filtered to determine the target display flow; based on the target display flow, real-time process monitoring data is parsed to generate display content.

[0106] The food processing flow can be an ordered sequence of cooking operations parsed from structured recipe data.

[0107] The target display process can be a subset of the process obtained by screening the food processing process through the traceability display project.

[0108] Specifically, the step fields in the structured recipe data are parsed by key-value matching, and the steps are extracted as the food processing flow (i.e., the sequential list of cooking operations). For example, the flow is determined to be chopping vegetables → stir-frying → plating.

[0109] Based on the traceability display item, a matching subset is selected from the food processing flow; then, the matching subset is defined as the target display flow. For example, when the traceability display item is the processing flow, the food processing flow is returned directly; when the traceability display item is the vegetable cutting step, the list of steps in the food processing flow is traversed and the vegetable cutting step is selected.

[0110] Based on the step identifiers (step names or IDs) in the target display process, the real-time process monitoring data is parsed. For example, if the target display process includes the step of chopping vegetables, the monitoring field for the corresponding step in the real-time process monitoring data is queried. Then, the step descriptions of the target display process are merged with the fields of the real-time monitoring data to generate the display content. For example, if the display item is a processing procedure, the step description is chopping vegetables, and the start time of the real-time monitoring data generation, operator, etc., are added.

[0111] This solution analyzes structured recipe data to determine ingredient processing flows, enabling dynamic generation of these flows. This reduces the risk of fragmented traceability content, improves the accuracy and efficiency of the entire traceability process, and supports end-to-end monitoring. Based on the traceability display project, it filters ingredient processing flows, determines the target display flow, reduces data redundancy, increases traceability transparency, enhances dynamic generation capabilities, and improves consumer trust and compliance. Based on the target display flow, it analyzes real-time monitoring data to generate display content, ensuring that the content includes the latest dynamic details and provides end-to-end transparency, thereby increasing trust and reducing the risk of decision-making delays.

[0112] In some embodiments, historical cooking datasets are obtained; the historical cooking datasets are parsed to determine the cooking logic of the dishes; the cost of purchased ingredients is determined based on order requirements and real-time ingredient inventory information; and structured recipe data is generated using an AI model based on the cooking logic of the dishes, order requirements, and real-time ingredient inventory information.

[0113] Historical cooking datasets can be collections of stored records of past cooking activities.

[0114] The cooking logic of a dish can be a sequence of structured cooking rules parsed from a historical cooking dataset.

[0115] The cost of purchasing ingredients can be a value that reflects the difference in cost between the order demand and the real-time ingredient inventory information.

[0116] Specifically, access the historical cooking database (which stores past cooking records, such as recipes, cooking steps, ingredient quantities, and time parameters, stored in the database and retrieved when needed); query the historical cooking database to obtain historical cooking datasets.

[0117] Iterate through each record in the historical cooking dataset (e.g., iterate over each object); and extract the step field from each record, mapping it to structured dish cooking logic.

[0118] The total amount of ingredients required for the order (the total quantity of ingredients required in the order) is compared with the available inventory in the real-time ingredient inventory information (the current available quantity of ingredients) to determine the difference in ingredients to be purchased (i.e., the total amount of ingredients minus the available inventory). Based on the difference in ingredients and the unit price field in the real-time ingredient inventory information (the unit price of ingredients, i.e., the cost value of each unit of ingredients), a multiplication operation is performed to determine the cost of purchasing ingredients.

[0119] AI models integrate recipe cooking logic, order requirements, and real-time ingredient inventory information. For example, the cooking rules (sequence of cooking steps and parameter settings, such as operation order and time parameters) in the recipe cooking logic are combined with the total amount of ingredients required by the order, and the freshness of ingredients (an indicator of ingredient freshness reflecting the real-time status of the ingredients) and available inventory are referenced in the real-time ingredient inventory information. Based on the integration results, structured recipe data is generated.

[0120] This solution acquires historical cooking datasets, allowing them to be directly used to determine the cooking logic for dishes and reducing data query latency. It analyzes these datasets to identify the cooking logic for each dish, ensuring that the logic is based on historical best practices and can be directly used in the recipe generation process. Based on order demand and real-time ingredient inventory information, it determines the cost of purchased ingredients, optimizes ingredient procurement quantities, and avoids oversupply or shortages. Based on the cooking logic, order demand, and real-time ingredient inventory information, an AI model generates structured recipe data, improving the accuracy and adaptability of the recipes.

[0121] In some embodiments, real-time food inventory information is analyzed to determine the food's shelf life; the food's freshness is determined based on the shelf life data; the actual cooking requirements are determined based on the food's freshness and order demand; the cooking time and temperature of the cooking nodes are determined using an AI model based on the dish's cooking logic and the actual cooking requirements; and the food's freshness, cooking time, and cooking temperature are converted into structured recipe data.

[0122] Food usage period data can represent the remaining usable time of food ingredients.

[0123] The freshness of ingredients can be measured by a spoilage index value mapped from the expiration date of the ingredients.

[0124] Actual cooking needs can be adjusted demand objects generated by combining the freshness of ingredients and order requirements.

[0125] A cooking node can be an operation point in the cooking process.

[0126] Cooking time can be a parameter for operation duration in minutes.

[0127] Cooking heat can be expressed as a thermal parameter in units of high, medium, and low heat levels or temperature.

[0128] Specifically, the spoilage index (a numerical indicator quantifying the real-time status of ingredients) is read from real-time food inventory information and directly mapped to the food's shelf life data. The shelf life data is then directly converted into food freshness.

[0129] A preset threshold for the spoilage index is determined using statistical methods (to compare the freshness of ingredients). Then, the freshness of ingredients is compared with the preset threshold (if the freshness of ingredients is higher than the preset threshold, it means that the freshness of ingredients is low), and the allocation of ingredients in the order demand is dynamically adjusted (for example, if the freshness of ingredients is low, low-freshness ingredients are allocated first to reduce waste) to determine the actual cooking needs.

[0130] Through AI models, the cooking rules in the recipe's cooking logic are analyzed. Combined with actual cooking needs (such as adjusted ingredient allocation), cooking nodes are dynamically optimized. For example, when the ingredients are not fresh (high spoilage index), the AI ​​model shortens the cooking time or lowers the cooking temperature to compensate for the spoilage; if the demand is high, the temperature is optimized to improve efficiency. This determines the cooking time and temperature for each cooking node. Ingredient freshness, cooking time, and cooking temperature are integrated to form structured recipe data.

[0131] This solution analyzes real-time ingredient inventory information to determine ingredient expiration dates, providing a real-time time constraint basis for the entire recipe generation process. Based on expiration date data, ingredient freshness is determined, ensuring consistency in ingredient condition and avoiding data conversion overhead. Based on ingredient freshness and order demand, actual cooking needs are determined, optimizing ingredient allocation and reducing spoilage risk. Based on the dish's cooking logic and actual cooking needs, an AI model determines the cooking time and temperature for each cooking node, improving recipe adaptability and the accuracy of processing instructions. Ingredient freshness, cooking time, and cooking temperature are structured into recipe data, enhancing traceability transparency and consumer trust.

[0132] In some embodiments, a processing service instruction is generated based on cooking time and cooking temperature; available ingredient information is determined based on ingredient freshness; ingredient requirement information is determined based on order requirements; procurement information is determined based on available ingredient information and ingredient requirement information; and an ingredient procurement instruction is generated based on the procurement information.

[0133] Available ingredient information can be a list of available ingredients that meet freshness requirements, including the types and quantities of available ingredients.

[0134] Ingredient requirements information can be a list of ingredients needed to complete a user's order.

[0135] The procurement information can be a list of ingredients that need to be purchased additionally.

[0136] Specifically, the cooking time (e.g., 5 minutes) and cooking temperature (e.g., high heat) are analyzed, and combined with the dish's cooking logic, an executable processing service instruction is generated. For example, if the cooking time is 5 minutes and the cooking temperature is high heat, the processing service instruction generated is to heat on high heat for 5 minutes at the stir-fry node.

[0137] Based on safety standards (such as food safety regulations), preset thresholds are manually set (to determine whether ingredients are usable). The freshness of the ingredients is then compared against these preset thresholds. If the freshness is below the preset threshold, the ingredient is considered fresh and marked as usable; if the freshness is above or equal to the preset threshold, it is considered fresh and excluded from the usable range. Based on the comparison results, all usable ingredient types (categories of ingredients that are safe to use after freshness screening, such as beef and spinach) and the number of usable ingredients (the number of ingredients that are safe to use after freshness screening) are selected, forming an information object of usable ingredients.

[0138] Based on order requirements and the dish cooking logic, the dishes are broken down into standard ingredient requirements (ingredient requirements after decomposition by the dish cooking logic); all ingredient requirements are aggregated to generate ingredient requirement information. The difference between the ingredient requirement information and the available ingredient information is calculated; if the difference is greater than zero, the purchase quantity is the difference (indicating replenishment is needed); if the difference is less than or equal to zero, the purchase quantity is zero (indicating no purchase is needed); the purchase quantities of several ingredient types are summarized to generate purchase information. The types and quantities of ingredients to be purchased are extracted from the purchase information and converted into ingredient purchase instructions.

[0139] This solution generates processing service instructions based on cooking time and temperature, improving recipe adaptability and avoiding processing errors caused by temperature or time variations. It determines available ingredients based on freshness, reducing waste from using spoiled ingredients and optimizing inventory utilization. It identifies ingredient demand based on order requirements, preventing response delays caused by coupling and ensuring demand analysis is independent of procurement logic. Based on available and demand information, it determines procurement information, preventing ingredient shortages or surpluses. Finally, it generates ingredient procurement instructions based on procurement information, improving supply chain responsiveness.

[0140] In some embodiments, food storage environment data is acquired; based on food shelf life data and storage environment data, a real-time spoilage index is determined through spoilage rate model analysis; and the freshness level of the food is determined based on the real-time spoilage index.

[0141] Food storage environment data can be a set of real-time parameters of the physical environment in which the food is located, obtained through temperature and humidity sensors.

[0142] The storage environment data can be a predefined static analysis rule used to determine the corruption index.

[0143] A corruption rate model can be a quantitative indicator of the degree of corruption output by a corruption rate model.

[0144] The real-time corruption index can be a discrete classification label for the real-time corruption index.

[0145] The freshness level of ingredients can be structured data that indicates the timeliness of ingredients.

[0146] Specifically, temperature and humidity sensors deployed within the physical space where food is stored acquire data about the food storage environment.

[0147] The baseline rate of food spoilage is adjusted based on storage environment data (such as temperature and humidity) (e.g., high temperatures accelerate spoilage); the degree of spoilage progression is quantified by combining food shelf-life data (e.g., the fewer days remaining, the higher the risk of spoilage); and a real-time spoilage index is determined by combining the baseline rate of food spoilage and the degree of spoilage progression through a spoilage rate model.

[0148] A preset threshold is constructed using historical cooking datasets (used to compare with real-time spoilage indices to determine the freshness level of ingredients). The real-time spoilage index is compared with the preset threshold to determine the freshness level of ingredients. For example, if most real-time spoilage indices are lower than or equal to the preset threshold, the freshness level of ingredients is determined to be high (meaning the ingredients are usable); if the real-time spoilage index is higher than the preset threshold, the freshness level of ingredients is determined to be low (meaning the ingredients are unusable).

[0149] This solution acquires data on food storage environment to capture the dynamic impact of the environment on spoilage rate. Based on food shelf-life data and storage environment data, a spoilage rate model is used to determine a real-time spoilage index, enabling a numerical representation of the food's spoilage status. Based on the real-time spoilage index, the freshness level of the food is determined, providing a decision-level output on food freshness and ensuring that only usable ingredients are included in the processing flow.

[0150] In some embodiments, the freshness of ingredients is analyzed to determine the emergency replenishment threshold; if the available amount of ingredients is lower than the required amount and the freshness of the ingredients reaches the emergency replenishment threshold, historical performance data of the supplier is obtained; based on the real-time spoilage index and the historical performance data of the supplier, the optimal supplier is determined through a weighted scoring model; and based on the optimal supplier, an ingredient procurement order is generated.

[0151] An emergency replenishment threshold can be a logical trigger condition used to determine whether an emergency replenishment process needs to be initiated.

[0152] The available ingredient quantity can be the current number of available ingredients.

[0153] The demand for ingredients can be the total amount of ingredients required.

[0154] Supplier historical performance data can be obtained from the supplier database.

[0155] A weighted scoring model can be an analytical model used to calculate supplier priority scores.

[0156] The optimal supplier is the supplier with the highest priority score after analysis using a weighted scoring model.

[0157] Specifically, based on the freshness of the ingredients, the emergency replenishment threshold conditions are read from a preset rule base (which stores the mapping rules between ingredient freshness and emergency replenishment thresholds, used to automatically determine whether the emergency replenishment threshold has been reached when analyzing ingredient freshness) built according to the mapping relationship between the real-time spoilage index and ingredient freshness. For example, when the ingredient freshness is low, it means that the emergency replenishment threshold has been reached; when the ingredient freshness is high or medium, it means that the emergency replenishment threshold has not been reached.

[0158] A food inventory database is built by regularly and manually updating the available and required quantities of food ingredients (storing the available and required quantities for extraction and comparison during inventory checks). Then, the available and required quantities of food ingredients are extracted from the food inventory database in real time. The available and required quantities of food ingredients are compared. If the available quantity of food ingredients is lower than the required quantity of food ingredients, and the freshness of the food ingredients reaches the emergency replenishment threshold, the supplier database (storing historical supplier fulfillment data for accessing and obtaining supplier reliability information when the emergency replenishment threshold is reached) is accessed to obtain the supplier's historical fulfillment data.

[0159] A weighted scoring model is used to calculate a score for each supplier by weighting the real-time corruption index and the supplier's historical performance data (for example, a higher weight for the real-time corruption index indicates that the degree of corruption has a greater impact on supplier selection, while the weight for the supplier's historical performance data indicates the reliability of performance). The supplier with the highest score is selected as the optimal supplier. Based on the optimal supplier and the type and quantity of ingredients to be purchased, a food procurement order is generated.

[0160] This solution analyzes ingredient freshness to determine emergency replenishment thresholds, avoiding the use of unusable ingredients and reducing waste and safety hazards caused by spoilage. If the available ingredient quantity is lower than the required quantity and the ingredient freshness reaches the emergency replenishment threshold, historical supplier performance data is retrieved to ensure efficient resource utilization and avoid processing supplier data in non-emergency scenarios. Based on real-time spoilage index and historical supplier performance data, a weighted scoring model is used to determine the optimal supplier, reducing order execution delays and improving procurement success rates. Based on the optimal supplier, ingredient procurement orders are generated, improving procurement efficiency and ingredient quality, and reducing the risk of supply chain disruptions.

[0161] In some embodiments, real-time process monitoring data is analyzed to determine operator information, including the operator's facial data. Based on the processing service instruction, the operator information is analyzed to determine whether the operator is a member of a preset whitelist. If the operator is not a member of the preset whitelist, the platform review data is retrieved based on the operator information. The platform review data is matched with the operator information. If the operator information matches any item in the platform review data, the platform review data that matches the operator information is reviewed based on the facial data to determine whether to lock the current processing equipment and suspend instruction execution.

[0162] Operator information can be data related to the operator's identity, including the operator's facial data.

[0163] The operator can be a person who performs the processing task.

[0164] Facial data can be the facial feature information of the operator.

[0165] The preset whitelist members can be a list of operators who have uploaded their health certificates and chef's licenses in advance and passed the platform's review and storage area. The list is stored in the database in advance and is called up when needed.

[0166] The platform's review data can be whitelist data that is currently under review.

[0167] The current processing equipment can be the equipment that is executing processing service instructions.

[0168] Specifically, facial data of operators is identified and extracted from real-time monitoring data sources to determine operator information. A pre-defined whitelist database (containing a list of operators approved by the platform, including health certificates and chef's licenses) is then accessed based on the health certificates and chef's licenses uploaded by the operators. The operator information is compared with the whitelist records in the pre-defined whitelist database. If a match is found (i.e., the operator information matches any record in the pre-defined whitelist), the operator is identified as a member of the pre-defined whitelist; otherwise, they are identified as a member outside the pre-defined whitelist.

[0169] When it is determined that the operator is not a member of the preset whitelist, the system retrieves the matching record from the platform's audit data storage area (which contains whitelist data under review, such as document information that has not been finally approved) built based on the operator's submitted document information to be reviewed, and retrieves the platform's audit data.

[0170] Each record in the platform's review data (such as document type or review status) is checked one by one and logically matched with the operator's information. If the operator's information matches any record in the platform's review data (such as document number or facial features), it is considered a match; otherwise, it is considered a mismatch.

[0171] After determining that the operator's information matches any item in the platform's audit data, facial data is used as the verification basis to audit the platform's audit data that matches the operator's information. If the audit is successful (i.e., the facial data matches the platform's audit data), the normal processing procedure is executed. If the audit fails (i.e., the facial data does not match the platform's audit data), the current processing equipment is locked, the execution of instructions is suspended, and an alarm is triggered (e.g., a siren sounds for 30 seconds).

[0172] In the specific implementation, if there are blank or missing timestamps on the screen during the analysis of real-time monitoring data, an alarm can be triggered to alert relevant personnel that the monitoring has failed. The alarm method is, for example, a siren sounds for 30 seconds.

[0173] This solution analyzes real-time monitoring data to identify operator information, including facial data, ensuring accurate identification of operators in the real-time processing environment. Based on processing service instructions, the system analyzes operator information to determine if the operator is on a pre-defined whitelist, enabling rapid screening of authorized personnel and preventing unverified individuals from directly operating processing equipment, thus enhancing operational security. If the operator is not on the whitelist, the system retrieves platform verification data based on the operator information to expand the verification scope, covering potentially authorized but not yet verified operators. This ensures that some operators awaiting verification can still be processed in emergencies, reducing the possibility of operational interruptions. The system matches platform verification data with operator information. If any item matches, the system verifies the matching data based on facial data to determine whether to lock the current processing equipment and suspend instruction execution. This ensures that only authorized or verified personnel can operate the equipment, effectively preventing unauthorized operations and improving the security of the processing process.

[0174] In some embodiments, a preset tracking target is determined according to the processing service instruction; the processing service instruction is analyzed to determine the tracking timestamp; based on the tracking timestamp, real-time process monitoring data is analyzed to determine a recommended target profile; the recommended target profile is matched with the preset tracking target, and if they do not match, the instruction execution is paused and an alarm is issued.

[0175] Preset tracking targets can be pre-defined criteria for analyzing whether processing service instructions are being followed. Different instructions may have different criteria.

[0176] The recommended target profile can be the profile corresponding to the target that is most likely to be the basis for verification at the corresponding timestamp.

[0177] The tracking timestamp can be the moment when the corresponding processing service instruction should occur under the expected state.

[0178] Specifically, since processing service instructions have been set, they need to be followed in order to better meet customer requirements. However, in many cases, due to dereliction of duty in supervision, there may be situations where instructions are not executed as required but are not detected. Therefore, in order to make it easier to distinguish whether certain processing service instructions have been executed normally, corresponding tracking targets can be assigned to them based on the characteristics of the processing service instructions. When this tracking target appears in the screen where the corresponding instruction is located, it can be considered that the processing service instruction has been executed.

[0179] For example, when cooking, oil is needed. Since oil is essential, a recognizable oil dispenser is first set up. Then, the outline of the recognizable oil dispenser is used as the preset tracking target. That is, under the command to add oil, the outline of the recognizable oil dispenser must appear for the processing service command to be considered to be executed normally. If it does not appear, it means that it has not been executed normally.

[0180] In the specific implementation process, the processing service instructions are analyzed to determine the timestamp at which the preset tracking target should appear under the expected state, and this timestamp is used as the tracking timestamp. Specifically, since processing personnel may have different levels of proficiency, error-prone periods can be set based on this timestamp to avoid erroneous analysis results caused by differences in proficiency.

[0181] At this point, the real-time monitoring data corresponding to the tracking timestamps or error-prone periods is analyzed to determine the contour in the image that most closely resembles the preset tracking target as the recommended target contour. This contour is then matched with the preset target contour to determine the matching degree. If the matching degree is higher than the preset matching degree (e.g., 90%, which can be set based on the substitutability of the preset tracking target; the easier it is to be replaced, the higher the matching degree should be), then the processing service instruction can be considered to have been executed normally.

[0182] If the process is not carried out correctly, it can be considered that the processing personnel are operating in violation of regulations. In this case, subsequent steps can be suspended and an alarm can be issued to alert the processing personnel to make corrections.

[0183] This solution binds processing service instructions to preset tracking targets (such as the outline of an oil can), enabling real-time monitoring of instruction execution. If the tracking target is not found, it indicates the instruction was not executed correctly, thus improving the transparency and traceability of the production process. Analyzing real-time monitoring data recommends the most likely outline as the target; if the match is below a set threshold, it is considered a violation, automatically pausing subsequent steps and issuing an alarm for timely intervention. This prevents errors from escalating, ensures the production process meets expectations, and ultimately improves overall production efficiency and product quality.

[0184] In some embodiments, the system analyzes the supplier's historical performance data to determine the supplier's commitment record; obtains the supplier's bidding information at the current moment; parses the supplier's bidding information to determine the supply logistics characteristics; determines the longest delay time based on the real-time corruption index; and determines the optimal supplier based on the supplier's commitment record, supply logistics characteristics, and longest delay time through a weighted scoring model.

[0185] Supplier commitment records can be performance reliability indicators extracted from the supplier's historical performance data.

[0186] Supplier bidding information can be supplier supply information that is available at the current moment.

[0187] Supply logistics characteristics can include features such as the level of freshness related to the supplier's logistics.

[0188] The longest delay time can be the time value at which ingredients must be replaced before they completely spoil, as determined by the real-time spoilage index.

[0189] Specifically, the process involves analyzing historical supplier performance data, filtering out performance event records for each supplier, and applying statistical methods to extract supplier commitment records. The system also utilizes a supplier bidding platform (a platform for publishing, querying, and retrieving supplier bidding information) to obtain real-time supplier bidding information. Furthermore, the system performs text parsing on the supplier bidding information to identify and extract supply logistics characteristics (such as the freshness level of logistics vehicles, converting descriptive fields like refrigeration temperature or freshness level into quantitative values).

[0190] The maximum delay time that food must be replaced before it becomes completely rotten is calculated based on the real-time corruption index (the higher the real-time corruption index, the shorter the maximum delay time). For example, if existing vegetables are rotten, new vegetables must be provided to replace them before they become completely unusable; this is the maximum delay time.

[0191] The supplier's commitment record, supply logistics characteristics, and longest delay time are weighted and summed using a weighted scoring model (for example, the supplier's commitment record is multiplied by weight A, the supply logistics characteristics are multiplied by weight B, and combined with the longest delay time); then, the comprehensive scores of all suppliers are compared, and the supplier with the highest score is selected as the optimal supplier.

[0192] This solution analyzes historical supplier performance data to determine supplier compliance records, ensuring supplier selection is based on objective historical performance. It acquires current supplier bidding information to ensure data timeliness and avoid decision-making biases caused by outdated information. The solution analyzes supplier bidding information to determine supply logistics characteristics, converting logistics details into comparable quantifiable feature values ​​to facilitate analysis of supplier logistics capabilities. Based on a real-time spoilage index, it determines the longest delay time, reflecting the urgency of food spoilage and providing a time constraint for supplier selection, ensuring that replacement ingredients arrive before spoilage. Based on supplier compliance records, supply logistics characteristics, and the longest delay time, a weighted scoring model is used to determine the optimal supplier, enabling rapid and accurate selection of the best supplier and optimizing emergency replenishment decisions.

[0193] Figure 3 A schematic diagram of the structure of a catering management system based on supply chain separation and full-link traceability, as provided in an embodiment of this application, is shown below. Figure 3 As shown, the catering management system 300 based on supply chain separation and full-link traceability in this embodiment includes: an order analysis module 301, a recipe generation module 302, an instruction generation module 303, a data acquisition module 304, and a traceability display module 305.

[0194] Order analysis module 301 is used to obtain user orders; analyze the user orders, and determine order requirements;

[0195] The recipe generation module 302 is used to obtain real-time ingredient inventory information and generate structured recipe data through an AI model based on the order requirements and the real-time ingredient inventory information.

[0196] The instruction generation module 303 is used to generate ingredient procurement instructions and processing service instructions based on the structured recipe data;

[0197] Data acquisition module 304 is used to acquire real-time process monitoring data based on the food procurement instruction and processing service instruction;

[0198] The traceability display module 305 is used to determine the traceability display items according to the order requirements, and to parse the real-time process monitoring data and the structured recipe data according to the traceability display items to generate display content.

[0199] Optionally, when the traceability display module 305 generates display content by parsing the real-time process monitoring data and the structured recipe data according to the traceability display item, it is used to: parse the structured recipe data to determine the ingredient processing flow; filter the ingredient processing flow according to the traceability display item to determine the target display flow; and parse the real-time process monitoring data based on the target display flow to generate display content.

[0200] Optionally, when the recipe generation module 302 generates structured recipe data using an AI model based on the order requirements and the real-time ingredient inventory information, it is used to: obtain a historical cooking dataset; parse the historical cooking dataset to determine the cooking logic of the dishes; determine the cost of purchased ingredients based on the order requirements and the real-time ingredient inventory information; and generate structured recipe data using an AI model based on the cooking logic of the dishes, the order requirements, and the real-time ingredient inventory information.

[0201] Optionally, when the recipe generation module 302 generates structured recipe data using an AI model based on the dish cooking logic, the order requirements, and the real-time ingredient inventory information, it is used to: analyze the real-time ingredient inventory information to determine the ingredient shelf-life data; acquire ingredient storage environment data; determine a real-time spoilage index based on the ingredient shelf-life data and the storage environment data using a spoilage rate model; determine the ingredient freshness level based on the real-time spoilage index; determine the actual cooking requirements based on the ingredient freshness and the order requirements; determine the cooking time and cooking temperature of the cooking node using an AI model based on the dish cooking logic and the actual cooking requirements; and convert the ingredient freshness, the cooking time, and the cooking temperature into structured recipe data.

[0202] Optionally, when the instruction generation module 303 generates ingredient procurement instructions and processing service instructions based on the structured recipe data, it is used to: generate processing service instructions based on the cooking time and cooking temperature; determine available ingredient information based on the ingredient freshness; determine ingredient demand information based on the order requirements; determine procurement information based on the available ingredient information and the ingredient demand information; and generate ingredient procurement instructions based on the procurement information.

[0203] Optionally, the available ingredient information includes the quantity of available ingredients; the ingredient demand information includes the quantity of ingredients required; when the instruction generation module 303 generates an ingredient purchase instruction based on the purchase information, it is used to: analyze the freshness of the ingredients and determine the emergency replenishment threshold; if the quantity of available ingredients is lower than the quantity of ingredients required and the freshness of the ingredients reaches the emergency replenishment threshold, then obtain the supplier's historical performance data; based on the real-time spoilage index and the supplier's historical performance data, analyze and determine the optimal supplier through a weighted scoring model; and generate an ingredient purchase instruction based on the optimal supplier.

[0204] Optionally, the catering management system based on supply chain separation and full-link traceability further includes an instruction pause module 306, used for: analyzing the real-time process monitoring data to determine operator information; the operator information includes the operator's facial data; based on the processing service instruction, analyzing the operator information to determine whether the operator is a member of a preset whitelist; if not a member of the preset whitelist, retrieving platform review data based on the operator information; matching the platform review data with the operator information; if the operator information matches any item in the platform review data, reviewing the platform review data matching the operator information based on the facial data to determine whether to lock the current processing equipment and pause instruction execution.

[0205] Optionally, the catering management system 300 based on supply chain separation and full-link traceability further includes a target tracking module 307, used for: determining a preset tracking target according to the processing service instruction; analyzing the processing service instruction to determine a tracking timestamp; analyzing real-time process monitoring data based on the tracking timestamp to determine a recommended target profile; matching the recommended target profile with the preset tracking target, and if they do not match, pausing instruction execution and issuing an alarm.

[0206] Optionally, when the instruction generation module 303 determines the optimal supplier based on the real-time corruption index and the supplier's historical performance data through a weighted scoring model, it is configured to: analyze the supplier's historical performance data to determine the supplier's commitment record; obtain the supplier's bidding information at the current moment; parse the supplier's bidding information to determine the supply logistics characteristics; determine the longest delay time based on the real-time corruption index; and determine the optimal supplier based on the supplier's commitment record, the supply logistics characteristics, and the longest delay time through a weighted scoring model.

[0207] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. A catering management method based on supply chain separation and full-link traceability, characterized in that, include: Obtain user orders; analyze the user orders to determine order requirements; Obtain real-time ingredient inventory information, and generate structured recipe data using an AI model based on the order requirements and the real-time ingredient inventory information; Based on the structured recipe data, generate ingredient procurement instructions and processing service instructions; Based on the aforementioned food procurement instructions and processing service instructions, real-time process monitoring data is obtained; Based on the order requirements, determine the traceability display items, and based on the traceability display items, parse the real-time process monitoring data and the structured recipe data to generate display content.

2. The method according to claim 1, characterized in that, The step of parsing the real-time process monitoring data and the structured recipe data based on the traceability display project to generate display content includes: Analyze the structured recipe data to determine the ingredient processing flow; Based on the traceability display items, the food processing flow is selected to determine the target display flow; Based on the target display process, the real-time monitoring data of the process is analyzed to generate display content.

3. The method according to claim 1, characterized in that, The step of generating structured recipe data using an AI model based on the order requirements and the real-time ingredient inventory information includes: Obtain historical cooking datasets; Analyze the historical cooking dataset to determine the cooking logic of the dishes; The cost of purchasing ingredients is determined based on the order requirements and the real-time ingredient inventory information. Based on the cooking logic of the dishes, the order requirements, and the real-time ingredient inventory information, structured recipe data is generated through an AI model.

4. The method according to claim 3, characterized in that, The step of generating structured recipe data using an AI model based on the dish cooking logic, order requirements, and real-time ingredient inventory information includes: Analyze the real-time food inventory information to determine the food's shelf life. Obtain data on the food storage environment; Based on the food's shelf life data and the storage environment data, a real-time spoilage index is determined through a spoilage rate model analysis. The freshness level of the ingredients is determined based on the real-time spoilage index. The actual cooking requirements will be determined based on the freshness of the ingredients and the order details. Based on the cooking logic of the dish and the actual cooking requirements, the cooking time and cooking temperature of the cooking nodes are determined by an AI model. The freshness of the ingredients, the cooking time, and the cooking temperature are determined as structured recipe data.

5. The method according to claim 4, characterized in that, The step of generating ingredient procurement instructions and processing service instructions based on the structured recipe data includes: Based on the cooking time and the cooking temperature, a processing service instruction is generated; Based on the freshness of the ingredients, determine the available ingredient information; Based on the order requirements, determine the required ingredient information; Based on the available ingredient information and the ingredient demand information, the procurement information is determined; Based on the procurement information, a food procurement instruction is generated.

6. The method according to claim 5, characterized in that, The available food information includes the quantity of available food; the food demand information includes the quantity of food required; generating a food procurement instruction based on the procurement information includes: Analyze the freshness of the ingredients to determine the emergency replenishment threshold; If the available amount of ingredients is lower than the required amount of ingredients and the freshness of the ingredients reaches the emergency replenishment threshold, then obtain the supplier's historical fulfillment data; Based on the real-time corruption index and the supplier's historical performance data, the optimal supplier is determined through a weighted scoring model. Based on the optimal supplier, generate a food procurement instruction.

7. The method according to claim 1, characterized in that, The method further includes: Analyze the real-time monitoring data to determine the operator information; the operator information includes the operator's facial data. Based on the processing service instruction, analyze the operator information to determine whether the operator is a member of the preset whitelist; If the user is not a member of the preset whitelist, the platform's review data will be retrieved based on the operator information. The platform review data is matched with the operator information. If the operator information matches any item in the platform review data, the platform review data that matches the operator information is reviewed based on the facial data to determine whether to lock the current processing equipment and suspend instruction execution.

8. The method according to claim 1, characterized in that, The method further includes: Based on the processing service instruction, a preset tracking target is determined; Analyze the processing service instructions to determine the tracking timestamp; Based on the tracking timestamp, analyze real-time process monitoring data to determine the recommended target profile; The recommended target contour is matched with the preset tracking target. If they do not match, the command execution is paused and an alarm is issued.

9. The method according to claim 6, characterized in that, The step of determining the optimal supplier based on the real-time corruption index and the supplier's historical performance data through a weighted scoring model includes: Analyze the supplier's historical performance data to determine the supplier's commitment record; Obtain the supplier bidding information at the current moment; parse the supplier bidding information to determine the supply logistics characteristics; The maximum delay time is determined based on the real-time corruption index. The optimal supplier is determined by weighted scoring model analysis based on the supplier's commitment record, supply logistics characteristics, and longest delay time.

10. A catering management system based on supply chain separation and full-link traceability, characterized in that, The method applied to any one of claims 1-9 includes: The order analysis module is used to obtain user orders; analyze the user orders, and determine order requirements; The recipe generation module is used to obtain real-time ingredient inventory information and generate structured recipe data through an AI model based on the order requirements and the real-time ingredient inventory information. The instruction generation module is used to generate ingredient procurement instructions and processing service instructions based on the structured recipe data; The data acquisition module is used to acquire real-time process monitoring data based on the food procurement instructions and processing service instructions. The traceability display module is used to determine the traceability display items according to the order requirements, and to parse the real-time process monitoring data and the structured recipe data according to the traceability display items to generate display content.

Images