An airport meal assignment task dynamic scheduling system based on microservices and event driving

Abstract

The application relates to a micro-service and event-driven airport meal allocation task dynamic scheduling system, and relates to the field of data processing specially used for administrative, commercial, financial, management, supervision or prediction purposes. The system comprises meal allocation prediction devices for adopting a micro-service architecture, triggering an event-driven prediction model to perform a meal allocation prediction operation every preset time length when a target flight is delayed until the target flight actually takes off, so as to obtain predicted consumption quantities of various foods of the target flight at the time when the meal allocation prediction operation is performed; and dynamic scheduling devices for performing emergency scheduling of insufficient foods based on the predicted consumption quantities. Through the application, the technical problem that the dining demands of the same passenger group under different delay degrees of the same flight are difficult to meet is solved by adopting a micro-service architecture to respectively perform intelligent prediction processing on the dining demands of the same passenger group under different delay degrees of the target flight.

Description

Technical Field

[0001] This invention relates to the field of data processing specifically applicable to administrative, commercial, financial, management, supervisory, or predictive purposes, and in particular to a dynamic scheduling system for airport catering tasks based on microservices and event-driven approaches. Background Technology

[0002] Airport catering units and airline catering departments provide catering services to aircraft operating flights at each aircraft parking position, which is part of their daily work. Generally, these units allocate a fixed quantity of various food items for flights departing that day based on ticket sales, actual passenger numbers, and historical consumption data. Specifically, this allocation is carried out using food supply warehouses located at the airport. In other words, airport catering units and airline catering departments primarily rely on historical experience to allocate the fixed quantities of various food items for each flight. Typically, a backup quantity of meals is prepared before delivery. If, after handing over meals to the cabin crew, a shortage is discovered before takeoff, the airport catering unit or airline catering department will be contacted for replenishment to meet passenger dining needs.

[0003] For example, Chinese invention patent publication CN119740846A proposes an airport catering scheduling system based on big data analysis. The system includes a real-time data perception layer, a decision optimization core layer, and an execution monitoring layer. The real-time data perception layer collects flight dynamics, resource status, and environmental variable data using airport operation databases, IoT sensors, and RFID tags. The decision optimization core layer's fusion prediction module uses an improved deep learning model to predict food demand and resource consumption. The dynamic scheduling module uses a multi-objective genetic algorithm to generate resource allocation schemes. The path planning module applies an improved ant colony algorithm to determine the optimal transportation path. The execution monitoring layer monitors scheme execution through a digital twin visualization platform and traces abnormal events through a blockchain traceability unit. This invention can effectively improve the rationality of resource allocation, delivery efficiency, and management reliability of airport catering, strongly guaranteeing the quality of airport catering services and adapting to the needs of modern airport operations.

[0004] For example, Chinese invention patent publication CN118504950A proposes an airport catering scheduling system. This invention constructs a precise 3D model of the airport, clearly presenting its terrain, buildings, catering routes, and other features, providing accurate spatial reference for catering scheduling. The positioning module on the delivery vehicle can synchronize the vehicle's location to the airport's 3D model in real time, allowing the scheduling center to monitor delivery progress. The delivery planning module, based on the catering sequence model and food truck information, can rationally allocate delivery tasks, ensuring a balanced load on each delivery vehicle and preventing some vehicles from being overloaded while others are idle. The flight catering analysis module, through in-depth analysis of historical flight data and ticketing information, can accurately predict the number of diners on each flight and prepare ingredients based on the catering sequence model and subsequent time intervals' food demand, thus ensuring timely supply of necessary ingredients.

[0005] Clearly, the existing technologies mentioned above only address the specific implementation process of airport catering services, without considering the analysis and scheduling of catering needs for each flight at future times. This leads to a rudimentary catering model that still relies on historical experience and on-site manual coordination. This can easily result in a mismatch between the quantity of catering provided for a particular flight and the actual demand from passengers, or waste of catering resources, or insufficient catering supply, causing passenger dissatisfaction, especially when flights are delayed. Because different passengers have different dining habits and times, even on the same flight, the dining needs of the same passenger group will differ when there are different degrees of delay. How to dynamically obtain the different dining needs of the same passenger group under different degrees of delay, and provide key data for dynamic scheduling of airport catering tasks, is one of the technical problems that needs to be solved. Summary of the Invention

[0006] To address the technical problems in existing technologies, this invention provides a dynamic scheduling system for airport catering tasks based on microservices and event-driven mechanisms. This system employs a microservice architecture to trigger a customized event-driven prediction model every preset time interval when a target flight departs later than its scheduled departure time. This process continues until the target flight actually departs, obtaining the predicted consumption quantities of various food items at the time the prediction operation is executed. The preset time interval is greater than or equal to a threshold value to ensure sufficient time between adjacent prediction operations, allowing for adequate time for subsequent dynamic mobilization of ingredients and response time for ingredient preparation at the supply warehouse. By using a microservice architecture as the main operating mechanism and flight delays as the event driver, this system achieves dynamic scheduling of airport catering tasks. It simultaneously addresses the dual objectives of conserving limited ingredients during flight delays and meeting passenger needs, preventing waste of catering resources or insufficient catering supply.

[0007] According to the present invention, a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods is provided, the system comprising:

[0008] The information parsing device uses a microservice architecture to parse the on-time departure time, full number of passengers, aircraft type number, and distance from the configured aircraft parking position to the food supply warehouse of the target flight, so as to use the corresponding configuration information of the target flight.

[0009] The food capture device uses a microservice architecture to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to each day of the target flight includes the delay duration of the target flight, the actual number of passengers, and the consumption of various types of food on that day.

[0010] A model building device is used to perform multiple training operations on a feedforward neural network to obtain a feedforward neural network after multiple training operations and output it as an event-driven prediction model.

[0011] The meal preparation prediction device is connected to the information analysis device, the food capture device, and the model building device. When the target flight takes off later than the scheduled departure time of the day, the event-driven prediction model is triggered to perform a meal preparation prediction operation once every preset time length until the target flight actually takes off, so as to obtain the predicted consumption of various food items of the target flight at the time of execution of the meal preparation prediction operation. The preset time length is greater than or equal to the time length threshold.

[0012] The dynamic scheduling device, connected to the meal prediction device, is used to issue an emergency scheduling request for a certain type of food when the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on the machine and the difference between the two is greater than or equal to a set difference threshold.

[0013] According to a second aspect of the present invention, a dynamic scheduling system for airport catering tasks based on microservices and event-driven architecture is provided. The system includes a memory and multiple processors. The memory stores a computer program configured to be executed by the multiple processors to complete the following steps:

[0014] A microservice architecture is used to parse the on-time departure time, full number of passengers, aircraft type number, and distance from the configured aircraft parking position to the food supply warehouse of the target flight, so as to serve as the corresponding configuration information for the target flight;

[0015] A microservice architecture is used to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to each day of the target flight includes the delay time of the target flight, the actual number of passengers, and the consumption of various types of food on that day.

[0016] The feedforward neural network is trained multiple times to obtain a feedforward neural network after multiple training sessions, which is then used as the output of the event-driven prediction model.

[0017] When the target flight takes off later than the scheduled departure time of the day, the event-driven prediction model is triggered to perform a catering prediction operation once every preset time length until the target flight actually takes off, so as to obtain the predicted consumption of various food items of the target flight at the time of the catering prediction operation. The preset time length is greater than or equal to the time length threshold.

[0018] When the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on board, and the difference is greater than or equal to a set difference threshold, an emergency dispatch request for that type of food will be issued.

[0019] Therefore, it can be seen that the present invention has at least the following prominent substantive features:

[0020] Substantive Feature A: A microservice architecture is adopted to trigger an event-driven prediction model to perform a catering prediction operation every preset time interval when the target flight takes off later than the scheduled departure time of the day until the target flight actually takes off. This is to obtain the predicted consumption of various food items for the target flight at the time the catering prediction operation is executed. The preset time interval is greater than or equal to a time length threshold to ensure that the interval between two adjacent catering prediction operations is long enough to allow sufficient time for subsequent dynamic mobilization of ingredients and response time for ingredient preparation at the ingredient supply warehouse. Thus, with the microservice architecture as the main operation body and flight delay as the event driver, the dynamic scheduling of airport catering tasks is realized, while meeting the passenger needs in the case of flight delay.

[0021] Substantive Feature B: Each meal preparation prediction operation is completed using a custom-designed event-driven prediction model. The event-driven prediction model is a feedforward neural network that has been trained multiple times. The number of training cycles of the feedforward neural network is positively correlated with the delay duration of the target flight the day before, so as to ensure that the dynamic accuracy requirements of the event-driven prediction model are met. Through the above-mentioned custom structural design, the stability and effectiveness of each meal preparation prediction result of the event-driven prediction model are guaranteed.

[0022] Substantive Feature C: The introduction of various basic data to complete each catering prediction operation. Specifically, the basic data includes the execution time of the catering prediction operation, the configuration information of the target flight, and multiple sets of food-related data corresponding to the target flight over the past few days. More specifically, the configuration information of the target flight includes the on-time departure time, the number of passengers at full capacity, the aircraft type number, and the distance from the configured aircraft parking position to the food supply warehouse. The food-related data for each day of the target flight includes the delay duration of the target flight on that day, the actual number of passengers, and the consumption of various types of food. Crucially, the number of days selected over the past few days is positively correlated with the flight mileage of the target flight. The comprehensive and sufficient introduction of the above basic data further ensures the stability and effectiveness of the catering prediction results of the event-driven prediction model for each catering prediction.

[0023] Substantive Feature D: In each training iteration of the feedforward neural network, the time of execution of the catering prediction operation closest to the actual departure time and the on-time departure time of the target flight on a certain day is taken as the reference execution time. The known consumption quantity of various types of food of the target flight on a certain day is taken as the predicted consumption quantity of various types of food of the target flight at the reference execution time, and is taken as the single output of the feedforward neural network. The reference execution time, the configuration information corresponding to the target flight, and the multiple sets of food association data corresponding to the target flight on the previous days before the stated date are taken as the input of the feedforward neural network to complete the training of the feedforward neural network, thereby ensuring the training effect of the feedforward neural network in each iteration.

[0024] Substantive Feature E: The specific architecture of the feedforward neural network used is as follows: The feedforward neural network has a network architecture including multiple hidden layers, a single input layer and a single output layer. The multiple hidden layers are located between the single input layer and the single output layer, and the number of hidden layers of the feedforward neural network is positively correlated with the flight mileage of the target flight. At the same time, an information transformation formula is selected to represent the information transformation relationship between the number of hidden layers of the feedforward neural network and the flight mileage of the target flight. Attached Figure Description

[0025] The embodiments of the present invention will now be described with reference to the accompanying drawings, wherein:

[0026] Figure 1 This is a schematic diagram illustrating the working principle of a dynamic scheduling system for airport catering tasks based on microservices and event-driven approaches according to the present invention.

[0027] Figure 2 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, according to a first embodiment of the present invention.

[0028] Figure 3 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the second embodiment of the present invention.

[0029] Figure 4 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the third embodiment of the present invention.

[0030] Figure 5 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the fourth embodiment of the present invention.

[0031] Figure 6 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the fifth embodiment of the present invention.

[0032] Figure 7 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the sixth embodiment of the present invention. Detailed Implementation

[0033] like Figure 1 The diagram illustrates the working principle of a dynamic scheduling system for airport catering tasks based on microservices and event-driven approaches, according to the present invention.

[0034] The specific technical process of this invention is as follows:

[0035] Technical Process 1: To complete the meal preparation prediction operation triggered every preset time length in the case of a target flight being delayed, a customized event-driven prediction model is designed.

[0036] Specifically, the customized structural design of the event-driven prediction model is mainly reflected in the following four aspects:

[0037] Aspect 1: The event-driven prediction model is a feedforward neural network that has undergone multiple training iterations;

[0038] Aspect 2: The number of training iterations performed by the feedforward neural network is positively correlated with the delay duration of the target flight the day before, to ensure that the dynamic accuracy requirements of the event-driven prediction model are met;

[0039] Aspect 3: The specific architecture of the feedforward neural network used is as follows: The feedforward neural network has a network architecture including multiple hidden layers, a single input layer and a single output layer. The multiple hidden layers are located between the single input layer and the single output layer, and the number of hidden layers of the feedforward neural network is positively correlated with the flight mileage of the target flight. At the same time, an information transformation formula is selected to represent the information transformation relationship between the number of hidden layers of the feedforward neural network and the flight mileage of the target flight.

[0040] Aspect 4: In each training iteration of the feedforward neural network, the time of execution of the catering prediction operation closest to the actual departure time and the on-time departure time of the target flight on a certain day is taken as the reference execution time. The known consumption quantity of various types of food of the target flight on a certain day is taken as the predicted consumption quantity of various types of food of the target flight at the reference execution time, and is used as the single output of the feedforward neural network. The reference execution time, the configuration information corresponding to the target flight, and the multiple sets of food association data corresponding to the target flight on the previous days before the stated date are taken as the input of the feedforward neural network, thus completing the training of the feedforward neural network for this iteration, thereby ensuring the training effect of the feedforward neural network in each iteration.

[0041] In this way, the stability and effectiveness of the meal preparation prediction results of the event-driven prediction model are ensured through the above-mentioned customized structural design at each location.

[0042] Technical Process 2: To complete the meal preparation prediction operation triggered every preset time length in the case of a target flight being delayed, various basic data are introduced;

[0043] Specifically, such as Figure 1 As shown, the basic data introduced includes the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the association data of multiple ingredients corresponding to the target flight over the past few days.

[0044] More specifically, the configuration information corresponding to the target flight includes the target flight's on-time departure time, full number of passengers, aircraft type number, and distance from the configured aircraft parking position to the food supply warehouse. The food-related data for each day of the target flight includes the delay duration of the target flight on that day, the actual number of passengers, and the consumption of various types of food.

[0045] Crucially, the number of past days selected is positively correlated with the flight mileage of the target flight;

[0046] In this way, by fully introducing the above-mentioned basic data, the stability and effectiveness of the meal preparation prediction results of the event-driven prediction model are further guaranteed.

[0047] Technical Process 3: Using the event-driven prediction model with a customized structure designed in Technical Process 1, and based on the comprehensive and sufficient basic data introduced in Technical Process 2, a meal preparation prediction operation is completed every preset time length triggered when the target flight is delayed.

[0048] Specifically, a microservice architecture is used to trigger an event-driven prediction model to perform a catering prediction operation every preset time interval when the target flight departs later than its scheduled departure time, until the target flight actually takes off. This obtains the predicted consumption quantities of various food items for the target flight at the time the catering prediction operation is executed, such as... Figure 1 As shown;

[0049] Among them, the preset time length is greater than or equal to the time length threshold to ensure that the interval between two adjacent meal preparation prediction operations is long enough to allow sufficient time for subsequent dynamic mobilization of ingredients and response time for ingredient preparation at the ingredient supply warehouse.

[0050] Technical Process 4: Based on the prediction results of Technical Process 3, determine whether to implement emergency dispatch for a certain type of food with insufficient supply;

[0051] For example, in the prediction results of technical process three, if the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on board, and the difference is greater than or equal to a set difference threshold, an emergency dispatch request for that type of food will be issued, such as... Figure 1 As shown;

[0052] Conversely, in the prediction results of technical process three, if the predicted consumption of a certain type of food does not exceed the actual quantity of that type of food on board, or if it does, but the difference between the two is less than the set difference threshold, then there is no need to issue an emergency dispatch request for any type of food.

[0053] As can be seen, this invention uses a microservice architecture as the main operating entity and flight delays as the event-driven mechanism. It intelligently predicts the different dining needs of the same passenger group under different degrees of time delay, provides key data for dynamic scheduling of airport catering tasks, realizes dynamic scheduling of airport catering tasks, and simultaneously takes into account the two goals of saving limited ingredients in the case of flight delays and meeting passenger needs.

[0054] The key points of this invention are: a directional intelligent prediction mechanism for different degrees of time delays for the same flight and the same passenger group in the scenario of flight delays, the introduction of microservice architecture, multiple customized structural designs of the event-driven prediction model, and the comprehensive and full introduction of various basic data for intelligent prediction.

[0055] The following will describe in detail an airport catering task dynamic scheduling system based on microservices and event-driven methods according to the present invention through an embodiment.

[0056] First Embodiment

[0057] Figure 2 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, according to a first embodiment of the present invention.

[0058] like Figure 2 As shown, the airport catering task dynamic scheduling system based on microservices and event-driven methods includes the following components:

[0059] The information parsing device uses a microservice architecture to parse the on-time departure time, full number of passengers, aircraft type number, and distance from the configured aircraft parking position to the food supply warehouse of the target flight, so as to use the corresponding configuration information of the target flight.

[0060] Specifically, microservices (or microservice architecture) is a cloud-native architectural approach that includes numerous loosely coupled and independently deployable small components or services within a single application. These services typically have their own technology stack, including databases and data management models; communicate with each other through a combination of REST APIs, event streams, and message brokers; and are organized according to business capabilities, with service delimiters often referred to as bounded contexts.

[0061] More specifically, microservices are characterized by easier code updates, allowing new features or functionalities to be added directly without updating the entire application. Teams can use different technology stacks and programming languages ​​for different components. Components can scale independently of each other, thereby reducing the waste and cost associated with having to scale the entire application (because a single function may face excessive load).

[0062] The food capture device uses a microservice architecture to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to each day of the target flight includes the delay duration of the target flight, the actual number of passengers, and the consumption of various types of food on that day.

[0063] For example, a microservice architecture is used to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to each day of the target flight includes the delay duration of the target flight on that day, the actual number of passengers, and the consumption quantity of various foods. This includes using multiple processes under the microservice architecture to obtain multiple sets of food-related data corresponding to the target flight over the past few days.

[0064] A model building device is used to perform multiple training operations on a feedforward neural network to obtain a feedforward neural network after multiple training operations and output it as an event-driven prediction model.

[0065] For example, performing multiple training operations on a feedforward neural network to obtain a feedforward neural network after multiple training operations and outputting it as an event-driven prediction model includes: selecting a numerical simulation mode to complete the testing and simulation of the model building process of performing multiple training operations on the feedforward neural network to obtain a feedforward neural network after multiple training operations and outputting it as an event-driven prediction model.

[0066] The meal preparation prediction device is connected to the information analysis device, the food capture device, and the model building device. When the target flight takes off later than the scheduled departure time of the day, the event-driven prediction model is triggered to perform a meal preparation prediction operation once every preset time length until the target flight actually takes off, so as to obtain the predicted consumption of various food items of the target flight at the time of execution of the meal preparation prediction operation. The preset time length is greater than or equal to the time length threshold.

[0067] For example, when the target flight departs after its scheduled departure time, the event-driven prediction model is triggered to perform a catering prediction operation once every preset time length until the target flight actually takes off, in order to obtain the predicted consumption quantity of various food items of the target flight at the time the catering prediction operation is executed. The preset time length is greater than or equal to the time length threshold, including: if the scheduled departure time of the target flight is 12:00 noon every day, and the flight has not taken off by 12:00 noon on the same day, it is considered that the target flight has entered a delay scenario. In this case, the event-driven prediction model is triggered to perform a catering prediction operation once every 15 minutes until the target flight actually takes off, in order to obtain the predicted consumption quantity of various food items of the target flight at the time the catering prediction operation is executed. The preset time length of 20 minutes is greater than or equal to the time length threshold of 15 minutes.

[0068] Specifically, the design of setting the preset time length to be greater than or equal to the time length threshold is to ensure that the interval between two adjacent meal preparation prediction operations is long enough to allow sufficient time for subsequent dynamic mobilization of ingredients and response time for ingredient preparation at the ingredient supply warehouse. For example, the interval between two adjacent meal preparation prediction operations is 15 minutes.

[0069] The dynamic scheduling device, connected to the meal prediction device, is used to issue an emergency scheduling request for a certain type of food when the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on the machine and the difference between the two is greater than or equal to a set difference threshold.

[0070] For example, when the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on board and the difference between the two is greater than or equal to a set difference threshold, issuing an emergency dispatch request for that type of food includes: when the predicted consumption of beef set meals exceeds the actual quantity of beef set meals on board and the difference between the two is greater than or equal to a set difference threshold, issuing an emergency dispatch request for beef set meals.

[0071] Conversely, when the predicted consumption of each type of food does not exceed the actual quantity of each type of food on board, or even if such a quantity exists but the difference between the two is greater than or equal to a set difference threshold, there is no need to issue an emergency dispatch request for any type of food.

[0072] As a further example, when the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on board and the difference between the two is greater than or equal to a set difference threshold, issuing an emergency dispatch request for that type of food also includes: the set difference threshold is set to 10 units. When the difference is less than 10 units, it indicates that on-site coordination can be easily carried out on the aircraft, and there is no need to issue an emergency dispatch request for that type of food.

[0073] In each meal preparation prediction operation executed by the event-driven prediction model, the event-driven prediction model intelligently predicts the predicted consumption of various food items on the target flight at the time of the meal preparation prediction operation based on the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the multiple food ingredient association data of the target flight over the past few days.

[0074] For example, in the scenario where the target flight is delayed, the event-driven prediction model is triggered every 15 minutes to execute a catering prediction operation until the target flight actually takes off. It can be seen that there is more than one execution time for the catering prediction operation, and it gradually increases and changes. For example, the execution times for each catering prediction operation can be 12:20, 12:40, 13:00, 13:20, and 13:40. If the target flight takes off at 13:50, there is no catering prediction operation execution time at 14:00. If the target flight takes off at 14:15, there is a catering prediction operation execution time at 14:00.

[0075] Among them, the process of training the feedforward neural network multiple times to obtain the feedforward neural network after multiple training sessions and using it as the output of the event-driven prediction model includes: the number of training sessions performed by the feedforward neural network is positively correlated with the delay duration of the target flight the day before.

[0076] For example, a positive correlation between the number of training iterations of the feedforward neural network and the delay duration of the target flight the previous day includes: if the delay duration of the target flight the previous day is 15 minutes, the feedforward neural network is trained 600 times; if the delay duration of the target flight the previous day is 20 minutes, the feedforward neural network is trained 700 times; if the delay duration of the target flight the previous day is 25 minutes, the feedforward neural network is trained 800 times; if the delay duration of the target flight the previous day is 30 minutes, the feedforward neural network is trained 900 times, and so on.

[0077] In each training iteration of the feedforward neural network, the time of execution of the catering prediction operation closest to the actual departure time and the on-time departure time of the target flight on a certain day is taken as the reference execution time. The known consumption quantity of various types of food of the target flight on a certain day is taken as the predicted consumption quantity of various types of food of the target flight at the reference execution time, and is used as the single output of the feedforward neural network. The reference execution time, the configuration information corresponding to the target flight, and the multiple sets of food association data corresponding to the target flight on the previous days before the stated date are taken as the item-by-item input of the feedforward neural network to complete this training of the feedforward neural network.

[0078] Among them, a microservice architecture is used to capture multiple sets of food-related data corresponding to the target flight over multiple days. The food-related data corresponding to each day of the target flight includes the delay time of the target flight on that day, the actual number of passengers, and the consumption of various types of food. The selected number of days in the past is positively correlated with the flight mileage of the target flight.

[0079] For example, a positive correlation between the selected number of past days and the target flight's mileage includes: a target flight mileage of 500 km and a selected number of past days of 7 days; a target flight mileage of 600 km and a selected number of past days of 9 days; a target flight mileage of 700 km and a selected number of past days of 11 days; a target flight mileage of 800 km and a selected number of past days of 13 days; a target flight mileage of 900 km and a selected number of past days of 15 days, and so on.

[0080] Second Embodiment

[0081] Figure 3 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the second embodiment of the present invention.

[0082] like Figure 3 As shown, compared to Figure 2 The microservice- and event-driven dynamic scheduling system for airport catering tasks also includes:

[0083] An event response device, set up at the food supply warehouse and connected to a dynamic scheduling device, is used to receive emergency scheduling requests for a certain type of food from a target flight using a microservice architecture, and to determine the quantity of a certain type of food that needs to be urgently replenished to the target flight based on the difference between the predicted consumption quantity of the certain type of food and the actual quantity of the certain type of food on board.

[0084] For example, the process of receiving an emergency dispatch request for a certain type of food from a target flight using a microservice architecture, and determining the quantity of a certain type of food that needs to be urgently replenished to the target flight based on the difference between the predicted consumption quantity of the certain type of food and the actual quantity of the certain type of food on board, includes: a first process using a microservice architecture receiving the emergency dispatch request for a certain type of food from the target flight, and a second process using a microservice architecture determining the quantity of a certain type of food that needs to be urgently replenished to the target flight based on the difference between the predicted consumption quantity of the certain type of food and the actual quantity of the certain type of food on board.

[0085] The process of receiving an emergency dispatch request for a certain type of food from a target flight using a microservice architecture, and determining the quantity of a certain type of food that needs to be urgently replenished to the target flight based on the difference between the predicted consumption quantity of the certain type of food and the actual quantity of the certain type of food on board, includes: the emergency dispatch request for a certain type of food from the target flight includes the predicted consumption quantity of the certain type of food, the actual quantity of the certain type of food on board, and the flight number of the target flight.

[0086] Third Embodiment

[0087] Figure 4 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the third embodiment of the present invention.

[0088] like Figure 4 As shown, compared to Figure 3 The microservice- and event-driven dynamic scheduling system for airport catering tasks also includes:

[0089] The response execution device is set at the food supply warehouse and connected to the event response device. It is used to take the quantity of a certain type of food that needs to be urgently replenished to the target flight as the target product quantity. When the production time of a certain type of food with the target product quantity is less than the preset time length, the production of the certain type of food is started.

[0090] For example, the quantity of a certain type of food that needs to be urgently supplied to the target flight is taken as the target product quantity. When the preparation time of a certain type of food is less than the preset time length, the preparation of the certain type of food is initiated. This includes: when the difference between the predicted consumption quantity of beef set meals and the actual quantity of beef set meals on board is 15 pieces, that is, when the determined quantity of beef set meals that need to be urgently supplied to the target flight is 15 pieces, since the preparation time of 15 beef set meals is 18 minutes, which is less than the preset time length of 20 minutes, the preparation of beef set meal food can be initiated.

[0091] Conversely, if the preparation time for 15 beef set meals is 28 minutes, which is longer than the preset time of 20 minutes, then there is no need to start preparing the beef set meals. This is because the preparation has not yet been completed, and the next meal preparation prediction operation has been started again. The passenger's demand for various food items may have changed. In this case, continuing to prepare the beef set meals would likely be meaningless.

[0092] The response execution device, located at the food supply warehouse and connected to the event response device, is used to take the quantity of a certain type of food that needs to be urgently supplied to the target flight as the target product quantity. When the production time of a certain type of food with the target product quantity is less than a preset time length, the production of the certain type of food is initiated. The response execution device is connected to the event response device through a microservice architecture network.

[0093] The response execution device is also used to temporarily suspend the production of a certain type of food when the production time of a certain type of food with a target product quantity is greater than or equal to a preset time length, in order to wait for the intelligent prediction result of the next meal preparation prediction operation execution time.

[0094] Fourth embodiment

[0095] Figure 5 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the fourth embodiment of the present invention.

[0096] like Figure 5 As shown, compared to Figure 2 The microservice- and event-driven dynamic scheduling system for airport catering tasks also includes:

[0097] The timing server is connected to the meal preparation prediction device, the information analysis device, the food ingredient capture device, and the model building device, respectively, to provide the timing services required by each of them.

[0098] Specifically, the timing server is connected to the meal preparation prediction device, the information analysis device, the food ingredient capture device, and the model building device, respectively, and is used to provide the timing services required by the meal preparation prediction device, the information analysis device, the food ingredient capture device, and the model building device respectively. The timing server uses a built-in quartz oscillator component to provide reference clock pulses, which is used to provide the timing services required by the meal preparation prediction device, the information analysis device, the food ingredient capture device, and the model building device respectively.

[0099] Fifth embodiment

[0100] Figure 6This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the fifth embodiment of the present invention.

[0101] like Figure 6 As shown, compared to Figure 2 The microservice- and event-driven dynamic scheduling system for airport catering tasks also includes:

[0102] A model storage device, connected to a model building device, is used to receive and store the event-driven prediction model;

[0103] For example, the model storage device can be implemented using a TF memory chip, an MMC memory chip, or a FLASH memory chip to receive and store the event-driven prediction model;

[0104] The model storage device, connected to the model building device, is used to receive and store the event-driven prediction model, including: the model storage device completes the model storage of the event-driven prediction model by storing various model parameters of the event-driven prediction model.

[0105] Next, various embodiments of the present invention will be further described.

[0106] Optionally, within the above embodiments, in the microservices-based and event-driven dynamic scheduling system for airport catering tasks:

[0107] The positive correlation between the number of training iterations of the feedforward neural network and the delay duration of the target flight the previous day includes: using a numerical mapping function to represent the numerical mapping relationship between the number of training iterations of the feedforward neural network and the delay duration of the target flight the previous day;

[0108] For example, you can choose to use the MATLAB toolbox to test and simulate the numerical mapping process that uses a numerical mapping function to represent the positive correlation between the number of training operations performed by the feedforward neural network and the delay duration of the target flight the previous day.

[0109] The numerical mapping function used to represent the positive correlation between the number of training operations performed by the feedforward neural network and the delay duration of the target flight the previous day includes: in the numerical mapping function, the delay duration of the target flight the previous day is used as the input value of the numerical mapping function, and the number of training operations performed by the feedforward neural network that is positively correlated with the delay duration of the target flight the previous day is used as the output value of the numerical mapping function.

[0110] Among them, a microservice architecture is used to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to the target flight for each past day includes the delay time of the target flight on that day, the actual number of passengers, and the consumption of various types of food, including: the consumption of various types of food on that day includes the consumption of beef set meals, chicken set meals, mineral water bottles, and snacks.

[0111] Specifically, when the target flight departs after the scheduled departure time of the day, the event-driven prediction model is triggered to perform a catering prediction operation once every preset time length until the target flight actually takes off, so as to obtain the predicted consumption quantity of various food items of the target flight at the time of execution of the catering prediction operation. The preset time length is greater than or equal to the time length threshold.

[0112] And among them, when the predicted consumption of a certain type of food exceeds the actual on-board quantity of that type of food and the difference between the two is greater than or equal to a set difference threshold, issuing an emergency dispatch request for that type of food includes: using a microservice architecture to issue an emergency dispatch request for that type of food when the predicted consumption of a certain type of food exceeds the actual on-board quantity of that type of food and the difference between the two is greater than or equal to a set difference threshold.

[0113] And, optionally, in the above embodiments, in the microservices-based and event-driven dynamic scheduling system for airport catering tasks:

[0114] In each meal preparation prediction operation executed by the event-driven prediction model, the event-driven prediction model intelligently predicts the predicted consumption quantity of various food items of the target flight at the time of the meal preparation prediction operation execution based on the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the multiple sets of food item association data corresponding to the target flight over the past several days. This includes: In each meal preparation prediction operation executed by the event-driven prediction model, the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the multiple sets of food item association data corresponding to the target flight over the past several days are synchronously input into the event-driven prediction model to run the event-driven prediction model and obtain the predicted consumption quantity of various food items of the target flight at the time of the meal preparation prediction operation output by the event-driven prediction model.

[0115] For example, a process under a microservice architecture can be used to synchronously input the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the data of multiple food ingredients associated with the target flight over the past few days into the event-driven prediction model. Then, another process under the microservice architecture can run the event-driven prediction model to obtain the predicted consumption quantity of various foods for the target flight at the execution time of the meal preparation prediction operation.

[0116] In each meal preparation prediction operation executed by the event-driven prediction model, the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days are synchronously input into the event-driven prediction model to run the event-driven prediction model and obtain the predicted consumption quantity of various foods of the target flight at the execution time of the meal preparation prediction operation. This includes using synchronous control equipment to synchronously input the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days into the event-driven prediction model.

[0117] In each meal preparation prediction operation executed by the event-driven prediction model, the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days are synchronously input into the event-driven prediction model to run the event-driven prediction model and obtain the predicted consumption quantity of various foods for the target flight at the execution time of the meal preparation prediction operation. This also includes: before using the synchronous control equipment to synchronously input the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days into the event-driven prediction model, performing binary value conversion processing on the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days.

[0118] In each meal preparation prediction operation executed by the event-driven prediction model, the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days are synchronously input into the event-driven prediction model to run the event-driven prediction model and obtain the predicted consumption quantity of various foods of the target flight at the execution time of the meal preparation prediction operation. The predicted consumption quantity of various foods of the target flight at the execution time of the meal preparation prediction operation is also represented in binary numerical form.

[0119] Specifically, before using synchronous control equipment to synchronously input the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days into the event-driven prediction model, the binary numerical conversion processing of the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days includes: selecting a numerical conversion device to perform binary numerical conversion processing on the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days into the event-driven prediction model using synchronous control equipment.

[0120] The selection of a numerical conversion device for performing binary numerical conversion processing on the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food association data corresponding to the target flight over the past few days before synchronously inputting them into the event-driven prediction model using a synchronous control device includes: the numerical conversion device is connected to the synchronous control device and shares the same serial configuration bus;

[0121] For example, the numerical conversion device being connected to the synchronization control device and sharing the same serial configuration bus includes the following: the shared serial configuration bus can be an IIC configuration bus.

[0122] Sixth Embodiment

[0123] Figure 7 This is an internal structure diagram of a dynamic scheduling system for airport catering tasks based on microservices and event-driven methods, as shown in the sixth embodiment of the present invention.

[0124] like Figure 7 As shown, the multimedia content automatic generation and publishing system includes a memory and multiple processors. The memory stores a computer program, which is configured to be executed by the multiple processors to complete the following steps:

[0125] Step 701: Use a microservice architecture to parse the on-time departure time, full number of passengers, aircraft type number, and distance from the configured aircraft parking position to the food supply warehouse of the target flight, so as to use the corresponding configuration information of the target flight;

[0126] Specifically, microservices (or microservice architecture) is a cloud-native architectural approach that includes numerous loosely coupled and independently deployable small components or services within a single application. These services typically have their own technology stack, including databases and data management models; communicate with each other through a combination of REST APIs, event streams, and message brokers; and are organized according to business capabilities, with service delimiters often referred to as bounded contexts.

[0127] More specifically, microservices are characterized by easier code updates, allowing new features or functionalities to be added directly without updating the entire application. Teams can use different technology stacks and programming languages ​​for different components. Components can scale independently of each other, thereby reducing the waste and cost associated with having to scale the entire application (because a single function may face excessive load).

[0128] Step 702: Use a microservice architecture to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to each day of the target flight includes the delay duration of the target flight on that day, the actual number of passengers, and the consumption quantity of various types of food.

[0129] For example, a microservice architecture is used to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to each day of the target flight includes the delay duration of the target flight on that day, the actual number of passengers, and the consumption quantity of various foods. This includes using multiple processes under the microservice architecture to obtain multiple sets of food-related data corresponding to the target flight over the past few days.

[0130] Step 703: Perform multiple training operations on the feedforward neural network to obtain the feedforward neural network after multiple training operations and use it as the output of the event-driven prediction model;

[0131] For example, performing multiple training operations on a feedforward neural network to obtain a feedforward neural network after multiple training operations and outputting it as an event-driven prediction model includes: selecting a numerical simulation mode to complete the testing and simulation of the model building process of performing multiple training operations on the feedforward neural network to obtain a feedforward neural network after multiple training operations and outputting it as an event-driven prediction model.

[0132] Step 704: When the target flight takes off later than the scheduled departure time of the day, the event-driven prediction model is triggered to perform a catering prediction operation once every preset time length until the target flight actually takes off, so as to obtain the predicted consumption quantity of various food items of the target flight at the time of execution of the catering prediction operation. The preset time length is greater than or equal to the time length threshold.

[0133] For example, when the target flight departs after its scheduled departure time, the event-driven prediction model is triggered to perform a catering prediction operation once every preset time length until the target flight actually takes off, in order to obtain the predicted consumption quantity of various food items of the target flight at the time the catering prediction operation is executed. The preset time length is greater than or equal to the time length threshold, including: if the scheduled departure time of the target flight is 12:00 noon every day, and the flight has not taken off by 12:00 noon on the same day, it is considered that the target flight has entered a delay scenario. In this case, the event-driven prediction model is triggered to perform a catering prediction operation once every 15 minutes until the target flight actually takes off, in order to obtain the predicted consumption quantity of various food items of the target flight at the time the catering prediction operation is executed. The preset time length of 20 minutes is greater than or equal to the time length threshold of 15 minutes.

[0134] Specifically, the design of setting the preset time length to be greater than or equal to the time length threshold is to ensure that the interval between two adjacent meal preparation prediction operations is long enough to allow sufficient time for subsequent dynamic mobilization of ingredients and response time for ingredient preparation at the ingredient supply warehouse. For example, the interval between two adjacent meal preparation prediction operations is 15 minutes.

[0135] Step 705: When the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on board and the difference between the two is greater than or equal to a set difference threshold, an emergency dispatch request for that type of food is issued.

[0136] For example, when the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on board and the difference between the two is greater than or equal to a set difference threshold, issuing an emergency dispatch request for that type of food includes: when the predicted consumption of beef set meals exceeds the actual quantity of beef set meals on board and the difference between the two is greater than or equal to a set difference threshold, issuing an emergency dispatch request for beef set meals.

[0137] Conversely, when the predicted consumption of each type of food does not exceed the actual quantity of each type of food on board, or even if such a quantity exists but the difference between the two is greater than or equal to a set difference threshold, there is no need to issue an emergency dispatch request for any type of food.

[0138] As a further example, when the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on board and the difference between the two is greater than or equal to a set difference threshold, issuing an emergency dispatch request for that type of food also includes: the set difference threshold is set to 10 units. When the difference is less than 10 units, it indicates that on-site coordination can be easily carried out on the aircraft, and there is no need to issue an emergency dispatch request for that type of food.

[0139] In each meal preparation prediction operation executed by the event-driven prediction model, the event-driven prediction model intelligently predicts the predicted consumption of various food items on the target flight at the time of the meal preparation prediction operation based on the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the multiple food ingredient association data of the target flight over the past few days.

[0140] For example, in the scenario where the target flight is delayed, the event-driven prediction model is triggered every 15 minutes to execute a catering prediction operation until the target flight actually takes off. It can be seen that there is more than one execution time for the catering prediction operation, and it gradually increases and changes. For example, the execution times for each catering prediction operation can be 12:20, 12:40, 13:00, 13:20, and 13:40. If the target flight takes off at 13:50, there is no catering prediction operation execution time at 14:00. If the target flight takes off at 14:15, there is a catering prediction operation execution time at 14:00.

[0141] Among them, the process of training the feedforward neural network multiple times to obtain the feedforward neural network after multiple training sessions and using it as the output of the event-driven prediction model includes: the number of training sessions performed by the feedforward neural network is positively correlated with the delay duration of the target flight the day before.

[0142] For example, a positive correlation between the number of training iterations of the feedforward neural network and the delay duration of the target flight the previous day includes: if the delay duration of the target flight the previous day is 15 minutes, the feedforward neural network is trained 600 times; if the delay duration of the target flight the previous day is 20 minutes, the feedforward neural network is trained 700 times; if the delay duration of the target flight the previous day is 25 minutes, the feedforward neural network is trained 800 times; if the delay duration of the target flight the previous day is 30 minutes, the feedforward neural network is trained 900 times, and so on.

[0143] In each training iteration of the feedforward neural network, the time of execution of the catering prediction operation closest to the actual departure time and the on-time departure time of the target flight on a certain day is taken as the reference execution time. The known consumption quantity of various types of food of the target flight on a certain day is taken as the predicted consumption quantity of various types of food of the target flight at the reference execution time, and is used as the single output of the feedforward neural network. The reference execution time, the configuration information corresponding to the target flight, and the multiple sets of food association data corresponding to the target flight on the previous days before the stated date are taken as the item-by-item input of the feedforward neural network to complete this training of the feedforward neural network.

[0144] Among them, a microservice architecture is used to capture multiple sets of food-related data corresponding to the target flight over multiple days. The food-related data corresponding to each day of the target flight includes the delay time of the target flight on that day, the actual number of passengers, and the consumption of various types of food. The selected number of days in the past is positively correlated with the flight mileage of the target flight.

[0145] For example, a positive correlation between the selected number of past days and the target flight's mileage includes: a target flight mileage of 500 km and a selected number of past days of 7 days; a target flight mileage of 600 km and a selected number of past days of 9 days; a target flight mileage of 700 km and a selected number of past days of 11 days; a target flight mileage of 800 km and a selected number of past days of 13 days; a target flight mileage of 900 km and a selected number of past days of 15 days, and so on.

[0146] Furthermore, in a dynamic scheduling system for airport catering tasks based on microservices and event-driven mechanisms according to the present invention:

[0147] The positive correlation between the number of training operations performed by the feedforward neural network and the delay duration of the target flight the previous day includes: the feedforward neural network has a network architecture including multiple hidden layers, a single input layer and a single output layer, and the number of hidden layers of the feedforward neural network is positively correlated with the flight mileage of the target flight;

[0148] For example, the feedforward neural network has a network architecture including multiple hidden layers, a single input layer, and a single output layer, and the positive correlation between the number of hidden layers of the feedforward neural network and the flight mileage of the target flight includes: using an information transformation formula to represent the information transformation relationship between the number of hidden layers of the feedforward neural network and the flight mileage of the target flight.

[0149] Specifically, in the information conversion formula, the flight mileage of the target flight is the input information of the information conversion formula, and the number of hidden layers of the feedforward neural network that are positively correlated with the flight mileage of the target flight is the output information of the information conversion formula.

[0150] Furthermore, the feedforward neural network has a network architecture comprising multiple hidden layers, a single input layer, and a single output layer, and the number of hidden layers in the feedforward neural network is positively correlated with the flight mileage of the target flight, further comprising: in the feedforward neural network, the multiple hidden layers are located between the single input layer and the single output layer.

[0151] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0152] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software plus a general-purpose hardware platform, or of course, using hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device to execute the methods described in the various embodiments or some parts of the embodiments.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A dynamic scheduling system for airport catering tasks based on microservices and event-driven mechanisms, characterized in that, The system includes: The information parsing device uses a microservice architecture to parse the on-time departure time, full number of passengers, aircraft type number, and distance from the configured aircraft parking position to the food supply warehouse of the target flight, so as to use the corresponding configuration information of the target flight. The food capture device uses a microservice architecture to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to each day of the target flight includes the delay duration of the target flight, the actual number of passengers, and the consumption of various types of food on that day. A model building device is used to perform multiple training operations on a feedforward neural network to obtain a feedforward neural network after multiple training operations and output it as an event-driven prediction model. The meal preparation prediction device is connected to the information analysis device, the food capture device, and the model building device. When the target flight takes off later than the scheduled departure time of the day, the event-driven prediction model is triggered to perform a meal preparation prediction operation once every preset time length until the target flight actually takes off, so as to obtain the predicted consumption of various food items of the target flight at the time of execution of the meal preparation prediction operation. The preset time length is greater than or equal to the time length threshold. The dynamic scheduling device, connected to the meal prediction device, is used to issue an emergency scheduling request for a certain type of food when the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on the machine and the difference between the two is greater than or equal to a set difference threshold. In each meal preparation prediction operation executed by the event-driven prediction model, the event-driven prediction model intelligently predicts the predicted consumption of various food items on the target flight at the time of the meal preparation prediction operation based on the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the multiple food ingredient association data of the target flight over the past few days. Among them, the number of times the feedforward neural network performs training is positively correlated with the delay duration of the target flight the day before; In each training iteration of the feedforward neural network, the time of execution of the catering prediction operation closest to the actual departure time and the on-time departure time of the target flight on a certain day is taken as the reference execution time. The known consumption quantity of various types of food of the target flight on a certain day is taken as the predicted consumption quantity of various types of food of the target flight at the reference execution time, and is used as the single output of the feedforward neural network. The reference execution time, the configuration information corresponding to the target flight, and the multiple sets of food association data corresponding to the target flight on the previous days before the stated date are taken as the item-by-item input of the feedforward neural network to complete this training of the feedforward neural network. Among them, the number of past days selected is positively correlated with the flight mileage of the target flight; The feedforward neural network has a network architecture including multiple hidden layers, a single input layer, and a single output layer. The number of hidden layers is positively correlated with the flight mileage of the target flight. An information transformation formula is used to represent the information transformation relationship between the number of hidden layers and the flight mileage of the target flight.

2. The airport catering task dynamic scheduling system based on microservices and event-driven approach as described in claim 1, characterized in that, The system also includes: An event response device, set up at the food supply warehouse and connected to a dynamic scheduling device, is used to receive emergency scheduling requests for a certain type of food from a target flight using a microservice architecture, and to determine the quantity of a certain type of food that needs to be urgently replenished to the target flight based on the difference between the predicted consumption quantity of the certain type of food and the actual quantity of the certain type of food on board. The process of receiving an emergency dispatch request for a certain type of food from a target flight using a microservice architecture, and determining the quantity of a certain type of food that needs to be urgently replenished to the target flight based on the difference between the predicted consumption quantity of the certain type of food and the actual quantity of the certain type of food on board, includes: the emergency dispatch request for a certain type of food from the target flight includes the predicted consumption quantity of the certain type of food, the actual quantity of the certain type of food on board, and the flight number of the target flight.

3. The airport catering task dynamic scheduling system based on microservices and event-driven approach as described in claim 2, characterized in that, The system also includes: The response execution device is set at the food supply warehouse and connected to the event response device. It is used to take the quantity of a certain type of food that needs to be urgently replenished to the target flight as the target product quantity. When the production time of a certain type of food with the target product quantity is less than the preset time length, the production of the certain type of food is started. The response execution device, located at the food supply warehouse and connected to the event response device, is used to take the quantity of a certain type of food that needs to be urgently supplied to the target flight as the target product quantity. When the production time of a certain type of food with the target product quantity is less than a preset time length, the production of the certain type of food is initiated. The response execution device is connected to the event response device through a microservice architecture network. The response execution device is also used to temporarily suspend the production of a certain type of food when the production time of a certain type of food with a target product quantity is greater than or equal to a preset time length, in order to wait for the intelligent prediction result of the next meal preparation prediction operation execution time.

4. The airport catering task dynamic scheduling system based on microservices and event-driven approach as described in claim 1, characterized in that, The system also includes: The timing server is connected to the meal preparation prediction device, the information analysis device, the food ingredient capture device, and the model building device, respectively, to provide the timing services required by each of them.

5. The airport catering task dynamic scheduling system based on microservices and event-driven approach as described in claim 1, characterized in that, The system also includes: A model storage device, connected to a model building device, is used to receive and store the event-driven prediction model; The model storage device, connected to the model building device, is used to receive and store the event-driven prediction model, including: the model storage device completes the model storage of the event-driven prediction model by storing various model parameters of the event-driven prediction model.

6. A dynamic scheduling system for airport catering tasks based on microservices and event-driven methods as described in any one of claims 1-5, characterized in that: The positive correlation between the number of training iterations of the feedforward neural network and the delay duration of the target flight the previous day includes: using a numerical mapping function to represent the numerical mapping relationship between the number of training iterations of the feedforward neural network and the delay duration of the target flight the previous day; The numerical mapping function used to represent the positive correlation between the number of training operations performed by the feedforward neural network and the delay duration of the target flight the previous day includes: in the numerical mapping function, the delay duration of the target flight the previous day is used as the input value of the numerical mapping function, and the number of training operations performed by the feedforward neural network that is positively correlated with the delay duration of the target flight the previous day is used as the output value of the numerical mapping function. Among them, a microservice architecture is used to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to the target flight for each past day includes the delay time of the target flight on that day, the actual number of passengers, and the consumption of various types of food, including: the consumption of various types of food on that day includes the consumption of beef set meals, chicken set meals, mineral water bottles, and snacks. Specifically, when the target flight departs after the scheduled departure time of the day, the event-driven prediction model is triggered to perform a catering prediction operation once every preset time length until the target flight actually takes off, so as to obtain the predicted consumption quantity of various food items of the target flight at the time of execution of the catering prediction operation. The preset time length is greater than or equal to the time length threshold. Specifically, when the predicted consumption of a certain type of food exceeds the actual on-board quantity of that type of food and the difference between the two is greater than or equal to a set difference threshold, issuing an emergency dispatch request for that type of food includes: using a microservice architecture to issue an emergency dispatch request for that type of food when the predicted consumption of a certain type of food exceeds the actual on-board quantity of that type of food and the difference between the two is greater than or equal to a set difference threshold.

7. A dynamic scheduling system for airport catering tasks based on microservices and event-driven methods as described in any one of claims 1-5, characterized in that: In each meal preparation prediction operation executed by the event-driven prediction model, the event-driven prediction model intelligently predicts the predicted consumption quantity of various food items of the target flight at the time of the meal preparation prediction operation execution based on the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the multiple sets of food item association data corresponding to the target flight over the past several days. This includes: In each meal preparation prediction operation executed by the event-driven prediction model, the execution time of the meal preparation prediction operation, the configuration information of the target flight, and the multiple sets of food item association data corresponding to the target flight over the past several days are synchronously input into the event-driven prediction model to run the event-driven prediction model and obtain the predicted consumption quantity of various food items of the target flight at the time of the meal preparation prediction operation output by the event-driven prediction model. In each meal preparation prediction operation executed by the event-driven prediction model, the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days are synchronously input into the event-driven prediction model to run the event-driven prediction model and obtain the predicted consumption quantity of various foods of the target flight at the execution time of the meal preparation prediction operation. This includes using synchronous control equipment to synchronously input the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days into the event-driven prediction model. In each meal preparation prediction operation executed by the event-driven prediction model, the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days are synchronously input into the event-driven prediction model to run the event-driven prediction model and obtain the predicted consumption quantity of various foods for the target flight at the execution time of the meal preparation prediction operation. This also includes: before using the synchronous control equipment to synchronously input the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days into the event-driven prediction model, performing binary value conversion processing on the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days. In each meal preparation prediction operation executed by the event-driven prediction model, the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days are synchronously input into the event-driven prediction model to run the event-driven prediction model and obtain the predicted consumption quantity of various foods of the target flight at the execution time of the meal preparation prediction operation. The predicted consumption quantity of various foods of the target flight at the execution time of the meal preparation prediction operation is also represented in binary numerical form. Specifically, before using synchronous control equipment to synchronously input the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days into the event-driven prediction model, the binary numerical conversion processing of the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days includes: selecting a numerical conversion device to perform binary numerical conversion processing on the execution time of the meal preparation prediction operation, the configuration information corresponding to the target flight, and the multiple sets of food-related data corresponding to the target flight over the past few days into the event-driven prediction model using synchronous control equipment. The selection of a numerical conversion device for performing binary numerical conversion processing on the meal preparation prediction operation execution time, the configuration information corresponding to the target flight, and the multiple sets of food association data corresponding to the target flight over multiple days, before synchronously inputting them into the event-driven prediction model using a synchronous control device, includes: a scheduling system, the system comprising a memory and multiple processors, the memory storing a computer program configured to be executed by the multiple processors to complete the following steps: A microservice architecture is used to parse the on-time departure time, full number of passengers, aircraft type number, and distance from the configured aircraft parking position to the food supply warehouse of the target flight, so as to serve as the corresponding configuration information for the target flight; A microservice architecture is used to capture multiple sets of food-related data corresponding to the target flight over the past few days. The food-related data corresponding to each day of the target flight includes the delay time of the target flight, the actual number of passengers, and the consumption of various types of food on that day. The feedforward neural network is trained multiple times to obtain a feedforward neural network after multiple training sessions, which is then used as the output of the event-driven prediction model. When the target flight takes off later than the scheduled departure time of the day, the event-driven prediction model is triggered to perform a catering prediction operation once every preset time length until the target flight actually takes off, so as to obtain the predicted consumption of various food items of the target flight at the time of the catering prediction operation. The preset time length is greater than or equal to the time length threshold. When the predicted consumption of a certain type of food exceeds the actual quantity of that type of food on board, and the difference is greater than or equal to a set difference threshold, an emergency dispatch request for that type of food will be issued.

Images