Intelligent airport apron job scheduling method and system based on dynamic monitoring

Abstract

The invention relates to an intelligent airport apron operation scheduling method and system based on dynamic monitoring. The method comprises the following steps: acquiring airport apron operation data; based on the airport apron operation data, constructing a digital twinborn model reflecting the real-time operation state of the airport apron; based on the digital twinborn model, describing the sequence of operation links by adopting a directed acyclic graph, and calculating the earliest start time and the latest completion time of each link through a critical path method; based on the sequence of the operation links and the earliest start time and the latest completion time of each link, constructing a multi-objective optimization model of airport apron operation scheduling; and solving the multi-objective optimization model of airport apron job scheduling to obtain an airport apron job scheduling scheme. According to the invention, intelligent scheduling and dynamic optimization of airport apron operation are realized, the flight turnover efficiency is effectively improved, the operation safety risk is reduced, and powerful support is provided for airport operation management.

Description

Technical Field

[0001] The present invention relates to the technical field of apron operation scheduling, and in particular to a smart apron operation scheduling method and system based on dynamic monitoring. Background Art

[0002] Apron operation scheduling involves multiple links such as aircraft docking, refueling, loading and unloading, cleaning, and maintenance. There are time dependencies and spatial constraints between each link. Due to the limited space in the apron area and the frequent activities of various support vehicles and personnel, how to ensure the turnover rate of aircraft while taking into account safety and economy is a complex combinatorial optimization problem. The traditional scheduling model relies on manual experience and lacks scientific and quantitative means, making it difficult to cope with the challenges of increasing flight density and increasing support requirements.

[0003] With the construction of the apron management system, although a huge amount of business data has been accumulated, due to the lack of unified planning, inconsistent data formats and incompatible interfaces between systems, the data island phenomenon is serious, and data interconnection and integrated application cannot be achieved. In addition, due to the lack of real-time and efficient data collection methods, the acquired data has poor timeliness, making it difficult to reflect the dynamic changes of apron operations and provide a reliable basis for scheduling optimization.

[0004] Therefore, there is an urgent need to study a smart apron operation scheduling method based on dynamic monitoring to perceive the location and status information of apron personnel, vehicles, and equipment in real time, and assist managers in making scientific decisions. Summary of the invention

[0005] The purpose of the present invention is to propose a smart apron operation scheduling method and system based on dynamic monitoring to solve the problems existing in the above-mentioned prior art.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A smart apron operation scheduling method based on dynamic monitoring, comprising:

[0008] Collecting apron operation data; wherein the apron operation data includes: location information of apron operation vehicles and activity status data of personnel;

[0009] Based on the apron operation data, a digital twin model reflecting the real-time operation status of the apron is constructed;

[0010] Based on the digital twin model, a directed acyclic graph is used to describe the sequence of the operation links, and the earliest start time and the latest completion time of each link are calculated by the critical path method; wherein the directed acyclic graph is used to characterize the time dependency relationship between each operation link;

[0011] Based on the sequence of the operation links and the earliest start time and the latest completion time of each link, a multi-objective optimization model for apron operation scheduling is constructed;

[0012] The multi-objective optimization model of the apron operation scheduling is solved to obtain the apron operation scheduling plan.

[0013] Optionally, the directed acyclic graph includes: aircraft arrival time, passenger disembarkation time, luggage unloading time and refueling time;

[0014] Among them, aircraft arrival means passengers disembarking, passenger disembarking means luggage unloading, and luggage unloading means refueling.

[0015] Optionally, constructing a multi-objective optimization model for apron operation scheduling includes:

[0016] Obtaining historical apron operation scheduling data; wherein the historical apron operation scheduling data includes flight turnaround time and key indicators of security cost;

[0017] Based on the historical apron operation scheduling data, a nonlinear integer programming model for apron operation scheduling, i.e., a multi-objective optimization model, is established. The multi-objective optimization model includes the number of personnel, vehicles, equipment resources and operation time constraints;

[0018] Determining objective function coefficients and constraint condition parameters in the multi-objective optimization model by analyzing the historical apron operation scheduling data;

[0019] The weighted summation method is used to transform the two optimization objectives of minimizing flight turnaround time and minimizing guarantee cost in the multi-objective optimization model into a single-objective optimization problem; wherein, the minimization of flight turnaround time is obtained by the sequence of the operation links and the earliest start time and latest completion time of each link.

[0020] Optionally, solving the multi-objective optimization model for apron operation scheduling includes:

[0021] By using the improved genetic algorithm and integer coding, the scheduling plan of the personnel, vehicles and equipment of each flight is encoded into an integer string;

[0022] For the integer string, a new feasible solution is generated through a crossover mutation operation, and the quality of the feasible solution is evaluated through a fitness function;

[0023] Obtaining a set of Pareto optimal solution sets through iterative search of the improved genetic algorithm;

[0024] For the obtained Pareto optimal solution set, the fuzzy comprehensive evaluation method is used to combine the two indicators of minimizing flight turnaround time and minimizing guarantee cost to determine the optimal scheduling plan.

[0025] Optionally, obtaining the apron operation scheduling plan includes:

[0026] Based on the optimal scheduling scheme, the actual operation data of each guarantee link is obtained, and compared with the planned data in the scheduling scheme to determine whether the actual operation requirements are met; if the actual operation data of a link does not match the scheduling scheme, the dynamic adjustment of the scheduling scheme is triggered; by taking the actual operation data as a constraint condition, the multi-objective optimization model is re-solved to obtain a new scheduling scheme;

[0027] Repeat the comparison and adjustment process, and continuously revise the scheduling plan through iterative optimization until a feasible solution that meets the actual work requirements is found; determine the various parameters in the final scheduling plan; the parameters include: the start time and completion time of each support link, and the allocation plan of personnel, vehicles, and equipment.

[0028] A smart apron operation adjustment system based on dynamic monitoring, the system comprising: a collection module, a first construction module, a calculation module, a second construction module, and a solution generation module;

[0029] The acquisition module is used to collect apron operation data; wherein the apron operation data includes: location information of apron operation vehicles and activity status data of personnel;

[0030] The first construction module is used to construct a digital twin model reflecting the real-time operation status of the apron based on the apron operation data;

[0031] The calculation module is used to describe the sequence of operation links using a directed acyclic graph based on the digital twin model, and calculate the earliest start time and the latest completion time of each link by the critical path method; wherein the directed acyclic graph is used to characterize the time dependency relationship between each operation link;

[0032] The second construction module is used to construct a multi-objective optimization model for apron operation scheduling based on the sequence of the operation links and the earliest start time and the latest completion time of each link;

[0033] The solution generation module is used to solve the multi-objective optimization model of the apron operation scheduling to obtain the apron operation scheduling solution.

[0034] Optionally, the directed acyclic graph includes: aircraft arrival time, passenger disembarkation time, luggage unloading time and refueling time;

[0035] Among them, aircraft arrival means passengers disembarking, passenger disembarking means luggage unloading, and luggage unloading means refueling.

[0036] Optionally, the second building block includes:

[0037] Obtaining historical apron operation scheduling data; wherein the historical apron operation scheduling data includes flight turnaround time and key indicators of support cost;

[0038] Based on the historical apron operation scheduling data, a nonlinear integer programming model for apron operation scheduling, i.e., a multi-objective optimization model, is established. The multi-objective optimization model includes the number of personnel, vehicles, equipment resources and operation time constraints;

[0039] By analyzing the historical apron operation scheduling data, determining the objective function coefficients and constraint condition parameters in the multi-objective optimization model;

[0040] The weighted summation method is used to transform the two optimization objectives of minimizing flight turnaround time and minimizing guarantee cost in the multi-objective optimization model into a single-objective optimization problem; wherein, the minimization of flight turnaround time is obtained by the sequence of the operation links and the earliest start time and latest completion time of each link.

[0041] Optionally, the solution generation module includes:

[0042] By using the improved genetic algorithm and integer coding, the scheduling plan of the personnel, vehicles and equipment of each flight is encoded into an integer string;

[0043] For the integer string, a new feasible solution is generated through a crossover mutation operation, and the quality of the feasible solution is evaluated through a fitness function;

[0044] Obtaining a set of Pareto optimal solution sets through iterative search of the improved genetic algorithm;

[0045] For the obtained Pareto optimal solution set, the fuzzy comprehensive evaluation method is used to combine the two indicators of minimizing flight turnaround time and minimizing guarantee cost to determine the optimal scheduling plan.

[0046] Optionally, the solution generation module further includes:

[0047] Based on the optimal scheduling scheme, the actual operation data of each guarantee link is obtained, and compared with the planned data in the scheduling scheme to determine whether the actual operation requirements are met; if the actual operation data of a link does not match the scheduling scheme, the dynamic adjustment of the scheduling scheme is triggered; by taking the actual operation data as a constraint condition, the multi-objective optimization model is re-solved to obtain a new scheduling scheme;

[0048] Repeat the comparison and adjustment process, and continuously revise the scheduling plan through iterative optimization until a feasible solution that meets the actual work requirements is found; determine the various parameters in the final scheduling plan; the parameters include: the start time and completion time of each support link, and the allocation plan of personnel, vehicles, and equipment.

[0049] The beneficial effects of the present invention are:

[0050] The present invention discloses an intelligent scheduling method and system for apron operations based on dynamic monitoring. The apron operation data is collected in real time through Internet of Things devices, and a digital twin model reflecting the real-time operating status of the apron is constructed. On this basis, the dependencies between the operation links are analyzed, a multi-objective optimization model is established, and an improved genetic algorithm is used to generate the optimal scheduling plan. The scheduling plan is sent to the site through a mobile terminal, and the operation progress is tracked in real time. If deviations are found, timely warnings and adjustments are made. At the same time, operation data is continuously collected, and the scheduling model is adaptively optimized. The present invention realizes intelligent scheduling and dynamic optimization of apron operations, effectively improves flight turnover efficiency, reduces operation safety risks, and provides strong support for airport operation management. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0052] Figure 1 The present invention is a flowchart of a smart apron operation scheduling method based on dynamic monitoring according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0055] The traditional apron operation scheduling model mainly relies on manual experience and simple scheduling rules, lacking scientific quantitative means and real-time data support. The traditional apron operation scheduling model usually involves the following links: aircraft docking, refueling, loading and unloading, cleaning, maintenance, etc.

[0056] 1. Scheduling Process

[0057] 1. Scheduling preparation

[0058] Information collection: Dispatchers collect flight information, including estimated time of arrival, aircraft type, required services, etc., by telephone, radio or on-site observation.

[0059] Resource allocation: Based on flight information, the dispatcher pre-allocates the required ground service resources, such as tow trucks, boarding stairs, refueling trucks, etc.

[0060] Make a preliminary plan: The dispatcher makes a preliminary operation plan based on experience and flight information, including the approximate start and end time of each link.

[0061] 2. Implementation of scheduling

[0062] Flight arrival: After the plane arrives, the dispatcher adjusts the operation plan according to the actual situation, such as reallocating resources or adjusting the order of operations.

[0063] Operation execution: Ground service personnel perform various operations according to the dispatcher's instructions, such as guiding the aircraft to dock, refueling, loading and unloading luggage, etc.

[0064] Monitoring and Adjustment: The dispatcher continuously monitors the progress of the operation and makes adjustments based on the actual situation, such as handling delays and coordinating resource conflicts.

[0065] 3. Homework completed

[0066] Job confirmation: After each job is completed, the operator needs to report to the dispatcher so that the dispatcher can update the job status.

[0067] Flight Pushback: After all ground services are completed, the aircraft is pushed out and prepared for takeoff.

[0068] Recording and Feedback: The dispatcher records the completion of the job and optimizes future dispatch plans based on the feedback.

[0069] 2. Decision-making Basis

[0070] Rules of thumb: Dispatchers rely on personal experience and some simple rules to make decisions, such as "first come, first served", "large aircraft first", etc.

[0071] Resource availability: Dispatchers need to consider the availability of resources, such as vehicles, personnel, equipment, etc., to ensure that operations can proceed smoothly.

[0072] Safety regulations: Dispatchers must comply with airport safety regulations to ensure that all operations are carried out under safe conditions.

[0073] 3. Existing Limitations

[0074] Poor data dependence: The traditional dispatching model lacks real-time data support, and dispatching decisions mainly rely on the dispatcher's personal experience, which makes it difficult to cope with the complex and changeable apron operating environment.

[0075] Inefficiency: Due to the lack of scientific quantitative methods, traditional scheduling models are often inefficient and difficult to optimize resource allocation, resulting in resource waste and operation delays.

[0076] Poor adaptability: Faced with challenges such as increased flight density and higher security requirements, the traditional scheduling model is difficult to adapt quickly and difficult to achieve dynamic adjustment and optimization.

[0077] Safety hazards: Due to the lack of real-time monitoring and early warning mechanisms, the traditional dispatching model has safety hazards and it is difficult to detect and handle abnormal situations in a timely manner.

[0078] Specific implementation cases

[0079] Assume that a medium-sized airport has 100 flights taking off and landing every day and needs to schedule apron operations. The following are the specific implementation steps:

[0080] Scheduling preparation:

[0081] Information collection: Dispatchers obtain information such as the estimated arrival time of flights, aircraft models, and required services through the airport information system. At the same time, through on-site observation and telephone communication, they collect information on resource availability, such as tow trucks, boarding stairs, and refueling trucks.

[0082] Resource allocation: Based on flight information and resource availability, dispatchers pre-allocate resources. For example, for large aircraft, two tugs may be required; for flights that require a quick turnaround, resources are allocated first.

[0083] Make a preliminary plan: The dispatcher makes a preliminary operation plan based on experience and flight information. For example, for a flight expected to arrive at 10 a.m., it is planned to start refueling at 10:15 and start loading and unloading baggage at 10:30.

[0084] Implementation of Scheduling:

[0085] Flight arrival: After the plane arrives, the dispatcher adjusts the operation plan according to the actual situation. For example, if the plane arrives early, the dispatcher may need to allocate resources in advance; if the plane is delayed, the dispatcher may need to rearrange resources to avoid resource conflicts.

[0086] Operation execution: Ground service personnel perform various operations according to the dispatcher's instructions. For example, guiding the aircraft to stop at a designated location, refueling, loading and unloading luggage, etc.

[0087] Monitor and adjust: The dispatcher continuously monitors the progress of the operation and makes adjustments based on the actual situation. For example, if the refueling operation is delayed, the dispatcher may need to coordinate other operations to ensure that the aircraft can be pushed out on time.

[0088] Homework Completed:

[0089] Operation confirmation: After each operation is completed, the operator needs to report to the dispatcher so that the dispatcher can update the operation status. For example, after refueling is completed, the tanker driver reports to the dispatcher, and the dispatcher updates the status on the dispatch board.

[0090] Flight Pushout: After all ground services are completed, the aircraft is pushed out and prepared for takeoff. The dispatcher needs to ensure that all operations are completed and the aircraft meets the conditions for takeoff.

[0091] Recording and feedback: The dispatcher records the completion of the work and optimizes future scheduling plans based on the feedback. For example, if a certain link is found to be frequently delayed, the dispatcher may need to re-evaluate the resource allocation strategy.

[0092] The traditional apron operation dispatching mode has many limitations in actual operation, such as poor data dependence, low efficiency, poor adaptability and safety hazards. However, with the development of information technology, airports need to adopt data-driven intelligent dispatching systems to improve dispatching efficiency and safety.

[0093] In this regard, Figure 1 As shown, this embodiment proposes a smart apron operation scheduling method based on dynamic monitoring, including:

[0094] Collect apron operation data; the apron operation data includes: location information of apron operation vehicles and activity status data of personnel;

[0095] Based on the apron operation data, a digital twin model reflecting the real-time operation status of the apron is constructed;

[0096] Based on the digital twin model, a directed acyclic graph is used to describe the sequence of work links, and the earliest start time and the latest completion time of each link are calculated through the critical path method; among them, the directed acyclic graph is used to characterize the time dependency relationship between each work link;

[0097] Based on the sequence of operation links and the earliest start time and latest completion time of each link, a multi-objective optimization model for apron operation scheduling is constructed;

[0098] The multi-objective optimization model of apron operation scheduling is solved to obtain the apron operation scheduling plan.

[0099] Specifically, in this embodiment, collecting apron operation data includes:

[0100] The apron area is an important part of the airport. Its safe and efficient operation is crucial to ensuring flight punctuality and passenger service quality. Using technologies such as the Internet of Things, big data, and artificial intelligence to build an intelligent apron operation monitoring system can effectively improve the safety and efficiency of apron operations. First, it is necessary to reasonably deploy various IoT sensors in the apron area, such as high-definition cameras, radio frequency identification readers, Beidou positioning terminals, etc., to build a full-range, multi-angle, and no-dead-angle data collection network. For example, high-definition cameras are installed at key locations such as boarding bridges, baggage conveyor belts, and aircraft parking spaces to monitor the activities of people and vehicles in real time; radio frequency identification tags are installed on luggage carts to track the transportation trajectory of luggage in real time; Beidou positioning terminals are installed on apron operation vehicles to obtain vehicle location information in real time. This can fully perceive the dynamic information of apron operations and provide a basis for subsequent data analysis. Secondly, the massive heterogeneous data collected is pre-processed. For example, the video data collected by the camera needs to be image recognized and tracked to extract the location information of people and vehicles; the data collected by the radio frequency identification reader needs to be cleaned to remove duplicate and erroneous data; the data collected by the Beidou positioning terminal needs to be coordinate converted and time synchronized. Through pre-processing operations such as data cleaning and normalization, data noise and redundancy are eliminated, and the data is converted into a unified standard format to form a standardized apron operation monitoring data set. For example, different types of sensor data are uniformly converted into standard formats such as timestamps, location coordinates, and object types for unified analysis and processing. This is done to improve data quality and avoid the impact of dirty data on subsequent analysis results. Then, the spatiotemporal data mining algorithm is used to analyze the location change trend of apron operation vehicles and the activity trajectory pattern of personnel. For example, by analyzing the trajectory data of vehicles over a period of time, the operation rules of vehicles can be identified, and it can be determined whether there are abnormal behaviors such as speeding, reverse driving, and crossing the boundary. By analyzing the activity trajectory of personnel, the operation process of personnel can be identified, and it can be determined whether there are abnormal behaviors such as illegal operations and detention. Suppose a certain apron vehicle should travel from point A to point B within the specified time, but the actual trajectory shows that it detoured point C, which may be an abnormal behavior and requires further investigation. For example, if a staff member stays in a restricted area for too long, there may also be safety hazards. This can timely discover potential safety risks and avoid accidents. If abnormal behavior of vehicles or personnel is detected, the system will immediately trigger an early warning mechanism. For example, the system will display abnormal situations in real time on the large screen of the monitoring center, and push warning information to relevant managers through text messages, emails, etc. For example, if the system detects that a vehicle is speeding, it will immediately mark the location of the vehicle on the map and send a warning message to the apron dispatcher so that the dispatcher can take timely measures. This can shorten the emergency response time and minimize losses.

[0101] Specifically, in this embodiment, building a digital twin model that reflects the real-time operating status of the apron includes:

[0102] Using 3D modeling technology, a high-precision virtual apron 3D model is constructed based on real-time data of the apron scene; dynamic information such as vehicle location, personnel location, and equipment status is mapped to the virtual apron model to achieve information synchronization between the virtual model and the actual apron; apron operation monitoring data is continuously acquired to determine whether the actual apron environment has changed. If so, the dynamic update of the virtual model is triggered; through the dynamic update of the virtual model, the virtual model and the actual apron environment are kept in real time synchronization to ensure that the virtual model can accurately reflect the real-time operating status of the apron; based on the synchronously updated virtual apron model, digital twin technology is used to build a digital twin system associated with the physical apron mapping; machine learning algorithms are used to analyze historical monitoring data, extract apron operation rules, optimize virtual model parameters, and improve the accuracy and predictive ability of the digital twin system.

[0103] Furthermore, the directed acyclic graph includes: aircraft arrival time, passenger disembarkation time, baggage unloading time, and refueling time;

[0104] Among them, aircraft arrival means passengers disembarking, passenger disembarking means luggage unloading, and luggage unloading means refueling.

[0105] Specifically, in this embodiment, the directed acyclic graph can clearly show the dependency relationship between each operation link. For example, passengers can only disembark after the plane arrives, luggage can only be unloaded after passengers are unloaded, and refueling can only be started after luggage is unloaded. These operation links can be used as nodes in the graph, and the dependency relationship between them can be represented by directed edges, for example, "plane arrival" points to "passenger disembarkation", "passenger disembarkation" points to "luggage unloading", and "luggage unloading" points to "refueling". With the directed acyclic graph model, the earliest start time and the latest completion time of each operation link can be calculated. For example, assuming that the time of "plane arrival" is 8 o'clock, "passenger disembarkation" takes 30 minutes, "luggage unloading" takes 45 minutes, and "refueling" takes 1 hour. Then, the earliest start time of "passenger disembarkation" is 8 o'clock, and the "latest completion time" is 8:30. The earliest start time of "luggage unloading" is 8:30, and the latest completion time is 9:15. The earliest start time of "refueling" is 9:15, and the latest completion time is 10:15. By calculating the earliest start time and the latest finish time of each node, the critical path method can be used to determine the critical path in the apron operation process. In the above example, all links on the path of "aircraft arrival" -> "passenger disembarkation" -> "baggage unloading" -> "refueling" are critical paths, because any delay in any link will cause a delay in the entire process. After determining the critical path, you can focus on optimizing the operation links on the critical path.

[0106] Furthermore, the multi-objective optimization model for apron operation scheduling is constructed including:

[0107] Obtain historical apron operation scheduling data; among which, historical apron operation scheduling data includes flight turnaround time and key indicators of support cost;

[0108] Based on the historical apron operation scheduling data, a nonlinear integer programming model for apron operation scheduling, namely a multi-objective optimization model, is established. The multi-objective optimization model includes the number of personnel, vehicles, equipment resources and operation time constraints;

[0109] By analyzing the historical apron operation scheduling data, the objective function coefficients and constraint condition parameters in the multi-objective optimization model are determined;

[0110] The weighted summation method is used to transform the two optimization objectives of minimizing flight turnaround time and minimizing guarantee cost in the multi-objective optimization model into a single-objective optimization problem; among them, the minimization of flight turnaround time is achieved through the sequence of operation links and the earliest start time and latest completion time of each link.

[0111] Specifically, in this embodiment, a multi-objective optimization model for apron operation scheduling is constructed, and historical apron operation scheduling data is obtained, the historical apron operation scheduling data including key indicators of flight turnaround time and security cost; based on the historical apron operation scheduling data, a nonlinear integer programming model for apron operation scheduling is established, the nonlinear integer programming model including the number of personnel, vehicles, equipment resources and operation time constraints; by analyzing the historical apron operation scheduling data, the objective function coefficients and constraint condition parameters in the nonlinear integer programming model are determined; the weighted sum method is used to transform the two optimization objectives of minimizing flight turnaround time and minimizing security cost in the nonlinear integer programming model into a single-objective optimization problem.

[0112] Furthermore, solving the multi-objective optimization model of apron operation scheduling includes:

[0113] The improved genetic algorithm is used to encode the scheduling plan of personnel, vehicles and equipment of each flight into an integer string by integer coding; wherein the improved genetic algorithm is: introducing the elite retention strategy and adaptive crossover mutation probability into the genetic algorithm;

[0114] For integer strings, new feasible solutions are generated through crossover and mutation operations, and the quality of feasible solutions is evaluated through fitness functions;

[0115] Through iterative search of improved genetic algorithm, a set of Pareto optimal solutions is obtained;

[0116] For the obtained Pareto optimal solution set, the fuzzy comprehensive evaluation method is used to determine the optimal scheduling plan by combining the two indicators of minimizing flight turnaround time and minimizing guarantee cost.

[0117] Specifically, a genetic algorithm is an optimization algorithm that simulates the natural evolution process. The introduction of elite retention strategy and adaptive crossover mutation probability in the improved genetic algorithm can improve the algorithm's search efficiency and solution quality. The elite retention strategy refers to directly copying the best individuals in each generation to the next generation to ensure that the algorithm does not lose the excellent solutions that have been found. The adaptive crossover mutation probability refers to dynamically adjusting the probability of crossover and mutation according to the algorithm's search process to balance the algorithm's global search ability and local search ability. In the encoding method of the improved genetic algorithm, integer encoding is used. For example, the scheduling plan for each flight can be encoded as an integer string, where each integer represents the allocation of a resource. Through the crossover mutation operation, a new feasible solution can be generated. For example, by exchanging some genes of two different scheduling plans, a new scheduling plan can be generated. The fitness function is used to evaluate the pros and cons of feasible solutions. The higher the value of the fitness function, the better the scheduling plan. Through the iterative search of the improved genetic algorithm, a set of Pareto optimal solutions can be obtained. The Pareto optimal solution set refers to a set of solutions that cannot be improved any further, and no other goal can be improved without sacrificing one goal. For the Pareto optimal solution set obtained, the fuzzy comprehensive evaluation method can be used to combine the two indicators of flight turnaround time and guarantee cost to determine the final optimal scheduling plan. Fuzzy comprehensive evaluation method is a comprehensive evaluation method that considers multiple factors. It can combine qualitative and quantitative factors to comprehensively evaluate different plans. Finally, the optimal scheduling plan is converted into an executable work plan. By connecting with the apron operation management system, detailed work tasks and time schedules for personnel, vehicles, and equipment can be automatically generated. For example, the system can automatically generate work instructions for each tractor, including which parking position to go to, which flight to tow, and the estimated completion time.

[0118] Obtaining the apron operation scheduling plan includes:

[0119] Based on the optimal scheduling plan, the actual operation data of each guarantee link is obtained, and compared with the planned data in the scheduling plan to determine whether it meets the actual operation requirements; if the actual operation data of a link does not match the scheduling plan, the dynamic adjustment of the scheduling plan is triggered; by using the actual operation data as a constraint condition, the multi-objective optimization model is re-solved to obtain a new scheduling plan;

[0120] Repeat the comparison and adjustment process, and continuously revise the scheduling plan through iterative optimization until a feasible solution that meets the actual work requirements is found; determine the various parameters in the final scheduling plan; the parameters include: the start time and completion time of each support link, and the allocation plan of personnel, vehicles, and equipment.

[0121] Specifically, in this embodiment, during the execution of the initial scheduling scheme, the collection of actual operation data is crucial. These data include the start and completion time of each operation link, as well as the actual number of resources used. By comparing these actual data with the planned data, the management team can evaluate the efficiency of the operation and the accuracy of resource allocation. For example, if the data shows that a certain operation link is delayed, it may be necessary to adjust the subsequent operation plan or increase resource investment. When the actual operation data does not match the plan, the scheduling scheme needs to be dynamically adjusted. This adjustment is achieved by re-entering the actual data into the optimization model and running the improved genetic algorithm again. In this way, new scheduling schemes can be quickly generated to cope with changing operating conditions. For example, if the unloading operation of a flight is delayed due to weather reasons, the system can automatically reallocate the ground handling team and equipment to ensure that the flight resumes normal operation progress as soon as possible. Through continuous iterative optimization, the scheduling scheme gradually approaches the optimal solution, and finally forms a scheduling scheme that meets the actual operation requirements and optimizes resource utilization. This scheme will specify the time arrangement and resource allocation of each guarantee link in detail, providing a clear execution guide for ground operations. For example, the final plan may indicate that in order to meet the flight density during the morning peak period, additional ground handling vehicles and personnel need to be dispatched during a specific period of time. During the execution of the operation, continuous data monitoring and real-time comparison are essential. This not only helps monitor the progress of the operation, but also responds quickly when problems arise. For example, if the monitoring system finds that the execution speed of a key operation link is lower than expected, the system can immediately notify the manager and recommend adjustments to resource allocation or operation methods. In order to support this complex scheduling system, it is very useful to establish a detailed database. This database records all relevant parameters and results of each scheduling, providing data support for future scheduling decisions. By analyzing historical data, machine learning algorithms can be used to discover potential improvements in the scheduling model, thereby continuously improving the efficiency and accuracy of the scheduling algorithm. For example, by analyzing historical scheduling data, it is possible to find which resource configurations are most effective under specific conditions, so that these configurations can be given priority in future scheduling.

[0122] This embodiment also proposes a smart apron operation adjustment system based on dynamic monitoring, the system includes: a collection module, a first construction module, a calculation module, a second construction module, and a solution generation module;

[0123] The collection module is used to collect apron operation data; the apron operation data includes: location information of apron operation vehicles and activity status data of personnel;

[0124] The first construction module is used to build a digital twin model reflecting the real-time operation status of the apron based on the apron operation data;

[0125] The calculation module is used to describe the sequence of operation links using a directed acyclic graph based on the digital twin model, and calculate the earliest start time and the latest completion time of each link through the critical path method; the directed acyclic graph is used to characterize the time dependency relationship between each operation link;

[0126] The second building module is used to build a multi-objective optimization model for apron operation scheduling based on the sequence of operation links and the earliest start time and latest completion time of each link;

[0127] The solution generation module is used to solve the multi-objective optimization model of apron operation scheduling and obtain the apron operation scheduling solution.

[0128] Furthermore, the directed acyclic graph includes: aircraft arrival time, passenger disembarkation time, baggage unloading time, and refueling time;

[0129] Among them, aircraft arrival means passengers disembarking, passenger disembarking means luggage unloading, and luggage unloading means refueling.

[0130] Furthermore, the second building block includes:

[0131] Obtain historical apron operation scheduling data; among which, historical apron operation scheduling data includes flight turnaround time and key indicators of support cost;

[0132] Based on the historical apron operation scheduling data, a nonlinear integer programming model for apron operation scheduling, namely a multi-objective optimization model, is established. The multi-objective optimization model includes the number of personnel, vehicles, equipment resources and operation time constraints;

[0133] By analyzing the historical apron operation scheduling data, the objective function coefficients and constraint condition parameters in the multi-objective optimization model are determined;

[0134] The weighted summation method is used to transform the two optimization objectives of minimizing flight turnaround time and minimizing guarantee cost in the multi-objective optimization model into a single-objective optimization problem; among them, the minimization of flight turnaround time is achieved through the sequence of operation links and the earliest start time and latest completion time of each link.

[0135] Furthermore, the solution generation module includes:

[0136] By using the improved genetic algorithm and integer coding, the scheduling plan of the personnel, vehicles and equipment of each flight is encoded into an integer string;

[0137] For integer strings, new feasible solutions are generated through crossover and mutation operations, and the quality of feasible solutions is evaluated through fitness functions;

[0138] Through iterative search of improved genetic algorithm, a set of Pareto optimal solutions is obtained;

[0139] For the obtained Pareto optimal solution set, the fuzzy comprehensive evaluation method is used to determine the optimal scheduling plan by combining the two indicators of minimizing flight turnaround time and minimizing guarantee cost.

[0140] Furthermore, the solution generation module also includes:

[0141] Based on the optimal scheduling plan, the actual operation data of each guarantee link is obtained, and compared with the planned data in the scheduling plan to determine whether it meets the actual operation requirements; if the actual operation data of a link does not match the scheduling plan, the dynamic adjustment of the scheduling plan is triggered; by using the actual operation data as a constraint condition, the multi-objective optimization model is re-solved to obtain a new scheduling plan;

[0142] Repeat the comparison and adjustment process, and continuously revise the scheduling plan through iterative optimization until a feasible solution that meets the actual work requirements is found; determine the various parameters in the final scheduling plan; the parameters include: the start time and completion time of each support link, and the allocation plan of personnel, vehicles, and equipment.

[0143] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A smart apron operation scheduling method based on dynamic monitoring, characterized in that: include: Collecting apron operation data; wherein the apron operation data includes: location information of apron operation vehicles and activity status data of personnel; Based on the apron operation data, a digital twin model reflecting the real-time operation status of the apron is constructed; Based on the digital twin model, a directed acyclic graph is used to describe the sequence of the operation links, and the earliest start time and the latest completion time of each link are calculated by the critical path method; wherein the directed acyclic graph is used to characterize the time dependency relationship between each operation link; Based on the sequence of the operation links and the earliest start time and the latest completion time of each link, a multi-objective optimization model for apron operation scheduling is constructed; The multi-objective optimization model of the apron operation scheduling is solved to obtain the apron operation scheduling plan.

2. The intelligent apron operation scheduling method based on dynamic monitoring according to claim 1 is characterized in that: The directed acyclic graph includes: aircraft arrival time, passenger disembarkation time, luggage unloading time and refueling time; Among them, aircraft arrival means passengers disembarking, passenger disembarking means luggage unloading, and luggage unloading means refueling.

3. The intelligent apron operation scheduling method based on dynamic monitoring according to claim 1 is characterized in that: The multi-objective optimization model for apron operation scheduling includes: Obtaining historical apron operation scheduling data; wherein the historical apron operation scheduling data includes flight turnaround time and key indicators of support cost; Based on the historical apron operation scheduling data, a nonlinear integer programming model for apron operation scheduling, i.e., a multi-objective optimization model, is established. The multi-objective optimization model includes the number of personnel, vehicles, equipment resources and operation time constraints; By analyzing the historical apron operation scheduling data, determining the objective function coefficients and constraint condition parameters in the multi-objective optimization model; The weighted summation method is used to transform the two optimization objectives of minimizing flight turnaround time and minimizing guarantee cost in the multi-objective optimization model into a single-objective optimization problem; wherein, the minimization of flight turnaround time is obtained by the sequence of the operation links and the earliest start time and latest completion time of each link.

4. The intelligent apron operation scheduling method based on dynamic monitoring according to claim 3 is characterized in that: Solving the multi-objective optimization model for the apron operation scheduling includes: The improved genetic algorithm is used to encode the scheduling plan of personnel, vehicles and equipment of each flight into an integer string by integer coding; wherein the improved genetic algorithm is: introducing the elite retention strategy and adaptive crossover mutation probability into the genetic algorithm; For the integer string, a new feasible solution is generated through a crossover mutation operation, and the quality of the feasible solution is evaluated through a fitness function; Obtaining a set of Pareto optimal solution sets through iterative search of the improved genetic algorithm; For the obtained Pareto optimal solution set, the fuzzy comprehensive evaluation method is used to combine the two indicators of minimizing flight turnaround time and minimizing guarantee cost to determine the optimal scheduling plan.

5. The intelligent apron operation scheduling method based on dynamic monitoring according to claim 4 is characterized in that: Obtaining the apron operation scheduling plan includes: Based on the optimal scheduling scheme, the actual operation data of each guarantee link is obtained, and compared with the planned data in the scheduling scheme to determine whether the actual operation requirements are met; if the actual operation data of a link does not match the scheduling scheme, the dynamic adjustment of the scheduling scheme is triggered; by taking the actual operation data as a constraint condition, the multi-objective optimization model is re-solved to obtain a new scheduling scheme; Repeat the comparison and adjustment process, and continuously revise the scheduling plan through iterative optimization until a feasible solution that meets the actual work requirements is found; determine the various parameters in the final scheduling plan; the parameters include: the start time and completion time of each support link, and the allocation plan of personnel, vehicles, and equipment.

6. A smart apron operation adjustment system based on dynamic monitoring, characterized in that: Used to implement the intelligent apron operation scheduling method based on dynamic monitoring as described in any one of claims 1 to 5, the system includes: a collection module, a first construction module, a calculation module, a second construction module, and a solution generation module; The acquisition module is used to collect apron operation data; wherein the apron operation data includes: location information of apron operation vehicles and activity status data of personnel; The first construction module is used to construct a digital twin model reflecting the real-time operation status of the apron based on the apron operation data; The calculation module is used to describe the sequence of operation links using a directed acyclic graph based on the digital twin model, and calculate the earliest start time and the latest completion time of each link by the critical path method; wherein the directed acyclic graph is used to characterize the time dependency relationship between each operation link; The second construction module is used to construct a multi-objective optimization model for apron operation scheduling based on the sequence of the operation links and the earliest start time and the latest completion time of each link; The solution generation module is used to solve the multi-objective optimization model of the apron operation scheduling to obtain the apron operation scheduling solution.

7. The intelligent apron operation dispatching system based on dynamic monitoring according to claim 6 is characterized in that: The directed acyclic graph includes: aircraft arrival time, passenger disembarkation time, luggage unloading time and refueling time; Among them, aircraft arrival means passengers disembarking, passenger disembarking means luggage unloading, and luggage unloading means refueling.

8. The intelligent apron operation dispatching system based on dynamic monitoring according to claim 6 is characterized in that: The second building block comprises: Obtaining historical apron operation scheduling data; wherein the historical apron operation scheduling data includes flight turnaround time and key indicators of support cost; Based on the historical apron operation scheduling data, a nonlinear integer programming model for apron operation scheduling, i.e., a multi-objective optimization model, is established. The multi-objective optimization model includes the number of personnel, vehicles, equipment resources and operation time constraints; By analyzing the historical apron operation scheduling data, determining the objective function coefficients and constraint condition parameters in the multi-objective optimization model; The weighted summation method is used to transform the two optimization objectives of minimizing flight turnaround time and minimizing guarantee cost in the multi-objective optimization model into a single-objective optimization problem; wherein, the minimization of flight turnaround time is obtained by the sequence of the operation links and the earliest start time and latest completion time of each link.

9. The intelligent apron operation dispatching system based on dynamic monitoring according to claim 8 is characterized in that: The solution generation module includes: By using the improved genetic algorithm and integer coding, the scheduling plan of the personnel, vehicles and equipment of each flight is encoded into an integer string; For the integer string, a new feasible solution is generated through a crossover mutation operation, and the quality of the feasible solution is evaluated through a fitness function; Obtaining a set of Pareto optimal solution sets through iterative search of the improved genetic algorithm; For the obtained Pareto optimal solution set, the fuzzy comprehensive evaluation method is used to combine the two indicators of minimizing flight turnaround time and minimizing guarantee cost to determine the optimal scheduling plan.

10. The intelligent apron operation dispatching system based on dynamic monitoring according to claim 9 is characterized in that: The solution generation module also includes: Based on the optimal scheduling scheme, the actual operation data of each guarantee link is obtained, and compared with the planned data in the scheduling scheme to determine whether the actual operation requirements are met; if the actual operation data of a link does not match the scheduling scheme, the dynamic adjustment of the scheduling scheme is triggered; by taking the actual operation data as a constraint condition, the multi-objective optimization model is re-solved to obtain a new scheduling scheme; Repeat the comparison and adjustment process, and continuously revise the scheduling plan through iterative optimization until a feasible solution that meets the actual work requirements is found; determine the various parameters in the final scheduling plan; the parameters include: the start time and completion time of each support link, and the allocation plan of personnel, vehicles, and equipment.

Images