A civil aviation airport ground support resource scheduling method, device and medium

Abstract

The application discloses a kind of civil aviation airport ground support resource scheduling method, equipment and medium, it is related to artificial intelligence technical field, including, obtain real-time flight plan, based on real-time flight plan and pre-defined support process rule base constructs support resource coupling net;Conflict prediction and conflict influence analysis are carried out to the support operation of future flight by traversing support resource coupling net, and structured conflict early warning report is generated;According to structured conflict early warning report triggers the on-demand scheduling mechanism of virtual resource pool of available resource on-demand query and distribution, available shared support resources in virtual resource pool are inquired, and directly executable conflict resolution scheme is generated;Conflict resolution scheme is executed and the execution result is monitored through Internet of Things data, and the execution result is fed back to support resource coupling net, and the time length parameter in support resource coupling net is calibrated.Support resource coupling net is constructed, and the foresight and collaborative robustness of ground support scheduling are improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method, equipment and medium for scheduling ground support resources at civil aviation airports. Background Technology

[0002] With the continuous expansion of civil aviation transport and the accelerating pace of operations, airport ground support faces complex collaborative challenges involving multiple flights with high concurrency, strong coupling of multiple resources, and strict time sequences for multiple processes. Support operations encompass multiple stages, including towing, catering, refueling, cleaning, and baggage handling. Each stage is highly sensitive to equipment, personnel, and time windows; resource mismatch or timing deviations at any node can trigger a chain reaction of delays. Under these circumstances, there is an urgent need for a scheduling mechanism that can dynamically perceive operational situations, accurately characterize resource dependencies, and possess proactive conflict intervention capabilities to support the evolution of the ground support system towards higher resilience, flexibility, and collaboration.

[0003] Existing scheduling methods generally lack the ability to structure and model resource competition among multiple flights when dealing with dynamic support needs. This makes it difficult to quantitatively assess the impact range and propagation path before conflicts occur, leading to delayed scheduling responses or a lack of targeted mitigation measures. To alleviate these problems, existing technologies typically employ preset buffer times or static resource redundancy strategies. While these can absorb local disturbances to some extent, they easily result in resource idleness or scheduling rigidity, making them unsuitable for real-time optimization needs in high-density operation scenarios. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for scheduling ground support resources at civil aviation airports to address the problem of difficulty in quantitatively assessing the scope and propagation path of impact before a conflict occurs.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for scheduling ground support resources at civil aviation airports, comprising,

[0008] Obtain real-time flight plans, and based on the real-time flight plans and a predefined support process rule base, construct a support resource coupling network that represents support events, support resources, and competition and dependency relationships;

[0009] By traversing the coupled network of support resources, conflict prediction and conflict impact analysis are performed on the support operations for future flights, generating a structured conflict early warning report;

[0010] Based on the structured conflict early warning report, the virtual resource pool is triggered by an on-demand scheduling mechanism that queries and allocates available resources as needed. The available shared protection resources in the virtual resource pool are queried, and a conflict resolution solution that can be executed directly is generated. The virtual resource pool aggregates shared resource information across protection units.

[0011] The conflict resolution scheme is implemented and the execution results are monitored through IoT data. The execution results are fed back to the resource protection coupling network, and the duration parameters in the resource protection coupling network are calibrated.

[0012] As a preferred embodiment of the civil aviation airport ground support resource scheduling method of the present invention, the construction of a support resource coupling network representing support events, support resources, and competition and dependency relationships is specifically as follows.

[0013] Real-time flight plans are obtained through standardized data interfaces, and semantic extraction is performed on the real-time flight plans to generate a standardized set of flight entity relationships.

[0014] By substituting a standardized set of flight entity relationships into a predefined support process rule base, and through rule matching, variable binding, and time window derivation, an instantiated support event sequence is generated for each flight.

[0015] By using multi-agent collaborative reasoning, the temporal conflict and spatial occupancy relationship of the guarantee events in the instantiated guarantee event sequence are analyzed, and the competition attribute and spatiotemporal dependency attribute are generated.

[0016] Competitive attributes and spatiotemporal dependency attributes are submitted to the Neo4j graph database to construct a resource coupling network.

[0017] As a preferred embodiment of the civil aviation airport ground support resource scheduling method of the present invention, the step of submitting competition attributes and spatiotemporal dependency attributes to the Neo4j graph database to construct a support resource coupling network is as follows.

[0018] Define the types of guaranteed event nodes, guaranteed resource nodes, and relation edge types representing contention attributes and spatiotemporal dependency attributes in the Neo4j graph database;

[0019] Create safeguard event nodes based on instantiated safeguard event sequences;

[0020] Based on the competition attribute, create competing relationship edges between guarantee event nodes that have conflicts, and based on the spatiotemporal dependency attribute, create spatiotemporal dependency relationship edges between guarantee event nodes that have constraints.

[0021] By organizing competitive relationships and spatiotemporal dependencies, a resource coupling network is formed to ensure resource security.

[0022] As a preferred embodiment of the civil aviation airport ground support resource scheduling method of the present invention, the generation of the structured conflict early warning report is specifically as follows.

[0023] Based on a preset future time window, extract the support operation network subgraph from the support resource coupling network;

[0024] Multiple graph propagation algorithms are executed in parallel on the support operations for future flights on the support operation network subgraph to simulate the propagation of initial disturbances and identify the initial conflict set;

[0025] Each conflict in the initial conflict set is taken as a disturbance source. The graph propagation algorithm is used to perform iterative disturbance propagation and conflict identification on the disturbance source until the graph propagation algorithm no longer identifies new conflicts, thus obtaining a complete set of potential conflicts.

[0026] The conflict data in the complete set of potential conflicts is parsed to generate conflict features, and the conflict features are mapped to predefined report entry fields as initial report entries;

[0027] The impact level of the initial report items is assessed according to the preset assessment rules, and the report items that have completed the impact level assessment are integrated and packaged according to the predefined structured data format to generate a structured conflict early warning report.

[0028] As a preferred embodiment of the civil aviation airport ground support resource scheduling method of the present invention, the simulated initial disturbance propagation refers to setting the planned time of future flights in the support operation network subgraph as the initial disturbance source, using the graph propagation algorithm to deduce the influence range of the initial disturbance source on the time and resource dependency relationship of adjacent support operation nodes on the support operation network subgraph, and identifying the set of affected support operations as the initial conflict set.

[0029] As a preferred embodiment of the civil aviation airport ground support resource scheduling method of the present invention, the step of obtaining a complete set of potential conflicts is specifically as follows.

[0030] Initialize the conflict propagation queue by adding conflict events from the initial conflict set to the conflict propagation queue and marking them as pending conflict events.

[0031] Take the pending conflict events from the conflict propagation queue as the current disturbance source, and identify the guarantee operation nodes affected by the current disturbance source as new conflict events through the graph propagation algorithm;

[0032] New conflict events are compared with the set of potential conflicts, and new conflict events not included in the set are added to the set of potential conflicts and the conflict propagation queue.

[0033] The system checks the status of the conflict propagation queue. If there are pending conflict events in the queue, it returns to the graph propagation algorithm for identification and continues iterating. If there are no pending conflict events in the queue, it outputs a complete set of potential conflicts.

[0034] As a preferred embodiment of the civil aviation airport ground support resource scheduling method of the present invention, the generation of a directly executable conflict resolution scheme is specifically as follows:

[0035] The conflict tasks are analyzed from the structured conflict warning report, and each conflict task is defined as a scheduling request object that includes resource requirement type, time window constraint and priority.

[0036] The time window constraint of the scheduling request object is used as the query condition. All shared guarantee resources available in the virtual resource pool within the time window are queried, and all available shared guarantee resources are aggregated to generate a feasible resource candidate set.

[0037] Multiple rounds of distributed negotiation are performed between conflicting tasks and feasible resource candidate sets. Conflicting tasks issue virtual bidding requests, resource nodes return feasible job time proposals, and conflicting tasks select the optimal proposal and confirm resource assignment based on optimization objectives.

[0038] Summarize the resource allocation results and corresponding job times of all conflicting tasks, verify global consistency, and then generate an executable conflict resolution solution.

[0039] As a preferred embodiment of the civil aviation airport ground support resource scheduling method of the present invention, the step of feeding back the execution result to the support resource coupling network and calibrating the duration parameter in the support resource coupling network is as follows.

[0040] Acquire real-time status data, synchronously execute conflict resolution schemes in the digital twin corresponding to the resource coupling network, inject real-time status data through IoT data interface, monitor execution results, and generate actual execution timeline;

[0041] By comparing the planned timeline in the digital twin with the actual execution timeline fed back by the IoT data interface, when a time deviation is detected, a causal discovery algorithm is activated to identify the causal relationship between the time deviation and the resource coupling network.

[0042] Based on causal relationships, the duration parameters in the resource coupling network are calibrated using targeted weighting.

[0043] Update the duration parameters after targeted weighted calibration to the resource coupling network to complete the parameter calibration.

[0044] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the civil aviation airport ground support resource scheduling method as described in the first aspect of the present invention.

[0045] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the civil aviation airport ground support resource scheduling method as described in the first aspect of the present invention.

[0046] The beneficial effects of this invention are as follows: By constructing a resource coupling network, explicit modeling of the dynamic coupling relationship between multiple flight support operations is realized. Flight plans, support rules, and resource occupancy status are uniformly mapped as nodes and relationship edges in a graph structure, supporting subsequent conflict cascading deduction and impact range quantification based on graph propagation. This provides a computable, traceable, and iteratively optimized structured decision-making basis for resource scheduling in high-density operation scenarios, achieving the ability to accurately identify potential chain disturbances and generate structured early warnings before conflicts occur, and significantly improving the foresight and collaborative robustness of ground support scheduling. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A flowchart for the scheduling method of ground support resources at civil aviation airports.

[0049] Figure 2 A flowchart for generating a structured conflict early warning report.

[0050] Figure 3 A flowchart for generating conflict resolution solutions.

[0051] Figure 4 A flowchart for implementing conflict resolution schemes and parameter calibration. Detailed Implementation

[0052] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0053] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0054] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0055] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a method for scheduling ground support resources at civil aviation airports, comprising the following steps:

[0056] S1. Obtain real-time flight plans. Based on the real-time flight plans and a predefined support process rule base, construct a support resource coupling network that represents support events, support resources, and competition and dependency relationships.

[0057] Real-time flight plans are obtained through a standardized data interface, and semantic extraction is performed on these plans to generate a standardized set of flight entity relationships. Specifically, the text content of the real-time flight plan is scanned line by line to identify specific elements such as flight number, departure and arrival times, airport codes, and resource requirements. Each identified element is labeled as an independent entity tag, and its position range in the text is recorded. The connection patterns between adjacent entity tags are checked according to preset entity association rules, such as direct adjacency between flight number and departure / arrival time or patterns connected by prepositions, to determine the dependency relationship between entities. The determined dependency relationship is converted into a triple form, such as flight number pointing to departure and arrival time. All entity tags and triple sets are then organized into a standardized set of flight entity relationships. The preset entity association rules are based on dependency parsing and pattern matching in natural language processing, including direct adjacency rules, preposition connection rules, and dependency relationship conversion rules, used to determine the dependency relationship between entities such as flight number and departure / arrival time.

[0058] A standardized set of flight entity relationships is substituted into a predefined support process rule base. Through rule matching, variable binding, and time window derivation, an instantiated support event sequence is generated for each flight. Specifically, after substituting the standardized set of flight entity relationships into the predefined support process rule base, each support rule in the predefined support process rule base is compared one by one with the entity tags and triples in the standardized set of flight entity relationships (i.e., the comparison is achieved by aligning the entity tags and triples in the standardized set of flight entity relationships with the placeholder structure of each support rule to check for position and relationship consistency). Completely matching rule entries are identified, such as the flight number entity tag matching the corresponding position of the flight number placeholder in the support rule, and the takeoff and landing time triples matching the dependency relationship of the time placeholder in the support rule. For each completely matching rule entry, the standardized set of flight entity relationships is further processed by aligning the entity tags and triples in the standardized set of flight entity relationships with the placeholder structure of each support rule. The entity values, such as the actual flight number and actual take-off and landing time, directly replace the corresponding placeholders in the rule entries to complete variable binding; the replaced rule entries are treated as bound rule entries; based on the time dependency instructions in the bound rule entries, subsequent event dependencies are sequentially linked starting from the actual take-off and landing time to expand the start and end ranges of each support event, forming a time window; all bound rule entries and corresponding time windows are sorted by flight number and chained together to form an instantiated support event sequence; the predefined support process rule base is set based on the requirements of civil aviation airport ground support (i.e., the operational conditions and sequence dependencies of flight support in airport ground services, used to ensure the orderliness and timeliness of resource allocation), including flight number matching rules, take-off and landing time placeholder rules, time dependency instruction rules, and event sequence sorting rules, which are used to generate instantiated support event sequences through rule matching, variable binding, and time window derivation.

[0059] Multi-agent collaborative reasoning is used to analyze the temporal conflicts and spatial occupancy relationships of security events in the instantiated security event sequence, generating competition attributes and spatiotemporal dependency attributes. Specifically, each security event is assigned to an agent to handle a specific security event in the instantiated security event sequence. The agent checks whether the start and end ranges of the security events in the instantiated security event sequence overlap with the start and end ranges of adjacent security events to identify temporal conflicts. For example, if the end range of security event A overlaps with the start range of security event B, it is marked as a temporal conflict. At the same time, the agent checks whether the resource occupancy positions of the security events in the instantiated security event sequence intersect with the resource occupancy positions of adjacent security events to identify spatial occupancy relationships. For example, if the boarding bridge position of security event A is the same as the boarding bridge position of security event B, it is marked as a spatial occupancy relationship. The agent exchanges the identified temporal conflicts and spatial occupancy relationships as preliminary attributes with another agent. The other agent verifies whether the exchanged preliminary attributes are consistent with the security events it is handling to confirm global consistency. The confirmed temporal conflicts are marked as competition attributes, and the spatial occupancy relationships are marked as spatiotemporal dependency attributes.

[0060] Competitive attributes and spatiotemporal dependency attributes are submitted to the Neo4j graph database to construct a resource coupling network.

[0061] Specifically, this involves defining the types of guaranteed event nodes, guaranteed resource nodes, and relation edge types representing contention attributes and spatiotemporal dependency attributes in the Neo4j graph database.

[0062] Assurance event nodes are created based on the instantiated assurance event sequence, and each assurance event node is configured with a unique identifier and time attribute. Specifically, each assurance event in the instantiated assurance event sequence is traversed and each assurance event is converted into an assurance event node using the Neo4j graph database creation command. At the same time, a unique identifier and time attribute are attached to each assurance event node to record the identity and time range of the assurance event.

[0063] Based on the competition attribute, competitive edges are created between guarantee event nodes with conflicts, and based on the spatiotemporal dependency attribute, spatiotemporal dependency edges are created between guarantee event nodes with constraints. Specifically: guarantee event nodes are traversed based on the competition attribute to identify pairs of guarantee event nodes with conflicts, and competitive edges are added between these pairs using the edge creation command of the Neo4j graph database; simultaneously, guarantee event nodes are traversed based on the spatiotemporal dependency attribute to identify pairs of guarantee event nodes with constraints, and spatiotemporal dependency edges are added between these pairs using the edge creation command of the Neo4j graph database.

[0064] By organizing competitive relationships and spatiotemporal dependencies, a resource coupling network is formed to ensure resource security.

[0065] S2. By traversing the resource coupling network, conflict prediction and conflict impact analysis are performed on the support operations for future flights, and a structured conflict early warning report is generated.

[0066] Based on a preset future time window, a subgraph of the support operation network is extracted from the support resource coupling network. The preset future time window is set based on the peak flight density and delay chain reaction cycle of civil aviation airports. For example, a 1-hour future window can capture resource conflicts in advance and optimize scheduling response. Setting it to 1 hour can ensure the timeliness of prediction and the balance of resource utilization. If a shorter time is set, the long-term chain effect will be ignored, resulting in a delayed response. If a longer time is set, irrelevant nodes will be expanded, increasing the analysis burden.

[0067] Multiple graph propagation algorithms are executed in parallel on the support operations for future flights in the support operation network subgraph to simulate the propagation of initial disturbances and identify the initial conflict set. Simulating the propagation of initial disturbances means setting the planned time of future flights in the support operation network subgraph as the initial disturbance source, and using graph propagation algorithms to deduce the impact range of the initial disturbance source on the time and resource dependencies of adjacent support operation nodes in the support operation network subgraph, thereby identifying the set of affected support operations as the initial conflict set. Among them, the graph propagation algorithm is an iterative propagation mechanism based on graph theory, which gradually spreads the initial signal or disturbance through the relationship between nodes to simulate the chain reaction in the network.

[0068] Each conflict in the initial conflict set is taken as a disturbance source. The graph propagation algorithm is used to perform iterative disturbance propagation and conflict identification on the disturbance source until the graph propagation algorithm no longer identifies new conflicts, thereby obtaining a complete set of potential conflicts.

[0069] Specifically, the process involves: initializing a conflict propagation queue, adding conflict events from the initial conflict set to the queue and marking them as pending conflict events; the conflict propagation queue is a temporary ordered list used to store conflict events from the initial conflict set and mark them as pending events to support the iterative identification of new conflict events by the graph propagation algorithm; it is an auxiliary tool customized based on the existing queue data structure.

[0070] The process involves retrieving pending conflict events from the conflict propagation queue as current disturbance sources, and using a graph propagation algorithm to identify support operation nodes affected by these current disturbance sources as new conflict events. Specifically, it involves sequentially traversing adjacent support operation nodes in the support operation network subgraph from the current disturbance source and checking whether the time and resource dependency relationship exceeds a preset time dependency threshold to mark the affected support operation nodes as new conflict events. The preset time dependency threshold is set based on the average delay response cycle and resource conflict sensitivity of civil aviation airport ground support operations. For example, a value of more than 5 minutes of time overlap or more than 80% resource occupancy can capture potential chain reactions in a timely manner and avoid oversensitivity. If a smaller value is used, it is easy to trigger false alarms and increase unnecessary adjustment burden. If a larger value is used, it may miss early conflicts and lead to an amplification of overall delays.

[0071] The process involves comparing new conflict events with the potential conflict set, and adding any new conflict events not already included to the potential conflict set and the conflict propagation queue. Specifically, this is done by checking the unique identifier of a new conflict event against the unique identifiers of existing conflict events in the potential conflict set to determine if it is already included, thereby adding any inconsistent new conflict events to the potential conflict set and the conflict propagation queue. The potential conflict set is formed by iteratively simulating the cascading impact of each conflict on the subgraph of the assurance operation network using a graph propagation algorithm, continuously identifying and accumulating newly generated conflicts.

[0072] The system checks the status of the conflict propagation queue. If there are pending conflict events in the queue, it returns to the graph propagation algorithm for identification and continues iterating. If there are no pending conflict events in the queue, it outputs a complete set of potential conflicts.

[0073] The conflict data in the complete set of potential conflicts is parsed to generate conflict features, which are then mapped to predefined report entry fields as initial report entries. Specifically, the type identifier, the list of involved support operation nodes, and the description of the scope of impact for each conflict event in the complete set of potential conflicts are extracted as basic attributes. These extracted basic attributes are combined into a structured conflict feature description. For example, the type identifier is linked to the list of involved support operation nodes and an impact scope description is appended to form a complete feature string. The type identifier in the conflict feature description is directly mapped to the conflict type classification position in the predefined report entry field; the list of involved support operation nodes is directly mapped to the positions of the involved flights and resource objects in the predefined report entry field; the impact scope description is directly mapped to the predicted occurrence time position in the predefined report entry field; and a unique sequence number is generated for each conflict event as a unique conflict identifier position. All the mapped positions are then filled into the initial report entries. The predefined report entry fields are set according to the standard format of structured conflict early warning reports and include a unique conflict identifier, conflict type classification, involved flights and resource objects, and a predicted occurrence time field, which are used to map conflict features to generate initial report entries.

[0074] The system assesses the impact level of initial report items according to preset assessment rules, and integrates and encapsulates the report items that have completed the impact level assessment in a predefined structured data format (such as JSON format) to generate a structured conflict early warning report containing a unique conflict identifier, conflict type classification, involved flights and resource objects, and predicted occurrence time. The preset assessment rules are set based on the severity and chain reaction risk of civil aviation airport ground support conflicts, including time delay level rules, resource impact range rules, flight priority rules, and overall delay cost rules, which are used to assess the impact level of initial report items to generate a structured conflict early warning report.

[0075] S3. Based on the structured conflict early warning report, trigger the on-demand scheduling mechanism of the virtual resource pool to query and allocate available resources as needed. Query the available shared protection resources in the virtual resource pool and generate a conflict resolution solution that can be executed directly. The virtual resource pool aggregates shared resource information across protection units.

[0076] The conflict tasks are analyzed from the structured conflict warning report, and each conflict task is defined as a scheduling request object that includes resource requirement type, time window constraint and priority.

[0077] The time window constraint of the scheduling request object is used as the query condition to query all shared guaranteed resources available in the virtual resource pool within the time window, and to aggregate all available shared guaranteed resources to generate a feasible resource candidate set. The virtual resource pool is a virtual collection of dynamically aggregated idle shared guaranteed resources, used to query and allocate available resources on demand to generate a feasible resource candidate set; it is a custom auxiliary structure based on existing resource management concepts.

[0078] A multi-round distributed negotiation process is performed between conflicting tasks and the feasible resource candidate set. Conflicting tasks issue virtual bidding requests with time windows, and resource nodes return feasible job time proposals. Conflicting tasks select the optimal proposal based on optimization objectives and confirm resource assignment. Specifically: each conflicting task broadcasts a virtual bidding request with time window constraints to all resource nodes in the feasible resource candidate set to solicit available time proposals. Each resource node, upon receiving a virtual bidding request, checks the intersection of its available time range and the time window constraint to generate a feasible job time proposal and returns it to the conflicting task. The conflicting task collects all returned feasible job time proposals and compares their delay, cost, and efficiency metrics according to task priority and preset global scheduling rules to select the proposal with the lowest delay, optimal cost, and highest efficiency as the optimal proposal. The conflicting task sends a confirmation message to the resource node providing the optimal proposal to lock the resource assignment and notify the others. Resource nodes cancel proposals, and multiple rounds of distributed negotiation repeat the above broadcast, return, comparison, and confirmation loop until all conflicting tasks have completed resource allocation to ensure a globally balanced distribution. The optimization objective refers to decision-making criteria determined based on task priorities and preset global scheduling rules, including delay minimization, cost optimization, and efficiency maximization. Task priorities are evaluated based on flight priority rules (i.e., classification rules based on historical delay statistics) in structured conflict warning reports, including factors such as flight type (e.g., cargo priority), time delay sensitivity, and chain reaction risk, forming a high, medium, and low three-level classification to guide resource allocation. Preset global scheduling rules are set based on the overall optimization objectives of civil aviation airport ground support and historical scheduling data, including delay minimization rules, resource utilization balance rules, cost control rules, and priority weighting rules, used to compare the delay degree, cost indicators, and efficiency indicators of proposals in distributed negotiation to achieve globally optimal allocation.

[0079] Summarize the resource assignment results and corresponding job times of all conflicting tasks, and after verifying global consistency (i.e., by checking whether the resource number is uniquely assigned in the resource assignment results of all conflicting tasks and whether the start and end ranges of the corresponding job times do not overlap), generate an executable conflict resolution scheme containing resource number, job start time and end time.

[0080] S4. Execute the conflict resolution scheme and monitor the execution results through IoT data, feed the execution results back to the resource protection coupling network, and calibrate the duration parameters in the resource protection coupling network.

[0081] Real-time status data is acquired, and conflict resolution schemes are synchronously executed in the digital twin corresponding to the resource coupling network. Real-time status data is injected through an IoT data interface to monitor execution results and generate an actual execution timeline. Specifically, conflict resolution schemes are synchronously executed in the digital twin corresponding to the resource coupling network. Simulation is initiated by mapping the resource number and job start and end times in the conflict resolution scheme to the corresponding guarantee event nodes in the digital twin. At the same time, real-time status data is injected through an IoT data interface to update the position and status changes of the guarantee event nodes in the digital twin in real time. Execution results are monitored, and the progress of the injected real-time status data and the synchronously executed guarantee event nodes in the digital twin are continuously compared to record the actual start and end times of each resource number, forming an actual execution timeline for subsequent deviation comparison.

[0082] By comparing the planned timeline in the digital twin with the actual execution timeline fed back by the IoT data interface, when a time deviation is detected, a causal discovery algorithm is activated to identify the causal relationship between the time deviation and the guarantee resource coupling network. Specifically, it checks whether the planned start and end times of each guarantee event in the planned timeline are completely consistent with the actual start and end times of the corresponding guarantee event in the actual execution timeline. For example, if the planned start time of guarantee event A in the planned timeline is 14:00, but the actual start time of guarantee event A in the actual execution timeline is 14:05, this is marked as a start time deviation. Similarly, if the planned end time of guarantee event A in the planned timeline is 14:30, but the actual end time of guarantee event A in the actual execution timeline is 14:35, this is marked as an end time deviation. When any start time deviation or end time deviation is marked, the set of all marked deviations is taken as the time deviation, and the causal discovery algorithm is activated to analyze how the time deviation is traced back to the influence chain of the upstream guarantee event node through the competitive relationship edge and the spatiotemporal dependency relationship edge in the guarantee resource coupling network, thereby identifying the causal relationship between the time deviation and the guarantee resource coupling network.

[0083] Based on causal relationships, targeted weighted calibration is performed on the duration parameters in the resource guarantee coupling network. Specifically, the competitive relationship edges and spatiotemporal dependency relationship edges in the resource guarantee coupling network are traversed according to causal relationships to locate the duration parameters affected by time deviations, and the weight values ​​of the affected duration parameters are adjusted to match the deviation direction in the actual execution timeline, thereby completing the targeted weighted calibration.

[0084] The expression for directional weighted calibration is:

[0085] ;

[0086] In the formula, The new duration parameter weights after calibration. The old duration parameter weights before calibration. Preset learning rate (e.g., 0 < <1, used to control the adjustment range), Causality strength factor (normalized influence coefficient calculated based on causal discovery algorithm, 0 ≤ ≤1 indicates the relative contribution of the deviation to propagation through the graph edges. This is a sign function used to ensure that the direction of weight adjustment is consistent with the sign of the deviation. For the edge (That is, the actual execution time deviation of the associated guarantee event (i.e., the competitive relationship edge or spatiotemporal dependency edge in the guarantee resource coupling network), usually in minutes, with positive values ​​indicating delays and negative values ​​indicating advances).

[0087] Update the duration parameters after targeted weighted calibration to the resource coupling network to complete the parameter calibration.

[0088] This embodiment also provides a computer device applicable to the civil aviation airport ground support resource scheduling method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the civil aviation airport ground support resource scheduling method proposed in the above embodiment.

[0089] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0090] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the method for scheduling ground support resources at civil aviation airports as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0091] In summary, this invention achieves explicit modeling of the dynamic coupling relationships between multiple flight support operations by constructing a resource coupling network. It maps flight plans, support rules, and resource occupancy status into nodes and relationship edges in a graph structure, supporting subsequent conflict cascading deduction and impact range quantification based on graph propagation. This provides a computable, traceable, and iteratively optimized structured decision-making basis for resource scheduling in high-density operation scenarios, achieving the ability to accurately identify potential chain disturbances and generate structured early warnings before conflicts occur, significantly improving the foresight and collaborative robustness of ground support scheduling.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for scheduling ground support resources of a civil airport, characterized in that: include, Obtain real-time flight plans, and based on the real-time flight plans and a predefined support process rule base, construct a support resource coupling network that represents support events, support resources, and competition and dependency relationships; By traversing the coupled network of support resources, conflict prediction and conflict impact analysis are performed on the support operations for future flights, generating a structured conflict early warning report; Based on the structured conflict early warning report, the virtual resource pool is triggered by an on-demand scheduling mechanism that queries and allocates available resources as needed. The available shared protection resources in the virtual resource pool are queried, and a conflict resolution solution that can be executed directly is generated. The virtual resource pool aggregates shared resource information across protection units. The conflict resolution scheme is implemented and the execution results are monitored through IoT data. The execution results are fed back to the resource protection coupling network, and the duration parameters in the resource protection coupling network are calibrated. The construction of the security resource coupling network, which represents security events, security resources, and competition and dependency relationships, is detailed below. Real-time flight plans are obtained through standardized data interfaces, and semantic extraction is performed on the real-time flight plans to generate a standardized set of flight entity relationships. By substituting a standardized set of flight entity relationships into a predefined support process rule base, and through rule matching, variable binding, and time window derivation, an instantiated support event sequence is generated for each flight. By using multi-agent collaborative reasoning, the temporal conflict and spatial occupancy relationship of the guarantee events in the instantiated guarantee event sequence are analyzed, and the competition attribute and spatiotemporal dependency attribute are generated. Submit the competition attributes and spatiotemporal dependency attributes to the Neo4j graph database to construct a resource coupling network to ensure resource security. The generation of the structured conflict early warning report is detailed below. Based on a preset future time window, extract the support operation network subgraph from the support resource coupling network; Multiple graph propagation algorithms are executed in parallel on the support operations for future flights on the support operation network subgraph to simulate the propagation of initial disturbances and identify the initial conflict set; Each conflict in the initial conflict set is taken as a disturbance source. The graph propagation algorithm is used to perform iterative disturbance propagation and conflict identification on the disturbance source until the graph propagation algorithm no longer identifies new conflicts, thus obtaining a complete set of potential conflicts. The conflict data in the complete set of potential conflicts is parsed to generate conflict features, and the conflict features are mapped to predefined report entry fields as initial report entries; The impact level of the initial report items is assessed according to the preset assessment rules, and the report items that have completed the impact level assessment are integrated and packaged according to the predefined structured data format to generate a structured conflict early warning report. The complete set of potential conflicts is obtained as follows: Initialize the conflict propagation queue by adding conflict events from the initial conflict set to the conflict propagation queue and marking them as pending conflict events. Take the pending conflict events from the conflict propagation queue as the current disturbance source, and identify the guarantee operation nodes affected by the current disturbance source as new conflict events through the graph propagation algorithm; New conflict events are compared with the set of potential conflicts, and new conflict events not included in the set are added to the set of potential conflicts and the conflict propagation queue. The system checks the status of the conflict propagation queue. If there are pending conflict events in the queue, it returns to the graph propagation algorithm for identification and continues iterating. If there are no pending conflict events in the queue, it outputs a complete set of potential conflicts.

2. The method of claim 1, wherein: The process of submitting competition attributes and spatiotemporal dependency attributes to the Neo4j graph database to construct a resource-guaranteed coupling network is detailed below. Define the types of guaranteed event nodes, guaranteed resource nodes, and relation edge types representing contention attributes and spatiotemporal dependency attributes in the Neo4j graph database; Create safeguard event nodes based on instantiated safeguard event sequences; Based on the competition attribute, create competing relationship edges between guarantee event nodes that have conflicts, and based on the spatiotemporal dependency attribute, create spatiotemporal dependency relationship edges between guarantee event nodes that have constraints. By organizing competitive relationships and spatiotemporal dependencies, a resource coupling network is formed to ensure resource security.

3. The method of claim 1, wherein: The simulated initial disturbance propagation refers to setting the planned time of future flights in the support operation network subgraph as the initial disturbance source, using a graph propagation algorithm to deduce the impact range of the initial disturbance source on the time and resource dependencies of adjacent support operation nodes on the support operation network subgraph, and identifying the set of affected support operations as the initial conflict set.

4. The method of claim 1, wherein: The generated, directly executable conflict resolution solution is as follows: The conflict tasks are analyzed from the structured conflict warning report, and each conflict task is defined as a scheduling request object that includes resource requirement type, time window constraint and priority. The time window constraint of the scheduling request object is used as the query condition. All shared guarantee resources available in the virtual resource pool within the time window are queried, and all available shared guarantee resources are aggregated to generate a feasible resource candidate set. Multiple rounds of distributed negotiation are performed between conflicting tasks and feasible resource candidate sets. Conflicting tasks issue virtual bidding requests, resource nodes return feasible job time proposals, and conflicting tasks select the optimal proposal and confirm resource assignment based on optimization objectives. Summarize the resource allocation results and corresponding job times of all conflicting tasks, verify global consistency, and then generate an executable conflict resolution solution.

5. The method for scheduling civil aviation airport ground support resources as described in claim 1, characterized in that: The execution results are then fed back to the resource guarantee coupling network, and the duration parameters in the resource guarantee coupling network are calibrated, as detailed below. Acquire real-time status data, synchronously execute conflict resolution schemes in the digital twin corresponding to the resource coupling network, inject real-time status data through IoT data interface, monitor execution results, and generate actual execution timeline; By comparing the planned timeline in the digital twin with the actual execution timeline fed back by the IoT data interface, when a time deviation is detected, a causal discovery algorithm is activated to identify the causal relationship between the time deviation and the resource coupling network. Based on causal relationships, the duration parameters in the resource coupling network are calibrated using targeted weighting. Update the duration parameters after targeted weighted calibration to the resource coupling network to complete the parameter calibration.

6. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the civil aviation airport ground support resource scheduling method according to any one of claims 1 to 5.

7. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by a processor to implement the steps of the civil aviation airport ground support resource scheduling method of any one of claims 1-5.

Images