A method, apparatus, equipment, and medium for generating flight missions based on causality

Abstract

This application provides a causal-based method, apparatus, device, and medium for generating flight tasks, applicable to the field of task generation technology, and capable of solving the problem of low execution efficiency in existing generated flight tasks. The method includes: mapping target flight data to obtain corresponding vectors and determining the flight task nodes corresponding to each vector; connecting any two flight task nodes to obtain an initial node graph, and performing independence checks on each connecting line in the initial node graph to obtain the check results; adjusting the connecting lines according to the check results and a preset connection rule base to obtain an adjusted initial node graph, and adjusting the node weights of the flight task nodes according to the target flight data and a preset data threshold to obtain a target node graph; and determining the flight tasks to be executed corresponding to the target flight data based on the target node graph. This improves the execution efficiency of the generated flight tasks.

Description

[0001] A method, apparatus, equipment, and medium for generating flight missions based on causality Technical Field

[0002] This application relates to the field of mission generation technology, and in particular to a method, apparatus, device and medium for generating flight missions based on causality. Background Technology

[0003] Flight support tasks refer to a series of ground operations and service activities that must be performed on a flight during its stopover at an airport to prepare it for takeoff again. These include guiding the aircraft to its dock, assisting passengers with boarding and disembarking, cabin cleaning and meal replenishment, aircraft security checks, refueling, baggage and cargo loading, unloading and transfer, and secondary cabin security checks. By determining the shortest task flow for the current flight, task execution efficiency can be improved.

[0004] Existing methods for generating flight missions are generally based on fixed templates, describing the support process through predefined standardized procedures and simple timing constraints. For example, fixed mission lists are generated based on different aircraft types, flight types, or turnaround times.

[0005] However, existing methods for generating flight missions suffer from low execution efficiency. Summary of the Invention

[0006] This application provides a causality-based method, apparatus, device, and medium for generating flight missions, in order to solve the problem of low execution efficiency in existing flight mission generation methods.

[0007] Firstly, this application provides a causal-based flight mission generation method, the method comprising:

[0008] Map the target flight data to obtain the corresponding vectors, and determine the flight task nodes corresponding to each vector;

[0009] Connect any two flight mission nodes to obtain an initial node graph, and perform an independence test on each connecting line in the initial node graph to obtain the test results;

[0010] Based on the test results and the preset connection rule base, the connection lines are adjusted to obtain the adjusted initial node graph. Then, based on the target flight data and the preset data threshold, the node weights of the flight task nodes are adjusted to obtain the target node graph.

[0011] Based on the target node diagram, determine the flight tasks to be executed corresponding to the target flight data.

[0012] In some embodiments of this application, target flight data is mapped to obtain corresponding vectors, including:

[0013] Acquire real-time task data, and perform missing value imputation, outlier handling, and normalization on the real-time task data to obtain target flight data;

[0014] The target flight data is mapped to obtain multiple vectors.

[0015] In some embodiments of this application, independence checks are performed on each connecting line in the initial node graph to obtain the check results, including:

[0016] For the target flight task node, other flight task nodes are combined in order to obtain multiple sets of node variables; the target flight task node is the two flight task nodes corresponding to the current connection line.

[0017] Based on the order of the node variable set, the node state of the flight task node in each node variable set is changed sequentially.

[0018] Based on the PC algorithm, the independence coefficients between target flight task nodes are calculated under different node states in each set of node variables, and the independent coefficients are compared with the preset independence threshold to obtain the comparison results.

[0019] Based on the comparison results, determine the inspection results corresponding to each connecting line.

[0020] In some embodiments of this application, the inspection result corresponding to each connection line is determined based on the comparison results, including:

[0021] Determine the comparison results;

[0022] If the comparison result shows that there is at least one independent coefficient that is less than the preset independent threshold, then the test result corresponding to the current connector is determined to be a non-associated connector.

[0023] If the comparison result shows that there is no independent coefficient less than the preset independent threshold, then the test result corresponding to the current connector is determined to be an associated connector.

[0024] Determine the test results for each connecting line.

[0025] In some embodiments of this application, the connection lines are adjusted based on the test results and a preset connection rule base to obtain an adjusted initial node graph, including:

[0026] Determine the connection rules corresponding to each connection line, and determine whether there is a target connection rule in each connection rule that is consistent with the preset connection rules in the preset connection rule library; the connection rules are used to represent the task logic between the corresponding flight task nodes;

[0027] If so, then the connecting line corresponding to the target connecting rule is determined as the target connecting line, and other connecting lines are determined as non-target connecting lines;

[0028] If not, then all connecting lines are determined to be non-target connecting lines;

[0029] Determine the test results for each non-target connection line, and delete the non-target connection lines whose test results are not associated connection lines to obtain the adjusted initial node diagram.

[0030] In some embodiments of this application, the node weights of flight task nodes are adjusted based on target flight data and preset data thresholds to obtain a target node graph, including:

[0031] Determine whether there is any task data to be adjusted in the target flight data that exceeds a preset data threshold;

[0032] If so, determine the flight task node to be adjusted corresponding to the task data to be adjusted, and calculate the adjusted node weight based on the node weight of the flight task node to be adjusted, the task data to be adjusted, and the preset data threshold multiplier, to obtain the target node graph.

[0033] In some embodiments of this application, determining the flight task to be executed corresponding to the target flight data based on the target node graph includes:

[0034] Based on the task start node and task end node in the target node graph, determine at least one task path, and determine the multiplication value between the preset task time and node weight of each flight task node in each task path.

[0035] Add up the multipliers to get the task duration for each task path, and determine the target task duration with the shortest duration among all task durations.

[0036] Based on the task path corresponding to the target task duration, determine the flight tasks to be executed.

[0037] Secondly, this application provides a flight mission generation device based on causality, the device comprising:

[0038] The mapping module is used to map the target flight data to obtain the corresponding vectors and determine the flight task nodes corresponding to each vector.

[0039] The connection module is used to connect any two flight mission nodes to obtain an initial node graph, and to perform independence checks on each connection line in the initial node graph to obtain the check results.

[0040] The adjustment module is used to adjust the connection lines according to the test results and the preset connection rule base to obtain the adjusted initial node graph, and to adjust the node weights of the flight task nodes according to the target flight data and the preset data threshold to obtain the target node graph.

[0041] The determination module is used to determine the flight tasks to be executed corresponding to the target flight data based on the target node graph.

[0042] Thirdly, this application provides an apparatus, including: a processor, and a memory communicatively connected to the processor;

[0043] The memory stores instructions that the computer executes;

[0044] The processor executes computer execution instructions stored in memory to implement the method of this application.

[0045] Fourthly, this application provides a computer-readable storage medium storing program code, which, when executed by a processor, is used to implement the method of this application.

[0046] This application provides a causal-based flight task generation method, apparatus, device, and medium. The method involves mapping target flight data to obtain corresponding vectors and determining the flight task nodes corresponding to each vector; connecting any two flight task nodes to obtain an initial node graph; performing independence checks on each connecting line in the initial node graph to obtain the check results; adjusting the connecting lines according to the check results and a preset connection rule base to obtain an adjusted initial node graph; and adjusting the node weights of the flight task nodes according to the target flight data and a preset data threshold to obtain a target node graph; and determining the flight tasks to be executed corresponding to the target flight data based on the target node graph.

[0047] Thus, in practical flight support applications, target flight data can be transformed into vectors to construct an initial node graph. Independence checks are then performed to remove false associations between nodes, and connection lines are further optimized using a pre-defined connection rule base. This effectively avoids the problems of unreasonable task order and missing implicit dependencies found in traditional fixed templates, ensuring that task association logic aligns with real-world support scenarios. Simultaneously, dynamically adjusting flight task node weights based on target flight data and pre-defined thresholds allows high-priority tasks to receive priority resource allocation during scheduling, better responding to real-time dynamic changes. Determining flight tasks to be executed based on the target node graph improves the efficiency and rationality of these tasks, significantly reducing flight delay risks, enhancing support efficiency and passenger experience, and improving the accuracy, flexibility, and adaptability of flight support task generation. Attached Figure Description

[0048] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0049] Figure 1 A flowchart illustrating a causality-based flight mission generation method provided in this application embodiment;

[0050] Figure 2 A flowchart illustrating another causal-based flight mission generation method provided in this application embodiment;

[0051] Figure 3 A schematic diagram of a causality-based flight mission generation device provided in this application embodiment;

[0052] Figure 4 This is a structural block diagram of an apparatus for performing a causality-based flight mission generation method according to an embodiment of this application. Detailed Implementation

[0053] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0054] Flight support refers to a series of standardized and procedural ground service activities completed collaboratively by multiple entities, including airports, airlines, and ground handling, throughout the entire ground cycle of a single flight from landing to takeoff. In practical application scenarios of flight support, in order to save resources and improve the actual execution efficiency of flight support tasks, it is possible to coordinate the execution sequence of tasks, allocate resources, and control time nodes to ensure that all necessary preparations are completed efficiently and safely during the flight's stay on the ground, ultimately ensuring that the flight takes off as scheduled.

[0055] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0056] Figure 1 This is a flowchart illustrating a causal-based flight mission generation method provided in an embodiment of this application. Figure 1 As shown, this causality-based flight mission generation method may include the following steps:

[0057] S110. Map the target flight data to obtain the corresponding vectors, and determine the flight task nodes corresponding to each vector.

[0058] Among them, target flight data refers to real-time task data after data processing, which can meet the standard data format, such as data that has undergone missing value filling and normalization processing; flight data can be data related to flight support tasks, such as real-time flight data such as aircraft number, aircraft type, route, arrival and departure port identification, arrival and departure airports, arrival and departure times, flight delays and transit times, passenger data such as passenger numbers, load factor, transfer rate, probability of passenger reduction, and special service needs of passengers, baggage data such as the quantity, weight, type, handling status, and special baggage that requires special attention, and historical flight support data that has been executed previously.

[0059] Mapping is an operation used to convert target flight data into corresponding vectors. It can be understood as a mathematical transformation process that transforms the discrete features (such as task type, resource ID, and status) in high-dimensional, sparse, and non-numerical "target flight data" into low-dimensional, dense, and continuous numerical vector representations through embedding techniques.

[0060] Flight task nodes are graphical elements in a task flowchart used to represent task execution steps or key states. They are a mapping of the task execution process and can characterize a specific activity, event, or state change during the execution of a flight support task.

[0061] Based on this, by determining the target flight data in a standard format, the flight data is mapped to obtain corresponding vectors. Since each numerical dimension of the vector corresponds to a core business feature of the flight data, by mapping the target flight data to the corresponding vector, the corresponding flight features are determined based on the numerical dimension features of the vector. This further determines the flight task nodes corresponding to the vectors. Based on multiple flight task nodes, the flight support task is divided into multiple sub-tasks, i.e., flight task nodes, so that the connection rules between each flight task node can be determined according to the execution logic between the sub-tasks. Based on the execution logic, the flight task nodes are connected to obtain the target node graph, so as to determine the target flight task to be executed with the shortest execution time.

[0062] S120. Connect any two flight mission nodes to obtain an initial node graph, and perform an independence test on each connecting line in the initial node graph to obtain the test results.

[0063] The initial node graph refers to the basic graphical structure formed after connecting any two flight task nodes without distinction. It includes flight task nodes and connecting lines between nodes. It can be understood as an undirected graph where nodes are completely connected. That is, it consists of several flight task nodes, and there is an undirected connecting line (edge) between any two different nodes. There is no graphical structure where any two nodes are isolated and unconnected.

[0064] The independence test is a test method based on preset task association indicators to check whether there is a non-random association between two flight task nodes connected by any connection line in the initial node graph. This determines whether the two nodes corresponding to this connection line are independent or related in terms of business logic or data characteristics. For example, if the relationship between node A and node B is that node A must be completed before node B can be executed, it indicates that there is a logical association between node A and node B, and they are not independent nodes.

[0065] The test result refers to the test conclusion on whether there is a relationship between the two flight task nodes corresponding to a single connecting line in the initial node graph after performing an independence test on that connecting line.

[0066] Based on this, by connecting any two nodes, a fully connected undirected initial node graph is obtained. Then, the independence of each connection line is further checked to determine whether there is execution logic between the two flight task nodes corresponding to each connection line, that is, whether the two flight task nodes are related or independent. This allows for subsequent adjustments to each connection line in the initial node graph, i.e., deleting connection lines that do not conform to the actual execution logic, ensuring that the adjusted connection lines all conform to the actual execution logic, thereby improving the accuracy and practicality of task generation.

[0067] S130. Based on the test results and the preset connection rule base, adjust the connection lines to obtain the adjusted initial node graph. Then, based on the target flight data and the preset data threshold, adjust the node weights of the flight task nodes to obtain the target node graph.

[0068] The preset connection rule library is a predefined set of standardized rules for determining whether a connection line in the initial node graph should be adjusted. These connection rules are pre-defined based on business logic, task characteristics, or data association patterns, thus ensuring that the adjusted connection line meets the actual execution logic. For example, if there is a data transmission dependency between two flight task nodes, such as the order payment node needing to transmit payment success data to the inventory deduction node, then there is an actual execution logic relationship between the two nodes. The connection rules between the two nodes can be stored in the preset connection rule library. If it can be subsequently determined that there is a connection line between the order payment node and the inventory deduction node in the initial node graph, then it can be determined that the connection line meets the execution logic, and no further adjustment to the connection line is needed.

[0069] Adjustment refers to adding or deleting connections to ensure that the adjusted connections satisfy the execution logic. For example, if two flight task nodes are completely independent execution tasks, such as customer information entry and logistics route planning, there is no data or time sequence relationship between these two nodes, then the connection between the two nodes needs to be adjusted.

[0070] The preset data threshold is a pre-set quantitative standard used to determine the importance of flight task nodes in the current actual application scenario.

[0071] Node weight is a numerical indicator used to quantify the importance, priority, or contribution of a single flight task node in the entire node graph. The higher the weight, the greater the impact of the node on the target task and the higher its priority.

[0072] The target node graph is obtained after adjusting the connecting lines and assigning node weights. In terms of connection relationships, the target node graph only retains connecting lines that conform to preset rules and have been verified to be valid, while invalid connecting lines are deleted. On the nodes, each flight task node has a quantified weight value, which can intuitively distinguish core nodes (high weight) and secondary nodes (low weight) for subsequent path planning and task generation.

[0073] Based on this, the initial node graph contains many redundant connections between nodes that do not actually execute logic. Therefore, further adjustments are needed to retain valid connections and remove redundant ones, making the adjusted initial node graph concise, clear, and consistent with the actual execution logic. This can be achieved by adjusting the connections using the independence test results and a pre-defined connection rule base. Furthermore, to improve the accuracy of each flight task node in the adjusted initial node graph and make it more consistent with the current scenario, the target flight data corresponding to each flight task node can be compared with the values ​​of a pre-defined data threshold. Based on the comparison results, the node weights of the flight task nodes can be further adjusted to obtain the target node graph.

[0074] S140. Based on the target node diagram, determine the flight tasks to be executed corresponding to the target flight data.

[0075] Among them, the flight task to be executed refers to the task with the shortest execution time determined from all tasks covered by the target flight data, based on the node connection relationships and node attributes of the target node graph.

[0076] Based on this, by using a better task sorting strategy, the flight task with the shortest execution time in the current scenario can be determined, so as to maximize the overall task flow execution efficiency, resource utilization and target achievement speed.

[0077] Based on the feasible implementation of S110 described above, this application further provides mapping of target flight data to obtain corresponding vectors, including:

[0078] Acquire real-time task data, and perform missing value imputation, outlier handling, and normalization on the real-time task data to obtain target flight data;

[0079] The target flight data is mapped to obtain multiple vectors.

[0080] Real-time task data refers to raw data that is collected in real time during the execution of a task and is directly related to the task. For example, it could be multiple real-time flight data related to flight support.

[0081] Missing value imputation refers to a preprocessing operation that fills in missing values ​​for some fields in real-time task data that are empty or not collected (i.e., missing values) by using reasonable statistical methods or business rules, so as to make the data dimensions complete. Examples include fill in preceding and following items and fill in the mode.

[0082] Outlier handling refers to the preprocessing operation that identifies abnormal data (i.e., outliers) in real-time task data that deviates from the normal range, does not conform to business logic or statistical regularity, and reduces their interference with subsequent data processing and analysis by means of deletion, correction or isolation.

[0083] Normalization refers to the preprocessing operation that maps task data, after missing value filling and outlier handling, to a specific numerical range according to a unified rule, thereby eliminating differences in units and numerical ranges between different fields.

[0084] Based on this, multiple data preprocessing operations are performed on the collected real-time task data to obtain target flight data that meets the data standard format.

[0085] Thus, in flight support applications, by converting target flight data into vectors and constructing a node graph, the shortest-running flight task can be determined based on the node graph, thereby improving the efficiency of flight support and saving resources. Furthermore, in addition to flight support scenarios, this can also be applied to other scenarios such as logistics warehousing, sorting and distribution, and intelligent manufacturing production lines. By acquiring target logistics data or target production data and converting it into vectors, the shortest-running logistics or production task can be determined based on the node graph corresponding to the vector. In other words, it is not limited to flight support applications; it is feasible for any application scenario that requires determining the shortest-running target task or meeting other practical needs based on real-time collected data.

[0086] Based on the feasible implementation of S120 described above, this application further provides an independence test for each connecting line in the initial node diagram, and obtains the test results, including:

[0087] For the target flight task node, other flight task nodes are combined in order to obtain multiple sets of node variables; the target flight task node is the two flight task nodes corresponding to the current connection line.

[0088] Based on the order of the node variable set, the node state of the flight task node in each node variable set is changed sequentially.

[0089] Based on the PC algorithm, the independence coefficients between target flight task nodes are calculated under different node states in each set of node variables, and the independent coefficients are compared with the preset independence threshold to obtain the comparison results.

[0090] Based on the comparison results, determine the inspection results corresponding to each connecting line.

[0091] In this context, sequential combination refers to combining the remaining flight task nodes according to their order (i.e., the number of nodes in the set) from low to high to form node sets of different sizes. For example, assuming the target flight task nodes are A and B, and the other flight task nodes are C, D, and E, then the first-order combination is {C}, {D}, {E} (each set contains 1 other node); the second-order combination is {C,D}, {C,E}, {D,E} (each set contains 2 other nodes); and the third-order combination is {C,D,E} (containing 3 other nodes). By combining nodes sequentially, the computational chaos caused by directly handling large-scale combinations is avoided, ensuring the orderliness of the analysis.

[0092] A set of node variables refers to a collection of flight task nodes generated through successive combinations. This set can serve as a condition variable set to analyze whether the relationship between two target flight task nodes is influenced by the nodes within the set, i.e., whether there is a logical connection between the target flight task nodes. The order of the set refers to the number of flight task nodes in the set; sets of different orders correspond to different conditional dimensions.

[0093] Node state change refers to adjusting the attribute values ​​(i.e., states) of each flight task node in the set, resulting in multiple different state combinations of the node. This is used to subsequently test the stability of the relationship between the target flight task nodes under different conditions. For example, for the aircraft door node in flight data, "door open" and "door closed" are two different node states corresponding to the aircraft door node. By determining the correlation between two target flight task nodes under these two different node states, it can be determined whether there is a correlation between the two target flight task nodes.

[0094] The Propensity-Based (PC) algorithm is used to automatically discover causal relationships between variables from observed data. Through rigorous statistical testing, it eliminates spurious associations between variables, leaving only the most likely true causal connections. In practical applications, the PC algorithm can construct a node causal graph and perform independence analysis on node variables. For each pair of node variables, the PC algorithm considers whether the two node variables are independent given other sets of node variables. For example, in a flight support scenario, to verify whether there is a correlation between the refueling completion time (X) and the boarding start time (Y), the corresponding set of node variables can be determined as the cabin door status flight task node. The PC algorithm calculates the two independence coefficients for nodes X and Y in the two different cabin door statuses: cabin door open and cabin door closed. Based on the independence coefficients, it can be determined whether nodes X and Y are correlated in different cabin door status states. If they are correlated, it indicates that the correlation between nodes X and Y is not affected by the cabin door status flight task node, meaning that nodes X and Y have execution logic and are correlated with each other.

[0095] The independence coefficient is a numerical indicator used to quantify the degree of statistical independence between two target flight task nodes under a given set of node variables. The magnitude of the independence coefficient directly reflects the strength of the correlation between the target nodes. That is, the closer the coefficient is to the independence judgment value, the more independent the two nodes are under that condition; conversely, the smaller the coefficient is, the stronger the correlation. The PC algorithm can calculate the partial correlation coefficient and other independence coefficients between variables and compare them with preset thresholds to determine whether there is conditional independence between variables.

[0096] The preset independence threshold refers to a pre-set numerical standard used to determine whether two target flight mission nodes are independent.

[0097] Based on this, the independence of each connection line in the initial node graph can be checked using the PC algorithm. For each connection line, the two flight task nodes connected by the current connection line are first identified as target flight task nodes. Then, other flight task nodes are combined sequentially to obtain multiple sets of node variables. If the two target flight task nodes have execution logic, it means that the two target flight task nodes are completely related and will not be affected by the node states of other flight task nodes. In other words, regardless of the node states of other flight task nodes, the two target flight task nodes are related. This relationship can be determined by the independence coefficient, i.e., by using the PC algorithm to calculate the independence coefficient between the two target flight task nodes. If the independence coefficient is greater than the preset independence threshold in all other node states of the task, then the two target flight task nodes can be determined to be related and non-independent flight task nodes that satisfy the execution logic. The node states of other flight task nodes can be determined by the set of node variables. By changing the node states in the set of node variables one by one in ascending order of order, if there exists any set of node variables where the calculated independence coefficient between the two target flight task nodes is less than the preset independence threshold when the node state is changed, then it indicates that the two target flight task nodes are affected by the nodes in the set of node variables at this time. In this case, the target flight task nodes are not completely related, and the connection line between them is a redundant connection line that can be deleted.

[0098] Based on the feasible implementation of S120 described above, this application further provides a method for determining the inspection results corresponding to each connection line based on the comparison results, including:

[0099] Determine the comparison results;

[0100] If the comparison result shows that there is at least one independent coefficient that is less than the preset independent threshold, then the test result corresponding to the current connector is determined to be a non-associated connector.

[0101] If the comparison result shows that there is no independent coefficient less than the preset independent threshold, then the test result corresponding to the current connector is determined to be an associated connector.

[0102] Determine the test results for each connecting line.

[0103] Based on this, if two flight task nodes are related, the PC algorithm is used to calculate the independence coefficient between the two target flight task nodes. The coefficient is greater than the preset independence threshold in other node states of the task. Therefore, by determining the numerical comparison results between each independence coefficient and the preset independence threshold, the connection line test results corresponding to the independence coefficients less than the preset independence threshold are determined as non-related connection lines, and the connection line test results corresponding to the independence coefficients not less than the preset independence threshold are determined as related connection lines.

[0104] Based on the feasible implementation of S130 described above, this application further provides a method for adjusting the connection lines according to the test results and a preset connection rule base to obtain an adjusted initial node diagram, including:

[0105] Determine the connection rules corresponding to each connection line, and determine whether there is a target connection rule in each connection rule that is consistent with the preset connection rules in the preset connection rule library; the connection rules are used to represent the task logic between the corresponding flight task nodes;

[0106] If so, then the connecting line corresponding to the target connecting rule is determined as the target connecting line, and other connecting lines are determined as non-target connecting lines;

[0107] If not, then all connecting lines are determined to be non-target connecting lines;

[0108] Determine the test results for each non-target connection line, and delete the non-target connection lines whose test results are not associated connection lines to obtain the adjusted initial node diagram.

[0109] The connection rules can be understood as the logical relationship between two flight mission nodes.

[0110] Based on this, in flight support task prediction, the PC algorithm, as a classic method for causal discovery, can automatically generate an initial causal graph structure from historical flight data. However, the initial causal graph may not meet the business rules and constraints specific to the aviation support field. Therefore, it needs to be calibrated in conjunction with the rules of the civil aviation field to obtain a reliable node causal graph that conforms to the actual business logic. Therefore, by identifying a preset connection rule library that stores multiple preset connection rules, it is possible to determine whether there are connection lines in the initial node graph that are consistent with the preset connection rules. For example, if civil aviation regulations clearly require that "the cabin door can only be closed after the security check is completed", then if there are security check nodes and cabin door closing nodes in the initial node graph, it can be determined that the connection rule between these two nodes is the target connection rule that is consistent with the preset connection rule in the preset connection rule library. Therefore, the connection line between the two can be identified as the target connection line, thereby further identifying non-target connection lines. In this case, the non-target connection line whose corresponding test result is a non-associated connection line is identified. This indicates that the non-target connection line is an independent connection line, and there is no execution logic between the two corresponding flight task nodes. Therefore, this redundant connection line needs to be deleted.

[0111] Based on the feasible implementation of S130 described above, this application further provides a method for adjusting the node weights of flight task nodes according to target flight data and preset data thresholds to obtain a target node graph, including:

[0112] Determine whether there is any task data to be adjusted in the target flight data that exceeds a preset data threshold;

[0113] If so, determine the flight task node to be adjusted corresponding to the task data to be adjusted, and calculate the adjusted node weight based on the node weight of the flight task node to be adjusted, the task data to be adjusted, and the preset data threshold multiplier, to obtain the target node graph.

[0114] Based on this, by combining real-time data, the weight values ​​of each task are adjusted according to the actual scenario, so that the resulting target node graph can better meet the actual needs. For example, when the system receives real-time data such as "previous flight delayed by 30 minutes" and "thunderstorm weather", the dynamic adjustment mechanism in the cause-effect graph responds immediately. On the one hand, it activates the "rapid cleaning" node (increases the execution weight to 0.8), and on the other hand, it triggers the "de-icing" node. Thus, based on real-time data, by adjusting the node weights of flight task nodes and appropriately adding or deleting flight task nodes, a target node graph adapted to the actual scenario is determined.

[0115] Based on the feasible implementation of S140 described above, this application further provides a method for determining the flight tasks to be executed corresponding to the target flight data according to the target node graph, including:

[0116] Based on the task start node and task end node in the target node graph, determine at least one task path, and determine the multiplication value between the preset task time and node weight of each flight task node in each task path.

[0117] Add up the multipliers to get the task duration for each task path, and determine the target task duration with the shortest duration among all task durations.

[0118] Based on the task path corresponding to the target task duration, determine the flight tasks to be executed.

[0119] The task start node refers to the starting point node of a certain task process in the target node graph, which is the initial trigger point of the entire task logic; this node has no preceding dependent tasks, that is, no other flight task nodes point to it through connecting lines.

[0120] The task termination node refers to the endpoint node in the target node diagram that marks the complete termination of a certain task process. It is the final closing point of the entire task logic. This node has no subsequent tasks, that is, there are no connecting lines extending from it to other flight task nodes.

[0121] A task path refers to a complete sequence of nodes in the target node graph that starts from the task start node, follows the valid connection lines between flight task nodes (which conform to logical connection rules), and finally reaches the task end node. A task path represents a feasible solution to complete the flight task to be executed. There may be multiple parallel task paths in the target node graph (corresponding to different execution solutions).

[0122] The preset task time refers to the standard time required to complete the task undertaken by each specific flight task node in the target node diagram. It is a fixed value defined in advance based on historical data, task complexity, or business rules, and is used to quantify the execution time of a single flight task node. For example, in the e-commerce order fulfillment process, the preset task time for each flight task node can be set as follows: 2 hours for picking in a regular warehouse (goods in a regular warehouse are scattered, and picking takes a long time); 30 minutes for picking in a forward warehouse (forward warehouses are close to users and goods are concentrated); and 12 hours for shipping via ZTO Express.

[0123] Task duration refers to the total time taken for a given task path. It is calculated by multiplying the preset task time of all flight task nodes on that path by their corresponding weights and then summing all the multipliers. It represents the total time cost required to execute this task path (i.e., the solution) and is a core indicator for comparing the merits of different task paths.

[0124] Based on this, the target node graph generates multiple task paths, which are task plans for multiple flight tasks to be executed. Furthermore, based on the preset task time and node weight of each flight task node in the task path, the total time corresponding to each task path is determined in order to identify the target task path with the shortest time, and thus the task plan corresponding to the target task path is identified as the flight task to be executed.

[0125] Please refer to Figure 2 , Figure 2 A flowchart illustrating another causality-based flight mission generation method provided in this application embodiment; as shown Figure 2 As shown, the process begins with the arrival of the preceding flight. If the preceding flight is delayed, the entire process is extended by 25 minutes, and the delay factor has a weight of 0.8. Under normal circumstances, the cabin door is opened, which is mandatory. After the cabin door is opened, baggage unloading begins, and cabin cleaning is carried out simultaneously if there are sufficient cleaning personnel. After baggage unloading, baggage sorting is carried out. After cabin cleaning, quick cleaning is performed or other operations are carried out through the right-side passage of the gate. After baggage sorting, baggage loading is carried out. After quick cleaning, security checks (mandatory) and refueling truck positioning are carried out. After security checks, baggage loading is carried out, and refueling is carried out after the refueling truck is in position. If it is a false cause, meal preparation is carried out. After refueling, if it is winter, de-icing is carried out. After meal preparation, if there is thunderstorm weather, the cabin door is closed. After de-icing, the cabin door is closed to ensure that flight support work is carried out in an orderly and efficient manner.

[0126] In some embodiments of this application, the target flight data is first mapped and transformed into corresponding vectors, and the flight task node corresponding to each vector is further determined. Any two flight task nodes are then indiscriminately connected to form an initial node graph containing all flight task nodes and connecting lines between nodes. For each connecting line in the initial node graph, an independence test is performed, such as calculating the independence coefficient by combining the node variable set and comparing it with a preset independence threshold, thereby determining the association between the target flight task nodes at both ends of the connecting line and obtaining the independence test result for each connecting line. Then, based on the test result and a preset connection rule base, the connecting lines in the initial node graph are adjusted by retaining, deleting, or modifying them to obtain an adjusted initial node graph. Simultaneously, based on the core features of the target flight data and a preset data threshold, the node weights of each flight task node in the adjusted initial node graph are recalculated and adjusted, ultimately forming a target node graph containing accurate connecting lines and reasonable node weights. Finally, based on this target node graph, and combining information such as the effective connection relationships between nodes and the task importance reflected by the node weights, the target task path with the shortest execution time is determined, thereby determining the flight task to be executed corresponding to the target flight data.

[0127] In this way, by converting target flight data into vectors and constructing an initial node graph, and then removing false associations between nodes through independence checks, and further optimizing the connection lines by combining with a preset connection rule base, the problems of unreasonable task order and omission of implicit dependencies in traditional fixed templates can be effectively avoided, ensuring that the task association logic fits the real support scenario. At the same time, by dynamically adjusting the weight of flight task nodes based on target flight data and preset thresholds, high-priority tasks (such as rapid cleaning of delayed flights and service support for special passengers) can receive priority resource allocation in scheduling, thereby better responding to real-time dynamic changes (such as sudden weather changes and previous delays). Determining the flight tasks to be executed based on the target node graph improves the execution efficiency and rationality of the flight tasks to be executed, significantly reduces the risk of flight delays, improves support efficiency and passenger experience, and enhances the accuracy, flexibility and adaptability of flight support task generation.

[0128] Figure 3 This is a schematic diagram of a causality-based flight mission generation device 300 provided in an embodiment of this application. Figure 3 As shown, the causality-based flight mission generation device 300 includes: a mapping module 310, a connection module 320, an adjustment module 330, and a determination module 340; wherein:

[0129] The mapping module 310 is used to map the target flight data to obtain the corresponding vectors and determine the flight task nodes corresponding to each vector.

[0130] The connection module 320 is used to connect any two flight mission nodes to obtain an initial node diagram, and to perform independence checks on each connection line in the initial node diagram to obtain the check results.

[0131] The adjustment module 330 is used to adjust the connection lines according to the test results and the preset connection rule library to obtain the adjusted initial node graph, and to adjust the node weights of the flight task nodes according to the target flight data and the preset data threshold to obtain the target node graph.

[0132] The determination module 340 is used to determine the flight tasks to be executed corresponding to the target flight data based on the target node graph.

[0133] In this embodiment of the application, the mapping module 310 can also be specifically used for:

[0134] Acquire real-time task data, and perform missing value imputation, outlier handling, and normalization on the real-time task data to obtain target flight data;

[0135] The target flight data is mapped to obtain multiple vectors.

[0136] In this embodiment of the application, the connection module 320 can also be specifically used for:

[0137] For the target flight task node, other flight task nodes are combined in order to obtain multiple sets of node variables; the target flight task node is the two flight task nodes corresponding to the current connection line.

[0138] Based on the order of the node variable set, the node state of the flight task node in each node variable set is changed sequentially.

[0139] Based on the PC algorithm, the independence coefficients between target flight task nodes are calculated under different node states in each set of node variables, and the independent coefficients are compared with the preset independence threshold to obtain the comparison results.

[0140] Based on the comparison results, determine the inspection results corresponding to each connecting line.

[0141] In this embodiment of the application, the connection module 320 can also be specifically used for:

[0142] Determine the comparison results;

[0143] If the comparison result shows that there is at least one independent coefficient that is less than the preset independent threshold, then the test result corresponding to the current connector is determined to be a non-associated connector.

[0144] If the comparison result shows that there is no independent coefficient less than the preset independent threshold, then the test result corresponding to the current connector is determined to be an associated connector.

[0145] Determine the test results for each connecting line.

[0146] In this embodiment of the application, the adjustment module 330 can also be specifically used for:

[0147] Determine the connection rules corresponding to each connection line, and determine whether there is a target connection rule in each connection rule that is consistent with the preset connection rules in the preset connection rule library; the connection rules are used to represent the task logic between the corresponding flight task nodes;

[0148] If so, then the connecting line corresponding to the target connecting rule is determined as the target connecting line, and other connecting lines are determined as non-target connecting lines;

[0149] If not, then all connecting lines are determined to be non-target connecting lines;

[0150] Determine the test results for each non-target connection line, and delete the non-target connection lines whose test results are not associated connection lines to obtain the adjusted initial node diagram.

[0151] In this embodiment of the application, the adjustment module 330 can also be specifically used for:

[0152] Determine whether there is any task data to be adjusted in the target flight data that exceeds a preset data threshold;

[0153] If so, determine the flight task node to be adjusted corresponding to the task data to be adjusted, and calculate the adjusted node weight based on the node weight of the flight task node to be adjusted, the task data to be adjusted, and the preset data threshold multiplier, to obtain the target node graph.

[0154] In this embodiment of the application, the determining module 340 can also be specifically used for:

[0155] Based on the task start node and task end node in the target node graph, determine at least one task path, and determine the multiplication value between the preset task time and node weight of each flight task node in each task path.

[0156] Add up the multipliers to get the task duration for each task path, and determine the target task duration with the shortest duration among all task durations.

[0157] Based on the task path corresponding to the target task duration, determine the flight tasks to be executed.

[0158] Figure 4 This is a schematic diagram of the structure of an apparatus for performing a causality-based flight mission generation method according to an embodiment of this application. Figure 4 As shown, the device 400 includes:

[0159] The device 400 may include a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a communication component 403, and other components. The processor 401, memory 402, and communication component 403 are connected via a bus 404.

[0160] In the specific implementation process, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to execute the above-described causal flight mission generation method.

[0161] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0162] Furthermore, the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this application can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0163] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0164] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0165] In some embodiments, a computer program product is also provided, comprising a computer program or instructions that, when executed by a processor, implement the steps in any of the causality-based flight mission generation methods described above.

[0166] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0167] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0168] Therefore, embodiments of this application provide a computer-readable storage medium storing a plurality of program codes that can be loaded by a processor to execute steps in any of the causality-based flight mission generation methods provided in embodiments of this application.

[0169] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0170] According to one aspect of this application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium.

[0171] Since the instructions stored in the storage medium can execute the steps in any of the causal-based flight mission generation methods provided in the embodiments of this application, the beneficial effects that any of the causal-based flight mission generation methods provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0172] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the appended claims.

[0173] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A flight mission generation method based on causality, characterized in that, The method comprises: mapping target flight data to obtain corresponding vectors, and determining flight task nodes corresponding to each of the vectors; connecting any two of the flight task nodes to obtain an initial node graph, and performing independence testing on each connection line in the initial node graph to obtain a testing result; adjusting the connection lines according to the testing result and a preset connection rule library to obtain an adjusted initial node graph, and adjusting node weights of the flight task nodes according to the target flight data and a preset data threshold to obtain a target node graph; wherein the adjusting of the node weights of the flight task nodes according to the target flight data and the preset data threshold to obtain the target node graph comprises: determining whether there is to-be-adjusted task data greater than the preset data threshold in the target flight data; if yes, determining a to-be-adjusted flight task node corresponding to the to-be-adjusted task data, and calculating an adjusted node weight according to a product of the node weight of the to-be-adjusted flight task node, the to-be-adjusted task data and the preset data threshold to obtain the target node graph; determining a to-be-executed flight task corresponding to the target flight data according to the target node graph; wherein the determining of the to-be-executed flight task corresponding to the target flight data according to the target node graph comprises: determining at least one task path according to a task start node and a task end node in the target node graph, and determining a product between a preset task time and the node weight of each flight task node in each of the task paths; adding each of the products to obtain a task duration corresponding to each of the task paths, and determining a target task duration with the shortest duration among each of the task durations; determining the to-be-executed flight task according to the task path corresponding to the target task duration.

2. The method of claim 1, wherein, The mapping of the target flight data to obtain corresponding vectors comprises: obtaining real-time task data, and performing missing value filling, abnormal value processing and normalization processing on the real-time task data to obtain the target flight data; mapping the target flight data to obtain a plurality of the vectors.

3. The method of claim 1, wherein, The independence testing on each connection line in the initial node graph to obtain a testing result comprises: for a target flight task node, combining other flight task nodes in stages to obtain a plurality of node variable sets; the target flight task node is two flight task nodes corresponding to the current connection line; changing node states of the flight task nodes in each of the node variable sets according to the order size of the node variable set; calculating independence coefficients between target flight task nodes in different node states of each of the node variable sets according to a PC algorithm, and comparing each of the independence coefficients with a preset independence threshold to obtain a comparison result; determining the testing result corresponding to each of the connection lines according to the comparison result.

4. The method of claim 3, wherein, The determining of the testing result corresponding to each of the connection lines according to the comparison result comprises: determining the comparison result. If the comparison result is that there is at least one of the independence coefficients less than the preset independence threshold value, it is determined that the test result corresponding to the current connection line is a non-associated connection line. If the comparison result is that there is no independence coefficient less than the preset independence threshold value, it is determined that the test result corresponding to the current connection line is an associated connection line. The test results corresponding to each of the connection lines are determined.

5. The method of claim 1, wherein, The connection lines are adjusted according to the test results and a preset connection rule library, to obtain an adjusted initial node graph, including: The connection rules corresponding to each of the connection lines are determined, and it is judged whether there is a target connection rule consistent with a preset connection rule in the preset connection rule library in each of the connection rules; the connection rule is used to represent the task logic between the corresponding flight task nodes; If yes, the connection line corresponding to the target connection rule is determined as a target connection line, and other connection lines are determined as non-target connection lines; If no, the connection lines are all determined as non-target connection lines; The test results corresponding to each of the non-target connection lines are determined, and the non-target connection lines with the test result of a non-associated connection line are deleted, to obtain the adjusted initial node graph.

6. A flight mission generation apparatus based on causality, characterized by, The device comprises: A mapping module is configured to map target flight data to obtain corresponding vectors, and determine flight task nodes corresponding to each of the vectors; A connection module is configured to connect any two flight task nodes to obtain an initial node graph, and perform independence test on each connection line in the initial node graph to obtain test results; An adjustment module is configured to adjust the connection lines according to the test results and a preset connection rule library to obtain an adjusted initial node graph, and adjust node weights of the flight task nodes according to target flight data and a preset data threshold to obtain a target node graph; wherein the adjustment of the node weights of the flight task nodes according to the target flight data and the preset data threshold to obtain the target node graph includes: It is judged whether there is to-be-adjusted task data greater than the preset data threshold in the target flight data; If yes, a to-be-adjusted flight task node corresponding to the to-be-adjusted task data is determined, and an adjusted node weight is calculated by multiplying the node weight of the to-be-adjusted flight task node, the to-be-adjusted task data and the preset data threshold, to obtain the target node graph; A determination module is configured to determine to-be-executed flight tasks corresponding to the target flight data according to the target node graph; wherein the determination of the to-be-executed flight tasks corresponding to the target flight data according to the target node graph includes: At least one task path is determined according to a task start node and a task end node in the target node graph, and a product of a preset task time and the node weight of each flight task node in each of the task paths is determined. The multiplication values are added to obtain a task duration corresponding to each of the task paths, and a target task duration with the shortest duration among the task durations is determined; According to the task path corresponding to the target task duration, the flight task to be executed is determined.

7. An apparatus, comprising: Comprise: One or more processors; Memory; One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method of any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The program code is stored in the computer readable storage medium, and the program code can be called and executed by the processor to execute the method of any one of claims 1 to 5.

Images