Flight delay reason layering judgment method and system based on multi-task measurement

Abstract

The invention belongs to the technical field of air transportation management, and particularly relates to a flight delay reason layering judgment method and system based on multi-task measurement. The method comprises the following steps: acquiring civil aviation flight operation data and delay reason hierarchical labels; preprocessing the data; constructing a nested feature vector space based on a hypersphere, and expressing multi-scale semantics layer by layer; constructing a multi-task measurement and hierarchical classification model by utilizing multi-level nesting; training to obtain a delay reason classification model based on multi-task measurement and a hierarchical classification model; and performing hierarchical judgment on flight delay reasons through the delay reason classification model. By integrating multi-task measurement and hierarchical reasoning technologies, the stability and robustness of a delay reason judgment system are remarkably improved, efficient and accurate delay data analysis is achieved, and scientific decision support is provided for civil aviation management departments.

Description

Technical Field

[0001] The present invention belongs to the technical field of air transport management, and specifically relates to a method and system for hierarchical determination of flight delay causes based on multi-task metrics. Background Art

[0002] Flight delays are one of the key challenges in air transport management, and have a profound impact on airlines, airport management departments, and passengers. With the rapid development of the air transport industry, the flight operating environment has become increasingly complex, and delays have occurred frequently. The reasons usually involve many factors, such as weather, air traffic control, airport resource scheduling, and airline operation plans. These factors are intertwined to form a multi-level and multi-scale complex feature, making the intelligent determination and analysis of the cause of delays a huge challenge.

[0003] At present, the identification of flight delay causes mainly relies on manual experience and judgment. Although these methods have certain applicability, their efficiency and accuracy are difficult to meet actual needs. In recent years, flight delay cause determination methods based on machine learning and deep learning have gradually emerged, and the intelligence level of the determination process has been improved through feature extraction and modeling technology. However, existing deep learning models mostly focus on single-level feature analysis and lack effective characterization of the multi-level and multi-scale characteristics of delay causes. At the same time, many methods use fixed feature dimensions and classification mechanisms in the reasoning stage, and fail to dynamically adjust the classification accuracy and efficiency in combination with task requirements, making it difficult to provide reliable determination results in complex operating environments. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system, electronic device and storage medium for hierarchical determination of flight delay causes based on multi-task metrics.

[0005] In a first aspect, the present invention provides a method for hierarchical determination of flight delay causes based on multi-task metrics, comprising:

[0006] Obtain civil aviation flight operation data and hierarchical labels of delay reasons;

[0007] Data preprocessing;

[0008] Construct a nested feature vector space based on a hypersphere to express multi-scale semantics layer by layer;

[0009] Utilize multi-level nesting to build multi-task measurement and hierarchical classification models;

[0010] Based on multi-task measurement and hierarchical classification model, a delay cause classification model is trained;

[0011] The delay cause classification model is used to make hierarchical judgments on the causes of flight delays.

[0012] In a second aspect, the present invention provides a flight delay cause hierarchical determination system based on multi-task metrics, comprising:

[0013] Data collection module, used to obtain civil aviation flight operation data and delay stratification labels;

[0014] A data preprocessing module, used for preprocessing data;

[0015] Nested feature construction module, used to construct nested feature vector space to achieve multi-level semantic nesting of flight delay reasons;

[0016] Multi-task measurement and hierarchical classification model building module, used to build multi-task measurement and hierarchical classification models using multi-level nesting;

[0017] The delay cause classification model acquisition module is based on multi-task measurement and hierarchical classification model to train and obtain the delay cause classification model;

[0018] The delay cause identification module uses the delay cause classification model to make hierarchical judgments on the causes of flight delays.

[0019] In a third aspect, the present invention provides an electronic device comprising: a processor and a memory; the processor is used to execute computer execution instructions stored in the memory, and the processor runs the computer execution instructions to execute the above-mentioned method for hierarchical determination of flight delay causes based on multi-task metrics.

[0020] In a fourth aspect, the present invention provides a storage medium, comprising a readable storage medium and a computer program stored in the readable storage medium, wherein the computer program is used to implement the method for hierarchical determination of causes of flight delays based on multi-task metrics as described above.

[0021] The beneficial effects of the present invention are that the method and system for hierarchical determination of flight delay causes based on multi-task metrics of the present invention obtain civil aviation flight operation data and hierarchical labels of delay causes; preprocess data; construct a nested feature vector space based on a hypersphere to express multi-scale semantics layer by layer; use multi-level nesting to construct a multi-task metric and hierarchical classification model; based on the multi-task metric and hierarchical classification model, train a delay cause classification model; and perform hierarchical determination of flight delay causes through the delay cause classification model. It realizes efficient, automatic and accurate identification of specific causes of flight delays, which can effectively guide subsequent refined management and optimization decisions, and improve the safety and orderliness of the transportation system. At the same time, by constructing a multi-task metric and hierarchical classification model, the adaptability and determination accuracy of the model at multiple scales and multiple semantic levels are significantly enhanced, ensuring the reliability and stability of the system in a complex operating environment.

[0022] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 It is a principle flow chart of the flight delay cause hierarchical determination method based on multi-task measurement involved in the present invention;

[0026] Figure 2 It is a schematic diagram of the process of the hypersphere-based nested eigenvector space model involved in the present invention;

[0027] Figure 3 It is a schematic diagram of the multi-task measurement and hierarchical classification model process under multi-level nesting involved in the present invention;

[0028] Figure 4 It is a specific flow chart of the method for hierarchical determination of flight delay causes based on multi-task measurement involved in the present invention;

[0029] Figure 5 The present invention is a block diagram of the flight delay cause hierarchical determination system based on multi-task measurement. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] Example

[0032] like Figure 1As shown, the first embodiment of the present invention provides a method for hierarchical determination of flight delay causes based on multi-task metrics, comprising the following steps:

[0033] S110: Obtain civil aviation flight operation data and hierarchical labels of delay reasons;

[0034] S120: data preprocessing;

[0035] S130: Construct a nested feature vector space based on a hypersphere to express multi-scale semantics layer by layer;

[0036] S140: Using multi-level nesting to build multi-task measurement and hierarchical classification models;

[0037] S150: Based on multi-task metrics and hierarchical classification models, a delay cause classification model is trained;

[0038] S160: Perform hierarchical determination on flight delay causes through a delay cause classification model.

[0039] In this embodiment, the civil aviation flight operation data obtained in step S110 specifically includes: basic flight information, actual take-off and landing time, expected and actual arrival time, delay time, delay cause hierarchical labels, etc.; the delay cause hierarchical labels obtained divide the flight delay causes from coarse to fine levels, where the delay causes are classified into airport reasons, air traffic control reasons, airline reasons, traffic reasons and other reasons at the coarse level to cover the main delay categories. On this basis, it is further refined to fine-level labels, for example, airport reasons can be specifically divided into ground service delays, taxiway congestion, etc., and airline reasons can be divided into mechanical failures, crew taxiing errors, etc., and finally a progressive label system is formed, which provides clear semantic guidance for the accurate determination and detailed classification of delay causes.

[0040] After acquiring the data, you can also check the integrity of the data and supplement the data for missing time periods or routes.

[0041] In this embodiment, step S120 specifically includes: processing the collected civil aviation flight operation data, detecting and filtering outliers and duplicate data in the data, and obtaining cleaned data; using a dimensionality reduction method, combined with the hierarchical characteristics of flight delay causes, selecting key features related to flight delay cause judgment; labeling the hierarchical delay cause labels of data set samples; constructing a flight delay cause judgment data set, and dividing the data set into training and validation data sets; standardizing or normalizing numerical features to eliminate dimensional effects, and digitizing categorical features.

[0042] like Figure 2As shown, in this embodiment, step S130 specifically includes: constructing a nested feature vector space, dividing the features of the flight operation data into multiple subspaces, wherein the feature vector of each sample is represented as:

[0043]

[0044] Among them, d represents the total dimension of the feature vector. The strategy of gradually arranging features from coarse to fine is adopted. It is defined as the coarsest-grained feature of a sample, which usually contains important global information shared among multiple categories. Located at the end of the feature vector as the most refined features, used to capture subtle differences between specific categories. This strategy places features that are common across categories at the front of the vector, while more discriminative and refined features are placed at the end of the vector.

[0045] Furthermore, for the semantic expression of hierarchical delay causes, each subspace is defined to correspond to a specific level:

[0046]

[0047] Among them, H represents the total number of semantic levels; h represents the current level; Represents the feature subvector of the hth layer, including the subvectors from the 1st to the Dimensional characteristics.

[0048] Coarse-grained features are a subset of fine-grained features, and features at each level satisfy the following nested relationship:

[0049]

[0050] Through this progressive relationship, coarse-grained features provide contextual support for fine-grained features, ensuring the semantic consistency of multi-level features.

[0051] Furthermore, based on the nested eigenvectors, hypersphere mapping is used for geometric optimization, and the eigenvectors are normalized to the surface of the hypersphere to obtain the unit vector:

[0052]

[0053] Feature Satisfaction , avoiding the influence of feature amplitude differences on classification judgment, the contribution of all hierarchical features is determined only by directionality, making the distribution of features in geometric space more regular. Through the above construction of nested feature vector space and hypersphere mapping optimization, efficient and hierarchical feature input is obtained.

[0054] like Figure 3As shown, in step S140 of this embodiment, a dynamic measurement mechanism based on cosine similarity is introduced on the basis of the nested feature vector space to measure the similarity between feature vectors. The core calculation formula is:

[0055]

[0056] in, and Represents the nested feature vector after hypersphere normalization. Through the directional calculation of cosine similarity, the interference of feature amplitude on similarity measurement can be effectively avoided, while ensuring the directional consistency of features.

[0057] Furthermore, in order to integrate multi-level nested features into the classification model, a hierarchical classification task framework is constructed, and a loss function is designed independently for each level of task. For each level h, the classification goal of the model is to maximize the confidence of the category to which the sample belongs. A multi-level loss function based on hypersphere normalization is used:

[0058]

[0059] in, The unit vector representing category c; It represents the dynamic scale factor, which regulates the judgment range of each level; Represents the normalized feature vector of sample i at level h; Indicates category The weight vector of Represents the set of categories at level h.

[0060] In order to achieve collaborative optimization of multi-level tasks, the loss functions of each level are added together to form a total loss function:

[0061]

[0062] like Figure 4 As shown, in step S150 of this embodiment, based on the multi-task metric and hierarchical classification model, that is, using the nested feature space constructed in the training phase, the multi-level features of the flight delay causes are gradually nested in the same vector, and combined with the hierarchical multi-task learning framework, the coarse-grained and fine-grained classification tasks are optimized respectively, and the multi-scale classification information is integrated in the final result to provide a complete flight delay cause determination; the delay cause classification model is trained on the flight operation training data set with multi-scale labels; the model classification performance is evaluated on the independent validation set, and the hierarchical classification effect of the model in determining the flight delay causes is calculated.

[0063] In this embodiment, step S160 specifically includes: constructing an independent verification set, including multiple feature scenarios of flight operation and delay data, covering delay cause labels of different granularities; mapping the flight operation data to the constructed nested feature vector space, and extracting its corresponding multi-scale features; verifying the classification ability of the model at all levels from the full classification task range from coarse-grained to fine-grained; dynamically calling the hierarchical classification logic, and completing the determination of the cause of delay layer by layer from low dimension to high dimension; at the coarse-grained level, quickly identifying the coarse-level causes of delay; calling higher-dimensional feature vectors for further refined classification; and finally realizing a hierarchical determination system for flight delay causes that can dynamically adapt to different scenarios in an actual complex application environment.

[0064] Based on the above embodiments, the second aspect of the present invention provides a flight delay cause hierarchical determination system based on multi-task metrics, including: a data acquisition module for acquiring civil aviation flight operation data and delay hierarchical labels; a data preprocessing module for preprocessing data; a nested feature construction module for constructing a nested feature vector space to achieve multi-level semantic nesting of flight delay causes; a multi-task metric and hierarchical classification model construction module for constructing a multi-task metric and hierarchical classification model using multi-level nesting; a delay cause classification model acquisition module for training a delay cause classification model based on the multi-task metric and hierarchical classification model; a delay cause identification module for performing hierarchical determination of flight delay causes through the delay cause classification model.

[0065] In this embodiment, the flight delay cause hierarchical determination system based on multi-task measurement can be implemented using the flight delay cause hierarchical determination method based on multi-task measurement as described above; the implementation steps of each module can refer to the above and will not be repeated here.

[0066] Based on the above embodiments, a third aspect of the present invention provides an electronic device, comprising: a processor and a memory; the processor is used to execute computer execution instructions stored in the memory, and the processor runs the computer execution instructions to execute the above-mentioned method for hierarchical determination of flight delay causes based on multi-task metrics.

[0067] Based on the above embodiments, a third aspect of the present invention provides a storage medium, including a readable storage medium and a computer program stored in the readable storage medium, wherein the computer program is used to implement the above-mentioned method for hierarchical determination of flight delay causes based on multi-task metrics.

[0068] In summary, the method and system for hierarchical determination of flight delay causes based on multi-task metrics of the present invention obtains civil aviation flight operation data and hierarchical labels of delay causes; preprocesses data; constructs a nested feature vector space based on a hypersphere to express multi-scale semantics layer by layer; utilizes multi-level nesting to construct a multi-task metric and hierarchical classification model; based on the multi-task metric and hierarchical classification model, the delay cause classification model is trained to achieve hierarchical automated determination of flight delay causes. It realizes efficient, automatic and accurate identification of specific causes of flight delays, which can effectively guide subsequent refined management and optimization decisions, and improve the safety and orderliness of the transportation system. At the same time, by constructing a multi-task metric and hierarchical classification model, the adaptability and determination accuracy of the model at multiple scales and multiple semantic levels are significantly enhanced, ensuring the reliability and stability of the system in a complex operating environment.

[0069] In several embodiments provided in the present application, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0070] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0071] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0072] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A multi-task metric-based method for determining the causes of flight delays, characterized in that: include: Obtain civil aviation flight operation data and hierarchical labels of delay reasons; Data preprocessing; Construct a nested feature vector space based on a hypersphere to express multi-scale semantics layer by layer; Utilize multi-level nesting to build multi-task measurement and hierarchical classification models; Based on multi-task measurement and hierarchical classification model, a delay cause classification model is trained; The delay cause classification model is used to make hierarchical judgments on the causes of flight delays.

2. The flight delay cause hierarchical determination method based on multi-task metrics according to claim 1, characterized in that: The method for obtaining civil aviation flight operation data and delay cause hierarchical labels includes: Determine the types of data required, including basic flight information, actual take-off and landing times, expected and actual arrival times, delay time, and hierarchical labels for delay reasons; Check the completeness of the data and supplement the data for missing time periods or routes.

3. The flight delay cause hierarchical determination method based on multi-task metrics according to claim 2, characterized in that: The data preprocessing method comprises: Process the collected flight operation data, detect and filter outliers and duplicate data in the data, and obtain cleaned data; By using the dimensionality reduction method and combining the hierarchical characteristics of the reasons for flight delays, we select key features related to the judgment of the reasons for flight delays; Label the hierarchical delay reasons of the dataset samples; Construct a flight delay cause determination dataset and divide the dataset into training and validation datasets; Standardize or normalize numerical features to eliminate the impact of dimensions, and quantify categorical features.

4. The flight delay cause hierarchical determination method based on multi-task metrics according to claim 3, characterized in that: The method of constructing a hypersphere-based nested feature vector space to express multi-scale semantics layer by layer includes: A nested feature vector space is constructed to divide the features of the flight operation data into multiple subspaces, where the feature vector of each sample is expressed as: ； Among them, d represents the total dimension of the feature vector; The strategy of gradually arranging features from coarse to fine is adopted; Defined as the coarsest-grained feature of a sample, it contains important global information shared among multiple categories; Located at the end of the feature vector, it is the most detailed feature and is used to capture subtle differences between specific categories; For the semantic expression of hierarchical delay causes, each subspace is defined to correspond to a specific level: ； Where H represents the total number of semantic levels, h represents the current level, Represents the feature subvector of the hth layer, including the subvectors from the 1st to the Characteristics of dimensions; Coarse-grained features are a subset of fine-grained features, and features at each level satisfy the following nested relationship: ； Through this progressive relationship, coarse-grained features provide contextual support for fine-grained features; Based on the nested eigenvectors, hypersphere mapping is used for geometric optimization, and the eigenvectors are normalized to the hypersphere surface to obtain the unit vector: ； Feature Satisfaction .

5. The flight delay cause hierarchical determination method based on multi-task metrics according to claim 4, characterized in that: The method for constructing a multi-task measurement and hierarchical classification model based on multi-level nesting includes: On the basis of the nested feature vector space, a dynamic measurement mechanism based on cosine similarity is introduced to measure the similarity between feature vectors; its core calculation formula is: ； in, and represents the nested eigenvector after hypersphere normalization; In the multi-level nested feature space, samples of similar categories are gathered more tightly, while the distance between samples of different categories is widened; the constraints of multi-level nested features in classification are: ； This formula indicates that in the h-th level feature space, for Sample and , their similarity is higher than that of Sample and ; Integrate multi-level nested features into the classification model, build a hierarchical classification task framework, and independently design loss functions for each level of task; for each level h, the classification goal of the model is to maximize the confidence of the category to which the sample belongs; adopt a multi-level loss function based on hypersphere normalization: ； in, The unit vector representing category c; It represents the dynamic scale factor, which regulates the judgment range of each level; Represents the normalized feature vector of sample i at level h; Indicates category The weight vector of represents the set of categories at level h; The loss functions of each layer are added together to form a total loss function: 。 6. The flight delay cause hierarchical determination method based on multi-task metrics according to claim 5, characterized in that: The method for training and obtaining a delay cause classification model based on multi-task metrics and a hierarchical classification model includes: Using multi-task metrics and hierarchical classification models, that is, using the nested feature space constructed in the training phase, the multi-level features of flight delay causes are gradually nested in the same vector, and combined with a hierarchical multi-task learning framework, the coarse-grained and fine-grained classification tasks are optimized respectively, and multi-scale classification information is integrated in the final result to provide a complete determination of flight delay causes; Train a delay cause classification model on a flight operation training dataset with multi-scale labels; The model classification performance was evaluated on an independent validation set, and the hierarchical classification effect of the model in determining the causes of flight delays was calculated.

7. The flight delay cause hierarchical determination method based on multi-task metrics according to claim 6, characterized in that: The method for hierarchically determining the causes of flight delays by using a delay cause classification model includes: Using the delay cause classification model, i.e. Build an independent validation set that includes multiple feature scenarios of flight operation and delay data, and covers delay reason labels of different granularities; Map the flight operation data into the constructed nested feature vector space and extract its corresponding multi-scale features; Dynamically call the hierarchical classification logic to determine the cause of the delay layer by layer from low dimension to high dimension; at the coarse-grained level, quickly identify the coarse-level cause of the delay; call higher-dimensional feature vectors for further refined classification; Ultimately, it is possible to dynamically adapt to different scenarios in actual complex application environments to hierarchical determination of flight delay causes.

8. A flight delay cause hierarchical determination system based on multi-task metrics using the flight delay cause hierarchical determination method based on multi-task metrics as described in any one of claims 1 to 7, characterized in that: include: Data collection module, used to obtain civil aviation flight operation data and delay stratification labels; A data preprocessing module, used for preprocessing data; Nested feature construction module, used to construct nested feature vector space to achieve multi-level semantic nesting of flight delay reasons; Multi-task measurement and hierarchical classification model building module, used to build multi-task measurement and hierarchical classification models using multi-level nesting; The delay cause classification model acquisition module is based on multi-task measurement and hierarchical classification model to train and obtain the delay cause classification model; The delay cause identification module uses the delay cause classification model to make hierarchical judgments on the causes of flight delays.

9. An electronic device, characterized in that: include: Processor and memory; The processor is used to execute the computer-executable instructions stored in the memory, and the processor runs the computer-executable instructions to execute the flight delay cause hierarchical determination method based on multi-task measurement as described in any one of claims 1-7.

10. A storage medium, characterized in that: It comprises a readable storage medium and a computer program stored in the readable storage medium, wherein the computer program is used to implement the flight delay cause hierarchical determination method based on multi-task metrics as described in any one of claims 1 to 7.

Images