Aircraft delivery management method, device, equipment and medium

Abstract

The invention discloses an aircraft delivery management method and device, equipment and a medium, and relates to the technical field of data processing, and the method comprises the steps: calling a basic system through an operation system, so as to obtain aircraft delivery problem information; obtaining a disposal scheme draft through the artificial intelligence system according to the aircraft delivery problem information and a historical problem database; performing risk assessment processing on the disposal scheme draft through the application system to obtain a risk assessment result corresponding to the disposal scheme draft; and through the application system, according to the risk assessment result and feedback data corresponding to the disposal scheme draft, updating a preset delivery acceptance standard to obtain a target delivery acceptance standard. The invention aims to improve the airplane delivery management capability of an enterprise and the airplane delivery satisfaction degree of a client by updating the airplane delivery acceptance standard.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to an aircraft delivery management method, device, equipment and medium. BACKGROUND

[0002] An aircraft is a large and complex equipment system. The handover process between the main manufacturing unit and the user includes multiple links such as delivery plan formulation and control, delivery technical state inspection and briefing, user ground static acceptance inspection of the aircraft, user dynamic flight inspection of the aircraft, delivery of support equipment and technical data, aircraft and product history file acceptance, delivery problem recording and disposal, aircraft transfer, and signing of handover procedures. The basic characteristics of such a complex handover process are multiple inspection and acceptance procedures, multiple plan disturbance factors, multiple delivery problems, multiple user requirements, complex delivery technical states, and multiple support equipment items. The complex handover procedures and huge workload are a difficult challenge for both parties, especially the efficient cooperation between the departments of the aircraft main manufacturing unit and the effective information transmission.

[0003] However, the current aircraft delivery management technology has the following shortcomings: after a fault is found on site, a paper document needs to be manually carried to the technical department; the problem state depends on manual update, and the form flow is easy to lose or stagnate; paper records are difficult to structure and archive, and historical problem analysis relies on manual review; the material department needs to repeatedly confirm the allocation requirements by telephone, and the industrial control site is separated from the management park; the current aircraft delivery process is inefficient and heavily relies on on-site personnel's judgment. With the rapid development of aviation equipment and technology, customers have increasingly high requirements for high-quality and efficient aircraft delivery. However, the current aircraft delivery management level has low customer satisfaction. SUMMARY

[0004] The main purpose of the present application is to provide an aircraft delivery management method, device, equipment and medium, which aims to improve the aircraft delivery management capability of an enterprise and the customer satisfaction with aircraft delivery by updating the aircraft delivery acceptance standards.

[0005] To achieve the above purpose, the present application provides an aircraft delivery management method applied to an aircraft delivery management information system, the aircraft delivery management information system comprising a basic system, an operation system, an artificial intelligence system and an application system, the basic system being used for collecting aircraft delivery related information, the operation system being used for managing aircraft delivery related information, delivery plan related information and delivery technology related information, the artificial intelligence system being used for providing a preset neural network model and a preset algorithm, and the application system being used for processing delivery problems and displaying the processing process of the delivery problems. The method comprises: The operation system is used to call the basic system to obtain aircraft delivery problem information; The artificial intelligence system is used to obtain a treatment scheme draft according to the aircraft delivery problem information and a historical problem database; The application system is used to perform risk assessment processing on the treatment scheme draft to obtain a risk assessment result corresponding to the treatment scheme draft; The application system is used to update a preset delivery acceptance standard according to the risk assessment result and feedback data corresponding to the treatment scheme draft to obtain a target delivery acceptance standard.

[0006] Specifically, the aircraft delivery problem information includes delivery problem description text, delivery problem on-site images, and delivery problem voice records, the treatment scheme draft includes treatment steps, a list of required resources, and a treatment expected period, and the historical problem database includes at least one historical delivery problem and at least one historical treatment scheme, and the historical delivery problem and the historical treatment scheme are in one-to-one correspondence. The artificial intelligence system is used to obtain a treatment scheme draft according to the aircraft delivery problem information and a historical problem database, including: Delivery problem feature vectors are constructed according to delivery problem description text, delivery problem on-site images, and delivery problem voice records; Based on the delivery problem feature vectors, the historical problem database is traversed to filter out a target historical delivery problem with the highest matching degree from the delivery problem feature vectors and a target historical treatment scheme corresponding to the target historical delivery problem; The target historical treatment scheme is determined as the treatment scheme draft.

[0007] Specifically, the delivery problem feature vectors are constructed according to delivery problem description text, delivery problem on-site images, and delivery problem voice records, including: A delivery problem description text feature vector is obtained according to the delivery problem description text through a preset BERT model; A delivery problem on-site image feature vector is obtained according to the delivery problem on-site images through a preset ResNet model; A delivery problem voice feature vector is obtained according to the delivery problem voice records through a preset Conformer-CTC model; The delivery problem description text feature vector, the delivery problem on-site image feature vector, and the delivery problem voice feature vector are subjected to splicing and fusion processing to obtain the delivery problem feature vectors.

[0008] Specifically, the highest degree of matching between the delivery problem feature vector and the target historical delivery problem is filtered out from the historical delivery problem database based on the delivery problem feature vector, and the target historical delivery problem corresponding to the target historical delivery problem is obtained, comprising: The cosine similarity between the delivery problem feature vector and each historical delivery problem corresponding vector is calculated respectively; The historical delivery problem corresponding to the maximum cosine similarity in the cosine similarity is determined as the target historical delivery problem, and the historical delivery problem corresponding to the target historical delivery problem is determined as the target historical delivery solution.

[0009] Specifically, the risk assessment result corresponding to the treatment scheme draft is obtained by performing risk assessment processing on the treatment scheme draft through the application system, comprising: At least two treatment technical indicators are extracted from the treatment scheme draft, wherein the treatment technical indicators are used to represent the condition of the technical means in the treatment scheme draft; Each index weight corresponding to each treatment technical indicator is calculated through a preset risk hierarchy structure; Each index weight is mapped to the corresponding fuzzy set through a preset multi-level fuzzy evaluation model to obtain a fuzzy evaluation result corresponding to the treatment scheme draft; A predicted risk evaluation result corresponding to the treatment scheme draft is obtained according to the treatment scheme draft through a preset predicted risk evaluation model; The fuzzy evaluation result and the predicted risk evaluation result are weighted and fused to obtain the risk assessment result.

[0010] Specifically, the target delivery acceptance standard is obtained by updating the preset delivery acceptance standard according to the risk assessment result and the feedback data corresponding to the treatment scheme draft through the application system, comprising: The target delivery acceptance standard is obtained according to the risk assessment result, the feedback data and the preset delivery acceptance standard through a preset optimized delivery acceptance standard model.

[0011] Specifically, the preset optimized delivery acceptance standard model comprises a feature extraction layer, a multi-modal fusion layer, a risk mapping layer and a standard parameter generation layer; The target delivery acceptance standard is obtained according to the risk assessment result, the feedback data and the preset delivery acceptance standard through a preset optimized delivery acceptance standard model, comprising: The risk assessment result input vector and the feedback data input vector are obtained according to the risk assessment result and the feedback data through the feature extraction layer; The multi-modal fusion layer is used for obtaining a fusion feature vector according to the risk assessment result input vector and the feedback data input vector. The risk mapping layer is used for obtaining a risk-sensitive parameter adjustment direction vector according to the fusion feature vector. The standard parameter generation layer is used for obtaining a parameter adjustment suggestion vector according to the risk-sensitive parameter adjustment direction vector and the preset delivery acceptance standard. Based on the parameter adjustment suggestion vector, a delivery acceptance parameter in the preset delivery acceptance standard is adjusted to obtain the target delivery acceptance standard.

[0012] To achieve the above object, the application further provides an aircraft delivery management device applied to an aircraft delivery management information system, wherein the aircraft delivery management information system comprises a basic system, an operation system, an artificial intelligence system and an application system; the basic system is used for collecting aircraft delivery related information; the operation system is used for managing aircraft delivery related information, delivery plan related information and delivery technology related information; the artificial intelligence system is used for providing a preset neural network model and a preset algorithm; and the application system is used for processing delivery problems and displaying a processing process of the delivery problems. The device comprises: A first unit is configured to call the basic system through the operation system to obtain aircraft delivery problem information. A second unit is configured to obtain a draft treatment scheme through the artificial intelligence system according to the aircraft delivery problem information and a historical problem database. A third unit is configured to perform risk assessment processing on the draft treatment scheme through the application system to obtain a risk assessment result corresponding to the draft treatment scheme. A fourth unit is configured to update a preset delivery acceptance standard according to a risk assessment result and feedback data corresponding to a draft treatment scheme through the application system to obtain a target delivery acceptance standard.

[0013] To achieve the above object, the application further provides a device comprising a memory storing a plurality of instructions; and a processor loading the instructions from the memory to perform steps in any of the methods provided by the application.

[0014] To achieve the above object, the application further provides a medium storing a plurality of instructions, wherein the instructions are suitable for being loaded by a processor to perform steps in any of the methods provided by the application.

[0015] This application provides an aircraft delivery management method, apparatus, equipment, and medium. First, the operating system calls the basic system to obtain aircraft delivery problem information. Then, the artificial intelligence system, based on the aircraft delivery problem information and a historical problem database, generates a draft handling plan. Next, the application system performs a risk assessment on the draft handling plan to obtain a corresponding risk assessment result. Finally, the application system updates the preset delivery acceptance standards based on the risk assessment result and the feedback data corresponding to the draft handling plan. This update improves the company's aircraft delivery management capabilities and customer aircraft delivery satisfaction. Attached Figure Description

[0016] Figure 1 A schematic diagram of the architecture of an artificial intelligence-based aircraft delivery management information system provided in an embodiment of this application; Figure 2 A flowchart illustrating the method provided in the embodiments of this application; Figure 3 A schematic diagram illustrating the application process of the preset optimized delivery and acceptance standard model provided in this application embodiment; Figure 4 This is a schematic diagram of the structure of the aircraft delivery management device provided in the embodiments of this application; Figure 5 This is a schematic diagram of the device provided in an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] The current aircraft delivery management technology suffers from the following shortcomings: after a fault is discovered on-site, paper documents must be manually carried and passed up the chain of command to the technical department; the status of the problem relies on manual updates, and forms are easily lost or delayed during the process; paper records are difficult to archive in a structured manner, and the analysis of historical problems relies on manual review; the materials department needs to repeatedly confirm allocation requirements by phone, and the industrial control site is disconnected from the management park; the current aircraft delivery process is inefficient and heavily relies on the on-site judgment of personnel. With the rapid development of aviation equipment and technology, customers have increasingly higher requirements for high-quality and efficient aircraft delivery, while customer satisfaction with the current level of aircraft delivery management is low.

[0019] Therefore, this application provides an aircraft delivery management method, apparatus, equipment, and medium to solve practical technical problems.

[0020] In some embodiments, the device may be integrated into an electronic device, such as a device, server, or similar device.

[0021] In some embodiments, the server may also be implemented as a device.

[0022] The server can be a standalone physical server, a server cluster or distributed device consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0023] The devices may include smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, etc., but are not limited to these. The devices and servers may be connected directly or indirectly via wired or wireless communication, and this application does not impose any restrictions on this connection.

[0024] The following sections provide detailed descriptions of each example. It should be noted that the sequence numbers of the following embodiments are not intended to limit the preferred order of the embodiments.

[0025] This application provides an aircraft delivery management method that can improve an enterprise's aircraft delivery management capabilities and customer aircraft delivery satisfaction by updating aircraft delivery acceptance standards.

[0026] The aircraft delivery management method is applied to an aircraft delivery management information system, which includes a basic system, an operation system, an artificial intelligence system, and an application system. The basic system is used to collect aircraft delivery-related information, the operation system is used to manage aircraft delivery-related information, delivery plan-related information, and delivery technology-related information, the artificial intelligence system is used to provide a preset neural network model and a preset algorithm, and the application system is used to handle delivery problems and display the handling process of delivery problems.

[0027] In some embodiments, the basic system includes infrastructure and a PAD-mounted hardware system. The infrastructure includes campus network, industrial control network, CPU, GPU, server, router, switch, and other equipment and facilities. The PAD-mounted hardware system is an essential facility for the problem recording and user confirmation stages in the delivery problem handling management process, and includes basic equipment such as portable mobile terminals (PADs), thin clients, intelligent management cabinets, industrial control network servers, and displays.

[0028] The aforementioned operating system primarily refers to the information system supporting delivery operations, mainly comprising five application modules: delivery plan management, delivery issue management, support equipment delivery management, delivery technical status management, and user information and requirements management. The delivery plan management module includes annual / monthly plan management, delivery progress node management, delivery operation plan management, and delivery work log management. The delivery issue management module includes delivery issue handling management, delivery inquiry management, delivery issue assessment management, and delivery issue optimization management. The support equipment delivery management module includes support equipment delivery list management, support equipment payable ledger management, support equipment matching shipment management, and support equipment handover certificate management. The delivery technical status management module includes delivery technical status cleanup, delivery technical status review, delivery legacy project ledger management, and handover memorandum ledger management. The user information and requirements management module includes survey and follow-up management, receiving user information management, receiving user requirements management, and user office supplies management.

[0029] The artificial intelligence system comprises two parts: an AI algorithm engine and AI computing power configuration. The application of AI is the result of the close integration of algorithms, computing power, and data. Algorithms serve as the working engine of AI, while computing power is its foundational configuration. Together, they analyze, mine, model, and compute structured data such as general data and business data provided by the business operation system, as well as unstructured data such as images / videos, documents / logs, etc., thereby enabling intelligent applications in aircraft delivery management scenarios.

[0030] The artificial intelligence algorithm engine includes the embedding and application of AI technologies such as neural networks, computer vision, natural language processing, deep learning, speech technology, and retrieval augmentation generation (RAG). The artificial intelligence computing power configuration includes the construction and configuration of AI chips, cloud computing, cloud services, intelligent computing platforms, intelligent servers, high-performance switching networks, and application programming interface (API) gateways.

[0031] The application system includes the construction of an intelligent delivery control center and the implementation of intelligent delivery management applications. The intelligent delivery control center, based on core business data of delivery management, focuses on visualizing data related to the overall delivery situation, product acceptance details, plan execution progress, problem handling, and the quantity of guaranteed deliveries, from both user-concerned and company-emphasized perspectives, thereby enhancing the transparency of delivery process management. Simultaneously, data processed with AI technology assists delivery managers in risk decision-making and indicator early warning. The intelligent delivery management application includes functions such as intelligent generation of delivery plans, intelligent guidance of acceptance guidelines, intelligent generation of problem handling plans, intelligent evaluation of typical problems, and intelligent optimization of acceptance standards.

[0032] The following section illustrates the aircraft delivery management information system through its architecture design in an application scenario: like Figure 1 The architecture of an AI-based aircraft delivery management information system mainly includes the following steps: Step 1: Construct the foundation layer, i.e., the basic system.

[0033] The base layer includes infrastructure and PAD hardware systems.

[0034] The infrastructure includes campus network, industrial control network, CPU, GPU, server, router, switch and other equipment and facilities.

[0035] The PAD hardware system is an essential facility for problem recording and user confirmation in the delivery problem handling management process, and includes basic equipment such as portable mobile terminals (PADs), thin clients, intelligent management cabinets, industrial control network servers, and monitors.

[0036] Specifically, the main function of the mobile terminal is to collect and digitally display structured information, images, electronic signatures, and other data related to delivery issues.

[0037] Specifically, the thin client's main function is to support the normal operation of the industrial control network server management system. The intelligent management cabinet's main function is to store, charge, and protect the connected PADs.

[0038] Specifically, the main functions of the industrial control network server are to parse and receive data on issues issued by the campus network, manage accounts, roles and basic information, and perform statistics and analysis on the delivered issue data.

[0039] Specifically, the display refers to a display device used in conjunction with a thin client to support the normal operation and use of the industrial control network server management system.

[0040] Step 2: Construct the operations layer, i.e., the operations system.

[0041] The operations layer mainly refers to the information system that supports the delivery business, which mainly includes five application modules: delivery plan management, delivery problem management, support equipment delivery management, delivery technical status management, and user information and demand management.

[0042] The delivery plan management module includes an annual / monthly plan management module, a delivery progress node management module, a delivery operation plan management module, and a delivery work log management module.

[0043] Specifically, the annual / monthly plan management module is used for issuing annual / monthly aircraft delivery quantity plans and recording actual conditions.

[0044] Specifically, the delivery progress node management module is used for key node status management during the delivery process of each aircraft.

[0045] Specifically, the delivery operation plan management module constructs an online control system for delivery operation plans based on the parent-child item management approach of delivery work packages and operation plan details, which is used for the issuance and execution feedback of daily operation plans for each aircraft.

[0046] Specifically, the delivery work log management module is used to record the daily actual work content and delivery issues during the delivery process of each aircraft, and automatically generate electronic documents.

[0047] The delivery issue management module includes a delivery issue handling management module, a delivery question consultation management module, a delivery issue assessment management module, and a delivery issue optimization management module.

[0048] Specifically, the delivery issue handling and management module is used for structured recording and handling of delivery issues raised by users. It mainly includes an industrial control network server-side management system, an industrial control network PAD installation issue recording system, and a campus network online delivery issue handling system. The industrial control network server-side management system is responsible for receiving new delivery issues raised by PAD users and transmitting them back to the campus network. It is also the basic system responsible for receiving and managing the execution results of PAD delivery issues. The industrial control network PAD installation issue recording system is responsible for recording delivery issue information and collecting user electronic signatures. Based on this, it constructs structured description rules centered on the problem product model, the location of the fault, and the nature of the fault (level three). Delivery issues are stored in a database in the form of structured modules, clearly defining key fields such as detailed problem description, date of submission, date of reset, submitting user, detailed description, and applicable standards, forming a structured database of delivery issues. The campus network online delivery issue handling system is responsible for initiating the online delivery issue handling process, including five stages: solution formulation, plan formulation, resource allocation, on-site handling, and inspection and acceptance.

[0049] Specifically, the installation and connection question consultation management module is used to realize the flow and interaction of relevant data, such as users raising installation and connection questions and experts responding.

[0050] Specifically, the delivery issue assessment and management module is based on the structured data of the delivery issue handling management module. It extracts data such as the number of repetitions, the degree of impact, and the handling cycle based on assessment dimensions such as "repetition", "security", and "handling cycle" to identify typical repetitive issues. It calculates the importance score of each typical issue and sorts them, thereby selecting typical delivery issues and forming a database.

[0051] Specifically, the delivery problem optimization management module is used to initiate optimization and improvement processes such as root cause analysis, learning from past mistakes, formulating measures, and evaluating the effects of typical delivery problems that have been selected.

[0052] The support equipment delivery management module includes a support equipment delivery list management module, a support equipment issue ledger management module, a support equipment matching shipment management module, and a support equipment handover certificate management module.

[0053] Specifically, the support equipment delivery list management module is used to establish delivery information such as the name, model, quantity, and price of all support equipment to be delivered in each type of equipment contract.

[0054] Specifically, the support equipment delivery ledger management module is used to establish a support equipment delivery ledger based on the equipment delivery quantity and delivery schedule.

[0055] Specifically, the support equipment matching and shipment management module is used to carry out various support equipment matching and shipment work based on the support equipment shipment ledger, and to record and provide feedback in real time on the quantity of support equipment missing, the reason for the missing equipment, and the estimated arrival time of each support equipment.

[0056] Specifically, the support equipment handover certificate management module is used to manage the accounting of support equipment handover certificates, including information such as handover time, the two parties involved in the handover, the model being handed over, and the quantity.

[0057] The delivery technology status management includes a delivery technology status cleanup module, a delivery technology status review module, a delivery legacy project ledger management module, and a handover memorandum ledger management module.

[0058] Specifically, the delivery technical status cleanup module is used to clean up the physical installation status of missing items, faulty items, and pending engineering changes, determine the differences between the physical status and the delivery technical status, and clarify the relevant work that needs to be performed in the delivery stage.

[0059] Specifically, the delivery technology status review module reviews whether the design and manufacturing requirements have been fully implemented based on the data in the delivery technology status clearing module, and clarifies the cycle, methods, materials, and responsible persons for the work to be performed in the delivery stage.

[0060] Specifically, the delivery legacy project ledger management module is used to manage the manufacturing basis and procedures that have not been reset after product delivery.

[0061] Specifically, the handover memorandum ledger management module manages the physical work that has not been performed after product delivery.

[0062] The user information and demand management module includes a survey and follow-up management module, a user information management module, a user demand management module, and a user office supplies management module.

[0063] Specifically, the survey and follow-up management module is used to manage the time, location, personnel, and matters of each user survey and user follow-up activity in a structured manner.

[0064] Specifically, the user information management module is used for structured management of information such as the user's name, position, and employer.

[0065] Specifically, the user demand management module is used to classify and categorize user demands during the installation process, including their urgency, solutions, responsible persons, and implementation status.

[0066] Specifically, the user office supplies management module is used for managing the ledger and approving processes for the office supplies needed by users on a daily basis.

[0067] Step 3: Construct the AI ​​layer, i.e., the artificial intelligence system.

[0068] The AI ​​layer mainly consists of two parts: an artificial intelligence algorithm engine and an artificial intelligence computing power configuration.

[0069] The artificial intelligence algorithm engine incorporates and utilizes AI technologies such as neural networks, computer vision, natural language processing, deep learning, speech technology, and retrieval augmentation generation (RAG).

[0070] The artificial intelligence computing power configuration includes the construction and configuration of AI chips, cloud computing, cloud services, intelligent computing platforms, intelligent servers, high-performance switching networks, application programming interface (API) gateways, etc.

[0071] Step 4: Construct the application layer, i.e., the application system.

[0072] The application layer includes the construction of an intelligent delivery control center and the implementation of intelligent delivery management applications. The intelligent delivery control center mainly visualizes business data to enhance the transparency of delivery process management; at the same time, it uses data processed by AI technology to assist delivery managers in risk decision-making and indicator early warning.

[0073] The intelligent delivery control center includes functions such as aircraft delivery status prediction, plan completion perception, problem handling status push, intelligent decision-making on delivery risks, and intelligent early warning of delivery indicators.

[0074] Specifically, the product delivery status prediction integrates data from the delivery progress node management module. Through data conversion and modeling, it visualizes the user units, regions, quantities, and times of delivery for various types of equipment, while also predicting future delivery trends to assist in the advance adjustment of delivery plans.

[0075] Specifically, the plan completion status perception integrates delivery operation plan and delivery work log data, and displays in detail the completed work packages and completion time of the receiving equipment, the work packages in progress, the unfinished work packages, and the overall delivery progress.

[0076] Specifically, the issue handling status push integrates data from the delivery issue handling management module. By matching the issue date with the current date, it extracts relevant information on delivery issues with a handling cycle exceeding 3 days and displays it in the control center, alerting the responsible unit and urging them to speed up the handling process.

[0077] Specifically, the intelligent decision-making mechanism for delivery risks integrates risk items such as missing parts and missing documentation that affect the execution of the delivery operation plan management module, as well as overdue issues in the delivery problem handling module. After analysis and mining by AI algorithms, the delivery risks are statistically analyzed and visualized according to the classification and grading method of high risk, medium risk and low risk, to assist managers in making risk decisions.

[0078] Specifically, the intelligent early warning system for delivery indicators integrates data from various modules regarding delivery cycles, delivery issues, and handover memos, performs monthly data statistics, and provides early warnings by comparing the data with early warning values.

[0079] The intelligent delivery management application includes functions such as intelligent generation of delivery plans, intelligent guidance of acceptance and installation guidelines, intelligent generation of problem handling plans, intelligent evaluation of typical problems, and intelligent optimization of acceptance and installation standards.

[0080] Specifically, the intelligent generation of the delivery plan refers to the management of the entire delivery process, including aircraft sealing, modification, painting, delivery, inspection, and transfer. This involves using existing acceptance inspection forms, technical documents, design documents, process documents, and meeting minutes as learning samples for artificial intelligence. The process includes analyzing the preceding and following processes, required resources, and the implementation requirements and impacts on other matters of each item in the delivery plan (both fixed and miscellaneous), as well as business data from the operations information system. Using AI algorithms such as neural networks and deep learning, the process analyzes key delay points and risk points in the execution of each aircraft delivery plan. Based on weighted boundary conditions, it completes the analysis of key disturbance factors and their impact cycles in the delivery plan, and intelligently generates delivery plan adjustment strategies under the preceding and following process conditions.

[0081] Specifically, the intelligent guidance for aircraft delivery refers to breaking down the aircraft delivery process into multiple modules, providing delivery users with a personalized, multimodal knowledge base and intelligent question-and-answer functionality. Based on computer vision technology, it achieves digital and visual operation and intelligent feedback, enabling dynamic demonstrations of delivery work and delivery plans. It also incorporates natural language processing, enhanced retrieval generation, and intelligent question-and-answer technology using voice. Simultaneously, it utilizes user feedback and learning data to provide management personnel with fundamental data support for continuously updating and improving the demonstration content.

[0082] Specifically, the intelligent generation of problem-solving solutions refers to using natural language processing and retrieval enhancement technologies to extract feature values ​​from all delivery problems based on a structured database of delivery problems, and constructing an intelligent matching model for delivery problems. When a newly proposed delivery problem is sent back to the online problem-solving system of the campus network, the system automatically identifies similar problems with the highest feature value matching degree based on the model's calculation results, and generates historical best solutions and existing risks, facilitating rapid handling of delivery problems and risk warnings.

[0083] Specifically, the intelligent assessment of typical problems refers to establishing a hierarchical evaluation model for delivery problems by using the analytic hierarchy process (AHP), multi-level fuzzy evaluation method, and technologies such as natural language processing, retrieval enhancement generation, and deep learning to calculate the problem level. Simultaneously, problem classification management is implemented, categorizing problems into fault-related problems, protection and maintenance problems, suggestion-related problems, and documentation problems. For each category, different handling and optimization feedback measures are adopted based on the problem-level priority, achieving intelligent hierarchical and classified management of delivery quality problems.

[0084] Specifically, the intelligent optimization of the acceptance standard refers to using a database of delivery issues, acceptance questions, and acceptance standards as a foundation. Through technologies such as natural language processing, retrieval enhancement generation, and deep learning, the database is adaptively learned and continuously improved. This generates corresponding optimization and revision strategies for supplementing, improving, or deleting standards, gradually converging and simplifying the acceptance inspection and acceptance process. Under the premise of meeting safety thresholds and quality requirements, the optimal acceptance standard that is acceptable to the user, selectable, and has a limited acceptance cycle is intelligently output.

[0085] like Figure 2 The specific process of the method provided in the embodiments of this application can be as follows: S110. The operating system calls the basic system to obtain information on aircraft delivery issues.

[0086] In some embodiments, the operating system acts as an information management hub, calling the multi-source data acquisition module of the basic system through standardized interfaces. The basic system, relying on an integrated PAD hardware system and IoT sensors, collects aircraft delivery problem information in real time, which may include: Structured information: The problem is categorized into three levels: product model, cabin location, and nature of the malfunction, along with the time the issue was raised, the user who raised the issue, and the level of urgency. Unstructured information: Problem description text (entered by maintenance personnel), on-site photos (taken by equipment), and voice recordings (audio of on-site communication).

[0087] The operating system cleans and stores the above information in a structured database of delivery issues, providing standardized input for subsequent processing.

[0088] S120. The artificial intelligence system generates a draft solution based on the aircraft delivery problem information and the historical problem database.

[0089] In some embodiments, the aircraft delivery problem information includes a description of the delivery problem, on-site images of the delivery problem, and voice recordings of the delivery problem; the draft handling plan includes handling steps, a list of resources required for handling, and an expected handling period; and the historical problem database includes at least one historical delivery problem and at least one historical handling plan, with each historical delivery problem corresponding to a historical handling plan.

[0090] Specifically, the process of obtaining a draft solution through the artificial intelligence system, based on the aircraft delivery problem information and the historical problem database, includes the steps A1 to A3 as shown below: A1. Construct a feature vector for the delivery problem based on the delivery problem description text, on-site images of the delivery problem, and voice recordings of the delivery problem; In some embodiments, the step of constructing a delivery problem feature vector based on the delivery problem description text, on-site images of the delivery problem, and voice recordings of the delivery problem includes the steps A11 to A14 shown below: A11. Using a preset BERT model, obtain the feature vector of the delivery problem description text based on the delivery problem description text.

[0091] In some embodiments, a preset BERT model is used to encode the delivery problem description text, and the contextual semantics are learned through a masked language model (MLM) to output a 768-dimensional feature vector of the delivery problem description text, capturing key information such as fault type (e.g., "hydraulic leak") and affected components (e.g., "right wing main landing gear").

[0092] A12. Using a preset ResNet model, obtain the feature vector of the delivery problem scene image based on the delivery problem scene image.

[0093] In some embodiments, a preset ResNet model can be used to process the on-site images, extract visual features (such as the appearance and texture of loose bolts and the degree of wear of components) through a 16-layer residual network, and output a 2048-dimensional on-site image feature vector.

[0094] A13. Using the preset Conformer-CTC model, obtain the delivery problem speech feature vector based on the delivery problem speech recording.

[0095] In some embodiments, the speech recording can be converted into text by a preset Conformer-CTC model and then encoded again by a BERT model, or the speech spectrum features can be directly extracted to output a 512-dimensional speech feature vector to identify key semantics in maintenance instructions (such as "check pipeline sealing").

[0096] A14. The feature vector of the delivery problem description text, the feature vector of the delivery problem scene image, and the feature vector of the delivery problem voice are concatenated and fused to obtain the feature vector of the delivery problem.

[0097] In some embodiments, the feature vector of the delivery problem description text, the feature vector of the delivery problem scene image, and the feature vector of the delivery problem speech are concatenated along the dimension axis (e.g., 768+2048+512=3328 dimensions), and the scale difference is eliminated by layer normalization to form a delivery problem feature vector containing multimodal information.

[0098] A2. Based on the delivery problem feature vector, traverse the historical problem database, and select the target historical delivery problem and the corresponding target historical handling plan from the historical delivery problems that have the highest degree of matching with the delivery problem feature vector. In some embodiments, the step of traversing the historical problem database based on the delivery problem feature vector, and selecting the target historical delivery problem and the corresponding target historical handling plan from the historical delivery problems that have the highest degree of matching with the delivery problem feature vector, includes the steps A21 to A22 as shown below: A21. Calculate the cosine similarity between the feature vector of the delivery problem and the vector corresponding to each historical delivery problem.

[0099] In some embodiments, each historical delivery problem in the historical problem database has its corresponding vector pre-generated through the same multimodal feature extraction process (which can be stored in a pre-set vector database, which can be set within the historical problem database). When traversing the vector database, the cosine similarity between the current delivery problem feature vector and all historical problem vectors is calculated.

[0100] A22. The historical delivery problem corresponding to the cosine similarity with the largest value among the cosine similarities is determined as the target historical delivery problem, and the historical handling scheme corresponding to the target historical delivery problem is determined as the target historical handling scheme.

[0101] In some embodiments, the historical problem with the highest cosine similarity (largest value) is selected as the target historical delivery problem, and its corresponding historical handling plan (including handling steps, resource list, and expected cycle) is associated with it.

[0102] A3. The proposed historical disposal plan for the target is defined as the draft disposal plan.

[0103] In some embodiments, the target historical handling plan can be directly output as a draft handling plan to ensure that the target historical handling plan contains historical experience most similar to the current problem. For example, if the handling steps of the historical case are "disassembly → inspection → replacement of seals", the draft will inherit the process and adjust the expected cycle according to the urgency of the current problem.

[0104] S130. The application system is used to perform a risk assessment on the draft disposal plan to obtain the risk assessment result corresponding to the draft disposal plan.

[0105] In some embodiments, the risk assessment of the draft disposal plan through the application system to obtain the risk assessment result corresponding to the draft disposal plan includes the steps B1 to B5 as shown below: B1. Extract at least two disposal technical indicators from the draft disposal plan, wherein the disposal technical indicators are used to characterize the status of the technical means in the draft disposal plan.

[0106] In some embodiments, key technical indicators are parsed from the draft disposal plan, for example: Operational complexity: the number of procedures in the handling process and the frequency of use of special tools; Resource dependence: procurement cycle of required spare parts, professional personnel qualification requirements; Technology maturity: the historical rework rate of this type of failure and whether it involves the application of new technologies.

[0107] In some embodiments, NLP models (such as LSTM+CRF) can be used to identify indicator keywords from the draft disposal plan text and map them to a preset indicator system.

[0108] B2. By pre-setting a risk hierarchy structure, the weights of each indicator corresponding to each disposal technical indicator are calculated.

[0109] In some embodiments, a three-tiered indicator system can be constructed based on the Analytic Hierarchy Process (AHP): Target layer: Risk level (high / medium / low); Criteria Level: Technological Risk, Cyclical Risk, Resource Risk; Indicator level: operational complexity, expected cycle deviation, spare parts scarcity, etc.

[0110] A pairwise comparison matrix can be formed through expert scoring, the weight of each indicator can be calculated (e.g., operational complexity weight 0.3, expected cycle weight 0.4), and the logical rationality can be ensured through consistency checks.

[0111] B3. By pre-setting a multi-level fuzzy evaluation model, the weights of each indicator are mapped to the corresponding fuzzy sets to obtain the fuzzy evaluation results corresponding to the draft disposal plan.

[0112] In some embodiments, the index values ​​are mapped to fuzzy sets (e.g., "high" corresponds to a membership degree ≥ 0.7, "medium" to 0.3~0.7, and "low" to ≤ 0.3), for example: If the disposal cycle exceeds the historical average cycle by 50%, then the membership degree of "cycle risk" to "high" is 0.8. If special tools need to be imported, the membership degree of "resource dependence" to "high" is 0.9.

[0113] The fuzzy evaluation results of each criterion layer (such as the technical risk membership vector [0.6, 0.3, 0.1]) can be synthesized using the Mamdani fuzzy inference rule.

[0114] B4. By using a preset risk assessment model, the risk assessment results corresponding to the draft disposal plan are obtained based on the draft disposal plan.

[0115] In some embodiments, historical disposal data can be used to train a risk assessment model (such as a random forest or LSTM) to obtain the preset predictive risk assessment model. Then, input the scheme characteristics (number of steps, resource list, historical rework rate) in the draft disposal plan, and output the risk level probability distribution (e.g., high risk probability 0.65, medium risk 0.25), which is the predicted risk assessment result.

[0116] B5. The fuzzy evaluation results and the predicted risk evaluation results are weighted and fused to obtain the risk assessment results.

[0117] In some embodiments, the fuzzy evaluation result (weight 0.6) and the predicted risk evaluation result (weight 0.4) are weighted and fused together, for example: Risk assessment result = 0.6 × fuzzy evaluation result + 0.4 × predicted risk assessment result A quantitative risk assessment result is obtained (e.g., a risk value of 0.78 triggers a high-risk warning).

[0118] S140. Through the application system, based on the risk assessment results and the feedback data corresponding to the draft disposal plan, the preset delivery and acceptance standards are updated to obtain the target delivery and acceptance standards.

[0119] In some embodiments, the step of updating the preset delivery and acceptance criteria through the application system based on the risk assessment results and the feedback data corresponding to the draft disposal plan to obtain the target delivery and acceptance criteria includes the following specific implementation process: By pre-setting and optimizing the delivery and acceptance standard model, the target delivery and acceptance standard is obtained based on the risk assessment results, the feedback data, and the pre-set delivery and acceptance standard.

[0120] In some embodiments, the preset optimized delivery and acceptance standard model includes a feature extraction layer, a multimodal fusion layer, a risk mapping layer, and a standard parameter generation layer.

[0121] Specifically, the step of obtaining the target delivery and acceptance standard by pre-setting an optimized delivery and acceptance standard model, based on the risk assessment results, the feedback data, and the pre-set delivery and acceptance standard, includes the following steps C1 to C5: C1. Through the feature extraction layer, based on the risk assessment result and the feedback data, the risk assessment result input vector and the feedback data input vector are obtained.

[0122] In some embodiments, the risk level (high / medium / low) is converted into one-hot encoding and concatenated with the risk scores of each dimension (e.g., technical risk 0.8, cycle risk 0.7) to obtain the risk assessment result input vector; the structured data (actual processing time, resource consumption) in the feedback data is standardized and input, and the unstructured data (inspection report text, customer feedback) is encoded into a feature vector through BERT, which is the feedback data input vector.

[0123] C2. Through the multimodal fusion layer, a fusion feature vector is obtained based on the risk assessment result input vector and the feedback data input vector.

[0124] In some embodiments, the association weights between risk and feedback data (such as key parameters in the detection report corresponding to high-risk issues) are calculated through a cross-modal attention mechanism, and then fused into a 512-dimensional fusion feature vector using the gated recurrent unit (GRU) in the multimodal fusion layer to capture the mapping relationship between risk and execution effect.

[0125] C3. Through the risk mapping layer, a risk-sensitive parameter adjustment direction vector is obtained based on the fused feature vector.

[0126] In some embodiments, a "risk-parameter" association graph can be constructed based on the graph neural network (GNN) in the risk mapping layer. Nodes represent risk types (e.g., "insufficient assembly accuracy") and acceptance parameters (e.g., "bolt torque threshold"), and edge weights represent the historical association strength. Through a message passing algorithm, a risk-sensitive parameter adjustment direction vector (e.g., the adjustment trend of "tightening torque upper limit + 10%)" is output.

[0127] C4. Through the standard parameter generation layer, a parameter adjustment suggestion vector is obtained based on the risk-sensitive parameter adjustment direction vector and the preset delivery and acceptance criteria.

[0128] In some embodiments, based on the current acceptance standard parameters in the preset delivery acceptance standards, specific adjustment suggestions are generated by the Transformer decoder in the standard parameter generation layer. For example, if the risk mapping layer points to "insufficient pressure detection accuracy of the hydraulic system", then a parameter adjustment suggestion vector is generated to "shorten the pressure sensor calibration cycle from 12 months to 6 months".

[0129] C5. Based on the parameter adjustment suggestion vector, adjust the delivery and acceptance parameters in the preset delivery and acceptance standard to obtain the target delivery and acceptance standard.

[0130] In some embodiments, the parameter adjustment suggestion vector can be used to adjust the values ​​or update the rules of parameters (such as threshold range and testing cycle) in the preset delivery and acceptance criteria. After verification by an expert knowledge graph (to ensure that it does not violate engineering specifications), the target delivery and acceptance criteria are formed, for example: Original standard: "Landing gear bolt torque 150±10 N•m"; After adjustment: "Landing gear bolt torque 150±8 N•m (high-risk scenario)".

[0131] In summary, this application provides an aircraft delivery management method that improves an enterprise's aircraft delivery management capabilities and customer aircraft delivery satisfaction by updating aircraft delivery acceptance standards.

[0132] To better implement the above methods, this application also provides an aircraft delivery management device, which can be integrated into an electronic device, such as a mobile phone, tablet computer, smart Bluetooth device, laptop computer, or personal computer; the server can be a single server or a server cluster composed of multiple servers.

[0133] For example, in this embodiment, the method of this application embodiment will be described in detail by taking the aircraft delivery management device specifically integrated into the equipment as an example.

[0134] For example, such as Figure 4 As shown, the aircraft delivery management device 400 may include a first unit 401, a second unit 402, a third unit 403, and a fourth unit 404, and is applied to an aircraft delivery management information system. The aircraft delivery management information system includes a basic system, an operation system, an artificial intelligence system, and an application system. The basic system is used to collect aircraft delivery-related information, the operation system is used to manage aircraft delivery-related information, delivery plan-related information, and delivery technology-related information, the artificial intelligence system is used to provide a preset neural network model and a preset algorithm, and the application system is used to handle delivery problems and display the handling process of delivery problems. The device includes: The first unit 401 is used to call the basic system through the operating system to obtain aircraft delivery problem information; The second unit 402 is used to obtain a draft solution through the artificial intelligence system based on the aircraft delivery problem information and the historical problem database; The third unit 403 is used to perform risk assessment processing on the draft disposal plan through the application system to obtain the risk assessment result corresponding to the draft disposal plan; The fourth unit 404 is used to update the preset delivery and acceptance standards through the application system based on the risk assessment results and the feedback data corresponding to the draft disposal plan, so as to obtain the target delivery and acceptance standards.

[0135] In practice, each of the above units can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units, please refer to the previous method embodiments, which will not be repeated here.

[0136] As can be seen from the above, the embodiments of this application can improve the company's aircraft delivery management capabilities and customer aircraft delivery satisfaction by updating the aircraft delivery and acceptance standards.

[0137] This application also provides an electronic device, which can be a device, a server, or other similar device. The device can be a mobile phone, tablet computer, smart Bluetooth device, laptop computer, personal computer, etc.; the server can be a single server or a server cluster composed of multiple servers, etc.

[0138] In some embodiments, the product processing device may also be integrated into multiple electronic devices, such as multiple servers, with the multiple servers implementing the aircraft delivery management method of this application.

[0139] In this embodiment, the electronic device of this embodiment will be used as an example for detailed description, such as... Figure 5 As shown, it illustrates a structural schematic diagram of the device 500 involved in the embodiments of this application. Specifically: The device 500 may include components such as a processor 501 with one or more processing cores, a memory 502 with one or more media, a power supply 503, an input module 504, and a communication module 505. Those skilled in the art will understand that... Figure 5 The structure of device 500 shown does not constitute a limitation on device 500, and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 501 is the control center of the device 500. It connects various parts of the device 500 via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 502, and by calling data stored in the memory 502, it performs various functions of the device 500 and processes data, thereby providing overall monitoring of the device 500. In some embodiments, the processor 501 may include one or more processing cores; in some embodiments, the processor 501 may integrate an application processor and a modem processor, wherein the application processor mainly handles operating the device, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor 501.

[0140] The memory 502 can be used to store software programs and modules. The processor 501 executes various functional applications and data processing by running the software programs and modules stored in the memory 502. The memory 502 may mainly include a program storage area and a data storage area. The program storage area may store application programs required for operating the device and at least one function (such as sound playback function, image playback function, etc.); the data storage area may store data created based on the use of the device 500. In addition, the memory 502 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 502 may also include a memory controller to provide the processor 501 with access to the memory 502.

[0141] The device 500 also includes a power supply 503 that supplies power to the various components. In some embodiments, the power supply 503 can be logically connected to the processor 501 through a power management device, thereby enabling functions such as managing charging, discharging, and power consumption through the power management device. The power supply 503 may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0142] The device 500 may also include an input module 504, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0143] The device 500 may also include a communication module 505. In some embodiments, the communication module 505 may include a wireless module. The device 500 can use the wireless module of the communication module 505 to perform short-range wireless transmission, thereby providing users with wireless broadband Internet access. For example, the communication module 505 can be used to help users send and receive emails, browse web pages, and access streaming media.

[0144] Although not shown, device 500 may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 501 in device 500 loads the executable files corresponding to the processes of one or more applications into memory 502 according to the following instructions, and the processor 501 runs the applications stored in memory 502 to realize various functions, as follows: The operating system calls the basic system to obtain information about aircraft delivery issues; The artificial intelligence system generates a draft solution based on the aircraft delivery problem information and the historical problem database. The application system is used to perform a risk assessment on the draft disposal plan to obtain the risk assessment result corresponding to the draft disposal plan. The application system updates the preset delivery and acceptance standards based on the risk assessment results and the feedback data corresponding to the draft disposal plan, thereby obtaining the target delivery and acceptance standards.

[0145] As can be seen from the above, the embodiments of this application can improve the company's aircraft delivery management capabilities and customer aircraft delivery satisfaction by updating the aircraft delivery and acceptance standards.

[0146] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be accomplished by instructions, or by instructions controlling related hardware. These instructions can be stored in a medium and loaded and executed by a processor.

[0147] Therefore, embodiments of this application provide a medium storing multiple instructions that can be loaded by a processor to execute steps in any of the aircraft delivery management methods provided in embodiments of this application. For example, the instructions can execute the following steps: The operating system calls the basic system to obtain information about aircraft delivery issues; The artificial intelligence system generates a draft solution based on the aircraft delivery problem information and the historical problem database. The application system is used to perform a risk assessment on the draft disposal plan to obtain the risk assessment result corresponding to the draft disposal plan. The application system updates the preset delivery and acceptance standards based on the risk assessment results and the feedback data corresponding to the draft disposal plan, thereby obtaining the target delivery and acceptance standards.

[0148] The medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0149] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a medium. A processor of a computer device reads the computer instructions from the medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations of the above embodiments.

[0150] Since the instructions stored in the medium can execute the steps in any of the aircraft delivery management methods provided in the embodiments of this application, the beneficial effects that any of the aircraft delivery management methods provided in the embodiments of this application can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0151] The above provides a detailed description of an aircraft delivery management method, apparatus, equipment, and medium provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An aircraft delivery management method, characterized in that, This is applied to an aircraft delivery management information system, which includes a basic system, an operation system, an artificial intelligence system, and an application system. The basic system is used to collect aircraft delivery-related information, the operation system is used to manage aircraft delivery-related information, delivery plan-related information, and delivery technology-related information, the artificial intelligence system is used to provide preset neural network models and preset algorithms, and the application system is used to handle delivery problems and display the handling process of delivery problems. The method includes: The operating system calls the basic system to obtain information about aircraft delivery issues; The artificial intelligence system generates a draft solution based on the aircraft delivery problem information and the historical problem database. The application system is used to perform a risk assessment on the draft disposal plan to obtain the risk assessment result corresponding to the draft disposal plan. The application system updates the preset delivery and acceptance standards based on the risk assessment results and the feedback data corresponding to the draft disposal plan, thereby obtaining the target delivery and acceptance standards.

2. The method as described in claim 1, characterized in that, The aircraft delivery problem information includes a delivery problem description text, on-site images of the delivery problem, and voice recordings of the delivery problem. The draft handling plan includes handling steps, a list of resources required for handling, and an expected handling cycle. The historical problem database includes at least one historical delivery problem and at least one historical handling plan, with each historical delivery problem corresponding to a historical handling plan. The process involves using the artificial intelligence system to generate a draft solution based on aircraft delivery issue information and a historical issue database, including: Based on the delivery problem description text, on-site images of the delivery problem, and voice recordings of the delivery problem, construct a feature vector for the delivery problem; Based on the delivery problem feature vector, the historical problem database is traversed to select the target historical delivery problem that matches the delivery problem feature vector the most and the target historical handling solution corresponding to the target historical problem. The proposed historical disposal plan is defined as the draft disposal plan.

3. The method as described in claim 2, characterized in that, The step of constructing a delivery problem feature vector based on the delivery problem description text, on-site images of the delivery problem, and voice recordings of the delivery problem includes: By using a pre-defined BERT model, a feature vector of the delivery problem description text is obtained based on the delivery problem description text. By using a pre-defined ResNet model, feature vectors of the delivery problem scene images are obtained based on the scene images of the delivery problem. By using the pre-defined Conformer-CTC model, the speech feature vector of the delivery problem is obtained based on the speech recording of the delivery problem; The feature vector of the delivery problem is obtained by concatenating and fusing the text feature vector describing the delivery problem, the image feature vector of the delivery problem site, and the voice feature vector of the delivery problem.

4. The method as described in claim 2, characterized in that, The step of traversing the historical problem database based on the delivery problem feature vector, and selecting the target historical delivery problem and its corresponding target historical handling plan from the historical delivery problems that have the highest degree of matching with the delivery problem feature vector, includes: The cosine similarity between the feature vector of the delivery problem and the vector corresponding to each historical delivery problem is calculated respectively. The historical delivery problem corresponding to the cosine similarity with the largest value among the cosine similarities is determined as the target historical delivery problem, and the historical handling scheme corresponding to the target historical delivery problem is determined as the target historical handling scheme.

5. The method as described in claim 1, characterized in that, The step of performing a risk assessment on the draft disposal plan through the application system to obtain the risk assessment result corresponding to the draft disposal plan includes: At least two disposal technical indicators are extracted from the draft disposal plan, wherein the disposal technical indicators are used to characterize the status of the technical means in the draft disposal plan; By pre-setting a risk hierarchy structure, the weights of each indicator corresponding to each disposal technical indicator are calculated. By pre-setting a multi-level fuzzy evaluation model, the weights of each indicator are mapped to the corresponding fuzzy sets to obtain the fuzzy evaluation results corresponding to the draft disposal plan. By using a pre-set predictive risk assessment model, the predictive risk assessment results corresponding to the draft disposal plan are obtained based on the draft disposal plan. The fuzzy evaluation results and the predicted risk evaluation results are weighted and fused to obtain the risk assessment results.

6. The method as described in claim 1, characterized in that, The process involves updating the preset delivery and acceptance standards through the application system based on the risk assessment results and feedback data corresponding to the draft disposal plan, to obtain the target delivery and acceptance standards, including: By pre-setting and optimizing the delivery and acceptance standard model, the target delivery and acceptance standard is obtained based on the risk assessment results, the feedback data, and the pre-set delivery and acceptance standard.

7. The method as described in claim 6, characterized in that, The preset optimized delivery and acceptance standard model includes a feature extraction layer, a multimodal fusion layer, a risk mapping layer, and a standard parameter generation layer; The step of obtaining the target delivery and acceptance standard by means of a preset optimized delivery and acceptance standard model, based on the risk assessment results, the feedback data, and the preset delivery and acceptance standard, includes: Through the feature extraction layer, the risk assessment result input vector and the feedback data input vector are obtained based on the risk assessment result and the feedback data. The multimodal fusion layer obtains a fused feature vector based on the risk assessment result input vector and the feedback data input vector. Through the risk mapping layer, a risk-sensitive parameter adjustment direction vector is obtained based on the fused feature vector; Through the standard parameter generation layer, a parameter adjustment suggestion vector is obtained based on the risk-sensitive parameter adjustment direction vector and the preset delivery and acceptance criteria. Based on the parameter adjustment suggestion vector, the delivery and acceptance parameters in the preset delivery and acceptance criteria are adjusted to obtain the target delivery and acceptance criteria.

8. An aircraft delivery management device, characterized in that, This is applied to an aircraft delivery management information system, which includes a basic system, an operation system, an artificial intelligence system, and an application system. The basic system is used to collect aircraft delivery-related information, the operation system is used to manage aircraft delivery-related information, delivery plan-related information, and delivery technology-related information, the artificial intelligence system is used to provide preset neural network models and preset algorithms, and the application system is used to handle delivery problems and display the handling process of delivery problems. The device includes: The first unit is used to call the basic system through the operating system to obtain information on aircraft delivery issues; The second unit is used to obtain a draft solution through the artificial intelligence system based on the aircraft delivery problem information and the historical problem database; The third unit is used to perform risk assessment processing on the draft disposal plan through the application system to obtain the risk assessment result corresponding to the draft disposal plan; The fourth unit is used to update the preset delivery and acceptance standards through the application system based on the risk assessment results and the feedback data corresponding to the draft disposal plan, so as to obtain the target delivery and acceptance standards.

9. A device, characterized in that, The method includes a processor and a memory, the memory storing multiple instructions; the processor loads instructions from the memory to perform the steps of the method as described in any one of claims 1 to 7.

10. A medium, characterized in that, The medium stores a plurality of instructions adapted for loading by a processor to execute the steps of the method according to any one of claims 1 to 7.

Images