An ai service unavailability emergency handover method and system

Abstract

The application relates to the technical field of data processing, and discloses an AI service emergency switching method and system. The method comprises the following steps: performing multidimensional health detection on a main AI service of an airport efficiency platform to obtain a real-time running state of the main AI service; performing service availability evaluation on the main AI service to obtain a state judgment result and trigger an emergency switching process of the airport efficiency platform; storing request images in a preset temporary message queue and obtaining a target standby AI service corresponding to a running guarantee service; starting return visit processing of all request images, obtaining service requests corresponding to the request images, and distributing the service requests to the target standby AI service for processing; continuously performing health detection on a running state of the target standby AI service, and periodically performing connection attempts on the main AI service to obtain a platform recovery switching strategy of the airport efficiency platform, so that AI service fault emergency switching is effectively realized, and airport running guarantee business continuity and stability are ensured.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to an emergency switching method and system for AI service unavailability. Background Technology

[0002] Airport operation support services encompass core operational aspects such as flight scheduling, gate allocation, and resource coordination. The stable operation of these services heavily relies on the main AI service on the airport efficiency platform to provide intelligent decision-making and execution support. Currently, the methods for monitoring the operational status of the main AI service in airport scenarios have significant limitations. They can only collect status data from a single dimension, such as interface connectivity and basic load, and cannot conduct multi-dimensional health checks based on the actual needs of airport operation support services. It is difficult to comprehensively and accurately identify the actual capabilities of the main AI service in supporting various support services. When the main AI service experiences operational anomalies, it is impossible to quickly form accurate service availability assessment conclusions and status judgment results, and it is difficult to initiate emergency response procedures in a timely manner. This can easily lead to execution delays and business interruptions in airport operation support services, failing to meet the core requirements of airport operation for the stability of intelligent services.

[0003] The existing AI service emergency switchover solution is not designed to adapt to the specific characteristics of airport operation support services. It lacks a secure temporary storage and complete restoration mechanism for service requests, which can easily lead to service request loss and incomplete business data in emergency situations. The selection of backup AI services is not matched with the airport operation support priority strategy, which cannot guarantee the priority processing of high-priority support services. The emergency switchover process lacks a civil aviation business compliance verification step, which cannot guarantee the security and compliance of the switchover operation. After the switchover is completed, the operation status of the backup AI service is not continuously monitored, and a smooth switchover mechanism is not established after the main AI service recovers. It is impossible to achieve a seamless connection between the emergency state and the recovery state, and it is difficult to ensure the continuous and stable execution of the entire airport operation support service process. This does not meet the design requirements of reliable operation and intelligent collaboration of the airport efficiency platform. Summary of the Invention

[0004] This invention provides an emergency switching method and system for AI service unavailability to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an emergency switching method for AI service unavailability, comprising: C1: Based on the airport operation support service request, perform multi-dimensional health detection on the main AI service of the airport efficiency platform to obtain the real-time operating status of the main AI service; C2: Based on the real-time operating status, perform a service availability assessment on the main AI service to obtain a status judgment result. When the status judgment result is unavailable, trigger the emergency switching process of the airport efficiency platform. C3: Based on the unique identifier of the request image corresponding to the airport operation support service request, store the request image in a preset temporary message queue, select the target backup AI service based on the preset airport operation support priority strategy, and obtain the target backup AI service corresponding to the operation support service. C4: Initiate the revisit processing of all the request images in the temporary message queue, restore each request image to obtain the service request to be processed corresponding to the request image, and distribute it to the target backup AI service for processing; C5: Perform continuous health checks on the operational status of the target backup AI service and periodically attempt to connect to the main AI service to obtain the platform recovery and switching strategy of the airport efficiency platform.

[0006] In a preferred embodiment, the step of performing multi-dimensional health checks on the main AI service of the airport efficiency platform based on airport operation support service requests to obtain the real-time operating status of the main AI service includes: Based on airport operation support service requests, multi-dimensional health checks are performed on the interface response status, task processing efficiency, and resource consumption of the main AI service of the airport efficiency platform to obtain the operating parameters and daily indicators of the main AI service in each dimension. The operating parameters and daily identifiers are integrated and summarized to obtain the real-time operating status of the main AI service.

[0007] In a preferred embodiment, based on the real-time operating status, the service availability of the main AI service is assessed to obtain a status judgment result. When the status judgment result is unavailable, the emergency switchover process of the airport efficiency platform is triggered, including: Based on the real-time operating status, the ability of the main AI service to carry airport operation support services is comprehensively evaluated to obtain the availability evaluation result of the main AI service. The availability assessment results are matched with the preset airport business carrying capacity standards to obtain the status judgment result of the main AI service; Based on the status judgment result, the civil aviation business compliance verification is performed on the service unavailability judgment conclusion. After the verification is passed, the emergency switching process of the airport efficiency platform is triggered, and the pre-emergency switching preparation is completed.

[0008] In a preferred embodiment, the step of comprehensively evaluating the capability of the main AI service to carry airport operation support services based on the real-time operating status, and obtaining the availability evaluation result of the main AI service, includes: Based on the preset availability calculation formula, the multi-dimensional detection parameters in the real-time running state are fused and calculated to obtain the comprehensive service availability score of the main AI service. The availability assessment result of the main AI service is generated based on the comprehensive service availability score. The availability calculation formula is expressed as follows: ; In the formula, This represents the overall service availability score of the main AI service. This indicates the health status of the main AI service's interface. This represents the task processing efficiency metric of the main AI service. This represents the resource load health index of the main AI service. Represented by natural constant An exponential function with base 0. Represented by natural constant A logarithmic function with base 0.

[0009] In a preferred embodiment, the step of matching the availability assessment result with a preset airport business capacity standard to obtain the status judgment result of the main AI service includes: Based on the preset airport business carrying standards, the availability assessment results are dimensionally aligned and matched to obtain the preliminary standard matching results of the main AI service; Based on the core business scenarios of the airport operation support service, the business support capabilities of the main AI service represented by the preliminary standard matching results are verified and judged to obtain the adaptability judgment result of the main AI service for the airport operation support service. The adaptability determination results are used to comprehensively confirm the service availability, and the status determination result of the main AI service is obtained.

[0010] In a preferred embodiment, the step of storing the request image in a preset temporary message queue based on the unique identifier of the request image corresponding to the airport operation support service request, and selecting a target backup AI service based on a preset airport operation support priority strategy to obtain the target backup AI service corresponding to the operation support service, includes: The airport operation support service request is mirrored to obtain a request image corresponding to the service request. A unique identifier is assigned to the request image to obtain a unique identifier for the request image. The request image carrying the unique identifier of the request image is stored in a preset temporary message queue to complete the temporary storage of the request image. Based on the preset airport operation support priority strategy, the service types corresponding to the requested images in the temporary message queue are prioritized to obtain the service priority ranking result corresponding to the operation support service. Based on the business priority ranking results, the backup AI services are selected through priority adaptation and fusion, and the service carrying capacity matching and business adaptability verification are completed simultaneously to obtain the target backup AI services adapted to the airport operation support services.

[0011] In a preferred embodiment, the step of initiating the revisit processing of all request images in the temporary message queue, restoring each request image to obtain the service request to be processed corresponding to the request image, and distributing it to the target backup AI service for processing includes: The request images in the temporary message queue are traversed and retrieved to obtain a set of request images to be revisited. A revisit operation is performed on each request image in the request image set to activate the business execution status of the request image. The activated request image is parsed and the fields are restored to obtain the service requests to be processed corresponding to the request image. The service requests to be processed are classified and integrated based on the airport operation support priority to generate a service request set with priority adaptation. The pending service requests in the service request set are distributed to the target backup AI service, triggering the target backup AI service to perform business processing on the pending service requests.

[0012] In a preferred embodiment, the continuous health monitoring of the target backup AI service and the periodic connection attempts to the primary AI service to obtain the platform recovery and switching strategy for the airport efficiency platform include: The operational status of the target backup AI service is continuously collected and anomaly monitored to obtain the real-time stable status of the target backup AI service; Based on a preset period, a communication handshake request is sent to the main AI service, the communication response status of the main AI service is detected and collected, an airport operation support service test instruction is sent to the main AI service, the execution effect of the business instruction of the main AI service is verified, and the recovery status information of the main AI service is obtained. The real-time stable state and the recovery state information are matched and analyzed collaboratively to obtain the state collaborative analysis results of the airport efficiency platform. Combined with the continuity requirements of airport operation support services, the platform recovery and switching strategy of the airport efficiency platform is generated.

[0013] In a preferred embodiment, the step of performing collaborative matching analysis on the real-time stable state and the recovery state information to obtain the state collaborative analysis result of the airport efficiency platform, and generating a platform recovery and switching strategy for the airport efficiency platform in conjunction with the continuity requirements of airport operation support services, includes: The real-time stable state and the recovery state information are subjected to state dimension alignment verification and business adaptation correlation analysis to obtain state collaboration analysis results. The state collaboration analysis results are then subjected to civil aviation operation business compliance verification to obtain the compliance verification results of the airport efficiency platform. Based on the compliance verification results, the timing of the switchover is determined in conjunction with the continuity requirements of airport operation support services, and a preliminary investigation of airport operation business conflicts is conducted to obtain the timing determination result of the switchover of the airport efficiency platform. Based on the handover timing determination result and the airport operation support priority rules, a platform recovery handover strategy for the airport efficiency platform is generated.

[0014] To address the aforementioned issues, this invention also provides an emergency switchover system for AI service unavailability. The system includes a main service health detection module, an availability assessment triggering module, an image storage filtering module, an image restoration and distribution module, and a monitoring and recovery generation module, wherein: The main service health detection module is used to perform multi-dimensional health detection on the main AI service of the airport efficiency platform based on airport operation support service requests, and obtain the real-time operating status of the main AI service. The availability assessment triggering module is used to assess the service availability of the main AI service based on the real-time operating status, obtain a status judgment result, and trigger the emergency switching process of the airport efficiency platform when the status judgment result is unavailable. The image storage filtering module is used to store the request image into a preset temporary message queue based on the unique identifier of the request image corresponding to the airport operation support service request, and to select the target backup AI service based on the preset airport operation support priority strategy to obtain the target backup AI service corresponding to the operation support service. The image restoration and distribution module is used to initiate the revisit processing of all the request images in the temporary message queue, restore each request image, obtain the service request to be processed corresponding to the request image, and distribute it to the target backup AI service for processing. The monitoring and recovery generation module is used to continuously monitor the operational status of the target backup AI service and periodically attempt to connect to the main AI service to obtain the platform recovery and switching strategy of the airport efficiency platform.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention performs multi-dimensional health checks on the main AI service based on airport operation support service requests. It can comprehensively collect various operational information of the main AI service supporting airport operation support services, forming a complete and accurate real-time operational status. The service availability assessment based on this operational status can accurately determine the actual support capability of the main AI service, quickly generate reliable status judgment results, and immediately trigger an emergency switchover process when the main AI service is determined to be unavailable. In conjunction with the civil aviation business compliance verification process, it ensures that the emergency switchover operation complies with airport operation management specifications, thereby avoiding business interruptions and execution anomalies in airport operation support services from the source. It continuously and stably provides intelligent support for airport operation support services, effectively improving the stability and security of the airport efficiency platform operation, and meeting the core requirement of uninterrupted operation of intelligent services in airport scenarios.

[0016] 2. This invention securely stores service requests by using a unique identifier in the request mirror. It selects target backup AI services based on airport operation support priority strategies, enabling precise matching between support and backup services. This ensures the orderly handling of various airport operation support services. Through the return processing and complete restoration of the request mirror, pending service requests can be fully recovered and distributed for execution, preventing service request loss and data incompleteness. Simultaneously, it continuously monitors the target backup AI service, periodically attempts to connect to the primary AI service, and generates platform recovery and switching strategies tailored to airport operation needs. This achieves seamless integration of emergency switching and service recovery, ensuring continuous execution of airport operation support services throughout the process. This enhances the emergency response capabilities and service recovery efficiency of the airport efficiency platform, optimizing the overall operational effectiveness of airport intelligent services. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an emergency switching method for AI service unavailability according to an embodiment of the present invention. Figure 2 This is a functional block diagram of an AI service unavailability emergency switching system provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for emergency switching when AI services are unavailable. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for emergency switching when AI services are unavailable can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be an independent server 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, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating an emergency switching method for AI service unavailability according to an embodiment of the present invention. In this embodiment, the emergency switching method for AI service unavailability includes: C1: Based on the airport operation support service request, perform multi-dimensional health detection on the main AI service of the airport efficiency platform to obtain the real-time operating status of the main AI service; In this embodiment of the invention, the step of performing multi-dimensional health checks on the main AI service of the airport efficiency platform based on airport operation support service requests to obtain the real-time operating status of the main AI service includes: Based on airport operation support service requests, multi-dimensional health checks are performed on the interface response status, task processing efficiency, and resource consumption of the main AI service of the airport efficiency platform to obtain the operating parameters and daily indicators of the main AI service in each dimension. The operating parameters and daily identifiers are integrated and summarized to obtain the real-time operating status of the main AI service.

[0021] It should be specifically explained that the airport operation support service request is an instruction received by the airport efficiency platform to initiate operation support-related tasks such as flight scheduling, gate allocation, and ground resource coordination. This request serves as the detection trigger to retrieve the interaction link corresponding to the main AI service within the airport efficiency platform, and sends a targeted detection instruction to the external interaction interface of the main AI service to collect the interface response feedback. It also collects the entire process time information of the main AI service in processing similar operation support service requests to determine task processing efficiency, and collects the actual usage of computing and storage resources occupied by the main AI service during operation to determine resource usage. Through the synchronous detection operation of interface response status, task processing efficiency, and resource usage, the operating parameters and daily indicators of the main AI service in various dimensions are obtained.

[0022] Furthermore, the operational parameters corresponding to the interface response status include the feedback latency information of the interaction interface between the main AI service and the airport operation support service request, the integrity information of the request instruction reception, and the communication link connectivity information. The daily identifiers corresponding to the interface response status include the normal response identifier, the instruction complete reception identifier, and the link smooth connectivity identifier of the interaction interface when the main AI service stably supports the airport operation support service.

[0023] The operational parameters corresponding to task processing efficiency include the full-process execution time information of the main AI service in processing airport operation support service requests, the completion status information of support business nodes, and the closed-loop status information of request processing. The daily indicators corresponding to task processing efficiency include the efficient execution indicator, the smooth completion indicator, and the business closed-loop indicator that the main AI service possesses when normally processing airport operation support services.

[0024] The operational parameters corresponding to resource usage include computing resource load information, storage resource consumption information, and data scheduling resource allocation information during the operation of the main AI service. The daily indicators corresponding to resource usage include reasonable load indicators, normal storage consumption indicators, and balanced resource allocation indicators when the main AI service stably supports airport operation and security services.

[0025] The main AI service's operational parameters and corresponding daily identifiers are categorized and collected according to interface response, task processing, and resource usage. Redundant information with no practical reference value in the parameters and identifiers is removed. The categorized information is then merged and sorted to obtain a real-time operational status that comprehensively reflects the current operational status of the main AI service.

[0026] Furthermore, the real-time operational status is a complete set of statuses formed by uniformly collecting, standardizing, and eliminating redundant and invalid content from the operational parameters and corresponding daily indicators of the three dimensions of main AI service interface response, task processing efficiency, and resource consumption. The interface response dimension integrates operational parameters such as interaction interface feedback latency, request instruction reception completeness, and communication link connectivity status, as well as daily indicators such as normal response, complete instruction reception, and smooth link connectivity. The task processing efficiency dimension integrates operational parameters such as service request full-process execution time, business node completion status, and request processing closed-loop status, as well as daily indicators such as efficient execution, smooth node completion, and business closed loop. The resource consumption dimension integrates operational parameters such as computing resource load, storage resource consumption, and data scheduling resource allocation, as well as daily indicators such as reasonable load, normal storage consumption, and balanced resource allocation. This status set comprehensively covers the full-dimensional operational status of the main AI service in connecting, processing, and supporting airport operation and support services. It can intuitively reflect the main AI service's connection and interaction capabilities, operational progress capabilities, and resource allocation status, providing a complete and accurate status basis for subsequent service availability assessment.

[0027] The beneficial effects are that by conducting multi-dimensional health checks on the interface response status, task processing efficiency, and resource usage of the main AI service based on airport operation support service requests, it is possible to comprehensively collect the main AI service's operating parameters and daily indicators in various dimensions. The real-time operating status formed by integrating and summarizing the above information can completely, accurately, and intuitively reflect the full-dimensional operation of the main AI service's interface interaction, operation progress, and resource allocation. This provides accurate and comprehensive status basis for subsequent service availability assessment, effectively improves the accuracy and reliability of status judgment, ensures timely identification of abnormal operating conditions of the main AI service, provides solid support for the accurate triggering of emergency switchover processes, and ensures the continuous and stable operation of airport operation support services.

[0028] C2: Based on the real-time operating status, perform a service availability assessment on the main AI service to obtain a status judgment result. When the status judgment result is unavailable, trigger the emergency switching process of the airport efficiency platform. In this embodiment of the invention, the step of assessing the service availability of the main AI service based on the real-time operating status to obtain a status judgment result, and triggering the emergency switching process of the airport efficiency platform when the status judgment result is unavailable, includes: Based on the real-time operating status, the ability of the main AI service to carry airport operation support services is comprehensively evaluated to obtain the availability evaluation result of the main AI service. The availability assessment results are matched with the preset airport business carrying capacity standards to obtain the status judgment result of the main AI service; Based on the status judgment result, the civil aviation business compliance verification is performed on the service unavailability judgment conclusion. After the verification is passed, the emergency switching process of the airport efficiency platform is triggered, and the pre-emergency switching preparation is completed.

[0029] Based on the real-time operating status, a comprehensive evaluation is performed on the capability of the main AI service to carry out airport operation support services, resulting in an availability evaluation result for the main AI service, including: Based on the preset availability calculation formula, the multi-dimensional detection parameters in the real-time running state are fused and calculated to obtain the comprehensive service availability score of the main AI service. The availability assessment result of the main AI service is generated based on the comprehensive service availability score. The availability calculation formula is expressed as follows: ; In the formula, This represents the overall service availability score of the main AI service. This indicates the health status of the main AI service's interface. This represents the task processing efficiency metric of the main AI service. This represents the resource load health index of the main AI service. Represented by natural constant An exponential function with base 0. Represented by natural constant A logarithmic function with base 0.

[0030] The process of matching the availability assessment results with preset airport business capacity standards to obtain the status judgment result of the main AI service includes: Based on the preset airport business carrying standards, the availability assessment results are dimensionally aligned and matched to obtain the preliminary standard matching results of the main AI service; Based on the core business scenarios of the airport operation support service, the business support capabilities of the main AI service represented by the preliminary standard matching results are verified and judged to obtain the adaptability judgment result of the main AI service for the airport operation support service. The adaptability determination results are used to comprehensively confirm the service availability, and the status determination result of the main AI service is obtained.

[0031] It should be specifically explained that, based on the multi-dimensional information of interface response, task processing efficiency, and resource consumption included in the real-time operating status, the interface operation health index, task processing efficiency index, and resource load health index of the main AI service are fused through exponential and logarithmic operations with the natural constant as the base, to complete the integrated calculation of multi-dimensional detection information and obtain the comprehensive service availability score of the main AI service.

[0032] Furthermore, the interface health index is dimensionless, with a normalized value range of [0,1]. The larger the value, the higher the interface health. It is obtained by extracting all information from the interface response dimension in the real-time running status. Specifically, the running parameters of interactive interface feedback latency, request instruction reception completeness, and communication link connectivity status are integrated and judged with the daily indicators of normal response (latency ≤200ms), complete instruction reception (completeness 100%), and smooth link connectivity (connectivity 100%). Single spikes and glitches in abnormal fluctuation information are removed, and finally, an interface health index that can intuitively reflect the main AI service interaction and docking capabilities is formed.

[0033] The interface health index is calculated by weighting three parts: feedback latency score (40%), instruction reception integrity score (30%), and communication link connectivity score (30%).

[0034] Feedback latency score calculation: Subtract the ratio of actual latency to 200 milliseconds from 1; instruction reception integrity score is calculated based on the actual reception integrity ratio; communication link connectivity score is calculated based on the actual connectivity ratio; add the three scores according to their corresponding weights to obtain the interface operation health index.

[0035] This indicator is used to characterize the stability of communication and command interaction between the main AI service and the airport efficiency platform. The higher the value, the smoother and more stable the interface interaction. When participating in the comprehensive availability score calculation, it will amplify the positive contribution through monotonically increasing quadratic convex function operation, directly improving the overall level of the comprehensive service availability score. If the interface health index is low, it will significantly lower the comprehensive score, intuitively reflecting the negative impact of interface anomalies on service availability.

[0036] The task processing efficiency index is dimensionless and has a normalized value range of [0,1]. The larger the value, the higher the processing efficiency. It is obtained by extracting all information of the task processing efficiency dimension in the real-time running status. Specifically, it integrates and judges the running parameters of the entire process execution time of the service request, the completion status of the business node, and the closed loop status of the request processing with the daily indicators of efficient execution (full process time ≤3s), smooth node completion (completion rate 100%), and business closure (closure rate 100%), and sorts them out to form a task processing efficiency index that can truly reflect the main AI service business processing capability.

[0037] The task processing efficiency index is calculated by weighting three parts: execution time (40%), business node completion rate (30%), and request processing closure rate (30%).

[0038] Execution time score calculation: Subtract the ratio of actual execution time to 3 seconds from 1; business node completion rate score is calculated based on the actual completion ratio; request processing closure rate score is calculated based on the actual closure ratio; add the three scores according to their corresponding weights to obtain the task processing efficiency index.

[0039] This indicator reflects the speed and completeness of the main AI service in processing airport operation support service requests. The higher the value, the more efficient and smooth the business processing. When participating in the comprehensive availability score calculation, it will release positive gain through logarithmic function calculation to steadily improve the comprehensive service availability score. If the task processing efficiency indicator is low, it will directly reduce the comprehensive score. ln(1+T) is monotonically increasing and growing slowly in the interval T∈[0,1], which can avoid small fluctuations in efficiency from causing drastic changes in the score, and adapt to the "stability first" requirement of airport business. It clearly reflects the restrictive effect of insufficient processing capacity on service availability.

[0040] The resource load health index is dimensionless, with a normalized value range of [0,1]. It represents the rationality of resource load and is obtained by extracting all information from the resource occupancy dimension in the real-time operating status. Specifically, it integrates and judges the operating parameters of computing resource load, storage resource consumption, and data scheduling resource allocation with daily indicators of reasonable load (CPU / storage utilization rate 30%-70%), normal storage consumption, and balanced resource allocation to form a resource load health index that can accurately represent the rationality of the main AI service resource usage.

[0041] The resource load health index is calculated by weighting three parts: computational resource load score accounts for 40%, storage resource consumption score accounts for 30%, and data scheduling resource allocation score accounts for 30%.

[0042] Resource load scoring rules: A utilization rate between 30% and 70% is recorded as full marks. A utilization rate below 30% (no load) or above 70% (overload) will result in a deduction of marks according to the actual deviation ratio. The scores of the three parts are added together according to their corresponding weights to obtain the resource load health index.

[0043] The healthy range for resource load is 0.3 to 0.7 after normalization; values ​​outside this range indicate an unhealthy state. The calculation formula for resource load is: the resource load health index minus 0.5 is used as the exponent of the natural constant; the result of this exponent is then squared after adding 1, and the reciprocal is taken. This formula outputs a higher score within the healthy range, and the score drops rapidly under no-load or overload conditions, accurately reflecting the rationality of resource load and possessing good numerical discrimination.

[0044] This indicator is used to measure the adaptability of resource allocation and load during the operation of the main AI service. When its value is in the healthy range, it can stably support the calculation results of the comprehensive availability score and provide a basic guarantee for the comprehensive score. If the resource load health index is too high or too low, it will damage the stability of resource operation and thus reduce the comprehensive service availability score, truly reflecting the interference and impact of resource anomalies on service availability.

[0045] The service availability comprehensive score is dimensionless, with a normalized value range of [0,1]. A score ≥0.7 indicates that the service is available, and a score <0.7 indicates that the service is unavailable. Based on the main AI service's operational capabilities reflected in the service availability comprehensive score, an availability assessment result that can fully characterize the main AI service's support level is formed.

[0046] Furthermore, the availability assessment results are based on a comprehensive service availability score, integrating three key indicators: interface operational health, task processing efficiency, and resource load health. These indicators comprehensively characterize the main AI service's communication stability, business processing efficiency, and resource load capacity in handling airport operation support service requests from three dimensions: basic interaction support, core business processing, and continuous resource capacity. This comprehensively presents the main AI service's actual support capabilities across the entire chain of request access, business execution, and stable operation, accurately reflecting whether the main AI service can meet the core requirements of basic access, efficient processing, and long-term stable operation for airport operation support services. This provides a true and comprehensive capability basis for subsequent matching with airport business capacity standards. A higher comprehensive service availability score indicates a higher overall support level for the main AI service in terms of interface interaction, business processing, and resource capacity, enabling it to stably support airport operation support services. Conversely, a lower score indicates a lower overall support level, failing to meet the support requirements of airport operation support services.

[0047] According to the preset airport business capacity standards, the various capability indicators in the availability assessment results are compared and matched with the corresponding dimensions of the standards one by one to obtain the preliminary standard matching results of the main AI service.

[0048] Furthermore, the pre-defined airport business capacity standards are standardized capability assessment benchmarks pre-configured within the airport efficiency platform and specifically adapted to airport operation and support service scenarios. They serve as the core basis for determining whether the main AI service can normally support airport operation and support services. These standards consist of three parts: interface interaction capacity standards, business processing execution standards, and resource operation adaptation standards. They also integrate the operational specifications of the civil aviation private network and the rules for multi-level collaborative decision-making at the airport. The interface interaction capacity standards specify the minimum interface response time, complete instruction reception, and stable communication link required for the main AI service to connect with support service requests. The business processing execution standards set the minimum execution efficiency, business node completion rate, and process closure requirements for the main AI service to handle core support services such as flight scheduling, gate allocation, and ground resource coordination. The resource operation adaptation standards define the reasonable load range and consumption limits for the computing, storage, and data scheduling resources required for the main AI service to operate. All of these standards are customized specifically for the airport and are used to match the availability assessment results one by one, accurately determining whether the actual support capability of the main AI service meets the stable execution requirements of airport operation and support services.

[0049] Civil aviation private network operation specifications: "Operation and Maintenance Procedures for Civil Aviation Communication, Navigation and Surveillance Systems" (MH / T4006-2018) and "Civil Aviation Dedicated Network Security Management Specifications" (MH / T1016-2021); Airport multi-level collaborative decision-making rules: Technical requirements for airport operation collaborative decision-making (A-CDM) system (MH / T5044-2017) and management rules for collaborative decision-making of civil airport flight operations (AC-140-FS-2021-028).

[0050] By combining the core business scenarios of flight scheduling, gate allocation, and ground resource coordination in airport operation support services, the effectiveness of the main AI service business support reflected in the preliminary results of the verification standard matching is obtained, and the adaptability judgment result of the main AI service for airport operation support services is obtained.

[0051] Based on the verification information in the adaptability assessment results, the final verification of the main AI service support capabilities is completed, and the status assessment result of the main AI service is obtained.

[0052] In accordance with the Civil Aviation Private Network Operation Specifications and Airport Operation Management Rules, a compliance check is conducted on the service unavailability judgment conclusion in the status assessment results. After confirming that the judgment conclusion meets the requirements of civil aviation business operation, the emergency switching process of the airport efficiency platform is triggered, and the preparatory work such as link connection and service preparation required for emergency switching is completed.

[0053] Furthermore, the Civil Aviation Private Network Operation Specifications refer to the unified technical and management requirements for network security, access control, data transmission, link stability, and fault handling formulated by the civil aviation industry for dedicated communication networks. These requirements ensure the secure transmission, compliant access, and stable communication of airport operational data within the private network, and ensure that system interactions comply with civil aviation network security standards.

[0054] Airport operation management rules refer to the on-site operation guidelines for airports formulated in accordance with civil aviation regulations. These rules standardize processes such as flight support, resource scheduling, emergency response, and multi-level collaborative decision-making, clarify operational responsibilities and requirements, and ensure the safe, orderly, and efficient operation of airports.

[0055] The beneficial effects include conducting a multi-dimensional comprehensive assessment of the main AI service's ability to support airport operations based on real-time operational status. A scientific and accurate availability assessment result is generated through pre-set formulas and calculations. This assessment result is then aligned and matched with airport business support standards and verified for compatibility with core business scenarios. This allows for precise determination of the actual support status of the main AI service and the acquisition of reliable status judgments. Simultaneously, compliance verification of service unavailability conclusions with civil aviation private network operation specifications and airport operation management rules ensures the accuracy and compliance of emergency switchover processes, effectively preventing misjudgments or unauthorized switches. This provides solid support for the airport efficiency platform to quickly respond to main AI service anomalies and smoothly initiate emergency response, effectively ensuring uninterrupted and undelayed airport operations support services.

[0056] C3: Based on the unique identifier of the request image corresponding to the airport operation support service request, store the request image in a preset temporary message queue, select the target backup AI service based on the preset airport operation support priority strategy, and obtain the target backup AI service corresponding to the operation support service. In this embodiment of the invention, the step of storing the request image in a preset temporary message queue based on the unique identifier of the request image corresponding to the airport operation support service request, and selecting a target backup AI service based on a preset airport operation support priority strategy to obtain the target backup AI service corresponding to the operation support service, includes: The airport operation support service request is mirrored to obtain a request image corresponding to the service request. A unique identifier is assigned to the request image to obtain a unique identifier for the request image. The request image carrying the unique identifier of the request image is stored in a preset temporary message queue to complete the temporary storage of the request image. Based on the preset airport operation support priority strategy, the service types corresponding to the requested images in the temporary message queue are prioritized to obtain the service priority ranking result corresponding to the operation support service. Based on the business priority ranking results, the backup AI services are selected through priority adaptation and fusion, and the service carrying capacity matching and business adaptability verification are completed simultaneously to obtain the target backup AI services adapted to the airport operation support services.

[0057] It needs to be specifically explained that the complete business content, instruction information and associated parameters of the airport operation support service request are completely extracted and encapsulated to form a request image that completely corresponds to the service request. According to the preset encoding rules, an independent and non-repeatable identification information is assigned to the request image to obtain a unique identifier for the request image.

[0058] The request image, which is bound to a unique identifier, is written into a pre-built temporary message queue according to the preset storage rules. The temporary message queue is used to securely store and arrange the request image in an orderly manner, thus completing the reliable temporary storage operation of the request image.

[0059] Based on the pre-set airport operation support priority strategy, the business types corresponding to each request image in the temporary message queue, such as flight scheduling, gate allocation, and ground resource coordination, are identified. Priorities are distinguished according to the importance and urgency of the business, and the business priority ranking results corresponding to the operation support services are obtained.

[0060] Furthermore, the airport operation support priority strategy is a priority determination and execution criterion pre-configured by the airport efficiency platform based on civil aviation operation management standards, airport on-site operation requirements, and multi-level collaborative decision-making rules. This strategy aims to ensure airport operational safety, uninterrupted core business operations, and priority handling of emergency tasks. It clarifies the priority levels, classification criteria, and adaptation requirements for different types of airport operation support services. The priority classification criteria include four core dimensions: the degree of business coreness, the degree of urgency, the level of safety risk, and the flight support stage. Specifically, airport operation support services are divided into three priority levels, with the first priority being flight takeoff and landing support and emergency fault handling. The policy prioritizes core emergency operations related to flight safety and operational bottom lines, such as gate allocation and large-scale flight delay coordination. Secondary priority includes routine operations that affect operational efficiency, such as gate allocation, ground resource scheduling, and ground service personnel dispatch. Tertiary priority includes non-real-time auxiliary operations such as operational data statistics, support report generation, and process compliance verification. The policy also stipulates that priority determination should be dynamically adjusted based on flight attributes, support time periods, and abnormal scenarios. In emergency scenarios, the priority of emergency support operations will be automatically increased. All priority classifications must comply with the Civil Aviation Private Network Operation Specifications and Airport Safety Management Requirements, providing a standardized and compliant priority basis for the adaptation and selection of backup AI services.

[0061] Based on the processing order determined by the business priority ranking results, the candidate backup AI services are screened and processed in accordance with the business priority. At the same time, it is verified whether the service carrying capacity of each backup AI service meets the business processing requirements, and the compatibility of each backup AI service with the airport operation support business is verified. Finally, the target backup AI service that is suitable for the airport operation support service is determined.

[0062] The beneficial effects are as follows: by mirroring and extracting airport operation support service requests and assigning unique identifiers, all business information and instruction content of the service requests can be completely preserved. Relying on temporary message queues, the request mirrors can be securely and orderly stored, effectively avoiding the loss of service requests or data incompleteness in emergency situations. Combined with the preset airport operation support priority strategy, the priority classification of business types can be completed, ensuring that high-priority core support businesses are allocated processing resources first. Then, through priority adaptation and fusion screening, service carrying capacity matching, and business adaptability verification, target backup AI services that are highly compatible with business needs can be accurately screened, ensuring the adaptability and reliability of backup services, and providing solid support for the uninterrupted and orderly handling of airport operation support services.

[0063] C4: Initiate the revisit processing of all the request images in the temporary message queue, restore each request image to obtain the service request to be processed corresponding to the request image, and distribute it to the target backup AI service for processing; In this embodiment of the invention, the step of initiating the revisit processing of all request images in the temporary message queue, restoring each request image to obtain the service request to be processed corresponding to the request image, and distributing it to the target backup AI service for processing includes: The request images in the temporary message queue are traversed and retrieved to obtain a set of request images to be revisited. A revisit operation is performed on each request image in the request image set to activate the business execution status of the request image. The activated request image is parsed and the fields are restored to obtain the service requests to be processed corresponding to the request image. The service requests to be processed are classified and integrated based on the airport operation support priority to generate a service request set with priority adaptation. The pending service requests in the service request set are distributed to the target backup AI service, triggering the target backup AI service to perform business processing on the pending service requests.

[0064] It should be specifically explained that, in accordance with the storage order of the temporary message queue, all request images in the internal storage are retrieved one by one. All retrieved request images are aggregated to form a set of request images to be revisited. Revisit instructions are sent to each request image in the request image set in turn according to the unique identifier of the request image. The revisit instructions are used to wake up the business execution attributes of the request image and complete the activation of the business execution status of the request image.

[0065] For request images that have completed the activation of business execution status, perform complete business data parsing and field restoration operations. Restore all business content, instruction information and associated parameters encapsulated in the request image to the original business request format, and obtain the pending service requests that correspond one-to-one with the request image. Based on the airport operation support priority strategy, classify, arrange and merge all pending service requests to generate a service request set that matches the business priority.

[0066] Furthermore, the airport operation support priority strategy is a standardized priority judgment and execution criterion pre-configured by the airport efficiency platform based on the civil aviation private network operation specifications, airport operation management rules, and multi-level collaborative decision-making mechanisms. With flight safety as the priority, core business as the priority, and emergency response as the priority, it is specifically designed for service request sorting, backup AI service matching, and request distribution and execution in emergency switching scenarios. It is the core basis for ensuring the orderly handling of airport operation support services and the uninterrupted operation of core businesses.

[0067] This strategy uses five dimensions—core business importance, urgency of execution, safety risk level, flight support stage, and operational scenario attributes—to classify all airport operation support services into three fixed priority levels, with support for dynamic adjustments in emergency scenarios: Level 1 is the highest priority, covering core emergency businesses involving flight safety and operational bottom lines, such as flight takeoff and landing command, emergency fault handling, large-scale flight delay coordination, and special flight and key flight support. It requires priority allocation of computing power and scheduling of the highest available target backup AI services. Level 2 is the medium priority, covering important routine businesses affecting airport operational efficiency, such as gate allocation, ground resource scheduling, dispatching ground service and security personnel, and adjustments to regular flight schedules. These are handled in order of priority. Level 3 is the basic priority, covering non-real-time auxiliary businesses such as operational data statistics, support report generation, process compliance verification, and administrative supervision. These are handled systematically when backup AI service resources are available. Meanwhile, the strategy clearly states that in emergency scenarios such as blizzards and equipment failures, the priority level of corresponding emergency support services will be automatically upgraded. During peak flight periods, priority ranking will be strictly followed to avoid crowding out core business resources. All priority determinations and executions will comply with civil aviation safety control requirements, ensuring that high-value, high-risk, and high-urgency airport operation support services are given priority and handled reliably during emergency switchovers.

[0068] According to the priority order within the service request set, each pending service request is sent to the corresponding target backup AI service through a preset communication link. A business processing start command is sent to the target backup AI service, triggering the target backup AI service to carry out complete business processing operations on the pending service request in accordance with the execution specifications of airport operation support services.

[0069] The beneficial effects include the ability to accurately recover the original pending service requests by traversing, retrieving, activating, and fully restoring the request images in the temporary message queue, thus avoiding the loss of business information or data distortion in emergency situations. Combined with the airport operation support priority strategy, the service requests are classified and integrated to ensure that high-priority core support services are processed first. Then, the service requests are distributed to the target backup AI service and the execution of business processing is triggered, achieving seamless acceptance and continuous handling of airport operation support services. This effectively ensures that the airport's core business is uninterrupted and the operation process is not interrupted, and greatly improves the smoothness of emergency switching and the orderliness and reliability of business processing.

[0070] C5: Perform continuous health checks on the operational status of the target backup AI service and periodically attempt to connect to the main AI service to obtain the platform recovery and switching strategy of the airport efficiency platform.

[0071] In this embodiment of the invention, the step of continuously monitoring the operational status of the target backup AI service and periodically attempting to connect to the primary AI service to obtain the platform recovery and switching strategy for the airport efficiency platform includes: The operational status of the target backup AI service is continuously collected and anomaly monitored to obtain the real-time stable status of the target backup AI service; Based on a preset period, a communication handshake request is sent to the main AI service, the communication response status of the main AI service is detected and collected, an airport operation support service test instruction is sent to the main AI service, the execution effect of the business instruction of the main AI service is verified, and the recovery status information of the main AI service is obtained. The real-time stable state and the recovery state information are matched and analyzed collaboratively to obtain the state collaborative analysis results of the airport efficiency platform. Combined with the continuity requirements of airport operation support services, the platform recovery and switching strategy of the airport efficiency platform is generated.

[0072] The process involves performing a collaborative matching analysis between the real-time stable state and the recovery state information to obtain the state collaborative analysis results of the airport efficiency platform. Combined with the continuity requirements of airport operation support services, a platform recovery and switching strategy for the airport efficiency platform is generated, including: The real-time stable state and the recovery state information are subjected to state dimension alignment verification and business adaptation correlation analysis to obtain state collaboration analysis results. The state collaboration analysis results are then subjected to civil aviation operation business compliance verification to obtain the compliance verification results of the airport efficiency platform. Based on the compliance verification results, the timing of the switchover is determined in conjunction with the continuity requirements of airport operation support services, and a preliminary investigation of airport operation business conflicts is conducted to obtain the timing determination result of the switchover of the airport efficiency platform. Based on the handover timing determination result and the airport operation support priority rules, a platform recovery handover strategy for the airport efficiency platform is generated.

[0073] It needs to be specifically explained that the system continuously collects operational information such as the interface response status, task processing progress, and resource usage of the target backup AI service. It performs real-time anomaly identification and fluctuation monitoring on the collected information to promptly detect abnormal issues such as service lag, processing interruption, and resource overload. The system integrates the normal operation data and anomaly investigation results obtained from the collection and monitoring to form a real-time stable status that can comprehensively reflect the support capabilities of the target backup AI service.

[0074] The system sends communication handshake requests to the main AI service at pre-set time intervals to detect and collect data on the main AI service's communication connectivity and interface response speed. At the same time, it sends standard airport operation support service test commands to the main AI service to observe and verify the main AI service's reception, processing, and closed-loop execution of the test commands. The communication detection results and business execution verification results are then summarized to obtain the recovery status information of the main AI service.

[0075] The real-time stable status of the target backup AI service is aligned and verified with the recovery status information of the main AI service in terms of interface operation, task processing, and resource carrying capacity. The degree of compatibility and correlation between the two in terms of airport operation support service capabilities is analyzed to form a status collaboration analysis result. Then, the status collaboration analysis result is verified for compliance in accordance with the civil aviation private network operation specifications and airport operation management rules to obtain the compliance verification result of the airport efficiency platform.

[0076] Based on the compliance verification results and the continuous requirements for uninterrupted execution of airport operation support services, the optimal time to switch back to the main service after the main AI service is restored is determined. At the same time, potential business interruptions and request conflicts that may occur during the switchover process are investigated and avoided in advance, resulting in the switchover timing determination result of the airport efficiency platform.

[0077] Based on the timing of the switchover and the requirements of prioritizing core business and ensuring uninterrupted emergency business in the airport operation support priority rules, the execution sequence, business migration method, and abnormal rollback mechanism for the recovery switchover are planned, ultimately generating a complete platform recovery switchover strategy for the airport efficiency platform that meets the requirements of airport operation management.

[0078] Furthermore, the platform recovery and switching strategy is a complete execution plan for the airport efficiency platform to smoothly switch back to the main AI service from the target backup AI service without interruption, conflict, compliance, and order, under the premise that the main AI service function is restored and the target backup AI service continues to operate stably. This strategy takes flight safety as the priority, uninterrupted core business, no disturbance during the switching process, and full civil aviation compliance as its core principles. It integrates the collaborative results of the main and backup service status, the continuity requirements of airport operation support, business priority rules, and civil aviation private network operation specifications to form the key basis for achieving a smooth platform reset after the main AI service is restored and ensuring the long-term stable operation of the airport.

[0079] The strategy clearly stipulates that the main AI service must meet the required indicators and pass the compliance verification, and the backup service must be stable and the main and backup services must be in coordinated status. The switching time should avoid peak and core business periods and be executed during off-peak hours. Business migration should be smoothly promoted from low to high priority, and temporary dual takeover of main and backup services should be adopted to avoid interruption. Circuit breakers and rollback mechanisms should be set up during the switching to prevent conflicts, and compliance audits should be carried out throughout the process. After the switching, the status of the main service should be verified, the backup service should be put into standby, and data archiving and platform reset should be completed.

[0080] The beneficial effects are that by continuously monitoring the target backup AI service and periodically probing the recovery status of the main AI service, the operation of the main and backup services can be grasped in real time, ensuring that the backup service can stably take over the business; through status coordination and matching, compliance verification and switching timing determination, a recovery switching strategy can be scientifically generated, realizing a smooth switchback without interruption or conflict after the main service is restored, effectively avoiding the risk of business disturbance, improving the platform's emergency recovery capability, and ensuring the continuous and reliable operation support services of the airport throughout the process.

[0081] like Figure 2 The diagram shown is a functional block diagram of an emergency switching system for unavailable AI services provided in an embodiment of the present invention.

[0082] The AI ​​service unavailability emergency switchover system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the AI ​​service unavailability emergency switchover system 100 may include a main service health detection module 101, an availability assessment triggering module 102, an image storage filtering module 103, an image restoration and distribution module 104, and a monitoring and recovery generation module 105. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0083] In this embodiment, the functions of each module are as follows: The main service health detection module 101 is used to perform multi-dimensional health detection on the main AI service of the airport efficiency platform based on the airport operation support service request, and obtain the real-time operating status of the main AI service. The availability assessment triggering module 102 is used to assess the service availability of the main AI service based on the real-time operating status, obtain a status judgment result, and trigger the emergency switching process of the airport efficiency platform when the status judgment result is unavailable. The image storage filtering module 103 is used to store the request image in a preset temporary message queue based on the unique identifier of the request image corresponding to the airport operation support service request, and to select the target backup AI service based on the preset airport operation support priority strategy to obtain the target backup AI service corresponding to the operation support service. The image restoration and distribution module 104 is used to initiate the revisit processing of all the request images in the temporary message queue, restore each request image, obtain the service request to be processed corresponding to the request image, and distribute it to the target backup AI service for processing. The monitoring and recovery generation module 105 is used to continuously monitor the operating status of the target backup AI service and periodically attempt to connect to the main AI service to obtain the platform recovery and switching strategy of the airport efficiency platform.

[0084] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0085] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0086] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0087] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0088] The embodiments of this application can acquire and process relevant data based on an artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for emergency switching when AI services are unavailable, characterized in that, The method includes: C1: Based on the airport operation support service request, perform multi-dimensional health checks on the main AI service of the airport efficiency platform to obtain the real-time operating status of the main AI service; C2: Based on the real-time operating status, perform a service availability assessment on the main AI service to obtain a status judgment result. When the status judgment result is unavailable, trigger the emergency switching process of the airport efficiency platform. C3: Based on the unique identifier of the request image corresponding to the airport operation support service request, store the request image in a preset temporary message queue, select the target backup AI service based on the preset airport operation support priority strategy, and obtain the target backup AI service corresponding to the operation support service. C4: Initiate the revisit processing of all the request images in the temporary message queue, restore each request image to obtain the service request to be processed corresponding to the request image, and distribute it to the target backup AI service for processing; C5: Perform continuous health checks on the operational status of the target backup AI service and periodically attempt to connect to the main AI service to obtain the platform recovery and switching strategy of the airport efficiency platform. 2.The AI service unavailability emergency handover method of claim 1, wherein, The process of performing multi-dimensional health checks on the main AI service of the airport efficiency platform based on airport operation support service requests to obtain the real-time operating status of the main AI service includes: Based on airport operation support service requests, multi-dimensional health checks are performed on the interface response status, task processing efficiency, and resource consumption of the main AI service of the airport efficiency platform to obtain the operating parameters and daily indicators of the main AI service in each dimension. The operating parameters and daily identifiers are integrated and summarized to obtain the real-time operating status of the main AI service. 3.The AI service unavailability emergency handover method of claim 1, wherein, Based on the real-time operating status, the service availability of the main AI service is assessed to obtain a status judgment result. When the status judgment result is unavailable, the emergency switchover process of the airport efficiency platform is triggered, including: Based on the real-time operating status, the ability of the main AI service to carry airport operation support services is comprehensively evaluated to obtain the availability evaluation result of the main AI service. The availability assessment results are matched with the preset airport business carrying capacity standards to obtain the status judgment result of the main AI service; Based on the status judgment result, the civil aviation business compliance verification is performed on the service unavailability judgment conclusion. After the verification is passed, the emergency switching process of the airport efficiency platform is triggered, and the pre-emergency switching preparation is completed. 4.The AI service unavailability emergency handover method of claim 3, wherein, Based on the real-time operating status, a comprehensive evaluation is performed on the capability of the main AI service to carry out airport operation support services, resulting in an availability evaluation result for the main AI service, including: Based on the preset availability calculation formula, the multi-dimensional detection parameters in the real-time running state are fused and calculated to obtain the comprehensive service availability score of the main AI service. The availability assessment result of the main AI service is generated based on the comprehensive service availability score. The availability calculation formula is expressed as follows: ； In the formula, This represents the overall service availability score of the main AI service. This indicates the health status of the main AI service's interface. This represents the task processing efficiency metric of the main AI service. This represents the resource load health index of the main AI service. Represented by natural constant An exponential function with base 0. Represented by natural constant A logarithmic function with base 0. 5.The AI service unavailability emergency handover method of claim 3, wherein, The process of matching the availability assessment results with preset airport business capacity standards to obtain the status judgment result of the main AI service includes: Based on the preset airport business carrying standards, the availability assessment results are dimensionally aligned and matched to obtain the preliminary standard matching results of the main AI service; Based on the core business scenarios of the airport operation support service, the business support capabilities of the main AI service represented by the preliminary standard matching results are verified and judged to obtain the adaptability judgment result of the main AI service for the airport operation support service. The adaptability determination results are used to comprehensively confirm the service availability, and the status determination result of the main AI service is obtained. 6.The AI service unavailability emergency handover method of claim 1, wherein, The process involves storing the request image in a preset temporary message queue based on the unique identifier of the request image corresponding to the airport operation support service request, and selecting a target backup AI service based on a preset airport operation support priority strategy to obtain the target backup AI service corresponding to the operation support service, including: The airport operation support service request is mirrored to obtain the request image corresponding to the service request. A unique identifier is assigned to the request image to obtain the unique identifier of the request image. The request image carrying the unique identifier of the request image is stored in a preset temporary message queue to complete the temporary storage of the request image. Based on the preset airport operation support priority strategy, the service types corresponding to the requested images in the temporary message queue are prioritized to obtain the service priority ranking result corresponding to the operation support service. Based on the business priority ranking results, the backup AI services are selected through priority adaptation and fusion, and the service carrying capacity matching and business adaptability verification are completed simultaneously to obtain the target backup AI services adapted to the airport operation support services. 7.The AI service unavailability emergency handover method of claim 1, wherein, The step of initiating the revisit processing of all request images in the temporary message queue, restoring each request image to obtain the service request to be processed corresponding to the request image, and distributing it to the target backup AI service for processing includes: The request images in the temporary message queue are traversed and retrieved to obtain a set of request images to be revisited. A revisit operation is performed on each request image in the request image set to activate the business execution status of the request image. The activated request image is parsed and the fields are restored to obtain the service requests to be processed corresponding to the request image. The service requests to be processed are classified and integrated based on the airport operation support priority to generate a service request set with priority adaptation. The pending service requests in the service request set are distributed to the target backup AI service, triggering the target backup AI service to perform business processing on the pending service requests. 8.The AI service unavailability emergency handover method of claim 1, wherein, The continuous health monitoring of the target backup AI service and the periodic connection attempts to the primary AI service, resulting in the platform recovery and switching strategy for the airport efficiency platform, include: The operational status of the target backup AI service is continuously collected and anomaly monitored to obtain the real-time stable status of the target backup AI service; Based on a preset period, a communication handshake request is sent to the main AI service, the communication response status of the main AI service is detected and collected, an airport operation support service test instruction is sent to the main AI service, the execution effect of the business instruction of the main AI service is verified, and the recovery status information of the main AI service is obtained. The real-time stable state and the recovery state information are matched and analyzed collaboratively to obtain the state collaborative analysis results of the airport efficiency platform. Combined with the continuity requirements of airport operation support services, the platform recovery and switching strategy of the airport efficiency platform is generated.

9. The emergency switching method for AI service unavailability as described in claim 8, characterized in that, The process involves performing a collaborative matching analysis between the real-time stable state and the recovery state information to obtain the state collaborative analysis results of the airport efficiency platform. Combined with the continuity requirements of airport operation support services, a platform recovery and switching strategy for the airport efficiency platform is generated, including: The real-time stable state and the recovery state information are subjected to state dimension alignment verification and business adaptation correlation analysis to obtain state collaboration analysis results. The state collaboration analysis results are then subjected to civil aviation operation business compliance verification to obtain the compliance verification results of the airport efficiency platform. Based on the compliance verification results, the timing of the switchover is determined in conjunction with the continuity requirements of airport operation support services, and a preliminary investigation of airport operation business conflicts is conducted to obtain the timing determination result of the switchover of the airport efficiency platform. Based on the handover timing determination result and the airport operation support priority rules, a platform recovery handover strategy for the airport efficiency platform is generated.

10. An AI service unavailability emergency handover system, characterized by, To implement the emergency switchover method for AI service unavailability as described in claim 1, the system includes a main service health detection module, an availability assessment triggering module, an image storage filtering module, an image restoration and distribution module, and a monitoring and recovery generation module, wherein: The main service health detection module is used to perform multi-dimensional health detection on the main AI service of the airport efficiency platform based on airport operation support service requests, and obtain the real-time operating status of the main AI service. The availability assessment triggering module is used to assess the service availability of the main AI service based on the real-time operating status, obtain a status judgment result, and trigger the emergency switching process of the airport efficiency platform when the status judgment result is unavailable. The image storage filtering module is used to store the request image into a preset temporary message queue based on the unique identifier of the request image corresponding to the airport operation support service request, and to select the target backup AI service based on the preset airport operation support priority strategy to obtain the target backup AI service corresponding to the operation support service. The image restoration and distribution module is used to initiate the revisit processing of all the request images in the temporary message queue, restore each request image, obtain the service request to be processed corresponding to the request image, and distribute it to the target backup AI service for processing. The monitoring and recovery generation module is used to continuously monitor the operational status of the target backup AI service and periodically attempt to connect to the main AI service to obtain the platform recovery and switching strategy of the airport efficiency platform.

Images