A closed-loop IT operation strategy optimization method fusing user feedback

Abstract

This invention relates to the field of IT operations and maintenance (O&M) technology, and more particularly to a closed-loop IT O&M strategy optimization method that integrates user feedback. The method includes: upon receiving user feedback regarding service unavailability, performance degradation, or functional limitations, determining that a user-perceptible anomaly exists, and attempting to determine the responsible party with repair or modification authority based on O&M responsibility rules; when the responsible party cannot be determined within a preset time or there is a responsibility conflict, marking the system operation state as a responsibility-frozen state, and switching to a responsibility-neutral minimal action strategy set in this state to execute O&M actions that are independent of responsibility attribution and can be rolled back; after continuously receiving user feedback and meeting the conditions for lifting the responsibility-frozen state, resuming the normal O&M strategy, and recording the O&M actions and corresponding user feedback results as the basis for O&M strategy optimization. This forms a closed-loop O&M strategy control method based on user feedback.

Description

Technical Field

[0001] This invention relates to the field of IT operations and maintenance technology, and in particular to a closed-loop IT operations and maintenance strategy optimization method that integrates user feedback. Background Technology

[0002] As information systems become increasingly larger and more complex, IT operations and maintenance activities have gradually evolved from a single-system, single-responsibility operation and maintenance model to a complex operation and maintenance scenario involving multi-system collaboration, multi-service coupling, and the participation of multiple operation and maintenance entities. In existing technologies, for abnormal situations such as service unavailability, service performance degradation, or functional limitations, the operation and maintenance process is usually triggered through alarm systems, work order systems, or automated operation and maintenance tools. Based on pre-configured responsibility division rules or operation and maintenance permission configurations, the corresponding operation and maintenance entity performs repair or change operations to restore the system to normal operation.

[0003] When an anomaly involves multiple service components or multiple operation and maintenance entities, it is common for the operation and maintenance responsibility to be unclear in a short period of time or for different operation and maintenance entities to disagree on the attribution of responsibility for the anomaly. As a result, before the responsibility is clear, there is no clear correspondence between the system's operating status and the operation and maintenance process. This makes it easy for the triggering and execution of operation and maintenance strategies to rely on manual coordination or ad-hoc judgment, making it difficult to form a stable and consistent operation and maintenance process. Summary of the Invention

[0004] To address the above shortcomings, this invention provides a closed-loop IT operations and maintenance strategy optimization method that integrates user feedback. This method aims to improve the problem that the triggering and execution of operations and maintenance strategies are prone to relying on manual coordination or ad-hoc judgments, making it difficult to form a stable and consistent operations and maintenance processing path.

[0005] This invention provides the following technical solution: a closed-loop IT operations and maintenance strategy optimization method that integrates user feedback, comprising the following steps: S1. When receiving feedback from the user side that the service is unavailable, the service performance is degraded, or the function is limited, it is determined that there is a user-perceptible anomaly in the current IT system. S2. Based on preset operation and maintenance responsibility rules, attempt to determine the responsible entity with the authority to perform repair or change operations for user-perceptible anomalies; S3. If no responsible entity with legitimate execution authority can be identified within the preset time window, or if multiple mutually exclusive claims of responsibility are detected, the current operating status of the IT system will be marked as a responsibility freeze state. S4. After entering the responsibility freeze state, the execution of routine operation and maintenance strategies that rely on the authorization of the responsible subject is prohibited, and the system switches to a predefined set of responsibility-neutral minimal action strategies. S5. Select and execute an operational action from the set of minimal action strategies that is independent of responsibility attribution, rollbackable, and does not introduce irreversible effects, in order to adjust the operating status of the IT system. S6. After executing a neutral responsibility operation and maintenance action, continuously receive user feedback information. When the user feedback meets the preset conditions for lifting the responsibility freeze, lift the responsibility freeze, restore the normal operation and maintenance strategy, and record this neutral responsibility operation and maintenance action and the corresponding user feedback results as the basis for operation and maintenance strategy optimization.

[0006] By adopting the above technical solution, when the responsibility for operation and maintenance cannot be clearly determined within a preset time or there is a dispute over responsibility, the operating status of the IT system is clearly marked as a responsibility freeze state. This responsibility freeze state is used as a unified trigger condition for switching operation and maintenance strategies and selecting operation and maintenance actions. This establishes a clear and stable correspondence between the system operating status and the operation and maintenance process, thereby avoiding the reliance on manual coordination or temporary judgment for the triggering and execution of operation and maintenance strategies before the responsibility is clear. It ensures that the operation and maintenance process can automatically enter the controlled processing path according to preset rules, forming a consistent and repeatable operation and maintenance strategy execution process.

[0007] Preferably, in S1, determining that the current IT system has a user-perceptible anomaly includes: Receive feedback information from the user side indicating that the service is unavailable, the service performance is degraded, or the function is limited, and classify and label the feedback information. Determine whether the feedback information matches a preset type of user-perceived anomaly; When the feedback information matches at least one type of user-perceptible anomaly, it is determined that there is a user-perceptible anomaly in the current IT system.

[0008] Preferably, in S2, the preset operation and maintenance responsibility rules include: Based on the pre-configured association between services and operation and maintenance entities in the IT system, candidate responsible entities are identified for anomalies that are perceptible to users. Verify the operational permissions of candidate responsible entities to determine whether they have the authority to perform repair or change operations; When a candidate responsible entity passes the operation permission verification, it is identified as a responsible entity with the authority to perform repair or change operations.

[0009] Preferably, in S3, the failure to identify any responsible entity with legitimate execution authority within the preset time window includes: After entering the process of determining the responsible party, a preset time window is activated; Within a preset time window, monitor whether any responsible party has received a confirmation instruction with the authority to perform repair or change operations. If no confirmation instruction from any responsible party is received by the end of the preset time window, it is determined that no responsible party with legitimate execution authority could be identified within the preset time window.

[0010] Preferably, in S3, the detection of multiple mutually exclusive liability claims includes: During the process of determining the responsible party, we receive liability statements from multiple candidate responsible parties. Determine whether there are conflicting liability attribution results among multiple liability statements; When multiple liability claims conflict with each other, it is determined that multiple mutually exclusive liability claims have been detected.

[0011] Preferably, in S3, marking the current operating state of the IT system as a responsibility-frozen state includes: A liability freeze flag is generated when it is determined that no liable entity with legal enforcement authority can be identified or when multiple mutually exclusive liability claims are detected. Write the responsibility freeze flag into the status record corresponding to the IT system's operating status, so that the IT system enters the responsibility freeze state.

[0012] Preferably, in S4, the switching to a predefined set of responsibility-neutral minimal action strategies includes: After marking the current operating status of the IT system as a responsibility freeze state, determine the operation and maintenance policy switching instruction corresponding to the responsibility freeze state; According to the operation and maintenance policy switching instruction, stop executing the regular operation and maintenance policy that depends on the authorization of the responsible party; Load the minimum set of responsibility-neutral action strategies corresponding to the responsibility freeze state from the pre-configured set of operation and maintenance strategies, and use it to replace the regular operation and maintenance strategies.

[0013] Preferably, in S5, the operational actions that are independent of attribution of responsibility, rollback-capable, and do not introduce irreversible effects include the following constraints: The operation and maintenance actions do not require confirmation of the identity or authorization status of any responsible party before execution, thus making the operation and maintenance actions independent of the attribution of responsibility; The operation and maintenance action corresponds to at least one pre-configured rollback operation when it is executed, so that the system can be restored to the state before the operation and maintenance action is executed. The aforementioned operations and maintenance actions will not make permanent modifications to the IT system's infrastructure, data structure, or interface definitions during the execution process.

[0014] Preferably, in S6, the step of "when user feedback meets the preset conditions for lifting the responsibility freeze state" includes: After performing a neutral operational action, obtain user feedback information received within a preset feedback observation period; Determine whether user feedback still contains information about service unavailability, degraded service performance, or limited functionality; If no feedback indicating service unavailability, service performance degradation, or functional limitations is detected within the preset feedback observation period, the user feedback is deemed to meet the preset conditions for lifting the responsibility freeze state.

[0015] Preferably, in S6, recording this neutral operational action and the corresponding user feedback results as the basis for operational strategy optimization includes: After the responsibility freeze is lifted, a strategy record identifier corresponding to this responsibility-neutral operation and maintenance action is generated; The strategy record identifier is associated with the execution information of responsibility-neutral operation and maintenance actions and stored accordingly; User feedback received during the execution of neutral operations and maintenance actions will be associated with policy record identifiers and stored to form policy record information for subsequent adjustments to operations and maintenance policies.

[0016] The present invention has the following beneficial effects: 1. In this invention, when maintenance responsibilities cannot be determined within a preset time window or when mutually exclusive responsibility claims arise, the IT system's operating status is explicitly marked as a responsibility-frozen state, and this state is written into the system's operating status record. This transforms situations where responsibility is uncertain, which originally relied on manual coordination or implicit handling, into a system-identifiable and manageable operating state. Consequently, subsequent maintenance strategy switching, maintenance action selection, and state recovery processes can all be scheduled and controlled around this unified responsibility-frozen state, avoiding the problems of inconsistent maintenance processing paths or scattered processing logic when responsibilities are unclear. 2. In this invention, by pre-configuring a set of responsibility-neutral minimum action strategies, and allowing only operation and maintenance actions that meet the constraints of not depending on responsibility attribution, being rollbackable, and not introducing irreversible impacts while in a responsibility-frozen state, the system can still make necessary adjustments to the IT system's operating status even when the responsible party is not yet clear or there is a conflict of responsibility. At the same time, it avoids triggering operation and maintenance operations that require responsibility authorization or may cause long-term impacts, thereby providing the system with a controlled and recoverable operation and maintenance execution path in responsibility conflict scenarios.

[0017] 3. In this invention, user feedback is used as the basis for determining the lifting of the responsibility freeze state. By continuously receiving user feedback within a preset feedback observation period and judging whether there are still user-perceptible anomalies, a systematic determination of the conditions for lifting the responsibility freeze state is achieved. After the responsibility freeze state is lifted, the responsibility-neutral operation and maintenance actions and their corresponding user feedback results are linked and recorded, so that the operation and maintenance strategy adjustment can be traced and referenced based on historical operation and maintenance actions and user perception results. Thus, without introducing complex models or real-time optimization mechanisms, a closed-loop convergence mechanism for operation and maintenance strategies with user feedback as the core is formed. Attached Figure Description

[0018] Figure 1 This is a flowchart of a closed-loop IT operations and maintenance strategy optimization method that integrates user feedback, as proposed in this invention. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] In a first embodiment of the present invention, the present invention provides a closed-loop IT operations and maintenance strategy optimization method that incorporates user feedback, such as... Figure 1 As shown, it includes the following steps: S1. When receiving feedback from the user side that the service is unavailable, the service performance is degraded, or the function is limited, it is determined that there is a user-perceptible anomaly in the current IT system. Furthermore, in S1, determining whether the current IT system has user-perceptible anomalies includes: Receive feedback from users regarding service unavailability, service performance degradation, or functional limitations, and categorize and label the feedback information. Determine whether the feedback information matches the preset user-perceived anomaly type; When the feedback information matches at least one type of user-perceptible anomaly, it is determined that there is a user-perceptible anomaly in the current IT system.

[0021] Specifically, the IT operations and maintenance system sets up a feedback access module to receive user-side feedback information. This module collects feedback information from users regarding service unavailability, service performance degradation, or functional limitations. User-side feedback can come from user-submitted work orders, client-built-in feedback interfaces, service call failure messages, or other feedback channels that reflect the user's actual experience. After entering the system, the feedback information is uniformly converted into a structured feedback record format and written into the feedback processing queue so that subsequent processing can process feedback information from different sources in a unified data format. After receiving feedback information, the system uses a configured feedback classification and processing module to classify and label the received feedback information. Specifically, the feedback classification and processing module parses the abnormal descriptions contained in the feedback information according to pre-configured feedback feature rules, and classifies and labels the feedback information based on the corresponding service identifier, function identifier, or access result status. The classification result of the feedback information can be expressed as follows: ; in Indicates the first The feature set of user feedback information This indicates the first [item] extracted from the feedback information. An anomaly feature is used to describe the anomaly type attribute contained in the feedback information; After completing the feature extraction and classification of the feedback information, the system matches the feedback information with preset user-perceptible anomaly types. Specifically, the system pre-maintains a set of user-perceptible anomaly types: ; in Indicates the first There are three types of user-perceptible anomalies, including at least service unavailability, service performance degradation, and functional limitation. Each type of user-perceptible anomaly corresponds to a set of pre-configured anomaly features. ; in Indicates the use of describing the first Abnormal characteristics of various abnormal types; When performing matching judgments, the system calculates the feature set of feedback information. With the set of features of anomaly types The matching degree between the two is used to determine whether the feedback information meets the criteria for identifying user-perceptible anomalies. The matching degree can be calculated in the following way: ; in, Indicates the first Feedback message and the first Users can perceive the degree of matching between anomaly types. This represents the number of intersection features between the feedback information feature set and the anomaly type feature set. Indicates the number of features contained in the feature set of an anomaly type; In matching degree Greater than or equal to the preset matching threshold In this case, the system determines that the feedback information matches the corresponding user-perceptible anomaly type, where the matching threshold is... The pre-set threshold parameter is used to control the sensitivity of anomaly detection. The matching threshold can be configured and adjusted according to the actual operating requirements of the system. When the feedback information meets the above matching conditions with at least one type of user-perceptible anomaly, the system determines that there is a user-perceptible anomaly in the current IT system and generates a corresponding user-perceptible anomaly identifier within the system to characterize the current anomaly determination result. This user-perceptible anomaly identifier then serves as the trigger condition for the subsequent responsibility determination process, enabling the system to directly enter the subsequent operation and maintenance process based on actual user feedback without relying on monitoring indicators or performance data, thereby realizing an anomaly determination mechanism triggered by user feedback. Through the above implementation method, this embodiment performs structured processing, feature extraction, and anomaly type matching and determination on user feedback information, introduces clear matching degree calculation rules and threshold determination mechanisms, so that the user-perceptible anomaly determination process has definite operating logic and feasibility, thereby providing a stable and reproducible anomaly determination basis for subsequent responsibility determination and operation and maintenance strategy switching.

[0022] S2. Based on preset operation and maintenance responsibility rules, attempt to determine the responsible entity with the authority to perform repair or change operations for user-perceptible anomalies; Furthermore, in S2, the preset operation and maintenance responsibility rules include: Based on the pre-configured association between services and operation and maintenance entities in the IT system, candidate responsible entities are identified for anomalies that are perceptible to users. Verify the operational permissions of candidate responsible entities to determine whether they have the authority to perform repair or change operations; When a candidate responsible entity passes the operation permission verification, it is identified as a responsible entity with the authority to perform repair or change operations.

[0023] Specifically, after the IT operations and maintenance system completes the determination of user-perceptible anomalies, it initiates the process of determining the responsible party. By setting the responsibility rule parsing module, it calls the preset operations and maintenance responsibility rules to realize the automatic determination of the responsible party with the authority to perform repair or change operations. The preset operations and maintenance responsibility rules are pre-configured by the system during the deployment or configuration phase and stored in the responsibility rule configuration unit. They are used to describe the relationship between various services in the IT system and the corresponding operations and maintenance parties, as well as the scope of the operations and maintenance parties' operation permissions. During the determination of responsible parties, the system first identifies a set of candidate responsible parties corresponding to user-perceptible anomalies based on the pre-configured associations between services and operations and maintenance (O&M) entities within the IT system. Specifically, the system maintains a service-O&M entity association mapping table, which describes the correspondence between each service identifier and one or more O&M entity identifiers. When a user-perceptible anomaly is associated with a target service identifier, the system queries the mapping table to obtain the set of O&M entity identifiers associated with that service identifier and uses this set of O&M entity identifiers as the candidate responsible party set. This candidate responsible party set can be represented as follows: ; in Represents the set of candidate responsible parties. Indicates the first One candidate operation and maintenance entity; After obtaining the set of candidate responsible entities, the system further verifies their operational permissions to determine whether they have the authority to perform repair or change operations. Specifically, the system pre-maintains a set of operation and maintenance permissions. ; in Indicates the first The system provides different types of operation and maintenance (O&M) permissions to describe repair operation permissions, change operation permissions, or other O&M-related permission categories. Additionally, the system configures a corresponding set of permission identifiers for each candidate responsible entity. ;in Indicates the candidate responsible entity The set of operation permissions possessed; when performing operation permission verification, the system determines the target set of operation permissions based on the maintenance operation requirements corresponding to the anomalies that the current user can perceive. ;in This represents the set of operation permissions required under the current abnormal scenario. The system then performs permission verification by determining whether the set of permission identifiers of the candidate responsible entity covers the target set of operation permissions. This determination process can be implemented using the following rules: ; in Indicates the candidate responsible entity The permission verification result, when This indicates that the candidate responsible entity has the authority to perform repair or change operations. This indicates that the candidate responsible party does not have the corresponding operational authority; After completing the operation permission verification of all candidate responsible entities, the system determines the candidate responsible entities that pass the permission verification as responsible entities with the authority to perform repair or change operations, and writes the responsible entity's identifier into the responsible entity determination result record as the responsible entity input condition for subsequent operation and maintenance processing. This enables the system to automatically determine the responsible entity based on preset operation and maintenance responsibility rules and permission verification logic without manual intervention. Through the above implementation method, this embodiment realizes the systematic operation of the responsibility subject determination process by pre-configured association between services and operation and maintenance subjects and deterministic verification rules based on permission sets. This ensures that the determination of the responsibility subject has clear source basis, clear operation logic and feasible technical path, thereby providing a reliable basis for responsibility determination when the responsibility subject cannot be determined, such as entering a responsibility freeze state or implementing a responsibility neutral operation and maintenance strategy.

[0024] S3. If no responsible entity with legitimate execution authority can be identified within the preset time window, or if multiple mutually exclusive claims of responsibility are detected, the current operating status of the IT system will be marked as a responsibility freeze state. Furthermore, in S3, if no responsible entity with legitimate execution authority can be identified within the preset time window, the following are included: After entering the process of determining the responsible party, a preset time window is activated; Within a preset time window, monitor whether any responsible party has received a confirmation instruction with the authority to perform repair or change operations. If no confirmation instruction from any responsible party is received by the end of the preset time window, it is determined that no responsible party with legitimate execution authority could be identified within the preset time window.

[0025] Furthermore, in S3, several mutually exclusive liability claims were detected, including: During the process of determining the responsible party, we receive liability statements from multiple candidate responsible parties. Determine whether there are conflicting liability attribution results among multiple liability statements; When multiple liability claims conflict with each other, it is determined that multiple mutually exclusive liability claims have been detected.

[0026] Furthermore, in S3, marking the current operating state of the IT system as a responsibility-frozen state includes: A liability freeze flag is generated when it is determined that no liable entity with legal enforcement authority can be identified or when multiple mutually exclusive liability claims are detected. Write the responsibility freeze flag into the status record corresponding to the IT system's operating status, so that the IT system enters the responsibility freeze state.

[0027] Specifically, after the IT operations and maintenance system completes the judgment of user-perceptible anomalies and enters the responsibility subject judgment process, it initiates the responsibility freeze state identification process. This is so that when the responsible subject cannot be determined within a preset time or when mutually exclusive responsibility claims arise, the IT system operation status is switched to the responsibility freeze state to drive the subsequent operations and maintenance strategy switching process. The responsibility freeze state identification process is executed by the freeze state judgment module. The freeze state judgment module establishes a data interaction relationship with the responsibility subject judgment module and the system status management module, so that the output of the responsibility subject judgment process can be used to trigger the freeze state mark writing operation. When the process for determining the responsible party is initiated, the frozen determination module records the initiation time of the process as the start time of the time window, which is denoted as [start time]. At the same time, based on the pre-configured time window length parameter Generation time window end time ,in The system timestamp indicating when the process for determining the responsible party begins. Indicates the preset time window length. This represents the end time of a preset time window, and satisfies the relationship... ; During the preset time window, the freeze-state determination module monitors the receipt status of confirmation commands. Confirmation commands represent the confirmation result of any responsible party having the authority to perform repair or change operations. Confirmation commands can be generated through internal interfaces of the operation and maintenance system, messages returned by the permission system, or other command formats that can be parsed by the system. The freeze-state determination module uses whether a confirmation command has been received within the time window as an indicator variable. ,in Indicates time At least one confirmation command has been received prior to this time. Indicates time No confirmation instruction was received before the specified time, and in Initialization at any time ; When the system time reaches the end of the time window At that time, the frozen state determination module determines whether there is a situation where the responsible party cannot be identified within the time window based on the confirmation instruction indicator variable. The determination result of not being able to identify the responsible party within the time window is recorded as follows. ,in This indicates that no responsible party with legitimate execution authority could be identified within the preset time window. This indicates that at least one responsible entity with legitimate execution authority has been identified within the preset time window, and the decision has been approved by the rules. Achieve, when When ,when When This enables a definitive determination of the responsible party with legitimate execution authority that cannot be identified within a preset time window; During the determination of responsible parties, the frozen state determination module also receives and checks the consistency of responsibility declaration information. This responsibility declaration information characterizes the candidate responsible party's declaration of attribution of anomaly responsibility. This information can be generated by the corresponding operation and maintenance component or responsibility rule parsing module of the candidate responsible party and is received by the operation and maintenance system in the form of structured records. The set of responsibility declaration information is denoted as […]. ,in Indicates the first Each candidate responsible entity outputs a liability statement record, and each liability statement record includes a liability attribution result field. The responsibility attribution result field is used to indicate the judgment result of the candidate responsible party on the attribution of responsibility for the anomaly; To detect mutually exclusive liability claims, the frozen state judgment module determines the consistency of the liability attribution result field in the liability claim information set, denoted as [the set of liability attribution results]. ,in Statement of Intent and Disclaimer of Liability Record The corresponding responsibility attribution results, the frozen state determination module defines conflict detection indicator variables ,in This indicates that multiple mutually exclusive liability claims have been detected. This indicates that no mutually exclusive liability claims were detected, and conflict determination is achieved by judging the number of different values ​​in the liability attribution result set. If the set... The number of different values ​​in the middle satisfies Then determine Otherwise, it will be judged. ,in Indicates from set Extract the set formed by different values, symbol Indicates the number of elements in the set; After completing the time window determination and conflict determination, the freeze state determination module generates a responsibility freeze identifier and writes it into the system status record to mark the IT system's operating status as a responsibility freeze state. Specifically, the system defines a responsibility freeze state determination result indicator variable. ,in This indicates that the conditions for entering the responsibility freeze state have been met. This indicates that the conditions for entering the responsibility freeze state are not met, and is implemented through logical conjunction. , where the symbol Represents a logical OR operation, when or When ,when and When ; when At that time, the system status management module generates a responsibility freeze flag. Among them, the liability freeze mark State labeling data used to characterize the responsibility freeze state, and responsibility freeze identifier. The status record is written to the status record corresponding to the IT system's operating status. The status record can be stored in a status management database, an in-memory status table, or other optional status storage media. After the write operation is completed, the system sets the current IT system operating status to a responsibility-frozen state, so that the subsequent operation and maintenance strategy switching module can trigger the switching to the responsibility-neutral minimum action strategy set based on the responsibility-frozen state. This transforms the operating scenario where the responsible party cannot be determined or the responsibility declaration is conflicting into a status event that the system can identify, record, and drive subsequent processing.

[0028] S4. After entering the responsibility freeze state, the execution of routine operation and maintenance strategies that rely on the authorization of the responsible party is prohibited, and the system switches to the predefined set of responsibility-neutral minimal action strategies. Furthermore, in S4, switching to a predefined set of responsibility-neutral minimal action strategies includes: After marking the current operating status of the IT system as a responsibility freeze state, determine the operation and maintenance policy switching instruction corresponding to the responsibility freeze state; According to the operation and maintenance policy switching instruction, stop executing the regular operation and maintenance policy that depends on the authorization of the responsible party; Load the minimum set of responsibility-neutral action strategies corresponding to the responsibility freeze state from the pre-configured set of operation and maintenance strategies, and use it to replace the regular operation and maintenance strategies.

[0029] Specifically, after the IT system status management module marks the IT system's operating status as a responsibility-frozen state, the IT operations and maintenance system initiates the operations and maintenance policy switching process. The operations and maintenance policy switching process is used to reconfigure the currently effective operations and maintenance policies under the responsibility-frozen state, so that the system prohibits the execution of regular operations and maintenance policies that depend on the authorization of the responsible subject, and switches to a predefined set of responsibility-neutral minimal action policies. The operations and maintenance policy switching process is completed collaboratively by the policy switching control module and the operations and maintenance policy execution module. The policy switching control module is used to generate and issue operations and maintenance policy switching instructions, and the operations and maintenance policy execution module is used to complete the specific execution of policy stopping and policy loading. During the determination of policy switching instructions, the system pre-maintains a mapping relationship between operating states and operation and maintenance policy types during the configuration phase. This mapping relationship describes the set of policy types that should take effect under different operating states. The operation and maintenance policy switching instruction corresponding to the responsibility-frozen state is denoted as... ,in This represents the control command data used to trigger the policy switching process. After detecting the responsibility freeze state flag, the policy switching control module generates the operation and maintenance policy switching command according to the mapping relationship. The operation and maintenance policy switching command is sent to the operation and maintenance policy execution module so that the operation and maintenance policy execution module enters the policy switching working mode that matches the responsibility freeze state; After receiving the operation and maintenance policy switching instruction, the operation and maintenance policy execution module first prohibits the execution of regular operation and maintenance policies that depend on the authorization of the responsible entity. The set of regular operation and maintenance policies is denoted as follows. ,in This refers to a set of policies that are effective in a non-responsibility-freeze state and whose execution process requires confirmation of the responsible entity's identity or authorization status. The operation and maintenance policy execution module can prohibit the regular operation and maintenance policy set by means of canceling pending policy tasks, suspending the currently executing policy process, or freezing periodic scheduling trigger conditions, in order to avoid continuing to trigger operation and maintenance operations that depend on the authorization of the responsible entity in the responsibility-freeze state. In order to ensure the consistency of the prohibition process, the system can optionally configure an execution status identifier for each policy in the regular operation and maintenance policy set and perform unified processing based on the execution status identifier during the prohibition process. After completing the prohibition processing of the regular operation and maintenance policy, the operation and maintenance policy execution module loads the set of responsibility-neutral minimal action policies corresponding to the responsibility freeze state from the pre-configured operation and maintenance policy set. The set of responsibility-neutral minimal action policies is denoted as... ,in This represents a set of operational actions that are effective under a responsibility freeze state, include actions that are independent of responsibility attribution, are rollbackable, and do not introduce irreversible effects. The operational strategy execution module reads the set of strategies from the strategy configuration storage unit. The policy description information is then registered as part of the currently active policy set, ensuring that subsequent operational and maintenance actions are scheduled based on this policy set. To complete the set of routine operation and maintenance strategies. The alternative; To ensure logical consistency during the policy switching process, the system can optionally verify the effective conditions and mutual exclusion relationships of the policy set. The effective conditions are used to limit the responsibility-neutral minimum action policy set to be effective only when the responsibility is frozen. The mutual exclusion relationships are used to limit the responsibility-neutral minimum action policy set and the regular operation and maintenance policy set to not be effective at the same time. After the policy switching control module completes the policy set replacement, it updates the policy status record of the operation and maintenance policy execution module so that the system maintains a single effective policy set during the responsibility freeze period. After the policy switch is completed, the system may optionally generate a policy switch completion identifier and write it into the system status record. The policy switch completion identifier is used to indicate that the current policy execution path has been switched from the regular operation and maintenance policy to the responsibility-neutral minimal action policy set. This makes it easier for the subsequent operation and maintenance action selection and execution process to directly call the operation and maintenance actions in the responsibility-neutral minimal action policy set based on the identifier, thereby forming a continuous operation link from policy switch to action execution under the responsibility-free state. Through the above implementation method, this embodiment uses the responsibility freeze state as the trigger condition, and drives the prohibition of conventional operation and maintenance policies and the loading and replacement of the responsibility-neutral minimum action policy set through the operation and maintenance policy switching command. This enables the system to complete the policy execution path switching in the operation scenario where the responsible subject cannot be determined or the responsibility statement conflicts, and provides a policy execution basis for the subsequent selection and execution of operation and maintenance actions within the responsibility-neutral policy set.

[0030] S5. Select and execute an operational action from the set of minimal action strategies that is independent of responsibility attribution, rollbackable, and does not introduce irreversible effects, in order to adjust the operating status of the IT system. Furthermore, in S5, operational actions that are independent of attribution of responsibility, rollbackable, and do not introduce irreversible effects include the following constraints: Before any operation and maintenance action is carried out, it is not necessary to confirm the identity or authorization status of any responsible party, so that the operation and maintenance action does not depend on the attribution of responsibility; Each operation and maintenance action corresponds to at least one pre-configured rollback operation, so that the system can be restored to its state before the operation and maintenance action was executed. During the operation and maintenance process, no permanent modifications are made to the IT system's infrastructure, data structure, or interface definitions.

[0031] Specifically, after the IT operations and maintenance system loads and activates the set of responsibility-neutral minimum action strategies in the responsibility-free state, it enters the operation and maintenance action selection and execution process. This process is used to select and execute operation and maintenance actions that meet the constraints from the set of responsibility-neutral minimum action strategies in order to adjust the operating status of the IT system. The operation and maintenance action selection and execution process is completed collaboratively by the operation and maintenance action scheduling module and the operation and maintenance strategy execution module. The operation and maintenance action scheduling module is used to determine and select the constraints of operation and maintenance actions within the strategy set, while the operation and maintenance strategy execution module is used to execute the selected operation and maintenance actions and record the execution process information. In the responsibility-neutral minimal action strategy set, the system establishes an operation and maintenance action description record for each operation and maintenance action during the configuration phase. This record provides information such as the execution target, execution steps, preconditions, and rollback information for the operation and maintenance action. The set of operation and maintenance actions is denoted as […]. ,in Indicates the first Each operation and maintenance action is associated with a set of constraint attribute fields, which are used to support the operation and maintenance action scheduling module to perform deterministic filtering at runtime. To ensure that executed maintenance actions do not depend on attribution of responsibility, the system configures a responsibility dependency constraint field for each maintenance action. This field indicates whether the identity or authorization status of the responsible party needs to be confirmed before the maintenance action is executed. The maintenance action scheduling module defines a responsibility dependency determination function when selecting maintenance actions. ,in; ; when This indicates that the operational actions meet the constraint of not depending on the attribution of responsibility. This indicates that the operation and maintenance action does not meet the constraint of not depending on the attribution of responsibility. The operation and maintenance action scheduling module only processes actions that meet the constraint. The maintenance and operation activities will proceed to the next screening process; To ensure the rollback capability of executed operational actions, the system pre-configures at least one rollback operation for each operational action. The rollback operation is used to restore the IT system to its state before the operational action was executed. Rollback operations can employ methods such as restoring configuration parameters, revoking temporary policies, resetting the operating mode, or other optional recovery methods. The system defines a set of rollback configurations. Indication and Operations The associated set of rollback operations, where To ensure that a set contains at least one rollback operation, the operations and maintenance action scheduling module defines a rollback capability determination function when selecting operations and maintenance actions. ,in: ; symbol Indicates the number of elements in the set, when This indicates that the operational action has at least one pre-configured rollback operation. This indicates that the operation and maintenance actions are not rollbackable; the operation and maintenance action scheduling module only handles actions that meet the following conditions. The maintenance and operation activities will proceed to the next screening process; To avoid irreversible impacts introduced by operational actions, the system configures constraints on the scope of changes made during these actions. The scope of changes characterizes the categories of objects involved in the execution of the operational action. These object categories include at least three types of objects: infrastructure, data structures, and interface definitions. Infrastructure includes, but is not limited to, compute node topology, network connectivity, or storage organization. Data structures include, but are not limited to, database table structures or data model definitions. Interface definitions include, but are not limited to, external service interface protocols or interface parameter conventions. The system configures an irreversible impact constraint field for operational actions to indicate whether the action permanently modifies the aforementioned object categories. The operational action scheduling module defines an irreversible impact determination function based on this constraint. ,in: ; when This indicates that the operational actions meet the constraint that they do not introduce irreversible effects. This indicates that the operation and maintenance action does not meet the constraint of not introducing irreversible effects. The operation and maintenance action scheduling module only processes actions that meet the constraint. The operational and maintenance actions are entered into the executable set; After completing the three types of constraint determination, the operation and maintenance action scheduling module comprehensively determines the executability of operation and maintenance actions, and identifies operation and maintenance actions that meet all constraints as responsibility-neutral executable operation and maintenance actions. The system defines an operation and maintenance action executability determination function. ,in: ; symbol Represents the logical AND operation, when This indicates that the operational actions simultaneously meet the constraints of being independent of liability attribution, being rollbackable, and not introducing irreversible effects. This indicates that at least one constraint condition of the operation and maintenance action is not met; During the operation and maintenance action execution phase, the operation and maintenance action scheduling module starts from meeting... During the operation and maintenance process, the pending operation and maintenance actions are identified and issued to the operation and maintenance policy execution module. The operation and maintenance policy execution module adjusts the IT system's operating status according to the execution steps recorded in the operation and maintenance action description record. Operating status adjustments may include adjusting operating configuration parameters, switching preset operating modes, or enabling temporary control policies. During execution, the operation and maintenance policy execution module records the start time, execution status, execution result, and association with rollback operations for each operation and maintenance action, and writes these records to the system status management module so that the rollback operation set can be invoked later when recovery is needed. The rollback operation in the middle completes the recovery process; Through the above implementation method, this embodiment makes deterministic determinations on the responsibility dependency of operation and maintenance actions, rollback configuration and irreversible impact constraints under the responsibility freeze state, so that the operation and maintenance actions in the responsibility-neutral minimum action strategy set can be screened and executed with clear rules, thereby forming an operation and maintenance action selection and execution process that meets the constraints, and providing traceable operation and maintenance action execution records for subsequent responsibility freeze state lifting and operation and maintenance strategy restoration based on user feedback.

[0032] S6. After executing a neutral responsibility operation and maintenance action, continuously receive user feedback information. When the user feedback meets the preset conditions for lifting the responsibility freeze, lift the responsibility freeze, restore the normal operation and maintenance strategy, and record this neutral responsibility operation and maintenance action and the corresponding user feedback results as the basis for operation and maintenance strategy optimization.

[0033] Furthermore, in S6, when user feedback meets the preset conditions for lifting the responsibility freeze state, it includes: After performing a neutral operational action, obtain user feedback information received within a preset feedback observation period; Determine whether user feedback still contains information about service unavailability, degraded service performance, or limited functionality; If no feedback indicating service unavailability, service performance degradation, or functional limitations is detected within the preset feedback observation period, the user feedback is deemed to meet the preset conditions for lifting the responsibility freeze state.

[0034] Furthermore, in S6, the results of this neutral operational action and the corresponding user feedback will be recorded as the basis for operational strategy optimization, including: After the responsibility freeze is lifted, a strategy record identifier corresponding to this responsibility-neutral operation and maintenance action is generated; The strategy record identifier is associated with the execution information of responsibility-neutral operation and maintenance actions and stored accordingly; User feedback received during the execution of neutral operations and maintenance actions will be associated with policy record identifiers and stored to form policy record information for subsequent adjustments to operations and maintenance policies.

[0035] Specifically, after the IT operations and maintenance system completes the execution of a neutral operations and maintenance action under a responsibility freeze state, it enters the user feedback closed-loop monitoring and freeze state release process. This process continuously receives user feedback information and determines whether to release the responsibility freeze state and restore the normal operations and maintenance strategy based on preset responsibility freeze state release conditions. After releasing the responsibility freeze state, it records the neutral operations and maintenance action and its corresponding user feedback results to form strategy record information for subsequent operations and maintenance strategy adjustments. The user feedback closed-loop monitoring and freeze state release process is completed collaboratively by the feedback observation module, the freeze state release judgment module, the strategy recovery module, and the strategy record module. Each module shares the responsibility freeze state flag, operations and maintenance action execution information, and user feedback reception results through the system status management module. After the execution of a neutral, responsibility-based maintenance operation, the feedback observation module initiates a preset feedback observation period to limit the scope of observation of user feedback information. The start time of the preset feedback observation period is denoted as [missing information]. The preset feedback observation period duration parameter is denoted as... The end time of the preset feedback observation period is denoted as And satisfy the relation ,in This represents the system timestamp indicating the completion of a neutral, responsibility-neutral maintenance operation. This indicates the preset feedback observation time period length parameter. Indicates the end time of the preset feedback observation period; During the preset feedback observation period, the feedback observation module continuously receives user feedback information and writes it into the feedback observation cache, which is used to store data within the time interval. The set of user feedback records received internally, denoted as: ; in Indicates the first Each user feedback record received within a preset feedback observation period contains a feedback type field, which is used to characterize whether the feedback belongs to at least one of service unavailability, service performance degradation, or functional limitation. The freeze-state unfreezing determination module checks the user feedback records in the feedback observation buffer to determine whether the user feedback information still contains feedback content indicating service unavailability, service performance degradation, or functional limitations. To achieve a deterministic expression of this detection process, the system defines an anomaly feedback indicator function. ,in: ; in This indicates that the user feedback record is an abnormal feedback record. This indicates that the user feedback record is not an abnormal feedback record; To determine the presence of abnormal feedback within a time period, the freeze-state unfreezing determination module aggregates and calculates the abnormal feedback indicator functions within a preset feedback observation time period. The aggregated abnormal feedback result is denoted as... ,in: ; in This indicates the number of abnormal feedback records detected within the preset feedback observation period. This indicates that no feedback regarding service unavailability, service performance degradation, or functional limitations was detected within the preset feedback observation period. This indicates that at least one abnormal feedback item was detected within the preset feedback observation period; After completing the abnormal feedback aggregation calculation, the freeze state release determination module makes a determination based on the preset responsibility freeze state release conditions, and the system defines an indicator variable for the responsibility freeze state release determination result. ,in: ; in This indicates that the user feedback meets the preset conditions for lifting the liability freeze state. This indicates that the user feedback does not meet the preset conditions for lifting the freeze state. The freeze state lifting determination module will then determine the conditions upon receiving the feedback. Then, a freeze state release command is sent to the system status management module, causing the system status management module to clear the responsibility freeze state flag and triggering the policy recovery module to enter the process of restoring the normal operation and maintenance policy; After the responsibility freeze is lifted, the policy recovery module, according to the policy recovery rules pre-configured by the system, switches the effective policy set of the operation and maintenance policy execution module from the responsibility-neutral minimum action policy set back to the regular operation and maintenance policy set. The policy recovery process may include restoring the regular operation and maintenance policy scheduling trigger conditions, re-enabling the execution queue of the regular operation and maintenance policy, or other optional recovery methods, so that the system returns to the regular operation and maintenance policy execution path after the responsibility freeze is lifted. After the responsibility freeze is lifted and the regular operation and maintenance policy is restored, the policy recording module generates a policy record identifier corresponding to this responsibility-neutral operation and maintenance action. The policy record identifier is denoted as […]. ,in To uniquely identify the record of this neutral operational action, the policy record module will use the policy record identifier. The execution information of responsibility-neutral operation and maintenance actions is associated and stored. The execution information of responsibility-neutral operation and maintenance actions may include operation and maintenance action identifier, execution start time, execution end time, execution status, and rollback configuration reference information, etc. The policy logging module further links user feedback received during the execution of responsibility-neutral operations and maintenance actions with policy logging identifiers. For associated storage, user feedback results can include a set of user feedback records within a preset feedback observation period. Or its detection results information, after being associated and stored, form policy record information. The policy record information is used to characterize the relationship between responsibility-neutral operation and maintenance actions and changes in user feedback under the responsibility freeze state, so as to support the tracing and recall of historical execution records when adjusting subsequent operation and maintenance strategies; Through the above implementation method, after the execution of the responsibility-neutral operation and maintenance action, this embodiment realizes the determination of the release of the responsibility freeze state and the restoration of the normal operation and maintenance strategy by receiving user feedback, detecting abnormal feedback and determining the release of the freeze state through a preset feedback observation period. After the release, the operation and maintenance action execution information and user feedback results are associated and stored through the policy record identifier to form policy record information that can be used for subsequent operation and maintenance strategy adjustments, so that user feedback forms a closed-loop operation link in the process of releasing the responsibility freeze state and policy recording.

[0036] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A closed-loop IT operations and maintenance strategy optimization method incorporating user feedback, characterized in that, Includes the following steps: S1. When receiving feedback from the user side that the service is unavailable, the service performance is degraded, or the function is limited, it is determined that there is a user-perceptible anomaly in the current IT system. S2. Based on preset operation and maintenance responsibility rules, attempt to determine the responsible entity with the authority to perform repair or change operations for user-perceptible anomalies; S3. If no responsible entity with legitimate execution authority can be identified within the preset time window, or if multiple mutually exclusive claims of responsibility are detected, the current operating status of the IT system will be marked as a responsibility freeze state. S4. After entering the responsibility freeze state, the execution of routine operation and maintenance strategies that rely on the authorization of the responsible subject is prohibited, and the system switches to a predefined set of responsibility-neutral minimal action strategies. S5. Select and execute an operational action from the set of minimal action strategies that is independent of responsibility attribution, rollbackable, and does not introduce irreversible effects, in order to adjust the operating status of the IT system. S6. After executing a neutral responsibility operation and maintenance action, continuously receive user feedback information. When the user feedback meets the preset conditions for lifting the responsibility freeze, lift the responsibility freeze, restore the normal operation and maintenance strategy, and record this neutral responsibility operation and maintenance action and the corresponding user feedback results as the basis for operation and maintenance strategy optimization.

2. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S1, determining that the current IT system has a user-perceptible anomaly includes: Receive feedback information from the user side indicating that the service is unavailable, the service performance is degraded, or the function is limited, and classify and label the feedback information. Determine whether the feedback information matches a preset type of user-perceived anomaly; When the feedback information matches at least one type of user-perceptible anomaly, it is determined that there is a user-perceptible anomaly in the current IT system.

3. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S2, the preset operation and maintenance responsibility rules include: Based on the pre-configured association between services and operation and maintenance entities in the IT system, candidate responsible entities are identified for anomalies that are perceptible to users. Verify the operational permissions of candidate responsible entities to determine whether they have the authority to perform repair or change operations; When a candidate responsible entity passes the operation permission verification, it is identified as a responsible entity with the authority to perform repair or change operations.

4. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S3, the failure to identify any responsible party with legitimate execution authority within the preset time window includes: After entering the process of determining the responsible party, a preset time window is activated; Within a preset time window, monitor whether any responsible party has received a confirmation instruction with the authority to perform repair or change operations. If no confirmation instruction from any responsible party is received by the end of the preset time window, it is determined that no responsible party with legitimate execution authority could be identified within the preset time window.

5. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S3, the detection of multiple mutually exclusive liability claims includes: During the process of determining the responsible party, liability statements from multiple candidate responsible parties are received; Determine whether there are conflicting liability attribution results among multiple liability statements; When multiple liability claims conflict with each other, it is determined that multiple mutually exclusive liability claims have been detected.

6. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S3, marking the current operating state of the IT system as a responsibility-frozen state includes: A liability freeze flag is generated when it is determined that no liable entity with legal enforcement authority can be identified or when multiple mutually exclusive liability claims are detected. Write the responsibility freeze flag into the status record corresponding to the IT system's operating status, so that the IT system enters the responsibility freeze state.

7. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S4, the switch to the predefined set of responsibility-neutral minimal action strategies includes: After marking the current operating status of the IT system as a responsibility freeze state, determine the operation and maintenance policy switching instruction corresponding to the responsibility freeze state; According to the operation and maintenance policy switching instruction, stop executing the regular operation and maintenance policy that depends on the authorization of the responsible party; Load the minimum set of responsibility-neutral action strategies corresponding to the responsibility freeze state from the pre-configured set of operation and maintenance strategies, and use it to replace the regular operation and maintenance strategies.

8. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S5, the operational actions that are independent of attribution of responsibility, rollbackable, and do not introduce irreversible effects include the following constraints: The operation and maintenance actions do not require confirmation of the identity or authorization status of any responsible party before execution, thus making the operation and maintenance actions independent of the attribution of responsibility; The operation and maintenance action corresponds to at least one pre-configured rollback operation when it is executed, so that the system can be restored to the state before the operation and maintenance action is executed. The aforementioned operations and maintenance actions will not make permanent modifications to the IT system's infrastructure, data structure, or interface definitions during the execution process.

9. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S6, the step of "when user feedback meets the preset conditions for lifting the responsibility freeze state" includes: After performing a neutral operational action, obtain user feedback information received within a preset feedback observation period; Determine whether user feedback still contains information about service unavailability, degraded service performance, or limited functionality; If no feedback indicating service unavailability, service performance degradation, or functional limitations is detected within the preset feedback observation period, the user feedback is deemed to meet the preset conditions for lifting the responsibility freeze state.

10. The closed-loop IT operation and maintenance strategy optimization method integrating user feedback as described in claim 1, characterized in that, In S6, recording this neutral-responsibility operation and maintenance action and the corresponding user feedback results as the basis for operation and maintenance strategy optimization includes: After the responsibility freeze is lifted, a strategy record identifier corresponding to this responsibility-neutral operation and maintenance action is generated; The strategy record identifier is associated with the execution information of responsibility-neutral operation and maintenance actions and stored accordingly; User feedback received during the execution of neutral operations and maintenance actions will be associated with policy record identifiers and stored to form policy record information for subsequent adjustments to operations and maintenance policies.

Images