A Method and System for Predicting Operation and Maintenance Technical Service Issues Based on Big Data Analytics

Abstract

This application provides a method and system for predicting operational and maintenance (O&M) technical service issues based on big data analysis, relating to the field of O&M service technology. First, it constructs an O&M service link dependency model including dependency relationship types, tracks the operational status of link units in real time, and identifies initial deviation units. Next, it tracks the deviation propagation path to form an O&M service deviation propagation trajectory, maps it to a historical case database in a scenario-based manner, and extracts matching problem types and development characteristics. Finally, based on the mapping results and historical O&M problem processing big data, it generates prediction results including problem type, affected link unit range, and problem development direction, and associates them with a set of O&M technical service problem intervention schemes. This invention improves the accuracy and timeliness of O&M problem prediction.

Description

Technical Field

[0001] This application relates to the field of operation and maintenance service technology, and more specifically, to a method and system for predicting operation and maintenance technical service problems based on big data analysis. Background Technology

[0002] In the field of operations and maintenance (O&M) technical services, accurately predicting potential problems is crucial for ensuring stable system operation and reducing failure losses. Currently, traditional O&M problem prediction methods mainly rely on the experience and judgment of O&M personnel and simple rule settings. O&M personnel observe and analyze the system's operational status based on their personal experience, and when they find certain indicators abnormal, they infer the possible types of failures based on past experience. Rule-based methods, on the other hand, predefine a series of fixed rules, and when system operating data meets these rules, corresponding fault warnings are triggered.

[0003] However, these traditional methods have significant limitations. The experience of operations and maintenance (O&M) personnel is subjective and limited; different personnel have vastly different levels of experience, and it's difficult to cover all possible O&M scenarios. Simple rule settings cannot adapt to complex and ever-changing O&M environments. As system scale and business complexity increase, rule formulation and maintenance become extremely difficult, and rule updates often lag behind system changes, resulting in insufficient accuracy and timeliness in predictions. Furthermore, traditional methods lack in-depth analysis of the dependencies between units in the entire O&M service process, making it impossible to effectively track the transmission of deviations between stages, and difficult to accurately predict the type, scope, and direction of problems in advance, thus failing to meet the demands of modern O&M technical services for efficient and accurate problem prediction. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method and system for predicting operation and maintenance technical service problems based on big data analysis.

[0005] According to a first aspect of this application, a method for predicting operational and maintenance technical service problems based on big data analysis is provided, the method comprising:

[0006] By identifying the pre-dependencies and collaborative dependencies of each unit in the entire operation and maintenance service process, and combining the interaction frequency and continuity of unit interactions from historical operation and maintenance service big data statistics, an operation and maintenance service process dependency model containing dependency relationship types is generated.

[0007] According to the operation and maintenance service link dependency model, the operating status of each link unit of the current operation and maintenance service is tracked in real time. Based on the historical normal operating status data in the operation and maintenance service link dependency model structure, the link unit whose operating status differs from the historical normal operating status is identified, and the link unit is marked as the initial deviation unit. The deviation manifestation of the initial deviation unit is recorded.

[0008] Starting from the initial deviation unit, based on the dependency relationship type in the operation and maintenance service link dependency model structure, the propagation path of the deviation between each link unit is tracked, the deviation manifestation form and deviation propagation order of each link unit affected by the deviation are recorded, and the operation and maintenance service deviation propagation trajectory is formed.

[0009] The deviation transmission trajectory of operation and maintenance services is mapped to the deviation transmission trajectory of cases in the historical operation and maintenance technical service problem case library in a scenario-based manner. Based on the big data of historical cases, the problem types and problem development characteristics corresponding to the cases that match the current trajectory are extracted to form a scenario mapping result.

[0010] Based on the scenario mapping results and historical big data on operation and maintenance issues, we determine the types of problems that may occur in the current operation and maintenance technical services, the scope of affected links and units, and the direction of problem development. We then generate the final operation and maintenance technical service problem prediction results and associate them with the corresponding set of operation and maintenance technical service problem intervention solutions.

[0011] According to a second aspect of this application, a big data analysis-based operation and maintenance technical service problem prediction system is provided. The big data analysis-based operation and maintenance technical service problem prediction system includes a machine-readable storage medium and a processor. The machine-readable storage medium stores machine-executable instructions. When the processor executes the machine-executable instructions, the big data analysis-based operation and maintenance technical service problem prediction system implements the aforementioned big data analysis-based operation and maintenance technical service problem prediction method.

[0012] Based on any of the above aspects, the technical effect of this application is as follows:

[0013] By identifying the pre-dependencies and collaborative dependencies of each unit in the entire operation and maintenance service process, and combining this with the interaction frequency and continuity statistics of historical big data, a dependency model for operation and maintenance service stages, including dependency types, is constructed, clearly presenting the relationships between each stage of the operation and maintenance service. Based on this model, the operational status of each stage unit is tracked in real time, and initial deviation units that differ from historical normal operating status can be quickly identified. Starting from the initial deviation unit, the deviation propagation path is traced, forming an operation and maintenance service deviation propagation trajectory, which can show the propagation process of deviations between each stage unit. The deviation propagation trajectory is mapped to a historical case database in a scenario-based manner, and the problem types and development characteristics can be extracted using historical case big data, improving the accuracy of problem prediction. Finally, based on the scenario mapping results and the problem prediction results generated from historical operation and maintenance problem handling big data, not only are the possible problem types clarified, but also the scope of affected stage units and the direction of problem development are determined, and a set of corresponding intervention solutions is associated. This achieves full-process intelligence from problem prediction to intervention solution formulation, effectively improving the initiative and accuracy of operation and maintenance technical services, reducing the impact of operation and maintenance failures on system operation, and ensuring the efficient and stable operation of operation and maintenance services. Attached Figure Description

[0014] Figure 1 A flowchart illustrating the problem prediction method for operation and maintenance technical services based on big data analysis provided in an embodiment of this application is shown.

[0015] Figure 2 This paper illustrates a schematic diagram of the component structure of the operation and maintenance technical service problem prediction system based on big data analysis provided in an embodiment of this application. Detailed Implementation

[0016] Figure 1 This paper illustrates a flowchart of a method and system for predicting operational and maintenance technical service problems based on big data analysis, as provided in an embodiment of this application. The detailed steps include:

[0017] Step S110: By identifying the pre-dependencies and collaborative dependencies of each unit in the entire operation and maintenance service process, and combining the interaction frequency and continuity of the unit in the historical operation and maintenance service big data statistics, an operation and maintenance service link dependency model containing dependency relationship types is generated.

[0018] This embodiment will focus on the intelligent operation and maintenance service scenario. In this scenario, the technology center acts as the brain, providing technical support to frontline technicians through a large model knowledge base and robot assistance. On-site engineers provide services by going to the nearest service point through a physical positioning system. This also involves the intelligent control of chain stores, such as air conditioning, store information dissemination, and IoT control. In this scenario, the entire operation and maintenance service process covers multiple stages, with complex dependencies between these stages. This step generates an operation and maintenance service stage dependency model.

[0019] Step S111: Obtain the full process record of operation and maintenance services. The full process record of operation and maintenance services includes the record of the operation and maintenance service request initiation, the record of the service resource allocation, the record of the service execution operation, the record of the service result verification, the record of the service delivery, and the record of the service follow-up support.

[0020] In a smart operation and maintenance service scenario, the service request initiation record may include records of air conditioner malfunction repair requests submitted by store staff through the system, and abnormal feedback records from the information release system; the service resource allocation record involves records of the technical center allocating nearby on-site engineers and the required repair equipment based on information such as the type of repair request and the store location; the service execution record includes a detailed record of the on-site engineer's inspection and repair operations on the air conditioner, information release system, etc., after arriving at the store; the service result verification record is a record of the technical center or store staff checking and confirming the operating status of the equipment after repair; the service delivery record is a record of the relevant information confirming the completion of the service by both parties after the repair is completed; and the service follow-up support record includes a tracking record of the equipment's operating status within a certain period after the repair and a record of any follow-up inquiries that the store may raise.

[0021] Step S112: Divide the process into units according to the order of execution of the operation and maintenance service. The entire process is divided into request receiving unit, resource allocation unit, operation execution unit, result verification unit, delivery confirmation unit, and support response unit. Each unit includes a unit function description, information on related units required for unit execution, and a description of unit output content.

[0022] Next, after obtaining the above-mentioned full process record, the process is broken down into stages and units according to the order of execution of the maintenance services. In the smart maintenance service scenario, the function of the request receiving unit is described as receiving various maintenance service requests submitted by stores and performing preliminary classification and organization; the associated unit information required for its execution may include information related to the terminal system that submitted the request by the store; the unit output is described as a classified service request list and a preliminary request information summary. The function of the resource allocation unit is to allocate resources based on the service request information output by the request receiving unit, combined with the real-time location, skill specialties, and inventory status of the on-site engineer; its associated units are the request receiving unit and the physical positioning system that records the engineer's location; the output includes the allocated engineer information, the list of equipment carried, and the estimated arrival time at the store. The operation execution unit is responsible for the actual operation and repair of the faulty equipment after the on-site engineer arrives at the store, according to the allocation instructions and repair plan; the associated units include the resource allocation unit and the faulty equipment system of the store; the output includes the repair operation process record, information on replaced parts, and the current operating parameters of the equipment. The result verification unit verifies the operational status of the equipment after maintenance by the operation execution unit; the associated units are the operation execution unit and the store's equipment monitoring system; the output is a result report indicating whether the verification passed or failed, along with relevant verification data. The delivery confirmation unit confirms service delivery with store staff after successful result verification; the associated units are the result verification unit and the store's confirmation terminal; the output is a service delivery form signed by both parties. The support response unit receives and provides support for any subsequent inquiries from the store after service delivery; the associated units are the delivery confirmation unit and the store's inquiry feedback system; the output includes records of inquiry answers and arrangements for subsequent support services.

[0023] Step S113: Analyze the prerequisite dependencies between each stage unit, identify the stage unit that can only be started after other stage units are completed, mark the dependency as a prerequisite dependency, and record the stage unit that provides prerequisite support and the stage unit that receives support in each prerequisite dependency.

[0024] After completing the breakdown of the process units, we begin to analyze the prerequisite dependencies between each process unit.

[0025] Step S1131: Review the associated unit information required for the execution of each stage unit, filter out the stage units that require other stage units to be completed before they can be started, mark the stage unit as the unit to be analyzed and accepted, and filter based on the startup condition description of the operation and maintenance service stage unit.

[0026] In intelligent operation and maintenance service scenarios, you can view the information of related units required for the execution of each stage unit. For example, the startup condition description of the resource allocation unit may explicitly state that it can only start after the request receiving unit has completed the classification and organization of service requests. In this case, the resource allocation unit is marked as a receiving unit to be analyzed. Similarly, the startup condition of the operation execution unit may depend on the resource allocation unit completing the allocation of engineers and equipment, so the operation execution unit will also be screened as a receiving unit to be analyzed.

[0027] Step S1132: For each receiving unit to be analyzed, check the historical startup conditions of the receiving unit to be analyzed in the full process record of operation and maintenance service, determine the other link units that must be completed before starting the receiving unit to be analyzed, and mark the other link units as the providing unit to be analyzed.

[0028] Taking the resource allocation unit, which is the unit to be analyzed, as an example, a review of the historical startup conditions reveals that a request receiving unit must complete the receipt and classification of store service requests before it can be started. Therefore, the request receiving unit is the unit to be analyzed for the resource allocation unit. For the operation execution unit, historical startup conditions show that it can only be started after the resource allocation unit has completed engineer allocation. Therefore, the resource allocation unit is the unit to be analyzed for the operation execution unit.

[0029] Step S1133: Review the historical interaction records between the providing unit and the receiving unit to be analyzed, and confirm whether the providing unit to be analyzed completed the execution before the receiving unit to be analyzed started in all historical service cycles.

[0030] Next, review the historical interaction records between the request receiving unit and the resource allocation unit to see if, in all past operation and maintenance service cases, the request receiving unit completed the receiving and classification of service requests before the resource allocation unit started. Similarly, review the historical interaction records between the resource allocation unit and the operation execution unit to confirm whether the resource allocation unit always completes resource allocation before the operation execution unit starts.

[0031] Step S1134: If the historical record shows that the unit to be analyzed always completes execution before the unit to be analyzed starts, and the start instruction of the unit to be analyzed contains the completion confirmation information of the unit to be analyzed, then it is determined that there is a prerequisite dependency relationship between the unit to be analyzed and the unit to be analyzed.

[0032] If, throughout all historical service cycles, the request receiving unit completes execution before the resource allocation unit starts, and the resource allocation unit's startup command explicitly includes confirmation that the request receiving unit has completed request classification, then it can be determined that the request receiving unit and the resource allocation unit have a prerequisite dependency relationship. Similarly, if the resource allocation unit always completes before the operation execution unit starts, and the operation execution unit's startup command includes confirmation that resource allocation is complete, then the resource allocation unit and the operation execution unit have a prerequisite dependency relationship.

[0033] Step S1135: Record the link units that provide preceding support and the link units that receive support in the preceding dependency relationship, and mark the dependency direction of the link units that provide preceding support and the link units that receive support.

[0034] For established prerequisite dependencies, such as a request receiving unit and a resource allocation unit, the unit providing prerequisite support is recorded as the request receiving unit, and the unit receiving support is recorded as the resource allocation unit, with the dependency direction marked as from the request receiving unit to the resource allocation unit. Similarly, the resource allocation unit is recorded as the unit providing prerequisite support, and the operation execution unit is recorded as the unit receiving support, with the dependency direction from the resource allocation unit to the operation execution unit.

[0035] Step S1136: For each determined prerequisite dependency, analyze the historical service records and record multiple historical intervals from the completion of the supporting link unit to the start of the receiving link unit; based on the recorded multiple historical intervals, calculate or describe its interval time characteristics.

[0036] Taking the pre-dependency relationship between the request receiving unit and the resource allocation unit as an example, we analyze historical service records to record the interval between the completion of each request receiving unit and the start of the resource allocation unit. Based on these historical intervals, we describe their interval characteristics. For example, these intervals are mostly concentrated in a relatively stable range, showing a certain regularity, that is, the interval characteristics are relatively stable and have small fluctuations.

[0037] Step S1137: Check if there is a situation where a supporting link unit depends on multiple supporting link units. If so, record the identifiers of all supporting link units and the differences in the support content provided by each supporting link unit to the supporting link unit.

[0038] In intelligent operation and maintenance service scenarios, a single support-receiving unit may depend on multiple supporting units. For example, a result verification unit might rely not only on the operation execution unit to complete the repair operation but also on the store's equipment monitoring system to provide historical operational data prior to the repair. In this case, the result verification unit is the support-receiving unit, while the operation execution unit and the unit related to the store's equipment monitoring system (let's call them the monitoring data provider) are the supporting units. Therefore, it's necessary to record the identifiers of the operation execution unit and the monitoring data provider, and describe the support content of the operation execution unit as the repair operation process and results, and the support content of the monitoring data provider as the historical operational data of the equipment, thus reflecting the differences in the supported content.

[0039] Step S1138: Organize all preceding dependency records to form a preceding dependency list. The preceding dependency list includes the identifier of the link unit that provides preceding support, the identifier of the link unit that receives support, the dependency direction, the average interval time characteristics, and the differences in the supported content.

[0040] All the preceding dependencies obtained from the above analysis, such as the request receiving unit and the resource allocation unit, the resource allocation unit and the operation execution unit, etc., are organized according to the information such as the link unit identifier that provides preceding support, the link unit identifier that receives support, the direction of dependency, the average interval time characteristics, and the differences in the supported content, to form a list of preceding dependencies, which provides data support for the subsequent construction of the operation and maintenance service link dependency model.

[0041] Step S114: Analyze the collaborative dependencies between each component, identify the component that needs to be executed simultaneously with other components to complete the service task, mark the dependency as a collaborative dependency, and record all components that are executed together in each collaborative dependency.

[0042] In intelligent operation and maintenance service scenarios, in addition to the pre-existing dependencies, there may also be collaborative dependencies between the various links and units, which need to be analyzed and identified.

[0043] Step S1141: Review the execution requirements of each service task in the full process record of operation and maintenance services, filter out the service tasks that require multiple links and units to be executed simultaneously, mark the service task as a collaborative task, and filter according to the execution requirements description in the service task specification.

[0044] Review the execution requirements of each service task in the full process record of operation and maintenance services. For example, in the service task of jointly debugging the air conditioning system and information release system of the store, the execution requirements description in the service task specification may clearly indicate that the sub-unit responsible for air conditioning debugging and the sub-unit responsible for information release system debugging in the operation execution unit need to perform the operation simultaneously. In this case, the joint debugging service task is marked as a collaborative task.

[0045] Step S1142: For each collaborative task, extract all the link units involved in the execution of the collaborative task, and mark all the link units as the collaborative unit group to be analyzed.

[0046] For the aforementioned joint debugging and coordination task, the sub-units responsible for air conditioning debugging and information release system debugging involved in its execution process are extracted, and these two sub-units are marked as the coordination unit group to be analyzed.

[0047] Step S1143: Review the historical execution time records of each unit in the collaborative unit group to be analyzed, and confirm whether the execution start time of each unit in the collaborative unit group to be analyzed is within the preset synchronization threshold and whether the execution end time of each unit in the collaborative unit group to be analyzed is within the preset synchronization threshold during the execution of all historical collaborative tasks.

[0048] Review the time records of the collaborative unit group to be analyzed during the execution of historical joint debugging collaborative tasks, and check whether the execution start time of the sub-unit responsible for air conditioning debugging and the sub-unit responsible for information release system debugging are within the preset synchronization threshold, such as whether they start within five minutes before or after the planned start time; at the same time, check whether their execution end time is also within a similar preset synchronization threshold.

[0049] Step S1144: If the historical records show that the start time of each unit in the collaborative unit group to be analyzed is within the preset synchronization threshold, the end time of each unit in the collaborative unit group to be analyzed is within the preset synchronization threshold, and the output content of each unit in the collaborative unit group to be analyzed needs to be integrated before the collaborative task can be completed, then it is determined that there is a collaborative dependency relationship between the units in the collaborative unit group to be analyzed.

[0050] If historical records show that in all joint debugging and collaborative tasks, the start and end times of the subunit responsible for air conditioning debugging and the subunit responsible for information publishing system debugging are both within the preset synchronization threshold, and the outputs of these two subunits, such as air conditioning debugging parameters and information publishing system debugging parameters, need to be integrated to complete the entire joint debugging and collaborative task, then it is determined that these two subunits have a collaborative dependency relationship.

[0051] Step S1145: Record the identifiers of all components jointly executed in the collaborative dependency relationship, and label the name of the collaborative task corresponding to the collaborative dependency relationship.

[0052] Record the identifiers of the sub-units responsible for air conditioning debugging and information publishing system debugging that have collaborative dependencies, and label the corresponding collaborative task as "Joint Debugging Task of Store Air Conditioning and Information Publishing System".

[0053] Step S1146: For each collaborative dependency, analyze the historical service records and record the start and end times of each component within the collaborative dependency in multiple historical executions; based on the recorded start and end times of multiple historical executions, calculate or describe its execution synchronization characteristics.

[0054] The study analyzed the records of this collaborative dependency in multiple historical joint debugging tasks, recording the start and end times of each execution of the sub-unit responsible for air conditioning debugging and the sub-unit responsible for information publishing system debugging. Based on these records, the study described its execution synchronization characteristics, such as the small difference between the start times of the two sub-units and their essentially synchronized end times, indicating a high degree of synchronization in the overall execution process.

[0055] Step S1147: Check if there is a situation where a link unit participates in multiple collaborative dependencies. If so, record the names of all collaborative tasks that the link unit participates in and the functional differences of the link unit in different collaborative tasks.

[0056] In intelligent operation and maintenance service scenarios, a single unit may participate in multiple collaborative dependencies. For example, the subunit responsible for network configuration within the operation execution unit may participate in both the joint debugging of the store's air conditioning system and information publishing system, and the integration of the store's IoT control system and the headquarters management system. In this case, it is necessary to record the names of the collaborative tasks in which the subunit responsible for network configuration participates, such as the "joint debugging task of the store's air conditioning and information publishing system" and the "integration task of the store's IoT system and the headquarters system," and describe its functional differences in different collaborative tasks. For example, in the former, it is mainly responsible for configuring the network connection between the air conditioning and information publishing system, while in the latter, it is responsible for setting the network communication parameters between the IoT system and the headquarters management system.

[0057] Step S1148: Organize all collaborative dependency records to form a collaborative dependency list. The collaborative dependency list includes a list of unit identifiers for each link within the collaborative dependency, the corresponding collaborative task name, execution synchronization characteristics, and functional differences of the link units.

[0058] All the collaborative dependencies obtained from the analysis, such as the collaborative dependencies between the subunit responsible for air conditioning debugging and the subunit responsible for information publishing system debugging, are organized according to the list of unit identifiers, corresponding collaborative task names, execution synchronization characteristics, and functional differences of each unit within the collaborative dependency relationship, forming a collaborative dependency list.

[0059] Step S115: Collect the interaction records of each unit in the historical operation and maintenance service process, and count the interaction frequency of each unit under each dependency based on the interaction data of multiple historical service cycles. Record the dependency relationships that have interaction in multiple consecutive service cycles and mark the dependency relationship as a continuous interaction dependency relationship.

[0060] The system collects interaction records of each unit in multiple historical service cycles within a smart operations and maintenance service scenario. This includes the number of interactions between the request receiving unit and the resource allocation unit in each service cycle, and the number of interactions between the resource allocation unit and the operation execution unit. Based on this interaction data, the system calculates the interaction frequency of each unit within each dependency relationship (including prerequisite dependencies and collaborative dependencies). Simultaneously, it records dependencies that interact across multiple consecutive service cycles. For example, if the request receiving unit and the resource allocation unit interact in every cycle of the past twelve service cycles, then the dependency between them is marked as a continuous interaction dependency.

[0061] Step S116: Statistically analyze the interaction interruption status of each link unit under each dependency relationship, filter out the dependencies that have not been interrupted in the historical interactions or whose number of interruptions is lower than the preset interruption number threshold, and mark the dependency relationship as a stable interaction dependency relationship.

[0062] The system statistically analyzes the interaction interruptions of each dependency's components. For example, it counts the number of times the request receiving unit and the resource allocation unit experienced interaction interruptions during historical interactions—that is, when one party sends an interaction request and the other party fails to respond within a specified time. Then, dependencies that did not experience any interruptions in historical interactions or whose number of interruptions is below a preset interruption threshold (e.g., the number of interruptions does not exceed 5% of the total number of interactions) are identified and marked as stable interaction dependencies. For instance, if the number of interaction interruptions between the request receiving unit and the resource allocation unit is below this threshold, then their dependency is marked as a stable interaction dependency.

[0063] Step S117: Integrate the preceding dependencies and collaborative dependencies, and combine the marking results of continuous interaction dependencies and stable interaction dependencies to construct the initial operation and maintenance service dependency framework.

[0064] By integrating the lists of prerequisite dependencies and collaborative dependencies obtained from the previous analysis, and combining them with the labeling results of continuous and stable interactive dependencies, an initial dependency framework for the operation and maintenance service stages is constructed. This framework includes the prerequisite and collaborative dependencies between each stage unit, as well as information such as whether these relationships belong to continuous and stable interactive dependencies.

[0065] Step S118: Delete dependencies in the initial framework that have an interaction frequency lower than the preset interaction frequency threshold and are not continuous interactions, and retain continuous and stable dependencies and their corresponding link units.

[0066] Based on a preset interaction frequency threshold (e.g., the average number of interactions per cycle across multiple service cycles is not less than a certain value), the dependencies in the initial operation and maintenance service process dependency framework are filtered. Dependencies with interaction frequencies below the threshold and not belonging to continuous interactions are deleted, and only dependencies with high interaction frequencies, continuous interactions, and stability, along with their corresponding process units, are retained. This ensures that the constructed operation and maintenance service process dependency model can accurately reflect the main and stable dependencies between each process unit.

[0067] Step S119: Label the dependency type for each retained dependency relationship. The dependency type includes prerequisite dependencies and collaborative dependencies, and supplement the description of typical interaction scenarios of the dependency relationship in multiple historical service cycles.

[0068] For the dependencies that have been filtered and retained, such as the pre-dependency relationship between the request receiving unit and the resource allocation unit, and the collaborative dependency relationship between the sub-unit responsible for air conditioning debugging and the sub-unit responsible for information publishing system debugging, their dependency types are marked as pre-dependencies or collaborative dependencies. At the same time, descriptions of typical interaction scenarios for this dependency in multiple historical service cycles are added. For example, a typical interaction scenario between the request receiving unit and the resource allocation unit is as follows: After receiving an air conditioning malfunction request from a store, the request receiving unit completes the classification within a specified time and transmits the request information to the resource allocation unit, which then begins the allocation of engineers and equipment.

[0069] Step S1110: Based on the labeled dependencies, link units, and typical interaction scenario descriptions, generate an operation and maintenance service link dependency model structure. The accuracy of the operation and maintenance service link dependency model structure is verified through historical operation and maintenance service big data. It includes complete information of all link units, the type of dependency relationship between units, and a description of typical interaction scenarios of the dependency relationship.

[0070] Based on the annotated dependencies, information on each unit (including unit function descriptions, information on associated units required for unit execution, and descriptions of unit output content), and descriptions of typical interaction scenarios, a dependency model structure for the operation and maintenance service stages is generated. After generation, the accuracy of this dependency model structure is verified using historical operation and maintenance service big data to ensure that the model accurately reflects the dependencies and interactions between units in each stage of the operation and maintenance service. This dependency model structure includes complete information on all units, the types of dependencies between units, and descriptions of typical interaction scenarios.

[0071] Step S120: Based on the operation and maintenance service link dependency model, track the operating status of each link unit of the current operation and maintenance service in real time. Based on the historical normal operating status data in the operation and maintenance service link dependency model structure, identify the link unit whose operating status differs from the historical normal operating status, mark the link unit as the initial deviation unit, and record the deviation manifestation of the initial deviation unit.

[0072] After constructing the operation and maintenance service dependency model, it is necessary to track the running status of each unit of the current operation and maintenance service in real time based on the model and identify the initial deviation unit.

[0073] Step S121: Extract all link unit information and historical normal operation status descriptions of each link unit from the operation and maintenance service link dependency model structure. The historical normal operation status descriptions include the execution time range of the link unit, the integrity requirements of the output content of the link unit, and the interaction rhythm between the link unit and related units.

[0074] Information about all units in the operation and maintenance service process dependency model structure is extracted, such as the request receiving unit and the resource allocation unit. Simultaneously, historical normal operation status descriptions of each unit are extracted. Taking the resource allocation unit as an example, its historical normal operation status description might include the execution time range for completing resource allocation within a certain time frame after receiving a service request. The completeness requirements for the output content include the engineer's name, contact information, equipment list, and estimated arrival time. The interaction rhythm with related units might involve providing progress updates to the request receiving unit at regular intervals, and immediately sending allocation result information to the operation execution unit upon completion.

[0075] Step S122: Collect the running data of each unit in the current operation and maintenance service execution process in real time. The running data includes the execution start time, execution end time, output details, interaction time points and interaction content with related units of the current unit.

[0076] During the execution of current intelligent operation and maintenance services, operational data from each unit in each stage is collected in real time. For example, for the resource allocation unit, the following data is collected in real time: execution start time (i.e., the time when the service request information is received from the request receiving unit), execution end time (the time when the engineer and equipment allocation is completed and the allocation instruction is issued), output details (specific content such as the actual engineer information, equipment list, and estimated arrival time), interaction time points with related units (such as the request receiving unit and the operation execution unit) (such as the time point when the allocation progress is fed back to the request receiving unit and the time point when the allocation result is sent to the operation execution unit), and interaction content (specific progress information fed back and content of the allocation instruction sent).

[0077] Step S123: Compare the execution time of the current process unit with the range of execution times of process units in the historical normal operation state, wherein the execution time is the difference between the execution end time and the execution start time of the current process unit.

[0078] Taking the resource allocation unit as an example, its current execution time is calculated, which is the difference between the execution end time and the execution start time. Then, this difference is compared with the execution time range of the link unit in the historical normal operation status description to determine whether the current execution time is within the normal range.

[0079] Step S124: Compare the output details of the current link unit with the completeness requirements of the output content of the link unit in the historical normal operation state, and record the missing output content items.

[0080] Step S1241: Extract the output content integrity requirements of the current link unit from the historical normal operation status description, and describe the content item name that the current link unit should output, the content format requirements of each content item, and the necessary data items of each content item.

[0081] Extract the completeness requirements for output content from the historical normal operation status description of the resource allocation unit. The output content items should include the engineer's name, engineer's contact information, list of equipment carried, and estimated arrival time. The format requirements for each content item are as follows: engineer's name should be the full Chinese name, contact information should be 11 digits, the list of equipment carried should be a list of equipment names and quantities, and the estimated arrival time should be in the format of year, month, day, hour, and minute. The necessary data items for each content item are: engineer's name has no additional necessary data items, engineer's contact information should include location information, the list of equipment carried should include the unique identification number of the equipment, and the estimated arrival time should be accurate to the minute.

[0082] Step S1242: Obtain the output details of the current stage unit, and compare them one by one according to the content item names in the integrity requirements to confirm whether the output of the current stage unit contains all the required content items.

[0083] Obtain the current output details of the resource allocation unit, and compare them one by one according to the above content item names to check whether they include the four content items: engineer's name, engineer's contact information, list of equipment to be carried, and estimated arrival time.

[0084] Step S1243: If the output of the current stage unit is missing one or more content items, record the name of the missing content item, and at the same time record the name of the content item already included in the output of the current stage unit.

[0085] If the output details of the current resource allocation unit are missing the estimated arrival time, then the name of the missing item will be recorded as the estimated arrival time, and the names of the included items will be recorded as the engineer's name, engineer's contact information, and a list of equipment to be carried.

[0086] Step S1244: For the content items already included in the output of the current stage unit, further compare the content format of each content item with the content format requirements in the completeness requirements. If the formats do not match, record the name of the content item with the mismatch and the specific mismatch.

[0087] For items containing engineer contact information, if the format is not 11 digits but other formats, such as containing letters or symbols, the item name for the mismatched format will be recorded as "Engineer Contact Information," and the specific mismatch is that the format is not 11 digits.

[0088] Step S1245: For content items with matching formats, compare whether the necessary data items within the content item are complete. If any necessary data items are missing, record the name of the content item and the name of the missing necessary data item.

[0089] Assuming the engineer's contact information is in the correct format (11 digits), but lacks the necessary location information, then the name of this information item should be "Engineer's Contact Information," and the name of the missing necessary data item should be "Location Information."

[0090] Step S1246: Compile the names of missing content items, the names of content items with mismatched formats and the mismatches, and the names of missing necessary data items and the missing data items into an output deviation record.

[0091] The missing estimated arrival times, mismatched engineer contact information and specific mismatches, and missing engineer contact information and missing data items from the above resource allocation units are compiled into an output deviation record.

[0092] Step S1247: If the output of the current stage unit fully meets the integrity requirements, record that the output content of the current stage unit has no deviation, and output the output deviation record. The output deviation record includes the current stage unit identifier, the output deviation type and the specific deviation content. The output deviation type includes missing items, format mismatch and missing data items.

[0093] If the output of the current stage unit fully meets the integrity requirements, and there are no missing items, format mismatches, or missing data items, then the output of the current stage unit is recorded as having no deviation, and the output deviation record is output in a format that includes the current stage unit identifier, the output deviation type (in this case, no deviation), and the specific deviation content (no specific deviation content).

[0094] Step S125: Compare the interaction time and content between the current link unit and the related unit with the interaction rhythm between the link unit and the related unit in the historical normal operation state to identify the situation of delayed interaction time or incomplete interaction content.

[0095] Taking the interaction between the resource allocation unit and the operation execution unit as an example, the interaction rhythm in the historical normal operation state is that the resource allocation unit should send the allocation result information to the operation execution unit immediately after completing the allocation. If the current resource allocation unit sends the information to the operation execution unit after a long time after completing the allocation, that is, the interaction time is delayed; or the allocation result information sent is missing the engineer's contact information, that is, the interaction content is incomplete, these situations need to be identified.

[0096] Step S126: Integrate differences in execution time, missing output content, and abnormal interaction rhythm to determine whether the current unit's operating status differs from its historical normal operating status.

[0097] The system integrates and analyzes the following factors: differences in execution time of the current process unit (e.g., the execution time of the resource allocation unit exceeds the historical normal range), missing output content (e.g., missing estimated arrival time), and abnormal interaction rhythm (e.g., delays in sending information to the operation execution unit). If any one or more of these factors are abnormal, it is determined that the current process unit's operating status differs from its historical normal operating status.

[0098] Step S127: Mark the link unit with the difference in operating status as the initial deviation unit, and record the identification information of the initial deviation unit.

[0099] When it is determined that the operating status of a certain link unit differs from its historical normal operating status, such as the resource allocation unit, it is marked as the initial deviation unit, and its identification information is recorded, such as "resource allocation unit 001".

[0100] Step S128: Record the deviation manifestation of the initial deviation unit, including the specific circumstances of the initial deviation unit execution time exceeding the range, the specific items of the initial deviation unit output content missing, the specific time points and abnormal content of the initial deviation unit interaction rhythm abnormal. All records are associated with the comparison difference between the current running data and the historical data.

[0101] The system meticulously records the deviations exhibited by initial deviation units (such as resource allocation units). Specific instances of execution time exceeding the range include: the current execution time exceeding the upper limit of the historical normal execution time range; missing output items include estimated arrival times; and abnormal interaction rhythms occur when the allocation result information is sent to the operation execution unit at a time delayed from the historical normal sending time, with the abnormal content being the missing engineer contact information in the sent message. Furthermore, all these records must be correlated with the differences between current and historical data, such as the execution time exceeding the limit or the delay in interaction time.

[0102] Step S129: Classify and organize the deviation manifestations of the initial deviation units, and divide the deviation items according to the categories of execution deviation, output deviation, and interaction deviation, thereby outputting the identification information of the initial deviation units and the corresponding classification deviation manifestation records.

[0103] The deviations of the initial deviation units are categorized and organized into execution deviations (such as execution time exceeding the range), output deviations (such as missing expected arrival time of output content, mismatched engineer contact information format, etc.), and interaction deviations (such as delayed interaction time, incomplete interaction content), forming a record of categorized deviation manifestations. The identification information of the initial deviation units is then output along with this record.

[0104] Step S130: Starting from the initial deviation unit, based on the dependency relationship type in the operation and maintenance service link dependency model structure, trace the transmission path of the deviation between each link unit, record the deviation manifestation form and deviation transmission order of each link unit affected by the deviation, and form the operation and maintenance service deviation transmission trajectory.

[0105] After marking the initial deviation unit, it is necessary to use it as the starting point to trace the transmission path of the deviation between each link unit, forming the operation and maintenance service deviation transmission trajectory.

[0106] Step S131: Obtain the identification information of the initial deviation unit, the classification deviation performance record of the initial deviation unit, and the dependency relationship type in the operation and maintenance service link dependency model structure.

[0107] Obtain the identification information of the initial deviation unit (such as resource allocation unit 001), its classification deviation performance record (including specific items of execution deviation, output deviation, and interaction deviation), and the dependency relationship type between the initial deviation unit and other link units in the operation and maintenance service link dependency model structure, such as the pre-dependency relationship between the resource allocation unit and the operation execution unit.

[0108] Step S132: Based on the dependency relationship type, determine the associated link units that have a direct dependency relationship with the initial deviation unit. The associated link units include the subsequent link units that serve as the preceding support units of the initial deviation unit, and the parallel link units that have a cooperative dependency with the initial deviation unit.

[0109] Based on the dependency type, identify the related link units that have a direct dependency on the initial deviation unit. For example, if the resource allocation unit is a front-end support unit and its subsequent link unit is the operation execution unit, then the operation execution unit is a related link unit. If the resource allocation unit also has collaborative dependencies with other link units, then those parallel link units are also considered related link units.

[0110] Step S133: Real-time acquisition of the operation data of the associated link unit after the deviation occurs in the initial deviation unit. The operation data includes the execution time of the associated link unit, the output content of the associated link unit, and the interaction between the associated link unit and other units.

[0111] After a deviation occurs in the initial deviation unit (resource allocation unit), the operation data of its related link units (such as operation execution units) are collected in real time, including the execution time of the operation execution unit (the time from receiving the allocation result information to completing the maintenance operation), output content (maintenance operation process record, current equipment operating parameters, etc.), and interaction with other units (such as result verification unit) (such as the time and content of sending maintenance completion information to the result verification unit).

[0112] Step S134: Compare the current operating data of the related link unit with the historical normal operating status of the related link unit to identify whether the related link unit has new deviation manifestations.

[0113] The current operating data of the operation execution unit is compared with its historical normal operating status. For example, the execution time is compared to see if it is within the historical normal range, whether the output content is complete, and whether the interaction with the result verification unit is in line with the historical interaction rhythm. This is to identify whether the operation execution unit has new deviations, such as extended execution time or missing key operating parameters of the equipment in the output content.

[0114] Step S135: Record the identification information of the associated link unit where a new deviation manifestation occurs, mark the associated link unit as a deviation transmission node, and record the deviation manifestation of the deviation transmission node to determine the correlation between the deviation manifestation of the deviation transmission node and the deviation manifestation of the initial deviation unit.

[0115] If a new deviation occurs in the operation execution unit, such as an extended execution time, its identification information (e.g., "Operation Execution Unit 002") is recorded, and it is marked as a deviation transmission node. Simultaneously, its deviation manifestation is recorded as an extended execution time. The correlation between this deviation manifestation and the deviation manifestation of the initial deviation unit (resource allocation unit) is analyzed and determined. For example, a delay in the resource allocation unit sending allocation result information to the operation execution unit leads to a delay in the operation execution unit's startup time, thus extending the execution time.

[0116] Step S136: Starting from the deviation transmission node, repeat the above steps to identify the next-level related link unit that has a direct dependency relationship with the deviation transmission node, collect the operation data of the next-level related link unit and identify the deviation of the next-level related link unit, and repeat the process until no new deviation transmission node appears.

[0117] Starting from the deviation transmission node (operation execution unit 002), repeat steps S132 to S135 to identify the next-level related unit that has a direct dependency on the operation execution unit, such as the result verification unit. Collect the operating data of the result verification unit, compare it with its historical normal operating status, and identify whether a deviation has occurred. If the result verification unit also has a deviation, mark it as a new deviation transmission node, and continue to repeat the above process until no new deviation transmission nodes appear.

[0118] Step S137: Record the occurrence time of the initial deviation unit, each deviation transmission node, and the transmission relationship between the initial deviation unit and each deviation transmission node, and between each deviation transmission node, in the order in which the deviations occur, to form a deviation transmission sequence chain.

[0119] For example, step S1371: extract the time when the deviation of the initial deviation unit is first identified, mark the time as the initial time point, and record the correspondence between the initial deviation unit identifier and the initial time point.

[0120] Extract the time when the deviation of the initial deviation unit (resource allocation unit 001) is first identified, such as "May 20, 2024, 10:30", mark it as the initial time point, and record the correspondence between the initial deviation unit identifier "resource allocation unit 001" and the initial time point.

[0121] Step S1372: Record the time when the deviation of each deviation transmission node is first identified, mark the time as the transmission time point, and record the correspondence between the identifier of each deviation transmission node and the corresponding transmission time point.

[0122] Record the time when the deviation is first identified at each deviation transmission node (such as operation execution unit 002, result verification unit 003, etc.), and mark them as transmission time points. For example, the transmission time point of operation execution unit 002 is "May 20, 2024, 11:15", and the transmission time point of result verification unit 003 is "May 20, 2024, 12:00", etc., and record the correspondence between each deviation transmission node identifier and the corresponding transmission time point.

[0123] Step S1373: Sort the initial time point and all transmission time points in chronological order to form a time series, and describe the deviation unit identifier corresponding to each time point.

[0124] The initial time point "May 20, 2024, 10:30", the transmission time point of operation execution unit 002 "May 20, 2024, 11:15", and the transmission time point of result verification unit 003 "May 20, 2024, 12:00" are sorted in chronological order to form a time series. In this time series, the deviation unit identifier corresponding to each time point is described; for example, 10:30 corresponds to resource allocation unit 001, 11:15 corresponds to operation execution unit 002, and 12:00 corresponds to result verification unit 003.

[0125] Step S1374: Analyze the dependency relationship between the deviation units corresponding to two adjacent time points, determine whether the previous deviation unit is a predecessor dependency unit or a cooperative dependency unit of the next deviation unit, and determine the trigger dependency type of deviation propagation.

[0126] Analyze the dependency relationship between deviation units corresponding to two adjacent time points, such as resource allocation unit 001 (10:30) and operation execution unit 002 (11:15). According to the operation and maintenance service link dependency model, the resource allocation unit is the preceding dependent unit of the operation execution unit. Therefore, the triggering dependency type of deviation propagation is determined to be preceding dependency propagation.

[0127] Step S1375: If the previous deviation unit is the predecessor dependency unit of the next deviation unit, record the transitivity as predecessor dependency transitivity, describe that the deviation of the next deviation unit is caused by the abnormal predecessor support of the previous deviation unit, and describe the impact based on the impact data of historical predecessor dependency transitivity.

[0128] Since resource allocation unit 001 is a prerequisite unit for operation execution unit 002, the transmission relationship between them is recorded as prerequisite dependency transmission. The deviation of operation execution unit 002 is described as being caused by the abnormality of the prerequisite support of resource allocation unit 001 (such as the delay in sending allocation result information). This impact description is based on relevant data on the impact of abnormalities of the prerequisite unit on subsequent units in similar historical prerequisite dependency transmissions.

[0129] Step S1376: If the previous deviation unit is a cooperative dependency unit of the next deviation unit, record the transitive relationship as cooperative dependency transitive, describing that the deviation of the next deviation unit is caused by the abnormal cooperative execution of the previous deviation unit.

[0130] If two adjacent deviation units have a cooperative dependency relationship, such as two cooperatively executed sub-units, and a deviation in the first sub-unit causes a cooperative execution anomaly, which in turn causes a deviation in the second sub-unit, then the transitive relationship is recorded as a cooperative dependency transitive relationship, and it is described that the deviation of the second deviation unit is caused by the cooperative execution anomaly of the first deviation unit.

[0131] Step S1377: Record the identifier of each deviation unit, the occurrence time of each deviation unit, and the type of transmission relationship between each deviation unit and the previous deviation unit in chronological order to form the basic data of the deviation transmission sequence chain.

[0132] In chronological order (10:30, 11:15, 12:00...), the identifier (resource allocation unit 001, operation execution unit 002, result verification unit 003...), occurrence time (corresponding time point), and the type of transmission relationship with the previous deviation unit (preceding dependency transmission or collaborative dependency transmission) are recorded sequentially to form the basic data of the deviation transmission sequence chain.

[0133] Step S1378: Add node numbers to the deviation transmission sequence chain, starting from the initial deviation unit, and then incrementing the numbers of subsequent deviation transmission nodes in the order of their appearance.

[0134] Add node numbers to the deviation propagation sequence chain. The initial deviation unit (resource allocation unit 001) is numbered 1, the subsequent deviation propagation node operation execution unit 002 is numbered 2, the result verification unit 003 is numbered 3, and so on.

[0135] Step S1379: In the deviation transmission sequence chain, each node is numbered and associated with the corresponding deviation unit identifier, occurrence time, and transmission relationship type with the previous node; at the same time, each deviation unit identifier is associated with its core deviation performance summary; based on the node number, deviation unit identifier and its associated information, a deviation transmission sequence chain is formed, which presents the time sequence of deviation transmission, deviation unit association, transmission type and deviation performance of each node.

[0136] In the deviation propagation sequence chain, node number 1 is associated with the deviation unit identifier "Resource Allocation Unit 001", the occurrence time "May 20, 2024, 10:30", and the propagation relationship type with the previous node (none) is (none). The core deviation manifestation summary is "execution time exceeds the range, output content is missing, and interaction is delayed". Node number 2 is associated with the deviation unit identifier "Operation Execution Unit 002", the occurrence time "May 20, 2024, 11:15", and the propagation relationship type with the previous node (number 1) is "pre-dependent propagation". The core deviation manifestation summary is "execution time extended", etc. Based on this information, a complete deviation propagation sequence chain is formed.

[0137] Step S138: Label the dependency type corresponding to each deviation propagation node. The dependency type includes pre-dependent propagation and collaborative dependency propagation. Describe the dependency triggering conditions for deviation propagation. The triggering conditions are determined based on the dependency triggering data in the dependency model structure of the operation and maintenance service link.

[0138] Each deviation propagation node is labeled with its corresponding dependency type. For example, the deviation propagation of operation execution unit 002 corresponds to the dependency type of pre-dependency propagation. The dependency triggering condition of deviation propagation is described. This triggering condition is determined based on the dependency triggering data in the dependency model structure of the operation and maintenance service links. For example, the triggering condition of pre-dependency propagation is that the pre-link unit fails to complete the task on time or outputs abnormal information, causing the subsequent link unit to fail to start or execute normally.

[0139] Step S139: Integrate the initial deviation unit information, the information of each deviation transmission node, the deviation transmission sequence chain, the dependent triggering conditions for deviation transmission, and the deviation manifestation forms of the initial deviation unit and each deviation transmission node.

[0140] The initial deviation unit information (identifier, deviation representation, etc.), the information of each deviation transmission node (identifier, deviation representation, correlation, etc.), the deviation transmission sequence chain, and the dependent triggering conditions of deviation transmission are integrated to form a dataset that comprehensively reflects the deviation transmission situation.

[0141] Step S1310: Construct a visualized deviation propagation trajectory for the operation and maintenance service according to the time sequence and dependency logic of deviation propagation. The deviation propagation trajectory includes node identifiers, node deviation performance, propagation relationships between nodes, and propagation triggering conditions. The deviation propagation trajectory is then output to present the propagation process of deviation from the initial deviation unit to each deviation propagation node and the deviation characteristics between the initial deviation unit and each deviation propagation node.

[0142] Based on the temporal sequence and dependency logic of deviation propagation, a visualized deviation propagation trajectory for operations and maintenance services is constructed using the integrated dataset described above. This trajectory clearly displays the identifiers of each node (e.g., resource allocation unit 001, operation execution unit 002), node deviation behaviors (e.g., execution time exceeding limits, execution time extension), propagation relationships between nodes (e.g., propagation of preceding dependencies), and propagation triggering conditions (e.g., preceding units failing to complete tasks on time). This visualized trajectory provides a clear overview of the deviation propagation process from the initial deviation unit to each deviation propagation node, as well as the deviation characteristics of each node.

[0143] Step S140: Map the deviation transmission trajectory of operation and maintenance services to the deviation transmission trajectory of cases in the historical operation and maintenance technical service problem case library in a scenario-based manner. Based on the big data of historical cases, extract the problem type and problem development characteristics corresponding to the cases that match the current trajectory to form a scenario mapping result.

[0144] After generating the deviation propagation trajectory of operation and maintenance services, it is necessary to perform scenario-based mapping with the deviation propagation trajectory of cases in the historical operation and maintenance technical service problem case library.

[0145] Step S141: Obtain a historical operation and maintenance technical service problem case library. The historical operation and maintenance technical service problem case library contains multiple historical problem cases. Each historical problem case includes the case deviation transmission trajectory, the operation and maintenance service problem type corresponding to the case, the development characteristics description of the problem in the case, and the final impact range of the case.

[0146] Access a historical O&M technical service issue case library, which contains multiple historical issue cases. In the context of intelligent O&M services, each historical issue case, such as "Delay in resource allocation for store air conditioner repair leading to service timeout," includes the following: the trajectory of the deviation (e.g., deviation from the resource allocation unit is transmitted to the operation execution unit, and then to the result verification unit); the corresponding O&M service issue type (service response timeout); a description of the development characteristics of the problem in the case (the deviation starts from the resource allocation unit and gradually propagates to multiple subsequent units, resulting in a significant extension of the overall service completion time); and a record of the final impact of the case (affecting the normal use of the store's air conditioner and reducing customer satisfaction).

[0147] Step S142: Extract key features of the case deviation propagation trajectory from each historical problem case. Key features include the initial deviation unit type of the case, the number of deviation propagation nodes in the case, the dependency relationship type of the case deviation propagation, and the deviation manifestation type of each node in the case.

[0148] Key features of the deviation propagation trajectory are extracted from each historical problem case. Taking the case of "delay in resource allocation for store air conditioner repair leading to service timeout" as an example, the initial deviation unit type of the case is the resource allocation unit; the number of deviation propagation nodes is 2 (operation execution unit and result verification unit); the dependency relationship type of the deviation propagation is pre-dependency propagation; the deviation manifestations of each node in the case are execution time anomalies (execution time of resource allocation unit exceeds the range, execution time of operation execution unit is extended) and interaction anomalies (delay in sending information from resource allocation unit to operation execution unit).

[0149] Step S143: Extract key features from the current operation and maintenance service deviation propagation trajectory. Key features include the current initial deviation unit type, the current number of deviation propagation nodes, the current deviation propagation dependency type, and the deviation manifestation type of each node.

[0150] Key features are extracted from the current deviation propagation trajectory of the operation and maintenance service (such as the trajectory of the resource allocation unit as the initial deviation unit and the operation execution unit and result verification unit as deviation propagation nodes in the previous example). The current initial deviation unit type is resource allocation unit; the current number of deviation propagation nodes is 2; the current deviation propagation dependency type is pre-dependency propagation; the deviation manifestation of each node is abnormal execution time (the execution time of the resource allocation unit exceeds the range, and the execution time of the operation execution unit is extended) and abnormal interaction (the information sending from the resource allocation unit to the operation execution unit is delayed).

[0151] Step S144: Compare the key features of the current trajectory with the key features of each historical case trajectory in a scenario-based manner, focusing on whether the initial deviation unit type is consistent, whether the dependency relationship type of deviation propagation is the same, and whether the deviation manifestation category of each node matches.

[0152] The key features of the current trajectory are compared with the key features of each case in the historical problem case library in a scenario-based manner. The comparison focuses on whether the initial deviation unit type is always a resource allocation unit, whether the deviation propagation dependency type is always a pre-dependency propagation, and whether the deviation manifestation type of each node includes execution time anomalies and interaction anomalies, etc.

[0153] Step S145: Filter out the historical cases with the most key feature matching items and mark the historical cases as candidate matching cases. If there are multiple candidate matching cases, further compare whether the deviation transmission order of the current trajectory and the candidate matching case trajectory is consistent and whether the time interval characteristics of the deviation of each node are similar.

[0154] By comparing the data, historical cases with the most matching key features are selected, such as "delay in resource allocation for store air conditioning repair leading to service timeout," and marked as candidate matching cases. If multiple candidate matching cases exist, such as "delay in resource allocation for store information publishing system," the current trajectory is further compared with the trajectory of these candidate matching cases to see if the deviation transmission order (e.g., both are resource allocation unit—operation execution unit—result verification unit) is consistent, and if the time interval characteristics of the deviation occurrence at each node (e.g., after the initial deviation unit appears, how long does it take to transmit to the first deviation transmission node, and then how long does it take to transmit to the next node) are similar.

[0155] Step S146: Determine the historical case with the highest matching degree with the current trajectory, and extract the operation and maintenance service problem type corresponding to the historical case, the development characteristics description of the problem in the historical case, and the final impact range of the historical case.

[0156] Further comparison revealed that the case of "delayed allocation of store air conditioning repair resources leading to service timeout" had the highest match with the current trajectory. The corresponding maintenance service issue type for this case was identified as service response timeout. The development characteristics of the problem in the case were described as follows: the deviation started from the resource allocation unit and gradually spread to multiple stages such as the operation execution unit and the result verification unit, resulting in abnormal execution times at each stage and the overall service completion time significantly exceeding expectations. The final impact of the case was recorded as affecting the normal operation of the store's air conditioning, causing customer complaints, and negatively impacting the company's service reputation.

[0157] Step S147: Analyze the correlation between the problem development characteristics in the matched case and the deviation transmission trend in the current trajectory, and determine whether the current deviation transmission may show the same development direction as the matched case.

[0158] Analyze the correlation between the problem development characteristics in the matched case (gradual propagation of deviations, affecting multiple stages, and service timeouts) and the deviation propagation trend in the current trajectory (currently propagated from the resource allocation unit to the operation execution unit and result verification unit, with abnormal execution times at each stage). Determine whether the current deviation propagation is likely to continue to propagate to more stage units, leading to a further extension of service completion time, i.e., exhibiting the same service response timeout development trend as the matched case.

[0159] Step S148: Record the problem type of the matching case, the development characteristics of the problem in the matching case, the final impact scope of the matching case, and the correlation analysis conclusion to form a scenario mapping result. The scenario mapping result includes the matching case identifier, the problem type corresponding to the matching case, the development characteristics of the problem in the matching case, the final impact scope of the matching case, and the correlation analysis conclusion.

[0160] Record the problem type of the matched case (service response timeout), the development characteristics of the problem in the matched case (deviation gradually propagates, affecting multiple stages, service timeout), the final impact scope of the matched case (store air conditioners cannot be used normally, customer complaints, damage to reputation), and the correlation analysis conclusion (the current deviation propagation may show the same development direction as the matched case, i.e., service response timeout), forming a scenario mapping result. This scenario mapping result includes the matched case identifier (such as "Case No. 2023005") and the above items.

[0161] Step S150: Based on the scenario mapping results and historical big data on operation and maintenance issues, determine the types of problems that may occur in the current operation and maintenance technical services, the scope of affected links and units, and the direction of problem development, generate the final operation and maintenance technical service problem prediction results, and associate them with the corresponding set of operation and maintenance technical service problem intervention solutions.

[0162] Based on the scenario mapping results and combined with historical big data on operation and maintenance issues, we can predict current operation and maintenance technical service problems and associate intervention solutions.

[0163] Step S151: Analyze the matching case problem types in the scenario mapping results, determine the possible problem types of the current operation and maintenance technical service, describe the typical manifestations of the problem type and how the problem type affects the service.

[0164] The matching case in the scenario mapping results is classified as a service response timeout, thus confirming that the current maintenance technical service may also experience a service response timeout. This type of problem typically manifests as the total time from receiving a request to its completion exceeding the promised service time. The impact on the service includes customer dissatisfaction due to prolonged equipment downtime, potentially leading to customer complaints, damaging the company's relationship with customers, and increasing service costs (such as extended engineer waiting times).

[0165] Step S152: Based on the case impact range record in the scenario mapping result, combined with the number of link units that have deviated in the current operation and maintenance service deviation transmission trajectory and the deviation transmission trend, determine the current affected link unit range. The affected link unit range includes the link units that have deviated and the link units that may be affected by subsequent transmission.

[0166] Based on the final impact range record of the matched cases in the scenario mapping results (affecting the resource allocation unit, operation execution unit, result verification unit, etc.), combined with the number of link units that have already deviated in the current operation and maintenance service deviation propagation trajectory (resource allocation unit, operation execution unit, result verification unit, a total of 3) and the deviation propagation trend (the deviation may continue to propagate), the scope of the currently affected link units is determined. The link units that have already deviated are the resource allocation unit, operation execution unit, and result verification unit; the link units that may be affected by subsequent propagation are the delivery confirmation unit (due to the deviation in the result verification unit, the delivery confirmation unit may be unable to perform delivery confirmation in a timely manner) and the support response unit (if delivery is delayed, it may affect subsequent support response services).

[0167] Step S153: Based on the description of the problem development characteristics and the conclusion of the correlation analysis in the scene mapping results, predict the possible development direction of the current problem. The possible development direction of the current problem includes whether the deviation will continue to be transmitted to more links and units, whether the deviation will be aggravated, and whether the current problem will cause other related service anomalies.

[0168] Based on the problem development characteristics described in the scenario mapping results (deviations propagate gradually, affecting multiple stages) and the correlation analysis conclusions (the current deviation propagation may show the same development direction as the matched case), the possible development direction of the current problem is predicted. The deviation may continue to propagate to more stages and units such as the delivery confirmation unit and the support response unit; the manifestation of the deviation may intensify, such as the execution time of the resource allocation unit may be further extended, and the number of missing items in the output content of the operation execution unit may increase; the current problem (service response timeout) may trigger other related service anomalies, such as a store being affected by a prolonged air conditioning malfunction, which would require additional emergency support services, leading to a shortage of related service resources.

[0169] Step S154: Integrate the types of problems that may occur in the current operation and maintenance technical services, the scope of the currently affected links and units, and the possible development direction of the current problems to form preliminary operation and maintenance technical service problem prediction results.

[0170] By integrating the types of problems that may occur in the current operation and maintenance technical services (service response timeout), the scope of the currently affected links and units (resource allocation unit, operation execution unit, result verification unit, delivery confirmation unit, support response unit), and the possible development direction of the current problems (continued propagation of deviations, aggravation of deviation manifestations, and triggering anomalies in related services), a preliminary prediction result of operation and maintenance technical service problems is formed.

[0171] Step S155: Extract the corresponding operation and maintenance technical service problem intervention plan from the historical operation and maintenance technical service problem case library. The operation and maintenance technical service problem intervention plan includes the handling measures for the initial deviation unit of the matching case, the control measures for the deviation transmission node in the matching case, and the blocking measures to prevent the deviation from being further transmitted in the matching case.

[0172] Intervention plans for the corresponding O&M technical service issues are extracted from the historical O&M technical service issue case database ("Delay in resource allocation for store air conditioning repair leading to service timeout"). The handling measures for the initial deviation unit (resource allocation unit) of the matching case may include immediately activating the backup resource allocation plan and urgently allocating available engineers from other areas. The control measures for the deviation propagation node (operation execution unit) in the matching case may include coordinating engineers in the operation execution unit to work overtime to shorten execution time. Measures to prevent further propagation of deviations in the matching case may include timely sending early warning information to the result verification unit, preparing for verification in advance, and ensuring that verification can be carried out immediately after the operation execution unit completes its work.

[0173] Step S156: Based on the actual configuration of the current operation and maintenance service and the specific situation of the current deviation transmission, adjust the operation and maintenance technical service problem intervention plan corresponding to the matching case, replace the measures in the operation and maintenance technical service problem intervention plan that are not applicable to the current service, and supplement the handling steps for the current specific deviation manifestation.

[0174] Based on the actual configuration of the current operation and maintenance service units (e.g., the availability of backup engineers in the current resource allocation unit differs from historical cases) and the specific circumstances of the current deviation propagation (e.g., the current result verification unit has not yet shown serious deviations), the extracted intervention plan is adjusted. For example, if the engineers involved in the backup resource allocation plan in historical cases are currently unavailable, they are replaced by calling other available backup engineer teams; additional processing steps are added for the current specific deviation manifestations, such as the current resource allocation unit also having a deviation where the output content is missing the estimated arrival time, adding the processing step of "immediately contact the allocated engineer, obtain its estimated arrival time, and complete the output content."

[0175] Step S157: Sort the adjusted operation and maintenance technical service problem intervention plans according to the implementation priority. The implementation priority is determined by the effectiveness of the measures in blocking the transmission of deviations, the time required for the implementation of the measures, and the degree of impact of the measures on normal services. Generate a set of operation and maintenance technical service problem intervention plans containing implementation priorities. Each operation and maintenance technical service problem intervention plan includes the measure name, implementation steps, implementing entity, resources required for implementation, and a description of expected effects.

[0176] The adjusted intervention plans for operation and maintenance technical service issues are prioritized according to their implementation priority. The criteria for prioritization primarily include the effectiveness of the measure in preventing the propagation of deviations (e.g., activating a backup resource allocation plan directly resolves deviations in the resource allocation unit, indicating high effectiveness), the time required for implementation (e.g., short time required to complete estimated arrival time information), and the degree of impact on normal services (e.g., low impact on other services due to coordinating engineers working overtime). For example, the highest priority measure is "Activate the backup resource allocation plan and call upon the available backup engineer team." Its name is "Activate Backup Resource Allocation," the implementation steps are: the technical center queries and filters available backup engineers through the system, sends an emergency allocation instruction to them, the implementing entity is the resource allocation department of the technical center, the required resources are the backup engineer team and related communication equipment, and the expected effect is described as completing engineer allocation within a certain timeframe, shortening the overall execution time of the resource allocation unit. Following this method, a set of operation and maintenance technical service issue intervention plans with implementation priorities is generated.

[0177] Step S158: Integrate the preliminary operation and maintenance technical service problem prediction results with the set of operation and maintenance technical service problem intervention plans to form the final operation and maintenance technical service problem prediction results. The final operation and maintenance technical service problem prediction results include the types of problems that may occur in the current operation and maintenance technical service, the scope of the currently affected links and units, the possible development direction of the current problem, the set of operation and maintenance technical service problem intervention plans and the implementation priority of the plans.

[0178] The preliminary operational technical service problem prediction results (problem type, affected unit scope, and development direction) are integrated with the set of operational technical service problem intervention plans (including the name of each measure, implementation steps, implementing entity, required resources, expected effects, and implementation priority) to form the final operational technical service problem prediction results. This final operational technical service problem prediction result comprehensively presents the potential problems currently facing operational technical services and the corresponding solutions.

[0179] The above embodiments involve the collection of data such as store information and engineer locations, which may include some privacy-sensitive data. To protect the privacy of this data and prevent leakage, the technical measures employed include encrypted data storage (encrypting the collected sensitive data before storing it in the database), access control (setting strict access permissions so that only authorized personnel can access the relevant data), and data anonymization (anonymizing sensitive data such as engineers' contact information in non-essential scenarios, retaining only key information). Through these technical measures, it is ensured that while achieving the function of predicting operational and maintenance technical service issues, users' privacy-sensitive data is protected from leakage and misuse.

[0180] Figure 2This application illustrates a big data analytics-based operation and maintenance (O&M) technical service problem prediction system 100, comprising a processor 1001, a memory 1003, and program code stored in the memory 1003. The processor 1001 executes the program code to implement the steps of the big data analytics-based O&M technical service problem prediction method. The processor 1001 and the memory 1003 are connected, for example, via a bus 1002. Optionally, the big data analytics-based O&M technical service problem prediction system 100 may further include a transceiver 1004, which can be used for data interaction between this big data analytics-based O&M technical service problem prediction system and other big data analytics-based O&M technical service problem prediction systems, such as sending and / or receiving data. It should be noted that in actual scheduling, the transceiver 1004 is not limited to one, and the structure of this big data analytics-based O&M technical service problem prediction system 100 does not constitute a limitation on the embodiments of this application.

[0181] The memory 1003 is used to store program code for executing the embodiments of this application, and its execution is controlled by the processor 1001. The processor 1001 is used to execute the program code stored in the memory 1003 to implement the steps shown in the foregoing method embodiments.

[0182] The above description is only an optional implementation method for some implementation scenarios of this application. It should be noted that for those skilled in the art, other similar implementation methods based on the technical concept of this application, without departing from the technical concept of this application, also fall within the protection scope of the embodiments of this application.

Claims

1. A method for predicting technical service problems based on big data analysis of operation and maintenance, characterized in that, The method includes: By identifying the pre-dependencies and collaborative dependencies of each unit in the entire operation and maintenance service process, and combining the interaction frequency and continuity of unit interactions from historical operation and maintenance service big data statistics, an operation and maintenance service process dependency model containing dependency relationship types is generated. According to the operation and maintenance service link dependency model, the operating status of each link unit of the current operation and maintenance service is tracked in real time. Based on the historical normal operating status data in the operation and maintenance service link dependency model structure, the link unit whose operating status differs from the historical normal operating status is identified, and the link unit is marked as the initial deviation unit. The deviation manifestation of the initial deviation unit is recorded. Starting from the initial deviation unit, based on the dependency relationship type in the operation and maintenance service link dependency model structure, the propagation path of the deviation between each link unit is tracked, the deviation manifestation form and deviation propagation order of each link unit affected by the deviation are recorded, and the operation and maintenance service deviation propagation trajectory is formed. The deviation transmission trajectory of operation and maintenance services is mapped to the deviation transmission trajectory of cases in the historical operation and maintenance technical service problem case library in a scenario-based manner. Based on the big data of historical cases, the problem types and problem development characteristics corresponding to the cases that match the current trajectory are extracted to form a scenario mapping result. Based on the scenario mapping results and historical big data on operation and maintenance issues, we determine the types of problems that may occur in the current operation and maintenance technical services, the scope of affected links and units, and the direction of problem development. We then generate the final operation and maintenance technical service problem prediction results and associate them with the corresponding set of operation and maintenance technical service problem intervention solutions. The process involves mapping the deviation propagation trajectory of operation and maintenance services to the deviation propagation trajectories of cases in a historical operation and maintenance technical service problem case library. Based on historical case big data, the problem types and development characteristics corresponding to cases matching the current trajectory are extracted to form a scenario mapping result, including: Obtain a historical operation and maintenance technical service problem case library. The historical operation and maintenance technical service problem case library contains multiple historical problem cases. Each historical problem case includes the case deviation transmission trajectory, the operation and maintenance service problem type corresponding to the case, the development characteristics description of the problem in the case, and the final impact scope of the case. Extract key features of the case deviation propagation trajectory from each historical problem case. Key features include the initial deviation unit type of the case, the number of deviation propagation nodes, the dependency relationship type of the deviation propagation, and the deviation manifestation type of each node in the case. Extract key features from the current operation and maintenance service deviation propagation trajectory. Key features include the current initial deviation unit type, the current number of deviation propagation nodes, the current deviation propagation dependency type, and the deviation manifestation type of each node. The key features of the current trajectory are compared with the key features of each historical case trajectory in a contextualized manner, with a focus on whether the initial deviation unit type is consistent, whether the dependency relationship type of deviation propagation is the same, and whether the deviation manifestation category of each node matches. The historical case with the most key feature matching items is selected and marked as a candidate matching case. If there are multiple candidate matching cases, the deviation transmission order between the current trajectory and the candidate matching case trajectory is compared to see if they are consistent and if the time interval features of the deviations at each node are similar. Identify the historical case that best matches the current trajectory, and extract the operation and maintenance service problem type corresponding to the historical case, the description of the development characteristics of the problem in the historical case, and the record of the final impact scope of the historical case. Analyze the correlation between the problem development characteristics in the matched cases and the deviation transmission trend in the current trajectory, and determine whether the current deviation transmission may show the same development direction as the matched cases; Record the problem type of the matched case, the development characteristics of the problem in the matched case, the final impact scope of the matched case, and the correlation analysis conclusion to form a scenario mapping result. The scenario mapping result includes the matched case identifier, the problem type corresponding to the matched case, the development characteristics of the problem in the matched case, the final impact scope of the matched case, and the correlation analysis conclusion. 2.The big data analysis based operation and maintenance technical service problem prediction method according to claim 1, characterized in that, The process involves identifying the pre-dependencies and collaborative dependencies of each unit in the entire operation and maintenance service process. Combined with historical operation and maintenance service big data statistics on the frequency and continuity of interactions between units, an operation and maintenance service stage dependency model containing dependency relationship types is generated, including: Obtain the full process record of operation and maintenance services, which includes the record of operation and maintenance service request initiation, service resource allocation, service execution, service result verification, service delivery, and service follow-up support. The entire process is divided into stages and units according to the order of execution of operation and maintenance services. The process is divided into request receiving unit, resource allocation unit, operation execution unit, result verification unit, delivery confirmation unit, and support response unit. Each stage and unit includes a unit function description, information on related units required for unit execution, and a description of unit output content. Analyze the prerequisite dependencies between each stage unit, identify the stage unit that can only start after other stage units have completed, mark the dependency as a prerequisite dependency, and record the stage unit that provides prerequisite support and the stage unit that receives support in each prerequisite dependency; Analyze the collaborative dependencies between the various components, identify the components that need to be executed simultaneously with other components to complete the service task, mark the dependency as a collaborative dependency, and record all components that are executed together in each collaborative dependency. Collect interaction records of each unit in the historical operation and maintenance service process, and count the interaction frequency of each unit under each dependency based on the interaction data of multiple historical service cycles. Record the dependency relationships that have interaction in multiple consecutive service cycles and mark the dependency relationship as a continuous interaction dependency relationship. The interaction interruption status of each link unit under each dependency relationship is counted. Dependencies that have not been interrupted or whose number of interruptions is less than the preset interruption threshold in the history of interaction are selected and marked as stable interaction dependencies. By integrating prior dependencies and collaborative dependencies, and combining the labeling results of continuous interaction dependencies and stable interaction dependencies, a dependency framework for the initial operation and maintenance service is constructed. Remove dependencies in the initial framework that have an interaction frequency lower than a preset interaction frequency threshold and are not continuous interactions, and retain continuous and stable dependencies and their corresponding link units. For each retained dependency, label the dependency type, including prerequisite dependencies and collaborative dependencies, and supplement with a description of typical interaction scenarios of the dependency in multiple historical service cycles; Based on the labeled dependencies, process units, and typical interaction scenario descriptions, an operation and maintenance service process dependency model structure is generated. The accuracy of the operation and maintenance service process dependency model structure is verified through historical operation and maintenance service big data. It includes complete information of all process units, dependency types between units, and typical interaction scenario descriptions of dependencies.

3. The method for predicting operational and maintenance technical service problems based on big data analysis according to claim 1, characterized in that, The process involves tracking the operational status of each unit in the current operation and maintenance service in real time based on the operation and maintenance service link dependency model. Based on historical normal operation status data within the operation and maintenance service link dependency model structure, it identifies link units whose operational status differs from historical normal operation status, marks these link units as initial deviation units, and records the deviation manifestations of the initial deviation units, including: Extract all process unit information and historical normal operation status descriptions of each process unit from the operation and maintenance service process dependency model structure. The historical normal operation status descriptions include the process unit execution time range, the process unit output content integrity requirements, and the interaction rhythm between the process unit and related units. The system collects operational data from each stage of the current operation and maintenance service in real time. The operational data includes the start time, end time, output details, interaction time points and content with related units of the current stage. Compare the execution time of the current stage unit with the range of execution times of stage units in the historical normal operation state, where the execution time is the difference between the end time and start time of the current stage unit; Compare the output details of the current stage unit with the completeness requirements of the output content of the stage unit in the historical normal operation state, and record the missing output content items; By comparing the interaction time and content between the current link unit and related units with the interaction rhythm between the link unit and related units in the historical normal operation state, the situation of delayed interaction time or incomplete interaction content can be identified. By integrating differences in execution time, missing output content, and abnormal interaction rhythm, it can be determined whether the current operational status of the unit differs from its historical normal operational status. The components with discrepancies in their operational status are designated as initial deviation units, and their identification information is recorded. Record the deviation manifestations of the initial deviation unit, including the specific circumstances of the initial deviation unit's execution time exceeding the range, the specific items of the initial deviation unit's output content being missing, the specific time points and abnormal content of the initial deviation unit's interaction rhythm being abnormal, and all records are associated with the comparison difference between the current running data and historical data. The deviation manifestations of the initial deviation units are classified and organized, and the deviation items are divided into categories such as execution deviation, output deviation, and interaction deviation. From this, the identification information of the initial deviation units and the corresponding classification deviation manifestation records are output.

4. The method for predicting operational and maintenance technical service problems based on big data analysis according to claim 1, characterized in that, Starting from the initial deviation unit, and based on the dependency relationship types in the operation and maintenance service link dependency model structure, the propagation path of the deviation between each link unit is traced, and the deviation manifestation form and deviation propagation order of each link unit affected by the deviation are recorded to form the operation and maintenance service deviation propagation trajectory, including: Obtain the identification information of the initial deviation unit, the classification deviation performance record of the initial deviation unit, and the dependency relationship type in the operation and maintenance service link dependency model structure; Based on the dependency relationship type, the associated link units that have a direct dependency relationship with the initial deviation unit are identified. The associated link units include the subsequent link units that serve as the preceding support units of the initial deviation unit, and the parallel link units that have a cooperative dependency with the initial deviation unit. Real-time acquisition of operational data of the associated link unit after the initial deviation unit has deviated. The operational data includes the execution time of the associated link unit, the output content of the associated link unit, and the interaction between the associated link unit and other units. By comparing the current operating data of the related link units with their historical normal operating status, we can identify whether any new deviations have occurred in the related link units. Record the identification information of the associated link unit where a new deviation occurs, mark the associated link unit as a deviation transmission node, and record the deviation manifestation of the deviation transmission node to determine the correlation between the deviation manifestation of the deviation transmission node and the deviation manifestation of the initial deviation unit. Starting from the deviation propagation node, repeat the above steps to identify the next-level related link unit that has a direct dependency on the deviation propagation node, collect the operation data of the next-level related link unit and identify the deviation of the next-level related link unit, and repeat the process until no new deviation propagation node appears. Record the occurrence time of the initial deviation unit and each deviation transmission node, as well as the transmission relationship between the initial deviation unit and each deviation transmission node, in the order in which the deviations occur, to form a deviation transmission sequence chain. Each deviation propagation node is labeled with the corresponding dependency type, which includes pre-dependent propagation and collaborative dependency propagation. The dependency triggering conditions for deviation propagation are described, and the triggering conditions are determined based on the dependency triggering data in the dependency model structure of the operation and maintenance service links. It integrates initial deviation unit information, deviation transmission node information, deviation transmission sequence chain, deviation transmission dependent triggering conditions, and deviation manifestation forms of initial deviation unit and each deviation transmission node; Based on the time sequence and dependency logic of deviation propagation, a visualized deviation propagation trajectory for operation and maintenance services is constructed. This trajectory includes node identifiers, node deviation behavior, propagation relationships between nodes, and propagation triggering conditions. The resulting deviation propagation trajectory is used to present the propagation process of deviations from the initial deviation unit to each deviation propagation node, as well as the deviation characteristics between the initial deviation unit and each deviation propagation node.

5. The method for predicting operational and maintenance technical service problems based on big data analysis according to claim 1, characterized in that, Based on the scenario mapping results and historical big data on operation and maintenance issues, the system determines the types of problems that may occur in the current operation and maintenance technical service, the scope of affected links and units, and the direction of problem development. It then generates a final prediction result for the operation and maintenance technical service issues and associates it with a corresponding set of intervention solutions, including: Analyze the matching case problem types in the scenario mapping results, determine the possible problem types of the current operation and maintenance technical service, describe the typical manifestations of the problem type and how the problem type affects the service; Based on the case impact range record in the scenario mapping results, combined with the number of link units that have already deviated and the deviation transmission trend in the current operation and maintenance service deviation transmission trajectory, the scope of the currently affected link units is determined. The scope of affected link units includes link units that have already deviated and link units that may be affected by subsequent transmission. Based on the problem development characteristics description and correlation analysis conclusions in the scenario mapping results, the possible development direction of the current problem is predicted. The possible development direction of the current problem includes whether the deviation will continue to be transmitted to more links and units, whether the manifestation of the deviation will be aggravated, and whether the current problem will cause other related service anomalies. By integrating the types of problems that may occur in the current operation and maintenance technical services, the scope of the currently affected links and units, and the possible development direction of the current problems, a preliminary prediction result of operation and maintenance technical service problems is formed; The system extracts the corresponding operation and maintenance technical service problem intervention plan from the historical operation and maintenance technical service problem case library. The operation and maintenance technical service problem intervention plan includes the handling measures for the initial deviation unit of the matching case, the control measures for the deviation transmission node in the matching case, and the blocking measures to prevent the deviation from further transmission in the matching case. Based on the actual configuration of the current operation and maintenance service and the specific situation of the current deviation transmission, the intervention plan for the operation and maintenance technical service problem corresponding to the matching case is adjusted, the measures in the operation and maintenance technical service problem intervention plan that are not applicable to the current service are replaced, and the handling steps for the current specific deviation manifestation are added. The adjusted operation and maintenance technical service problem intervention plans are sorted according to implementation priority. The implementation priority is determined by the effectiveness of the measures in preventing the propagation of deviations, the time required for implementation, and the degree of impact of the measures on normal services. A set of operation and maintenance technical service problem intervention plans containing implementation priorities is generated. Each operation and maintenance technical service problem intervention plan includes the measure name, implementation steps, implementing entity, resources required for implementation, and a description of expected results. The preliminary operational technical service problem prediction results are integrated with the set of operational technical service problem intervention plans to form the final operational technical service problem prediction results. The final operational technical service problem prediction results include the types of problems that may occur in the current operational technical service, the scope of the currently affected links and units, the possible development direction of the current problems, the set of operational technical service problem intervention plans and the implementation priority of the plans.

6. The method for predicting operational and maintenance technical service problems based on big data analysis according to claim 2, characterized in that, The analysis identifies prerequisite dependencies between various stage units, identifies stage units that require completion of other stage units to start, marks these dependencies as prerequisite dependencies, and records the stage units providing prerequisite support and the stage units receiving support in each prerequisite dependency, including: Examine the associated unit information required for the execution of each stage unit one by one, filter out the stage units that require the completion of other stage units before they can be started, mark the stage unit as the unit to be analyzed and accepted, and filter according to the start condition description of the operation and maintenance service stage unit; For each receiving unit to be analyzed, review the historical startup conditions of the receiving unit in the full process record of operation and maintenance services, determine the other link units that must be completed before starting the receiving unit to be analyzed, and mark the other link units as the providing unit to be analyzed; Review the historical interaction records between the providing unit and the receiving unit to be analyzed to confirm whether the providing unit to be analyzed completed the execution before the receiving unit to be analyzed started in all historical service cycles. If the historical records show that the unit to be analyzed always completes execution before the unit to be analyzed starts, and the start instruction of the unit to be analyzed contains the completion confirmation information of the unit to be analyzed, then it is determined that there is a prerequisite dependency between the unit to be analyzed and the unit to be analyzed. Record the link units that provide and receive support in the preceding dependency relationships, and mark the dependency direction of the link units that provide and receive support. For each defined prerequisite dependency, analyze historical service records and record multiple historical intervals from the completion of the supporting link unit to the initiation of the supporting link unit; based on the recorded multiple historical intervals, calculate or describe the interval time characteristics. Check if there is a situation where a receiving link unit depends on multiple link units that provide prior support. If so, record the identifiers of all link units that provide prior support and the differences in the support content provided by each link unit that provides prior support to the receiving link unit. Organize all preceding dependency records to form a preceding dependency list. The preceding dependency list includes the identifier of the link unit that provides preceding support, the identifier of the link unit that receives support, the dependency direction, the average interval time characteristics, and the differences in the supported content.

7. The method for predicting operational and maintenance technical service problems based on big data analysis according to claim 2, characterized in that, The analysis identifies the collaborative dependencies between various components, identifies components that require simultaneous execution with other components to complete the service task, marks these dependencies as collaborative dependencies, and records all components that execute jointly in each collaborative dependency, including: Review the execution requirements of each service task in the entire operation and maintenance service process record, filter out service tasks that require multiple stage units to be executed simultaneously, mark the service task as a collaborative task, and filter according to the execution requirement description in the service task specification; For each collaborative task, extract all the components involved in the execution of the collaborative task and mark all the components as the collaborative unit group to be analyzed; Review the historical execution time records of each unit in the collaborative unit group to be analyzed, and confirm whether the execution start time of each unit in the collaborative unit group to be analyzed is within the preset synchronization threshold and whether the execution end time of each unit in the collaborative unit group to be analyzed is within the preset synchronization threshold during the execution of all historical collaborative tasks. If the historical records show that the start time of each unit in the collaborative unit group to be analyzed is within the preset synchronization threshold, the end time of each unit in the collaborative unit group to be analyzed is within the preset synchronization threshold, and the output of each unit in the collaborative unit group to be analyzed needs to be integrated before the collaborative task can be completed, then it is determined that there is a collaborative dependency relationship between the units in the collaborative unit group to be analyzed. Record the identifiers of all components jointly executed in a collaborative dependency relationship, and label the name of the collaborative task corresponding to the collaborative dependency relationship; For each collaborative dependency, analyze historical service records and record the start and end times of each component within the collaborative dependency in multiple historical executions; based on the recorded start and end times of multiple historical executions, calculate or describe its execution synchronization characteristics; Check if there is a situation where a single unit participates in multiple collaborative dependencies. If so, record the names of all collaborative tasks that the unit participates in and the functional differences of the unit in different collaborative tasks. Organize all collaborative dependency records to form a collaborative dependency list. The collaborative dependency list includes a list of unit identifiers for each link in the collaborative dependency, the corresponding collaborative task name, execution synchronization characteristics, and functional differences of the link units.

8. The method for predicting operational and maintenance technical service problems based on big data analysis according to claim 3, characterized in that, The comparison of the output details of the current stage unit with the completeness requirements of the output content of the stage unit in the historical normal operation state is used to record the missing output content items, including: Extract the completeness requirements of the output content of the current stage unit from the historical normal operation status description, and describe the name of the content items that the current stage unit should output, the content format requirements of each content item, and the necessary data items of each content item; Obtain the output details of the current stage unit, and compare them one by one according to the content item names in the completeness requirements to confirm whether the output of the current stage unit contains all the required content items; If the output of the current stage unit is missing one or more content items, record the name of the missing content item, and also record the name of the content item already included in the output of the current stage unit; For the content items already included in the output of the current unit, further compare the content format of each content item with the content format requirements in the completeness requirements. If the formats do not match, record the name of the content item with the mismatch and the specific mismatch. For content items that match the format, check whether the necessary data items within the content item are complete. If any necessary data items are missing, record the name of the content item and the name of the missing necessary data item. Compile the names of missing content items, the names of content items with mismatched formats and the mismatches, and the names of missing necessary data items and the missing data items into an output deviation record. If the output of the current stage unit fully meets the integrity requirements, record that the output content of the current stage unit has no deviation, and output deviation record. The output deviation record includes the current stage unit identifier, the output deviation type and the specific deviation content. The output deviation type includes missing items, format mismatch and missing data items.

9. A system for predicting operational and maintenance technical service problems based on big data analysis, characterized in that, The invention includes a processor and a computer-readable storage medium storing machine-executable instructions, which, when executed by the processor, implement the big data analysis-based operation and maintenance technical service problem prediction method as described in any one of claims 1-8.

Images