Operation and maintenance process optimization method and system applied to operation and maintenance technology service

Abstract

The invention provides an operation and maintenance process optimization method and system applied to operation and maintenance technical services, and relates to the technical field of operation and maintenance technical services. Firstly, scene features, execution association relationships and resource demand attributes in an operation and maintenance task demand document are collected to form an operation and maintenance task information set; and meanwhile, collecting real-time operation state data and capability dimension information of the service nodes to construct dynamic capability portraits of the service nodes. And then, based on the scene features and the scene adaptation capability, executing bidirectional adaptation processing to generate a preliminary adaptation result, identifying and resolving resource supply conflicts in the preliminary adaptation result, and generating a final adaptation scheme. And finally, integrating an operation and maintenance task execution sequence and service node resource supply arrangement according to the final adaptation scheme to form an operation and maintenance process optimization scheme, and dynamically evolving and calibrating the scheme content according to real-time data in the execution process. The operation and maintenance task can be accurately allocated, the flexibility and adaptability of the operation and maintenance process are improved, and the operation and maintenance service quality and efficiency are improved.

Description

Technical Field

[0001] This application relates to the field of operation and maintenance technical service technology, and more specifically, to a method and system for optimizing operation and maintenance processes applied to operation and maintenance technical services. Background Technology

[0002] In the current field of operations and maintenance (O&M) technical services, as enterprise businesses continue to expand and become more complex, the number and types of O&M tasks are increasing, and the scenarios involved are becoming more diverse. Traditional O&M process management methods often adopt a static, fixed allocation model to arrange O&M tasks and service nodes. Specifically, after receiving an O&M task request, O&M personnel usually manually allocate tasks to various service nodes based on experience, and the capability assessment of service nodes is only based on their initial configuration information, lacking dynamic consideration of the real-time status of service nodes.

[0003] The aforementioned traditional methods have several drawbacks. Firstly, because they don't fully consider the business scenario category of the maintenance tasks and the collaborative execution requirements between tasks, task allocation may be unreasonable, leading to overload on some service nodes while others remain idle, hindering efficient resource utilization. Secondly, inaccurate assessment of service node capabilities prevents timely adjustments to task allocation based on real-time service node status. When a service node malfunctions or experiences performance degradation, it cannot promptly transfer related maintenance tasks to other suitable service nodes, impacting the efficiency and quality of maintenance task execution, and potentially causing business interruptions and significant losses to the enterprise. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method and system for optimizing operation and maintenance processes in operation and maintenance technical services.

[0005] According to a first aspect of this application, a method for optimizing operation and maintenance processes applied to operation and maintenance technical services is provided, the method comprising: Collect the scenario characteristics, execution relationships, and resource requirement attributes of each operation and maintenance task in the operation and maintenance task requirement document to form an operation and maintenance task information set. The scenario characteristics reflect the business scenario category to which the operation and maintenance task belongs, and the execution relationships reflect the collaborative execution requirements between operation and maintenance tasks. Collect real-time operational status data and capability dimension information of each service node, dynamically update capability dimension information based on real-time operational status data, and construct a dynamic capability profile of the service node. The capability dimension information includes the types of resources that the service node can provide, the resource supply mode, and the scenario adaptation capability. Based on the scenario features in the operation and maintenance task information set and the scenario adaptation capabilities in the dynamic capability profile of the service node, a two-way adaptation process is performed between the operation and maintenance task and the service node to generate a preliminary adaptation result between the operation and maintenance task and the service node. Identify the adaptation combinations with resource supply conflicts in the preliminary adaptation results, combine the execution relationship of operation and maintenance tasks with the resource supply mode of service nodes, resolve the resource supply conflicts, and generate the final adaptation scheme of operation and maintenance tasks and service nodes. Based on the final adaptation scheme, the execution sequence of operation and maintenance tasks and the resource supply arrangement of service nodes are integrated to form an operation and maintenance process optimization scheme. During the execution of the operation and maintenance process, the content of the operation and maintenance process optimization scheme is dynamically evolved and calibrated based on the changes in the real-time running status data of service nodes and the execution progress of operation and maintenance tasks.

[0006] According to a second aspect of this application, an operation and maintenance process optimization system for operation and maintenance technical services is provided. The operation and maintenance process optimization system for operation and maintenance technical services 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 operation and maintenance process optimization system for operation and maintenance technical services implements the aforementioned operation and maintenance process optimization method for operation and maintenance technical services.

[0007] According to a third aspect of this application, a computer-readable storage medium is provided, wherein computer-executable instructions are stored therein, and when the computer-executable instructions are executed, the aforementioned method for optimizing operation and maintenance processes applied to operation and maintenance technical services is implemented.

[0008] Based on any of the above aspects, the technical effect of this application is as follows: By collecting scenario features, execution relationships, and resource requirement attributes from operation and maintenance (O&M) task requirement documents to form an O&M task information set, and collecting real-time operational status data and capability dimension information of service nodes to construct a dynamic capability profile of service nodes, a two-way adaptation process is performed to generate preliminary adaptation results. Further, resource supply conflicts are identified, and conflicts are resolved by combining execution relationships and resource supply patterns to generate a final adaptation solution. This allows O&M tasks to be assigned to the most suitable service nodes, avoiding resource waste and low task execution efficiency caused by unreasonable task allocation. Based on the final adaptation solution, the O&M task execution sequence and service node resource supply arrangements are integrated to form an O&M process optimization plan. During execution, the plan content is dynamically evolved and calibrated based on real-time data, enabling the O&M process to be adjusted in real time according to actual conditions. This effectively improves the flexibility and adaptability of the O&M process, significantly enhances the quality and efficiency of O&M services, and reduces the risk of business interruption. Attached Figure Description

[0009] Figure 1 This paper illustrates a flowchart of an operation and maintenance process optimization method for operation and maintenance technical services provided in an embodiment of this application. Figure 2This paper illustrates a schematic diagram of the component structure of an operation and maintenance process optimization system for operation and maintenance technical services provided in an embodiment of this application. Detailed Implementation

[0010] Figure 1 This application provides a flowchart illustrating a method and system for optimizing operation and maintenance processes in technical services. The detailed steps include: Step S110: Collect the scenario characteristics, execution relationships, and resource requirement attributes of each operation and maintenance task in the operation and maintenance task requirement document to form an operation and maintenance task information set. The scenario characteristics reflect the business scenario category to which the operation and maintenance task belongs, and the execution relationships reflect the collaborative execution requirements between operation and maintenance tasks.

[0011] In the intelligent control and maintenance scenario of chain stores using smart operation and maintenance services, the technology center accesses maintenance task requirement documents through a pre-set document parsing tool. These documents originate from the chain stores' repair reporting platform, regular maintenance work orders, and internal task assignment records of the technology center. Taking a regional store cluster of a chain retail brand as an example, this cluster includes 20 stores, and the daily maintenance task requirement documents cover various types, such as air conditioning system failures, black screens on information display screens, and offline IoT sensors. Through batch processing of these documents, key information for each maintenance task is extracted.

[0012] Step S111: Read the text description content in the operation and maintenance task requirement document, and extract information describing the business background, execution objectives and involved business systems of the operation and maintenance task from the text description content. The information describing the business background, execution objectives and involved business systems of the operation and maintenance task together constitute the scenario characteristics of the operation and maintenance task. The text description content covers the introduction of the business background of the operation and maintenance task, the description of the execution objectives and the list of involved business systems.

[0013] The text stream of the operation and maintenance task requirement document is loaded using a document parsing tool. A BERT-based pre-trained model is used for word segmentation, filtering out stop words and retaining core semantic units. Taking the "store air conditioning system cooling failure" operation and maintenance task as an example, the business background description in the text is "During the summer promotion, the store's air conditioning system experienced an abnormal increase in outlet temperature after running continuously for 12 hours." The execution objective is "to restore the rated cooling capacity of the air conditioning system within 4 hours, ensuring the store temperature is maintained at 26℃±1℃." The list of involved business systems includes "Central Air Conditioning Control System V3.0, Store Power Monitoring System, and Environmental Temperature and Humidity Sensor Network." Named entity recognition technology is used to locate "summer promotion," "12 hours of operation," and "abnormal temperature increase" in the business background; "4-hour restoration," "rated cooling capacity," and "26℃±1℃" in the execution objective; and the specific system names in the business system list. This information is then concatenated into a scenario feature vector for the operation and maintenance task, with the vector dimensions corresponding to the three sub-dimensions of business background, execution objective, and business systems.

[0014] Step S112: Identify the collaborative execution requirements between operation and maintenance tasks from the text description. The collaborative execution requirements include the sequential execution requirements and synchronous execution requirements between operation and maintenance tasks. The sequential execution requirements and synchronous execution requirements between operation and maintenance tasks constitute the execution relationship between operation and maintenance tasks.

[0015] Dependency parsing techniques are used to extract associations from the text descriptions of multi-task requirement documents, constructing a directed graph model between tasks. For example, a store submits two maintenance tasks simultaneously: "Information publishing system software upgrade" (Task A) and "Network bandwidth expansion" (Task B). The text description states that "the information publishing system upgrade must be performed after the network bandwidth expansion is completed to avoid interruption of upgrade packet transmission." By identifying the time adverbial clause "must be performed after… is completed," it is determined that Task B and Task A have a sequential execution requirement, and a directed edge from Task B to Task A is added to the directed graph. In another scenario, the text descriptions of "Air conditioner filter replacement" (Task C) and "Air purifier filter replacement" (Task D) mention that "both maintenance tasks must be performed simultaneously to reduce store downtime." The semantic marker "perform simultaneously" indicates that Task C and Task D have a synchronous execution requirement, and a bidirectional dashed line is added to connect the two task nodes in the directed graph.

[0016] Step S113: Extract the types of resources, usage methods, and usage cycles required during the execution of the operation and maintenance task from the text description. The usage methods include the resource invocation and release methods, and the usage cycles include the start and end times of resource usage. The types of resources, usage methods, and usage cycles required during the execution of the operation and maintenance task constitute the resource requirement attributes of the operation and maintenance task.

[0017] An entity relation extraction model is used to extract resource-related triples from text descriptions. Taking the "IoT gateway firmware update" maintenance task as an example, the text description states that it "needs to call the regional cloud platform's OTA upgrade server, push the firmware image via HTTPS protocol, and release the server connection resources after the upgrade is completed. The usage period is non-business hours 22:00-02:00." The model identifies the resource type as "OTA upgrade server," the resource usage method as "HTTPS protocol push," the release method as "release connection after upgrade," and the resource usage period as "22:00" and "02:00." The above information is structured and stored as a resource requirement attribute dictionary, where key-value pairs correspond to resource type, calling method, release method, start time, and end time, respectively.

[0018] Step S114: For each operation and maintenance task, associate and bind the scenario characteristics, execution relationship and resource requirement attributes of the operation and maintenance task. By establishing the corresponding mapping relationship between the three, the scenario characteristics, execution relationship and resource requirement attributes of each operation and maintenance task form a complete correspondence.

[0019] Each maintenance task is assigned a unique TaskID. A relational database table is created with the TaskID as the primary key. Fields in the table correspond to the scenario feature vector, the adjacency matrix of the execution association graph, and a dictionary of resource requirement attributes. For example, for a maintenance task with TaskID TK-20230601-001, its scenario feature vector is stored as a JSON string, its execution associations are linked to other TaskIDs via row and column indices in the adjacency matrix, and its resource requirement attributes are stored as key-value pairs. A database transaction mechanism ensures the atomic binding of these three pieces of information, preventing data inconsistencies.

[0020] Step S115: Organize the associated binding information of all operation and maintenance tasks according to the submission order of the operation and maintenance tasks, and organize the associated binding information using a structured data format. The structured data format includes field definitions and data type descriptions to form a structured set of operation and maintenance task information. Each entry in the set of operation and maintenance task information includes the operation and maintenance task identifier and the scenario characteristics, execution association and resource requirement attributes corresponding to the operation and maintenance task.

[0021] The submission order of operation and maintenance tasks is determined by the timestamp field (SubmitTime) in the document, and the system sorts tasks in ascending order of SubmitTime. A structured data format is defined using the Protocol Buffers protocol, with fields including TaskID (string type), SceneFeature (nested message type, containing three sub-fields: BusinessBackground, ExecutionGoal, and InvolvedSystems), ExecutionRelation (message type, containing two repeating fields: PredecessorTasks and SynchronousTasks), and ResourceDemand (message type, containing three sub-fields: ResourceType, UsageMode, and TimePeriod). The data type specification clearly defines the value range and constraints for each field; for example, the ResourceType field is limited to an enumeration type, including options such as "computational resources," "storage resources," and "network resources." The sorted association binding information is serialized into a Protocol Buffers binary stream, forming an operation and maintenance task information collection file. The file header contains the total number of tasks and a checksum.

[0022] Step S116: Classify the scenario features in the operation and maintenance task information set, and group operation and maintenance tasks with similar scenario features into the same scenario group.

[0023] Clustering algorithms are used to group scene feature vectors in the operation and maintenance task information set. First, the scene feature vectors are standardized to eliminate differences in the dimensions of different sub-dimensions. Then, the DBSCAN algorithm is used, with cosine similarity as the distance metric, and a neighborhood radius of 0.6 and a minimum sample size of 3 are set. For example, scene feature vectors containing keywords such as "air conditioning system," "refrigeration failure," and "temperature control" are clustered into one group, while vectors containing "information publishing," "screen display," and "content update" are clustered into another. Each clustering result corresponds to a scene group, and tasks within the group share similar business scene features.

[0024] Step S1161: Extract the scenario features of all operation and maintenance tasks in the operation and maintenance task information set. Extract the business background, execution target and related business system information from each scenario feature. Decompose each scenario feature into business background keywords, execution target keywords and related business system keywords. Keyword extraction adopts a combination of word segmentation algorithm and keyword filtering rules. Keyword filtering rules include keyword length limit and semantic relevance requirements.

[0025] For each scene feature vector, Chinese word segmentation is performed using the Jieba word segmentation algorithm to obtain the original word sequence. The TF-IDF weight calculation method is used to filter the word sequence, retaining the top 20 words by weight as candidate keywords. In the keyword selection rules, the length limit is set to 2-8 Chinese characters, and semantic relevance requires calculating the cosine similarity between the candidate keywords and the core theme of the scene feature using the Word2Vec model, retaining words with a similarity greater than 0.7. Taking the scene feature "store air conditioning system cooling failure" as an example, business background keywords include "summer promotion," "overload," and "intermittent shutdown," execution target keywords include "continuous stable operation" and "temperature control," and keywords related to the business system include "air conditioning main control system" and "temperature sensor."

[0026] Step S1162: Calculate the similarity between business background keywords of different operation and maintenance tasks, and group operation and maintenance tasks containing the same or similar business background keywords into the same business background subset. Similar keywords are determined based on the association relationship of industry common terms, which is established through an industry terminology dictionary.

[0027] A dictionary of industry terms for operations and maintenance (O&M) is constructed, including synonym pairs such as "cooling failure" and "abnormal cooling," and "overload" and "overload limit." Business background keywords are vectorized using Word2Vec, and the cosine similarity of keyword vectors for different tasks is calculated. Keywords with a similarity greater than 0.85 are considered similar. Tasks containing the same or similar keywords are grouped into the same business background subset; for example, tasks related to "air conditioning cooling failure" and "air conditioning cooling abnormality" are both grouped into the "air conditioning temperature abnormality" business background subset.

[0028] Step S1163: In each business background subset, the execution target keywords are compared. The comparison process adopts the keyword semantic matching method. Operation and maintenance tasks with the same target direction described by the execution target keywords are grouped into the same target direction subset. The consistency of target direction is determined by the semantic similarity of the execution target keywords.

[0029] In the "abnormal air conditioning temperature" business context subset, the target keywords might include "restore cooling," "stable operation," and "temperature adjustment." The semantic embedding vectors of these target keywords are calculated using the BERT model, and cosine similarity is used for matching, with a similarity threshold set to 0.8. Tasks with semantic similarity higher than the threshold are grouped into the same target subset. For example, the tasks "restore cooling function" and "maintain stable temperature" are grouped into the same subset because their target direction is "ensuring the normal operation of the air conditioning system."

[0030] Step S1164: In each target direction subset, perform overlap analysis on the keywords of the business systems involved. The overlap analysis is calculated by the number of overlaps of the keywords of the business systems involved. The maintenance tasks with the number of overlaps of the keywords of the business systems involved reach the set threshold are classified into the same business system subset. The standard for the number of overlaps to reach the set threshold is determined by preset rules.

[0031] A threshold of 2 is set for the number of overlapping keywords related to business systems. This means that if two tasks share two or more identical keywords related to business systems, they are considered related. For example, if Task A involves "air conditioning main control system" and "temperature sensor network," and Task B involves "air conditioning main control system," "temperature sensor network," and "power monitoring subsystem," the two tasks share 2 overlapping keywords, reaching the threshold and are therefore classified as belonging to the same subset of business systems. The preset rules are defined through a configuration file, and the threshold can be dynamically adjusted based on the business scenario.

[0032] Step S1165: Determine the subset of business systems as the final scenario group, and assign a unique scenario group identifier to each scenario group. The scenario group identifier includes a combination of business background, target direction and abbreviation of business system.

[0033] After manual review and confirmation, a subset of business systems is officially designated as a scenario group. The scenario group identifier follows the format "Abbreviation of Business Background - Abbreviation of Target Direction - Abbreviation of Business System," for example, "Air Conditioning Temperature Anomaly - Operational Support - Air Conditioning System." Here, "Air Conditioning Temperature Anomaly" is the abbreviation for the business background of "abnormal air conditioning temperature," "Operational Support" is the abbreviation for the target direction, and "Air Conditioning System" is the abbreviation for the core business system. Once generated, the identifier is stored in the scenario group metadata table and associated with the task ID within the group.

[0034] Step S1166: Record the classification criteria for each scenario group. The classification criteria include keyword similarity analysis results, target direction comparison results, and business system overlap analysis results. Keyword similarity analysis results include similarity calculation values ​​and a list of matching keywords. Target direction comparison results include semantic matching results and a list of consistent keywords. Business system overlap analysis results include the number of overlaps and a list of overlapping keywords.

[0035] The classification criteria are stored in JSON format, for example: {"KeywordSimilarity": {"Score": 0.89,"MatchedKeywords": ["Air conditioning system", "Refrigeration failure"]},"GoalAlignment": {"Result": "Consistent", "ConsistentKeywords": ["Stable operation", "Temperature control"]},"SystemOverlap": {"Count": 2,"OverlapKeywords": ["Air conditioning main control system", "Temperature sensor network"]}}. The above data is stored in the scenario group audit log for subsequent traceability and optimization of classification rules.

[0036] Step S1167: In the operation and maintenance task information set, add a corresponding scenario group identifier for each operation and maintenance task. By establishing a mapping relationship between operation and maintenance task identifiers and scenario group identifiers, if any operation and maintenance task meets the classification conditions of multiple scenario groups at the same time, determine the scenario group to which the operation and maintenance task belongs by priority sorting. The priority sorting is determined according to the business importance of the scenario group.

[0037] Add a SceneGroupID field to the structured data of the operation and maintenance task information set. Populate the value of this field by querying the TaskID with the scene group metadata table. When a task meets the conditions of multiple scene groups at the same time (such as involving both air conditioning system and power system failures), prioritize them according to the importance of the business. The "core business system failure repair" scene group has a higher priority than the "auxiliary system maintenance" scene group to ensure that the task belongs to a unique scene group.

[0038] Step S1168: Summarize the scenario groups to form a scenario group list. The scenario group list includes a scenario group identifier, a list of maintenance tasks within the group, and a classification basis. The list of maintenance tasks within the group includes maintenance task identifiers and maintenance task names. The classification basis includes complete analysis process and result data.

[0039] The scenario group list is generated in Excel spreadsheet format, including columns such as SceneGroupID, GroupName, TaskList, and ClassificationBasis. The TaskList column separates TaskIDs with commas, and the ClassificationBasis column embeds JSON-formatted classification criteria data. The list is stored on a shared file server for technical center personnel to access.

[0040] Step S1169: Store the scene group list to the target data file, and at the same time establish the index relationship between the scene group identifier and the target data file.

[0041] The target data files utilize the Apache Parquet columnar storage format to improve query efficiency. An index table mapping SceneGroupIDs to file paths is created and stored in a relational database. This index table includes fields such as SceneGroupID, FilePath, FileSize, and LastModified. The index table allows for quick location of inventory data for specific scene groups, supporting fast retrieval by scene group dimension.

[0042] Step S117: Add a scenario group identifier to the operation and maintenance task information set. The scenario group identifier is generated in the form of a combination of letters and numbers. Each operation and maintenance task entry corresponds to a unique scenario group identifier. The association between operation and maintenance tasks and scenario groups is established through the scenario group identifier.

[0043] The rule for generating scene group identifiers is as follows: the first two letters represent the business background category (e.g., "KT" represents the air conditioning temperature category), the middle two digits represent the target direction number (e.g., "01" represents operation support), and the last three letters represent the business system code (e.g., "ACS" represents the air conditioning system). For example, "KT01ACS" uniquely identifies the scene group "Air Conditioning Temperature Anomaly - Operation Support - Air Conditioning System". A SceneGroupID field is added to the ProtocolBuffers structure of the operation and maintenance task information set, and the generated identifier is filled in. This field establishes a one-to-many association between tasks and scene groups (one scene group contains multiple tasks).

[0044] Step S118: Store the complete set of operation and maintenance task information into the target database. The target database adopts a partitioned storage strategy and partitions the set of operation and maintenance task information according to the scenario group identifier.

[0045] The target database is PostgreSQL, employing table partitioning technology with a range-partitioned table created using SceneGroupID as the partition key. Each scene group corresponds to one partition, with the partition name formatted as "task_info_p_scene-group identifier". For example, data for scene group "KT01ACS" is stored in the "task_info_p_KT01ACS" partition. Partitioned storage significantly improves the efficiency of querying by scene group and facilitates data lifecycle management, such as archiving historical scene group data.

[0046] Step S120: Collect real-time operating status data and capability dimension information of each service node, dynamically update the capability dimension information based on the real-time operating status data, and construct a dynamic capability profile of the service node. The capability dimension information includes the types of resources that the service node can provide, the resource supply mode, and the scenario adaptation capability.

[0047] Service nodes include cloud servers in the technology center, edge computing nodes, and mobile terminals carried by on-site engineers. Real-time operational status data is collected through agents deployed on each node and reported periodically using the MQTT protocol (reporting interval set to 30 seconds). Capability dimension information is initially extracted from the service node's configuration documents (such as hardware specifications and software function lists) and stored in a capability dimension metadata database. The dynamic update module consumes the real-time operational status data stream through a stream processing engine (such as Flink), calculates changes in resource carrying capacity and scenario adaptability in real time, updates the capability dimension information, and ultimately constructs a dynamic capability profile including timestamps.

[0048] Step S121: Obtain real-time operating status data of each service node, and extract initial capability dimension information from the configuration document of the service node. The configuration document includes the hardware configuration specification and software configuration specification of the service node. The initial capability dimension information includes the resource types designed for the service node, the maximum resource capacity, and the types of business scenarios supported. The resource types include processor resources, memory resources, storage resources, and network resources. The maximum resource capacity includes the maximum allocable amount of each type of resource. The types of business scenarios supported include the list of business scenarios that can be processed preset when the service node is designed.

[0049] Real-time operational status data of service nodes is collected through the Agent program, including metrics such as CPU utilization, memory usage, disk I / O speed, and network bandwidth utilization. The hardware configuration manual is in PDF format, with key information extracted using OCR technology, such as "CPU model: Intel Xeon E5-2680v4, number of cores: 14 cores" and "Memory capacity: 128GB DDR4". The software configuration manual is in XML format, parsed to obtain "Operating system: Linux CentOS 7.9" and "Supported services: Docker 20.10.12, Kubernetes 1.21". The resource types in the initial capability dimension information are determined based on the hardware configuration; for example, the resource types of the above nodes include processor resources (14-core CPU), memory resources (128GB), storage resources (2TB SSD), and network resources (10Gbps network card). The maximum resource capacity is the design peak, such as a maximum allocable CPU of 12 cores (with 2 cores reserved for system overhead). Supported business scenario types are extracted from the "Scenario Support List" field in the software configuration manual, such as "Remote diagnostics of air conditioning systems" and "Information publishing content rendering".

[0050] Step S122: Compare the real-time running status data of the service node with the resource carrying capacity description in the initial capability dimension information, compare the real-time resource usage with the maximum resource carrying capacity, and analyze the impact of real-time resource usage on resource carrying capacity. The impact analysis includes the ratio of real-time resource usage to maximum resource carrying capacity and the impact of resource usage exceeding the maximum allocatable amount threshold for a continuous set collection period on the attenuation of resource carrying capacity.

[0051] The resource usage (e.g., CPU utilization 80%) in real-time operational status data is divided by the maximum resource capacity (e.g., 12 cores) to obtain the usage ratio (80%). The data collection period is set to 5 cycles (5 x 30 seconds = 2.5 minutes), and the ratio threshold is set to 90%. When the usage of a certain type of resource (e.g., memory) exceeds 90% of the maximum allocatable amount for 5 consecutive cycles (e.g., 128GB x 90% = 115.2GB), a resource capacity degradation analysis is triggered. The degradation impact is calculated by looking up the resource performance degradation curve (pre-stored in the database, such as the bandwidth degradation coefficient under long-term high memory load). For example, if the threshold is exceeded for 10 consecutive minutes, the degradation coefficient is 0.95, meaning the actual load capacity drops to 115.2GB x 0.95 = 109.44GB.

[0052] Step S123: Adjust the resource carrying capacity related description in the initial capacity dimension information according to the changes in real-time resource usage. When the usage of any type of resource of the service node exceeds the set proportion threshold of the maximum allocable amount of that type of resource for a set number of consecutive collection cycles, update the supply description of that type of resource based on the duration of the continuous exceedance of the threshold for the usage of that type of resource and the degree of decay reflected by the resource performance decay data. The supply description includes the current supplyable resource quantity and the resource supply constraints.

[0053] When resource capacity degradation is triggered, the resource performance degradation data table is queried based on the duration (e.g., 15 minutes) to obtain the degradation level (e.g., 10%). The current available resource quantity = maximum allocable quantity × (1 - degradation level) = 115.2GB × 90% = 103.68GB. Resource supply constraints include "prioritize allocation of short-cycle tasks" and "prohibit new allocations exceeding 80% utilization," etc. These conditions are generated by the rule engine and stored in the Constraints field of the supply description. Update operations are performed through database transactions to ensure data consistency.

[0054] Step S124: Combine the device operating status in the real-time operating status data to correct the scenario adaptation capability description in the capability dimension information. The operating status of hardware components and software modules in the device operating status directly affects the service node's ability to process specific business scenarios. When the device operating status shows that the operating parameters of any scenario-related component deviate from the normal working range set percentage threshold for a continuous set of collection cycles, the adaptation capability description of the corresponding scenario is adjusted based on the magnitude of the deviation of the operating parameters of the scenario-related component from the normal working range set percentage threshold for a continuous set of collection cycles and the scenario processing requirements. The adaptation capability description includes the degree of support for scenario adaptation and processing efficiency.

[0055] The device's operating status includes the temperature, voltage, and fan speed of hardware components (such as CPU, memory, and network card), and the response time and error rate of software modules (such as database services and message queues). Normal operating ranges are obtained from the device manual; for example, the normal CPU temperature range is 35℃-85℃, with a set threshold of 10% (i.e., exceeding 85℃ × 110% = 93.5℃ triggers an anomaly). When the operating parameters (such as packet loss rate) of scenario-related components (such as the network module in an air conditioning control scenario) deviate from the normal range by more than 10% for five consecutive cycles (normal packet loss rate <1%, currently 2%), the deviation (1%) is calculated. Combined with the scenario's processing requirements (air conditioning control requires a network packet loss rate <1.5%), the scenario's support level is adjusted from "fully supported" to "limited support," and the processing efficiency is adjusted from "high" to "medium."

[0056] Step S1241: Extract device operation status information from the real-time operation status data of the service node. The device operation status information includes the operation parameters of each hardware component, the response status of the software module, and the data interaction status between components.

[0057] After parsing, real-time operational status data is used to extract device operational status information. Hardware component operating parameters are collected via the IPMI protocol, including CPU core temperature, memory voltage, and hard drive SMART metrics. Software module response status is obtained through API calls, such as database query latency and JVM GC counts. The data interaction status between components is analyzed through network packet capture to extract TCP connection establishment time and data transmission throughput. All of the above information is encapsulated into a device status object, containing fields such as ComponentID, ParameterName, CurrentValue, and Timestamp.

[0058] Step S1242: Analyze the operating parameters of each hardware component to identify whether there are any hardware components whose operating parameters deviate from the normal operating range. The normal operating range is determined by the service node hardware configuration manual.

[0059] The CurrentValue of a hardware component is compared with the normal operating range specified in the hardware configuration manual. For example, if the CPU CurrentValue is 95℃, and the upper limit of the normal range is 85℃, it is considered a deviation. The deviation identification result is stored in a hardware anomaly list, which includes ComponentID, ParameterName, and DeviationPercentage ((CurrentValue - Upper Limit) / Upper Limit × 100%).

[0060] Step S1243: Determine whether each hardware component is related to the processing of a specific business scenario by the correlation between the hardware components and the business scenario processing flow. If there are hardware components that are related to a specific business scenario and whose operating parameters deviate from the normal working range, check the scenario adaptation capability description of the business scenario in the service node capability dimension information. The scenario adaptation capability description includes the degree of support for scenario adaptation and processing efficiency. By comparing the difference between the current operating parameters of the hardware component and the normal working parameters, determine whether the current description still conforms to the actual situation.

[0061] The relationship between business scenario processing flow and hardware components is stored in the scenario-component mapping table. For example, the "Remote Diagnosis of Air Conditioner" scenario is associated with the "Network Adapter" and "CPU" components. If the network adapter has a packet loss rate deviation (CurrentValue=3%, normal range<1%), then check the scenario's compatibility description (currently "Support Level: High, Processing Efficiency: 90Mbps"). Through difference calculation (packet loss rate exceeds 2 percentage points), and referring to the preset impact assessment model (for every 1% increase in packet loss rate, processing efficiency decreases by 10%), it is determined that the processing efficiency in the current description is no longer consistent with reality (90Mbps×(1-2×10%)=72Mbps).

[0062] Step S1244: Evaluate the response of the software modules and identify whether there are software modules with response delays or functional abnormalities. Response delay is determined by the difference between the module response time and the normal response time reaching a set time threshold. Functional abnormalities are determined by the difference between the module error rate and the normal error rate reaching a set ratio threshold.

[0063] Software module response time is collected using monitoring tools (such as Prometheus's `http_request_duration_seconds` metric). Normal response time is obtained from the module performance benchmark report, with a set time threshold of 500ms. If the module's current response time is 1200ms, the difference = 1200 - 500 = 700ms > 500ms, which is considered a response delay. Functional anomalies are calculated using error rate metrics (such as `http_requests_total{status=~"5.."}`). The normal error rate is 0.1%, with a set threshold of 0.5%. The current error rate is 0.8%, and the difference = 0.8% - 0.1% = 0.7% > 0.5%, which is considered a functional anomaly.

[0064] Step S1245: Based on the role of the software module in the business scenario processing flow, determine whether the software module is the core module for processing a specific business scenario. If there is a core module response delay or functional abnormality, assess the impact of this situation on the processing efficiency of the corresponding business scenario. The impact assessment uses an efficiency decay model to calculate the degree of decay of processing efficiency. Processing efficiency includes the speed and quality of business scenario processing. Adjust the efficiency description in the scenario adaptation capability according to the degree of decay. The efficiency description includes specific indicators of processing speed and processing quality.

[0065] The core modules are determined through flowcharts of business scenarios. For example, in the "information publishing content rendering" scenario, the core module is "GPU rendering service." When this module experiences response latency, an efficiency decay model (such as an exponential decay function) is used to calculate the degree of decay. The model input is the latency time (Δt), and the output is the decay coefficient (λ). The processing speed = baseline speed × λ, and the processing quality is adjusted according to the increase in the error rate (e.g., for every 0.1% increase in the error rate, the quality score decreases by 1 point). The adjusted processing speed and quality metrics are then updated in the efficiency description of the scenario adaptation capability.

[0066] Step S1246: Check the data interaction status between hardware components, identify whether there is a data transmission interruption or transmission error in the interaction link. Data transmission interruption is determined by the difference between the data interaction frequency and the normal interaction frequency reaching a set frequency threshold. Transmission error is determined by the difference between the data interaction success rate and the normal success rate reaching a set proportion threshold.

[0067] Data interaction status is obtained by analyzing network traffic logs. The interaction frequency is the number of data packets per unit time. The normal interaction frequency is obtained from historical statistics, and a frequency threshold of 50% is set (i.e., if the current frequency is less than the normal frequency × 50%, it is considered an interruption). The data interaction success rate = number of successfully transmitted data packets / total number of data packets. The normal success rate is 99.9%, and a threshold of 1% is set (if the current success rate is less than 99.9% - 1% = 98.9%, it is considered a transmission error).

[0068] Step S1247: Determine whether the interaction link is a critical path for data flow in a specific business scenario by assessing its importance in the data flow process of the business scenario. If there is a data interaction problem in the critical path, determine the impact of the data interaction problem on the execution continuity of the specific business scenario. The impact determination uses a continuity interruption risk assessment model to calculate the interruption risk level. Based on the interruption risk level, add a restriction description to the scenario adaptability. The restriction description includes the preconditions for scenario execution and risk warnings.

[0069] The importance of the interaction link is determined through critical path analysis (CPM) of the business scenario data flow flowchart, with links on the critical path being designated as "high" in importance. When a data transmission error occurs on a critical path link (success rate = 98% < 98.9%), a continuous interruption risk assessment model is used. Input parameters include error rate, duration, and link redundancy, with an output risk level (1-5). If the risk level is level 3 (medium risk), a constraint description is added to the scenario adaptability: "Prerequisites: Backup link must be enabled; Risk warning: There is a 5% probability of data transmission delay."

[0070] Step S1248: Based on the analysis results of hardware components, software modules and data interaction status, correct the scenario adaptation capability description in the capability dimension information. The correction includes adjusting the scope of the adapted scenarios, updating the efficiency description or adding restrictive conditions. Adjusting the scope of the adapted scenarios includes adding adapted scenarios or removing unsuitable scenarios. Updating the efficiency description includes adjusting the processing speed and processing quality indicators. Adding restrictive conditions includes supplementing the preconditions and risk warnings for scenario execution.

[0071] Based on the analysis results of hardware, software, and data interaction, the descriptions of scene adaptability were revised. For example, due to a GPU module malfunction on a certain service node, the "3D content rendering" scene was removed from the adaptation scope; due to network link risks, the efficiency description of the "air conditioning data upload" scene was updated to "processing speed: original 80Mbps - 60Mbps," and the constraint "VPN encrypted channel required" was added. These corrections were recorded in the capability dimension change log, including Before and After snapshots.

[0072] Step S1249: Update the revised scenario adaptation capability description to the dynamic capability profile of the service node, and generate a correction log. The correction log includes the correction time, correction content, correction basis and correction executor.

[0073] The revised scene adaptation capability description is written to the dynamic capability profile database via API. This database uses a document-oriented database (such as MongoDB), with the service node ID as the primary key and the dynamic capability profile as the document content. The revision log uses a structured log format (JSON), for example: {"Timestamp":"2023-06-01T10:30:00Z","NodeID":"SN-001","Content":{"Scene":"Air Conditioner Remote Diagnosis","Field":"Processing Efficiency","OldValue":"High","NewValue":"Medium"},"Basis":"GPU Response Latency 1200ms","Operator":"system"}. The logs are stored in a distributed logging system (such as ELKStack) and retained for 90 days.

[0074] Step S125: Integrate the updated resource types, resource supply modes, and scenario adaptation capabilities to form a dynamic capability profile of the service node. The resource supply mode includes the resource allocation mode and scheduling mode. Each capability dimension in the dynamic capability profile of the service node is marked with the update basis. The update basis includes the corresponding real-time running status data fragment. The real-time running status data fragment includes the data collection time, data value, and data source.

[0075] The dynamic capability profile uses JSON-LD format and includes fields such as @context (defining terms), @id (unique identifier for service nodes), resourceTypes (array of resource types), supplyMode (resource supply mode object), and sceneAdaptability (scene adaptation capability object). Updates for each capability dimension are based on real-time runtime data fragments associated with the prov:wasDerivedFrom attribute, for example: "resourceTypes": [{"@type":"processor resources","currentSupply":"10 cores","prov:wasDerivedFrom":{"dataPoint":{"timestamp":"2023-06-01T10:25:00Z","value":"10 cores","source":"IPMI sensor"}}}]. The resource supply mode includes allocation modes such as "allocation on demand" and "pre-allocation," and scheduling modes include "FIFO scheduling" and "priority scheduling."

[0076] Step S126: Set an update cycle for the dynamic capability profile of each service node, and repeatedly perform real-time operation status data collection and capability dimension information update operations according to the update cycle. After each update, an update log is generated, which includes the update time, update content and update executor. The update cycle is determined according to the resource fluctuation frequency and device status change frequency of the service node.

[0077] The update cycle is dynamically adjusted using an adaptive algorithm, initially set to 5 minutes. Resource fluctuation frequency is calculated based on the standard deviation of CPU utilization, and device status change frequency is determined by the number of hardware anomaly events. When the resource fluctuation frequency of a service node exceeds a threshold (standard deviation > 20%), the update cycle is shortened to 2 minutes; when the device status is stable (no anomalies for 3 consecutive hours), the cycle is extended to 10 minutes. The update log records the field changes for each update, such as "2023-06-01T10:30:00 Update sceneAdaptability.Air Conditioner Remote Diagnosis. Processing Efficiency from High to Medium", with the default executor being "System Automatic".

[0078] Step S127: Classify and store the dynamic capability profiles of all service nodes according to the service node identifier, and establish an index mapping relationship between capability dimensions and profile content.

[0079] Service node identifiers (NodeIDs) use the format "SN-Geographic Location-Device Type-Serial Number" (e.g., "SN-SH-Server-001"). Dynamic capability profile files are stored in object storage services (e.g., S3) by NodeID, with the directory structure " / dynamic_profiles / SN-SH-Server-001 / 20230601 / ". Capability dimension indexes use an inverted index structure, where keywords (e.g., "air conditioning scenario" "10-core CPU") are mapped to a list of NodeIDs containing that dimension. The index is stored in Elasticsearch, supporting fast dimension retrieval.

[0080] Step S130: Based on the scenario features in the operation and maintenance task information set and the scenario adaptation capabilities in the dynamic capability profile of the service node, perform bidirectional adaptation processing between the operation and maintenance task and the service node to generate preliminary adaptation results between the operation and maintenance task and the service node.

[0081] The bidirectional adaptation process is implemented through a matching engine. The input is a set of operation and maintenance task information and a dynamic capability profile library of service nodes, and the output is the preliminary adaptation result. The matching engine first groups the tasks according to scenario characteristics, then searches for service nodes with corresponding scenario adaptation capabilities for each group of tasks, and then determines the candidate nodes for adaptation through resource requirement filtering and efficiency evaluation, finally forming a preliminary association between tasks and nodes.

[0082] Step S131: Extract the scenario features of each operation and maintenance task and the scenario group identifier corresponding to the operation and maintenance task from the operation and maintenance task information set. Obtain the business background, execution target and business system information involved in the scenario features through field filtering operation. At the same time, extract the resource types in the resource requirement attributes of each operation and maintenance task. The resource types include the names of various resources required in the execution of the operation and maintenance task.

[0083] Fields are extracted from the database table containing the operation and maintenance task information collection using Structured Query Language (SQL), for example: SELECT SceneFeature, SceneGroupID, ResourceDemand.ResourceType FROM task_info WHERE EtaskID='TK-001'. Information such as business background in the scene features is obtained through JSON parsing, and the resource type is extracted as a ResourceType array, such as ["Computational Resources", "Network Resources"].

[0084] Step S132: Extract the scenario adaptation capability and the types of resources that the service node can provide from the dynamic capability profile of the service node. The scenario adaptation capability includes the types of business scenarios that the service node is good at handling and the resource supply efficiency in the corresponding scenarios. The resource supply efficiency includes the response speed of resource supply, the stability of continuous supply and the ability to adjust the supply intensity. The types of resources that can be provided include the names of various resources that the service node can currently allocate.

[0085] Query the dynamic capability profile of service nodes using Elasticsearch, filtering based on whether the scene adaptability is not empty. For example: GET / dynamic_profiles / _search{"query":{"exists":{"field":"sceneAdaptability"}}}. Extract the business scene type (e.g., ["air conditioning system maintenance", "environmental monitoring"]) and resource supply efficiency indicators (response speed: 50ms, stability: 99.9%) from the sceneAdaptability field, and the available resource types (e.g., ["CPU", "memory", "bandwidth"]) from the resourceTypes field.

[0086] Step S133: Compare the scenario group identifier of the operation and maintenance task with the business scenario type in the scenario adaptation capability of the service node, and use the scenario type matching algorithm to match the two. Select the service nodes whose scenario adaptation capability contains the scenario type corresponding to the scenario group identifier as candidate service nodes. The candidate service node list contains the service node identifier and the scenario matching result.

[0087] The scenario type matching algorithm uses set inclusion judgment. If the service node's business scenario type set contains the scenario type corresponding to the task's scenario group identifier (e.g., scenario group identifier "KT01ACS" corresponds to scenario type "air conditioning system maintenance"), then a match is determined. For example, if the business scenario type of service node SN-001 is ["air conditioning system maintenance", "power monitoring"], and the task scenario group identifier corresponds to "air conditioning system maintenance", then SN-001 is selected as a candidate service node. The scenario matching result records the matched scenario type name and similarity score (1.0 for a complete match and 0.5 for a partial match).

[0088] Step S134: For each candidate service node, compare the resource types that the candidate service node can provide with the resource types in the resource requirement attributes of the operation and maintenance task. The comparison is performed using the principle of full matching of resource types. If there is a missing resource type, the candidate service node is removed from the candidate service node list.

[0089] The principle of full resource type matching requires that the set of resource types that a service node can provide includes all types of resource requirements for the task. For example, if the task requires resource types ["CPU", "Memory", "Network Bandwidth"], and candidate node SN-001 can provide ["CPU", "Memory"], but lacks "Network Bandwidth", then it is removed from the list. The comparison process is implemented through set operations (task resource type set - node resource type set == empty set).

[0090] Step S135: Analyze the resource supply efficiency in the corresponding scenario in the scenario adaptation capability of candidate service nodes, and determine the candidate service nodes whose resource supply efficiency meets the requirements of the operation and maintenance task execution rhythm as adaptation candidate nodes. The resource supply efficiency analysis includes the resource supply response speed analysis, continuous supply stability analysis and supply intensity adjustment capability analysis. The operation and maintenance task execution rhythm requirements include the resource usage start time requirements, continuous usage requirements and usage intensity change requirements.

[0091] Resource supply efficiency analysis employs a multi-indicator weighted scoring system, with weights determined using the Analytic Hierarchy Process (AHP): response speed (40%), continuous stability (30%), and adjustment capability (30%). Each indicator is compared to the task execution rhythm requirements. For example, the response speed requirement is <100ms, while the actual node response time is 50ms, resulting in a score of 100; the stability requirement is >99.5%, while the actual node response time is 99.8%, resulting in a score of 90; and the adjustment capability requirement is to support 5 levels of intensity adjustment, while the actual node support is 3 levels, resulting in a score of 60. The weighted total score is 100×40%+90×30%+60×30%=85 points. A threshold of 70 points is set, and nodes with scores ≥70 are identified as suitable candidate nodes.

[0092] Step S1351: Extract the resource usage rhythm description from the resource requirement attributes of the operation and maintenance task. The resource usage rhythm description includes the start time of resource usage, duration and usage intensity change trend. The start time of resource usage includes the time when the resource starts to be used and the preparation time in advance. The duration includes the shortest and longest duration of resource usage. The usage intensity change trend includes the rising phase, the stable phase and the falling phase of resource usage intensity.

[0093] The TimePeriod field in the resource requirement attribute includes the start time (StartTime: "2023-06-01T14:00:00", PrepareTime: 30 minutes) and duration (MinDuration: 2 hours, MaxDuration: 4 hours). The UsageMode field includes the usage intensity trend (rising phase: 0-30 minutes, intensity from 20% to 100%; stable phase: 30 minutes to 3 hours; declining phase: 3 hours to 4 hours, intensity from 100% to 20%). The above information is extracted by parsing the JSON structure.

[0094] Step S1352: Extract resource supply efficiency information for the corresponding scenario from the dynamic capability profile of the candidate service node. The resource supply efficiency information includes the response speed of resource supply, the stability of continuous supply, and the ability to adjust the supply intensity. The response speed of resource supply includes the preparation time and start-up time of resource supply. The stability of continuous supply includes the fluctuation range and interruption frequency of resource supply. The ability to adjust the supply intensity includes the adjustment range and adjustment speed of resource supply intensity.

[0095] Resource supply efficiency information for the corresponding scenario is extracted from the sceneAdaptability field of the dynamic capability profile. For example, the response speed of the "air conditioning scenario" is: preparation time 15 minutes, startup time 5 minutes; continuous supply stability: fluctuation range ±5%, interruption frequency 0.1 times / day; supply intensity adjustment capability: adjustment range 20%-100%, adjustment speed 5% / minute.

[0096] Step S1353: Compare the start time in the resource usage rhythm description with the response speed in the resource supply efficiency information. Use a time comparison algorithm to determine whether the response speed of resource supply can meet the start time requirements of resource use. That is, determine whether resource supply can be ready before the start time. The criterion for being ready is that the sum of the preparation time and start time of resource supply is less than the advance preparation time threshold of resource use.

[0097] The time comparison algorithm calculates the total node response time (preparation time + startup time = 15 + 5 = 20 minutes) and compares it with the task preparation time threshold (30 minutes). If 20 < 30, the startup timing requirement is met. If the total node response time exceeds the threshold, the efficiency analysis score for this item is 0.

[0098] Step S1354: Compare the duration in the resource usage rhythm description with the continuous supply stability in the resource supply efficiency information. Continuous supply stability includes the fluctuation range and interruption frequency of resource supply. By comparing the duration and analyzing the stability, it is determined whether the resource supply can remain stable without interruption or large fluctuations within the duration of resource usage. The standard for judging stability is that the interruption frequency of resource supply is lower than the preset interruption frequency threshold and the fluctuation range is less than or equal to the allowable fluctuation range threshold.

[0099] The task duration requirement is 2-4 hours, and the node's continuous supply stability indicators are: fluctuation range ±5% (allowable threshold ±10%) and interruption frequency 0.1 times / day (allowable threshold 1 time / day). Based on a stability analysis model (such as a Markov chain), the predicted interruption probability within 4 hours is <0.01%, and the fluctuation range is within the allowable range, thus meeting the continuous usage requirements.

[0100] Step S1355: Compare the trend of change in usage intensity in the description of resource usage rhythm with the supply intensity adjustment capability in the resource supply efficiency information. By comparing trends and analyzing adjustment capabilities, determine whether the resource supply intensity can be adjusted to follow the change in usage intensity so that the two trends are consistent. The criteria for consistent trends are that the adjustment range of resource supply intensity covers the change range of usage intensity and the adjustment rate of resource supply intensity is greater than or equal to the threshold of the change rate of usage intensity.

[0101] The task intensity variation trend is 20%–100% (rising phase over 30 minutes), with a node adjustment range of 20%–100% (coverage) and an adjustment rate of 5% / minute. The task intensity variation rate is (100%–20%) / 30 minutes ≈ 2.67% / minute, and the node adjustment rate of 5% ≥ 2.67%, indicating that the adjustment capability meets the requirements. Trend comparison is performed by plotting intensity variation curves and calculating curve similarity (e.g., dynamic time warping distance < 10).

[0102] Step S1356: If the response speed, continuous supply stability and supply intensity adjustment capability of resource supply all meet the needs of resource usage rhythm, mark the candidate service node as an adaptation candidate node by establishing the association between the candidate service node identifier and the adaptation identifier.

[0103] When all three indicators meet the requirements, add "Adaptation flag: Yes" to the candidate service node list and record the degree of satisfaction of each indicator (e.g., "Response speed satisfaction: Good (20 / 30)").

[0104] Step S1357: If any one or more items do not meet the requirements, record the specific aspects of non-compliance. The specific aspects of non-compliance include the response speed not meeting the start-up timing, the continuous supply stability not covering the duration, or the supply intensity adjustment capability not matching the changes in usage intensity. At the same time, record the specific data differences of non-compliance. The data differences include the difference between the actual value and the demand value.

[0105] For example, if the node adjustment rate is 3% / minute and the task demand rate is 4% / minute, the data difference is 3-4=-1% / minute, and the record is "supply intensity adjustment capacity mismatch: adjustment rate is 1% / minute lower".

[0106] Step S1358: For the service node determined as a candidate node for adaptation, record the specific performance of the resource supply efficiency of the service node in meeting the demand. The specific performance includes specific data on response speed meeting the start-up timing, specific data on continuous supply stability covering the duration, and specific data on supply intensity adjustment capability matching changes in usage intensity.

[0107] Record information such as "Response speed: 20 minutes (requirement 30 minutes); Continuous stability: fluctuation ±3% (requirement ±10%); Adjustment capability: 5% / minute (requirement 2.67% / minute)" will be used as the basis for subsequent sorting.

[0108] Step S1359: Organize all candidate nodes for adaptation to form a preliminary list of candidate nodes for adaptation for each operation and maintenance task, including service node identifiers and efficiency performance descriptions that meet the requirements. The efficiency performance descriptions that meet the requirements include complete data on the compliance conditions and analysis results.

[0109] The initial list of candidate nodes for adaptation is sorted by node identifier. Each entry includes NodeID, EfficiencyDescription (specific performance that meets the requirements), and Score (weighted score), such as "SN-001, response speed 20 minutes / continuous fluctuation ±3% / adjustment 5% / minute, score 85".

[0110] Step S136: For each operation and maintenance task, sort all candidate nodes for adaptation in descending order according to the scenario adaptation matching degree. The scenario adaptation matching degree is determined based on the degree of overlap between business scenario types and the degree of fit between resource supply efficiency. The degree of overlap between business scenario types is calculated by the number of scenario type matches, and the degree of fit between resource supply efficiency is calculated by the degree of fit between resource supply efficiency parameters and operation and maintenance task execution rhythm requirements parameters.

[0111] The formula for calculating scene adaptation matching degree is: Matching degree = 0.6 × Scene overlap degree + 0.4 × Efficiency fit degree. Scene overlap degree = Number of matching node supported scene types with task scene group identifiers / Total number of task scene group identifiers; Efficiency fit degree = Weighted score of resource supply efficiency / 100. For example, if scene overlap degree = 1 / 1 = 1, efficiency fit degree = 85 / 100 = 0.85, matching degree = 0.6 × 1 + 0.4 × 0.85 = 0.94, the list of candidate nodes for adaptation is sorted in descending order of matching degree.

[0112] Step S137: Associate each operation and maintenance task with the sorted candidate nodes. By establishing the association between the operation and maintenance task identifier and the list of candidate nodes, a preliminary adaptation relationship between the operation and maintenance task and the service node is formed. The preliminary adaptation relationship between each operation and maintenance task and the service node includes the operation and maintenance task identifier, the list of candidate nodes, and the sorting criteria. The sorting criteria include the calculation results of the degree of overlap of business scenario types and the calculation results of the degree of matching of resource supply efficiency.

[0113] The relationships are stored in a relational database, with a table structure of (TaskID, NodeID, MatchDegree, SceneOverlap, EfficiencyFit), where SceneOverlap represents the scene overlap calculation result, and EfficiencyFit represents the efficiency fit calculation result. Foreign key constraints ensure the validity of TaskID and NodeID.

[0114] Step S138: Integrate the preliminary adaptation relationships of all operation and maintenance tasks and service nodes. Organize the preliminary adaptation relationships of operation and maintenance tasks and service nodes using a batch data integration method to form a structured preliminary adaptation result. Each entry in the preliminary adaptation result contains preliminary adaptation relationship information of operation and maintenance tasks and service nodes.

[0115] Batch data integration is achieved through database transactions. All preliminary matching relationships are written to the result table at once, and a preliminary matching result file (CSV format) is generated, which includes fields such as TaskID, NodeID, MatchDegree, and Ranking for subsequent conflict detection.

[0116] Step S140: Identify the adaptation combinations with resource supply conflicts in the preliminary adaptation results, combine the execution relationship of the operation and maintenance tasks with the resource supply mode of the service nodes, resolve the resource supply conflicts, and generate the final adaptation scheme of the operation and maintenance tasks and service nodes.

[0117] Resource supply conflict identification is achieved through a conflict detection engine, which detects resource competition among tasks on the same node based on the resource supply description of the service node and the resource demand cycle of the task. Conflict resolution employs two strategies: adjusting the execution order and re-associating nodes. Priority is given to adjusting the order; if adjustment fails, the node is switched, ultimately ensuring that all compatible combinations are free of resource conflicts.

[0118] Step S141: Process each entry in the preliminary adaptation result using a loop traversal algorithm, count all operation and maintenance tasks associated with each service node, and the resource requirement attributes and execution association of the operation and maintenance tasks. The statistical results are presented in tabular form, including service node identifier, list of associated operation and maintenance tasks, list of resource requirement attributes, and list of execution association.

[0119] The iterative traversal algorithm processes the initial adaptation results by grouping them by NodeID. For each NodeID, it collects a list of associated TaskIDs, queries the resource demand attribute (ResourceDemand) and execution relation (ExecutionRelation) of each TaskID, and forms a statistical record, such as "NodeID: SN-001, Tasks: [TK-001, TK-002], ResourceDemands: [{'Type': 'CPU', 'Period': '14:00-16:00'}, ...], ExecutionRelations: [{'Pre-task': [], 'Synchronization task': [TK-002]}]".

[0120] Step S142: Analyze the resource requirement attributes of the operation and maintenance tasks associated with each service node, focusing on the resource usage cycle and resource usage amount, and determine whether the operation and maintenance tasks overlap in the resource usage cycle, and whether the usage demand for the same type of resources during the overlapping period exceeds the resource supply description in the dynamic capability profile of the service node. The resource supply description includes the amount of resources that the service node can supply in the corresponding period.

[0121] For tasks with the same NodeID, group them by resource type and analyze whether the usage periods (TimePeriod) of resources in each group overlap. For example, TK-001 and TK-002 both require CPU resources on NodeIDSN-001, with usage periods of 14:00-16:00 and 15:00-17:00 respectively, resulting in an overlap between 15:00-16:00. Calculate the total resource usage during the overlapping period (TK-001 requires 4 cores + TK-002 requires 4 cores = 8 cores), and compare it with the node's available CPU resources (6 cores). Since 8 > 6, a resource supply conflict is determined.

[0122] Step S143: The combination of service nodes with overlapping resource usage cycles and resource demand exceeding supply description and corresponding operation and maintenance tasks is identified as a resource supply conflict adaptation combination. Such combinations are marked by establishing a conflict identifier. The conflict identifier includes conflict type and conflict degree. Conflict type includes resource usage cycle overlap conflict and resource demand exceeding conflict. Conflict degree is determined by the proportion of resource demand exceeding the available resource quantity.

[0123] The conflict identification format is "Conflict Type-NodeID-TaskID List-Conflict Period", such as "Cycle Overlap + Demand Exceeds -SN-001-[TK-001,TK-002]-15:00-16:00". Conflict severity = (Total Demand - Available Supply) / Available Supply, for example (8-6) / 6=0.33, the conflict severity is 33%.

[0124] Step S144: For each operation and maintenance task in the resource supply conflict adaptation combination, extract the execution association of the operation and maintenance task, identify whether there is a collaborative execution requirement with adjustable execution order, which includes non-mandatory synchronous execution requirement and flexible sequential execution requirement.

[0125] The collaborative execution requirement in the execution relationship is determined by the ExecutionRelation field. If PredecessorTasks is empty and SynchronousTasks is marked as "non-mandatory", then there is a requirement for flexible execution order; if SynchronousTasks is marked as "asynchronous", then there is a non-mandatory synchronous execution requirement. For example, the execution relationship of TK-001 is "Predecessor task: None, Synchronous task: TK-002 (non-mandatory)", indicating a requirement for adjustable execution order.

[0126] Step S145: For maintenance tasks with adjustable execution order, refer to the resource supply description of the service node and the resource usage cycle of other maintenance tasks, adjust the resource usage cycle of the maintenance task so that the adjusted resource usage cycle is different from the resource usage cycle of other maintenance tasks, and record the adjusted resource usage cycle back into the resource requirement attribute of the maintenance task.

[0127] Step S1451: Extract the basis for adjusting the execution order from the operation and maintenance task execution association in the resource supply conflict adaptation combination. The basis for adjusting the execution order includes the order flexibility description and collaborative execution requirement description in the execution association. The order flexibility description includes whether the order of execution can be adjusted, and the collaborative execution requirement description includes whether it is mandatory synchronous execution.

[0128] The IsOrderFlexible field, which describes the order flexibility in the execution relationship, takes the value "Yes" or "No," while the IsMandatorySync field, which describes the collaborative execution requirements, also takes the value "Yes" or "No." For example, if IsOrderFlexible is "Yes" and IsMandatorySync is "No" for TK-001, then there is sufficient basis for adjusting the execution order.

[0129] Step S1452: Determine the current resource usage period of the operation and maintenance tasks whose execution order can be adjusted. Obtain the current resource usage period by querying the resource requirement attributes of the operation and maintenance tasks. The current resource usage period includes the start time and end time of resource usage.

[0130] Querying the ResourceDemand.TimePeriod field of TK-001, we find that the current resource usage period is "start time: 14:00, end time: 16:00".

[0131] Step S1453: Analyze the resource usage cycle of other operation and maintenance tasks on the same service node. By traversing the resource demand attributes of all operation and maintenance tasks associated with the same service node, obtain the resource usage cycle of other operation and maintenance tasks, find the time periods that overlap with the current resource usage cycle, and clarify the specific time range of the overlap through the time interval comparison algorithm. The specific time range includes the start time and end time of the overlapping time period.

[0132] The TimePeriod of TK-002 associated with SN-001 is "15:00-17:00". The time interval comparison algorithm calculates the intersection as "15:00-16:00", that is, the overlapping period starts at 15:00 and ends at 16:00.

[0133] Step S1454: Based on the execution relationship of the operation and maintenance tasks, determine the earliest start time and the latest end time of the operation and maintenance tasks whose execution order can be adjusted. The earliest start time is determined by the completion time of the preceding task in the execution relationship, and the latest end time is determined by the start time of the following task in the execution relationship.

[0134] The prerequisite task for TK-001 has no completion time (no prerequisite task), and the earliest start time is "00:00"; the start time for the subsequent task TK-003 is "18:00", and the latest end time is "18:00".

[0135] Step S1455: Between the earliest startable time period and the latest end time period, select a time range that does not overlap with the resource usage periods of other operation and maintenance tasks. Select a suitable time range from the available time range as the adjusted resource usage period. The adjusted resource usage period includes the new start time period and end time period.

[0136] Within the range of "00:00-18:00", excluding the overlapping period of "15:00-16:00", the selectable time range is either "00:00-15:00" or "16:00-18:00". Select "16:00-18:00", which is close to the original cycle length, as the adjusted resource usage cycle.

[0137] Step S1456: If there are multiple non-overlapping time ranges, prioritize the time range that is close to the original resource usage cycle time span. The criterion for judging that the time span is close is that the difference between the new time span and the original time span is less than the proportion of the original time span.

[0138] The original time span is 2 hours (16:00-14:00). The selectable time ranges are "00:00-12:00" (spanning 12 hours) and "16:00-18:00" (spanning 2 hours). The latter has a difference ratio of 0, which is less than the preset ratio threshold (20%), so the latter is selected.

[0139] Step S1457: Determine the start and end times of the adjusted resource usage period, update the resource usage period description in the resource requirement attribute of the operation and maintenance task, and record the resource usage period data before and after the update to form an update log. The update log includes the update time, operation and maintenance task identifier, original period data and new period data.

[0140] Update TK-001's ResourceDemand.TimePeriod to "Start time period: 16:00, End time period: 18:00", and the update log records "2023-06-01T11:00, TK-001, original period 14:00-16:00, new period 16:00-18:00".

[0141] Step S1458: Notify the relevant collaborative operation and maintenance tasks of the adjusted resource usage period. The notification method adopts a data push mechanism to push the adjusted resource usage period data to the associated data nodes of the relevant collaborative operation and maintenance tasks.

[0142] Send a notification message to the associated data node of TK-002 via a message queue (such as Kafka). The message content is "The resource usage period of TK-001 has been adjusted to 16:00-18:00. Please update the collaboration plan".

[0143] Step S1459: Record the adjustment content of the resource usage cycle. The adjustment content includes the cycle before adjustment, the cycle after adjustment, and the basis for adjustment. The basis for adjustment includes the results of the execution correlation analysis and the results of other operation and maintenance task cycle analysis. The results of the execution correlation analysis include the basis data for the adjustable order. The results of other operation and maintenance task cycle analysis include overlapping time period data and available time range data.

[0144] The adjustments are stored in the periodic adjustment record table, with fields including TaskID, OldPeriod, NewPeriod, and AdjustmentBasis (JSON format, containing the results of the relationship analysis and the periodic analysis).

[0145] Step S146: For operation and maintenance tasks that cannot resolve conflicts by adjusting the execution order, view the list of adaptation candidate nodes for the operation and maintenance task in the preliminary adaptation results, select the adaptation candidate node that is ranked later in the list of adaptation candidate nodes but has no conflict in resource supply status, re-associate the operation and maintenance task, and establish a new association mapping relationship during the re-association process.

[0146] When the execution order of maintenance tasks cannot be adjusted (such as forced synchronization tasks), the second-highest ranked node is selected from the list of candidate nodes for adaptation. For example, if TK-003 has a resource conflict with SN-001 and the order cannot be adjusted, and the second-ranked node in its candidate node list is SN-002, and SN-002 has no conflict within the resource usage period of TK-003, then TK-003 will be reassociated with SN-002, establishing a new TaskID-NodeID mapping relationship.

[0147] Step S147: For the re-associated operation and maintenance task, check again whether there is a resource supply conflict between the operation and maintenance task and other operation and maintenance tasks of the newly associated service node. The check process repeats the resource usage cycle and resource usage analysis steps.

[0148] After re-association, repeat the analysis process in step S142 to confirm whether the resource requirements of TK-003 overlap with those of other tasks within the resource usage cycle of SN-002 and exceed the supply. If conflicts still exist, continue to select the next candidate node until there are no conflicts or all candidate nodes have been traversed (manual intervention is required at this point).

[0149] Step S148: Record the resolution method for each resource supply conflict adaptation combination. The resolution method includes adjusting the execution order or re-associating service nodes. At the same time, record the resolution basis, which includes the execution association analysis results and the service node resource supply description. The recorded content includes conflict identifier, resolution method, resolution basis and resolution time.

[0150] The conflict resolution record is stored in the conflict resolution log table, for example, "Conflict ID: Overlapping cycles + Excessive demand -SN-001-[TK-001,TK-002]-15:00-16:00, Resolution method: Execution order adjustment, Basis: TK-001 can be adjusted and there are resources after SN-001 16:00, Time: 2023-06-01T11:05".

[0151] Step S149: Integrate all the reconciled adaptation combinations, organize the adaptation combinations to form the final adaptation scheme of operation and maintenance tasks and service nodes, and mark each adaptation combination in the final adaptation scheme as having no resource supply conflict.

[0152] The final adaptation solution is organized in XML format, with the root element being... <finaladaptationplan>, containing multiple <adaptationitem>The sub-elements contain TaskID, NodeID, ResourcePeriod, and ConflictStatus (with a value of "no conflict"). The scheme file is stored in a version control system, with version numbers in the format "VYYYYMMDDHHMM".

[0153] Step S150: Based on the final adaptation scheme, integrate the execution sequence of operation and maintenance tasks with the resource supply arrangement of service nodes to form an operation and maintenance process optimization scheme. During the execution of the operation and maintenance process, dynamically evolve and calibrate the content of the operation and maintenance process optimization scheme based on the changes in the real-time running status data of service nodes and the execution progress of operation and maintenance tasks.

[0154] The operation and maintenance process optimization plan is generated by the process orchestration engine, including a task execution network diagram and a resource allocation Gantt chart. During execution, the monitoring module tracks the node status and task progress in real time. When anomalies occur, a dynamic calibration mechanism is triggered to adjust the task sequence or resource allocation to ensure continuous process optimization.

[0155] Step S151: Based on the final adaptation scheme, integrate the execution sequence of operation and maintenance tasks with the resource supply arrangement of service nodes to form an operation and maintenance process optimization scheme.

[0156] The workflow orchestration engine reads the final adaptation scheme, constructs a directed acyclic graph (DAG) based on the execution relationships of TaskIDs, and determines the task execution sequence. Resource allocation is arranged by generating a Gantt chart based on the resource usage cycle of NodeIDs, indicating the resource allocation for each time period. The DAG and Gantt chart are integrated into an operations and maintenance workflow optimization scheme, including information such as task ID, execution order, node ID, resource allocation time period, and responsible person.

[0157] Step S152: During the execution of the operation and maintenance process, based on the changes in the real-time running status data of the service nodes and the progress of the operation and maintenance tasks, dynamically evolve and calibrate the content of the operation and maintenance process optimization plan.

[0158] For example, step S1521: During the execution of the operation and maintenance process, real-time operating status data of each service node is obtained. The real-time operating status data includes changes in resource usage, updates in device status, and changes in data transmission status. Changes in resource usage include changes in the usage of various resources. Updates in device status include updates in the status of hardware components and software modules. Changes in data transmission status include changes in data transmission rate, latency, and error rate.

[0159] Real-time operational status data is continuously collected by the Agent program, pushed to the monitoring platform in real time via a message queue, and stored in a time-series database. Changes in resource usage are calculated using a sliding window (e.g., changes in CPU utilization within 5 minutes), device status updates are recorded through status code changes (e.g., from "normal" to "warning"), and changes in data transmission status are reflected through time-series curves of network metrics.

[0160] Step S1522: Track the execution progress of each operation and maintenance task in real time. The execution progress includes completed execution steps, currently executing steps, and remaining unexecuted steps. The execution status of each step is recorded in detail, including the step start time, step end time, step execution result, and step resource consumption data.

[0161] Task execution progress is collected through event tracking logs. Each execution stage (such as "fault diagnosis" or "component replacement") records logs at the start and end, including stage ID, TaskID, StartTime, EndTime, Result (success / failure), and ResourceConsumption (CPU and memory usage). The monitoring platform aggregates and analyzes these logs to update the task's progress status in real time.

[0162] Step S1523: Analyze the real-time operating status data to identify whether there are service nodes with reduced resource supply capacity or abnormal equipment status. Reduced resource supply capacity is determined by the reduction in available resources reaching a set reduction threshold. Abnormal equipment status is determined by the abnormal status data of hardware components and software modules exceeding their respective set abnormal status thresholds. By comparing the changed data with the requirements of operation and maintenance tasks, it is determined whether it will affect the normal execution of operation and maintenance tasks on the service node.

[0163] Resource supply capacity reduction analysis calculates the ratio of the current available supply to the baseline value ((baseline value - current value) / baseline value). If the ratio exceeds a set reduction threshold (e.g., 20%), it is considered a reduction. Abnormal equipment status is determined through the fusion of multiple indicators (e.g., CPU temperature > 90℃ and fan speed < 1500 RPM). The task execution requirements at the abnormal node are compared with the current supply capacity. If the resource gap cannot meet the minimum requirements of the task, it is determined that normal execution will be affected.

[0164] Step S1524: For service nodes with reduced resource supply capacity or abnormal equipment status, check the execution progress of the maintenance tasks being executed on the service node. Obtain the current execution status by querying the maintenance task execution progress record, and determine whether the current execution progress is affected. The judgment criteria for the execution progress being affected are that the delay time of the execution link exceeds the preset delay time threshold or an error occurs in the execution link.

[0165] Query the execution progress record of the task on the abnormal node, calculate the delay time of the current stage (current time - planned end time), and if it exceeds the preset delay time threshold (e.g., 30 minutes), or the stage execution result is "failure", then it is determined that the execution progress is affected.

[0166] Step S1525: If the execution progress is affected, the affected operation and maintenance tasks on the service node are transferred to other service nodes whose dynamic capability profiles meet the requirements of the operation and maintenance tasks and whose real-time running status is stable. The transfer process is implemented using a task migration algorithm. At the same time, the execution steps of the operation and maintenance tasks after the transfer are adjusted. The adjustments include reordering the execution steps and redistributing the execution time.

[0167] The task migration algorithm first retrieves stable nodes (without any abnormal statuses in the past hour) that match the task scenario characteristics and resource requirements. Then, it transfers the task instance to the new node using container migration technology (such as Docker Swarm service migration). The execution process is then adjusted by reallocating the start time of each stage based on the resource availability of the new node, ensuring that execution dependencies are not violated.

[0168] Step S1526: Analyze the execution progress of the operation and maintenance tasks. If any operation and maintenance task completes its execution phase ahead of schedule, the criterion for early completion is that the completion time of the phase is earlier than the planned completion time threshold. Release the service node resources occupied by the operation and maintenance task. The release process includes resource reclamation and marking. Allocate the released resources to operation and maintenance tasks on the same service node whose execution progress is lagging or about to start. Resource allocation is implemented using a resource scheduling algorithm.

[0169] Early completion is determined by the difference between EndTime and the planned completion time. If the difference is positive (i.e., early) and exceeds the planned completion time threshold (e.g., 10 minutes), resource release is triggered. Resource reclamation is achieved by calling the node's resource management interface (e.g., Kubernetes' resource deletion API), and released resources are marked as "available". Resource scheduling algorithms (e.g., greedy algorithms) prioritize allocating available resources to delayed tasks (those with the longest delays) or high-priority tasks that are about to start.

[0170] Step S1527: If any operation and maintenance task is lagging behind, check if there are any additional resources available. The additional resources available include the idle resources of the service node and the resources released by other operation and maintenance tasks. If they exist, allocate additional resources to the operation and maintenance task.

[0171] Tasks with delayed execution are sorted by delay time. For tasks ranked higher, the available resources (such as unallocated CPU cores) of the nodes are queried and the resource pool is released. If the total meets the task requirements, the resource quota for the task is increased (e.g., from 2 CPU cores to 4 cores) through a resource scheduling algorithm.

[0172] Step S1528: Record the specific content of each evolution calibration. The specific content includes the service node status change analysis results, operation and maintenance task transfer status, resource reallocation status, and execution progress adjustment status. The service node status change analysis results include change data and impact assessment. The operation and maintenance task transfer status includes the service node identifier and task status before and after the transfer. The resource reallocation status includes resource type, allocation amount, and allocation object. The execution progress adjustment status includes the execution process arrangement and time plan before and after the adjustment.

[0173] Calibration records are stored in the audit database. Each record contains the calibration time, relevant TaskID / NodeID, status before the change, status after the change, and calibration basis. For example, "2023-06-01T15:30:00, SN-001 status abnormal, TK-001 transferred to SN-003, resources reallocated to 4 CPU cores, execution phase adjusted: original 14:00-16:00-15:30-17:30".

[0174] Step S1529: Integrate the calibration content into the operation and maintenance process optimization plan, update the operation and maintenance task execution sequence and service node resource supply arrangement in the operation and maintenance process optimization plan, and form the evolved operation and maintenance process optimization plan.

[0175] The calibration content is integrated into the original solution through a version control mechanism, generating a new solution version labeled "Dynamic Calibration V2". The updated execution sequence and resource provisioning arrangements are regenerated into a DAG and Gantt chart through a process orchestration engine and pushed to the execution terminal to guide operations personnel or automated scripts in execution.

[0176] Figure 2 This application illustrates an operation and maintenance (O&M) process optimization system 100 for O&M technical services, including 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 O&M process optimization method for O&M technical services. The processor 1001 and the memory 1003 are connected, for example, via a bus 1002. Optionally, the O&M process optimization system 100 may further include a transceiver 1004, which can be used for data interaction between this O&M process optimization system and other O&M process optimization 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 O&M process optimization system 100 does not constitute a limitation on the embodiments of this application.

[0177] 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.

[0178] 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.< / adaptationitem> < / finaladaptationplan>

Claims

1. An operation process optimization method applied to operation and maintenance technical services, characterized in that, The method comprises: Collecting the scene features, execution association relationships and resource demand attributes of each operation and maintenance task in the operation and maintenance task demand document to form an operation and maintenance task information set, the scene features reflecting the business scene category to which the operation and maintenance task belongs, and the execution association relationships reflecting the cooperative execution demand between the operation and maintenance tasks; Collecting real-time running state data and capability dimension information of each service node, dynamically updating the capability dimension information based on the real-time running state data, and constructing a dynamic capability portrait of the service node, the capability dimension information including the resource type, resource supply mode and scene adaptation capability that can be provided by the service node; Based on the scene features in the operation and maintenance task information set and the scene adaptation capability in the dynamic capability portrait of the service node, performing bidirectional adaptation processing of the operation and maintenance tasks and the service nodes to generate a preliminary adaptation result of the operation and maintenance tasks and the service nodes; Identifying an adaptation combination with resource supply conflicts in the preliminary adaptation result, resolving the resource supply conflicts by combining the execution association relationships of the operation and maintenance tasks and the resource supply mode of the service nodes, and generating a final adaptation scheme of the operation and maintenance tasks and the service nodes; According to the final adaptation scheme, integrating the execution sequence of the operation and maintenance tasks and the resource supply arrangement of the service nodes to form an operation and maintenance process optimization scheme, and dynamically evolving and calibrating the content of the operation and maintenance process optimization scheme based on the changes of the real-time running state data of the service nodes and the execution progress of the operation and maintenance tasks in the operation and maintenance process execution process. 2.The operation flow optimization method for operation and maintenance technical service of claim 1, wherein, The collecting the scene features, execution association relationships and resource demand attributes of each operation and maintenance task in the operation and maintenance task demand document to form an operation and maintenance task information set comprises: Reading the text description content in the operation and maintenance task demand document, extracting information describing the business background, execution target and involved business system of the operation and maintenance task from the text description content, the information describing the business background, execution target and involved business system of the operation and maintenance task collectively constituting the scene features of the operation and maintenance task, and the text description content covering the business background introduction, execution target explanation and involved business system list of the operation and maintenance task; Identifying the cooperative execution requirements between the operation and maintenance tasks from the text description content, the cooperative execution requirements including the execution sequence requirements and synchronous execution requirements between the operation and maintenance tasks, and the execution sequence requirements and synchronous execution requirements between the operation and maintenance tasks constituting the execution association relationships of the operation and maintenance tasks; Extracting the resource type, resource usage mode and resource usage period required in the execution process of the operation and maintenance task from the text description content, the resource usage mode including the calling mode and release mode of the resource, the resource usage period including the start period and end period of the resource usage, and the resource type, resource usage mode and resource usage period required in the execution process of the operation and maintenance task constituting the resource demand attributes of the operation and maintenance task; For each operation and maintenance task, the scene features, execution association relationships and resource demand attributes of the operation and maintenance task are associated and bound, and a corresponding mapping relationship between the three is established, so that the scene features, execution association relationships and resource demand attributes of each operation and maintenance task form a complete corresponding relationship; According to the submission sequence of the operation and maintenance tasks, the associated binding information of all operation and maintenance tasks is sorted, the associated binding information is organized in a structured data format, the structured data format includes field definition and data type description, a structured operation and maintenance task information set is formed, each entry in the operation and maintenance task information set includes operation and maintenance task identification and corresponding scene characteristics, execution association relationship and resource demand attributes of the operation and maintenance task; The scene characteristics in the operation and maintenance task information set are classified, and operation and maintenance tasks with similar scene characteristics are divided into the same scene group; Scene group identifiers are added to the operation and maintenance task information set, the scene group identifiers are generated in the form of a combination of letters and numbers, each operation and maintenance task entry corresponds to a unique scene group identifier, and the association relationship between the operation and maintenance task and the scene group is established through the scene group identifier; The complete operation and maintenance task information set is stored in a target database, the target database adopts a partition storage strategy, and the operation and maintenance task information set is partitioned according to the scene group identifier. 3.The operation flow optimization method for operation and maintenance technical service of claim 1, wherein, The real-time running state data and capability dimension information of each service node are collected, the capability dimension information is dynamically updated based on the real-time running state data, and a service node dynamic capability portrait is constructed, including: Real-time running state data of each service node is obtained, and initial capability dimension information is extracted from a configuration document of the service node, the configuration document includes hardware configuration instructions and software configuration instructions of the service node, the initial capability dimension information includes resource types, maximum resource carrying capacity and supported business scenario types designed for the service node, the resource types include processor resources, memory resources, storage resources and network resources, the maximum resource carrying capacity includes the maximum allocatable amount of each type of resource, and the supported business scenario types include a list of business scenarios that can be processed when the service node is designed; The real-time running state data of the service node is compared with the resource carrying capacity description in the initial capability dimension information, the real-time resource usage is compared with the maximum resource carrying capacity, the influence of the real-time resource usage on the resource carrying capacity is analyzed, and the influence analysis includes the proportion of the real-time resource usage to the maximum resource carrying capacity and the decay influence of the resource usage exceeding the maximum allocatable amount by a set proportion threshold for a continuous set of collection periods; According to the change of the real-time resource usage, the resource carrying related description in the initial capability dimension information is adjusted, when the usage of any type of resource of the service node exceeds the maximum allocatable amount of the type of resource by a set proportion threshold for a continuous set of collection periods, the available description of the type of resource is updated according to the duration of the continuous exceeding of the threshold by the usage of the type of resource, and the decay degree reflected by the resource performance decay data, the available description includes the current available resource amount and the limitation condition of resource supply; The device running state in the real-time running state data is combined to correct the scene adaptation capability description in the capability dimension information. The hardware component running state and the software module running state in the device running state directly affect the processing capability of the service node for a specific business scene. When the running parameter of any one type of scene-related component deviates from the normal working range by a set proportion threshold or more for a continuous set number of collection periods, the amplitude of the deviation and the scene processing demand are used to adjust the adaptation capability description for the corresponding scene. The adaptation capability description includes the support degree and the processing efficiency of scene adaptation. The updated resource type, resource supply mode, and scene adaptation capability are integrated to form a dynamic capability profile of the service node. The resource supply mode includes the allocation mode and the scheduling mode of the resource. Each capability dimension in the dynamic capability profile of the service node is marked with an update basis, which includes the corresponding real-time running state data segment. The real-time running state data segment includes the data collection time, the data value, and the data source. An update period is set for each dynamic capability profile of the service node. The real-time running state data collection and capability dimension information update operations are repeatedly performed according to the update period. An update log is generated after each update. The update log includes the update time, the update content, and the update executor. The update period is determined according to the resource fluctuation frequency and the device state change frequency of the service node. The dynamic capability profiles of all service nodes are classified and stored according to the service node identifier. An index mapping relationship between the capability dimension and the profile content is established.

4. The operation flow optimization method for operation and maintenance technical service of claim 1, wherein, Based on the scene features in the operation and maintenance task information set and the scene adaptation capability in the dynamic capability profile of the service node, the bidirectional adaptation processing of the operation and maintenance task and the service node is performed to generate a preliminary adaptation result of the operation and maintenance task and the service node, including: The scene features and the scene group identifier corresponding to each operation and maintenance task are extracted from the operation and maintenance task information set. The business background, execution target, and involved business system information in the scene features are obtained through field filtering operation. The resource types in the resource demand attributes of each operation and maintenance task are extracted. The resource types include various resource names required in the operation and maintenance task execution process. The scene adaptation capability of each service node and the resource types that can be provided by the service node are extracted from the dynamic capability profile of the service node. The scene adaptation capability includes the business scene types that the service node is good at processing and the resource supply efficiency under the corresponding scene. The resource supply efficiency includes the response speed, the continuous supply stability, and the supply intensity adjustment capability of resource supply. The resource types that can be provided include various resource names that can be allocated by the service node at present. The scene group identifier of the operation and maintenance task and the business scene type in the scene adaptation capability of the service node are compared. The scene type matching algorithm is used to match the two. The service nodes whose scene adaptation capability includes the scene type corresponding to the scene group identifier are selected as candidate service nodes. The candidate service node list includes the service node identifier and the scene matching result. For each candidate service node, compare the resource types that the candidate service node can provide with the resource types in the resource requirement attribute of the operation and maintenance task, and use the resource type full matching principle for comparison. If there is a missing resource type, remove the candidate service node from the candidate service node list; Analyze the resource supply efficiency of the candidate service node in the scene adaptation capability, and determine the candidate service node that meets the operation and maintenance task execution rhythm demand as the adaptive candidate node. The resource supply efficiency analysis includes response speed analysis, continuous supply stability analysis, and supply intensity adjustment capability analysis. The operation and maintenance task execution rhythm demand includes resource usage start time demand, continuous usage demand, and usage intensity change demand. For each operation and maintenance task, sort all adaptive candidate nodes in descending order according to the scene adaptation matching degree. The scene adaptation matching degree is determined according to the coincidence degree of business scene type and the matching degree of resource supply efficiency. The coincidence degree of business scene type is calculated by the number of scene type matches, and the matching degree of resource supply efficiency is calculated by the matching degree of resource supply efficiency parameters and operation and maintenance task execution rhythm demand parameters. Associate each operation and maintenance task with the sorted adaptive candidate nodes to form a preliminary adaptation relationship between the operation and maintenance task and the service node by establishing an association between the operation and maintenance task identifier and the adaptive candidate node list. Each preliminary adaptation relationship between the operation and maintenance task and the service node includes an operation and maintenance task identifier, an adaptive candidate node list, and a sorting basis. The sorting basis includes the coincidence degree calculation result of the business scene type and the coincidence degree calculation result of the resource supply efficiency. Integrate all preliminary adaptation relationships between operation and maintenance tasks and service nodes to organize the preliminary adaptation relationships in batch data integration mode to form a structured preliminary adaptation result. Each entry in the preliminary adaptation result includes preliminary adaptation relationship information between operation and maintenance tasks and service nodes. 5.The operation flow optimization method for operation and maintenance technical service of claim 1, wherein, The identification of the preliminary adaptation result that exists resource supply conflict adaptive combination, combined with the execution association relationship of the operation and maintenance task and the resource supply mode of the service node, resolves the resource supply conflict, and generates the final adaptation scheme of the operation and maintenance task and the service node, including: Use a loop traversal algorithm to process each entry in the preliminary adaptation result. Statistics of all operation and maintenance tasks associated with each service node, and the resource requirement attribute and execution association relationship of the operation and maintenance task. The statistical result is presented in table form. The table includes service node identifier, associated operation and maintenance task list, resource requirement attribute list, and execution association relationship list. Analyze the resource requirement attribute of the operation and maintenance task associated with each service node, focusing on the resource usage period and resource usage amount. Determine whether the operation and maintenance task has overlap in the resource usage period, and whether the usage demand of the same type of resource in the overlapping period exceeds the resource supply description in the service node dynamic capability profile. The resource supply description includes the available resource amount of the service node in the corresponding period. The combination of service nodes and corresponding operation and maintenance tasks that have overlapping resource usage periods and resource demand exceeding supply is determined as a resource supply conflict adaptation combination. This type of combination is marked by establishing a conflict identifier, which includes a conflict type and a conflict degree. The conflict type includes resource usage period overlap conflict and resource demand exceeding conflict. The conflict degree is determined by the proportion of resource demand exceeding amount to available resource amount. For each operation and maintenance task in the resource supply conflict adaptation combination, the execution association of the operation and maintenance task is extracted to identify whether there is a cooperative execution demand that can adjust the execution order. The cooperative execution demand that can adjust the execution order includes non-mandatory synchronous execution demand and flexible sequential execution demand. For the operation and maintenance task that has a cooperative execution demand that can adjust the execution order, the resource usage period of the operation and maintenance task is adjusted by referring to the resource supply description of the service node and the resource usage periods of other operation and maintenance tasks. The adjusted resource usage period is staggered with the resource usage periods of other operation and maintenance tasks. The adjusted resource usage period is recorded in the resource demand attribute of the operation and maintenance task. For the operation and maintenance task that cannot be resolved by adjusting the execution order, the adaptation candidate node list of the operation and maintenance task in the preliminary adaptation result is checked. An adaptation candidate node that is ranked last but has no resource supply conflict in the resource supply state is selected from the adaptation candidate node list. The operation and maintenance task is re-associated, and a new association mapping relationship is established during the re-association process. For the re-associated operation and maintenance task, it is checked again whether there is a resource supply conflict between the operation and maintenance task and other operation and maintenance tasks of the newly associated service node. The checking process repeats the resource usage period and resource usage amount analysis steps. The resolution of each resource supply conflict adaptation combination is recorded, including execution order adjustment or re-association of service nodes. The resolution basis is recorded, including execution association analysis results and service node resource supply description. The recording content includes conflict identifier, resolution method, resolution basis, and resolution time. All adaptation combinations after conflict resolution are integrated to organize the adaptation combinations and form a final adaptation scheme of operation and maintenance tasks and service nodes. Each adaptation combination in the final adaptation scheme is marked as having no resource supply conflict. 6.The operation flow optimization method for operation and maintenance technical service of claim 2, wherein, The scene features in the operation and maintenance task information set are classified, and operation and maintenance tasks with similar scene features are divided into the same scene group, including: The scene features of all operation and maintenance tasks in the operation and maintenance task information set are extracted. The business background, execution target, and information of the business system are extracted from each scene feature. Each scene feature is disassembled into business background keywords, execution target keywords, and business system keywords involved. The keywords are extracted by combining a word segmentation algorithm with keyword screening rules. The keyword screening rules include keyword length limitation and semantic correlation requirement. The similarity between the business background keywords of different operation and maintenance tasks is calculated. Operation and maintenance tasks containing the same or similar business background keywords are classified into the same business background subset. Similar keywords are determined based on industry general term association relationship, which is established through an industry term dictionary. In each business background subset, the execution target keyword is compared, and the comparison process adopts keyword semantic matching to classify the operation and maintenance tasks with consistent target direction described by the execution target keyword into the same target direction subset, and the target direction consistency is determined by the semantic similarity of the execution target keyword; In each target direction subset, the overlap degree of the business system keyword is analyzed, the overlap degree is calculated by the overlap number of the business system keyword, and the operation and maintenance tasks with the overlap number of the business system keyword reaching a set number threshold are classified into the same business system subset, and the standard of the overlap number reaching the set number threshold is determined by a preset rule; The business system subset is determined as the final scene group, a unique scene group identifier is assigned to each scene group, and the scene group identifier includes a combination of business background, target direction and business system abbreviation; The classification basis of each scene group is recorded, and the classification basis includes keyword similarity analysis result, target direction comparison result and business system overlap degree analysis result, the keyword similarity analysis result includes similarity calculation value and matching keyword list, the target direction comparison result includes semantic matching result and consistent keyword list, and the business system overlap degree analysis result includes overlap number and overlap keyword list; In the operation and maintenance task information set, a corresponding scene group identifier is added to each operation and maintenance task, a mapping relationship between the operation and maintenance task identifier and the scene group identifier is established, if any one operation and maintenance task meets the classification conditions of multiple scene groups at the same time, the scene group to which the operation and maintenance task belongs is determined through priority sorting, and the priority sorting is determined according to the business importance of the scene group; The scene groups are summarized to form a scene group list, and the scene group list includes scene group identifier, operation and maintenance task list in the group and classification basis, the operation and maintenance task list in the group includes operation and maintenance task identifier and operation and maintenance task name, and the classification basis includes complete analysis process and result data; The scene group list is stored in a target data file, and an index relationship between the scene group identifier and the target data file is established.

7. The operation flow optimization method for operation and maintenance technical service of claim 3, wherein, The device running state in the real-time running state data is combined to correct the scene adaptation capability description in the capability dimension information, including: Device running state information is extracted from the real-time running state data of the service node, and the device running state information includes running parameters of each hardware component, response of software modules and data interaction state between components; The running parameters of each hardware component are analyzed to identify whether there is a hardware component with running parameters deviating from the normal working range, and the normal working range is determined by the service node hardware configuration manual; It is judged whether each hardware component is related to the processing of a specific business scene through the association relationship between the hardware component and the business scene processing flow, if there is a hardware component related to the specific business scene and deviating from the normal working range, the scene adaptation capability description of the business scene in the service node capability dimension information is checked, the scene adaptation capability description includes support degree and processing efficiency of scene adaptation, and whether the current description still conforms to the actual situation is judged by comparing the difference between the current running parameters of the hardware component and the normal working parameters; The response of the software module is evaluated to identify whether there is a response delay or a functionally abnormal software module, the response delay is determined by the difference between the module response time and the normal response time reaching a set time threshold, and the functionally abnormal is determined by the difference between the module error rate and the normal error rate reaching a set proportion threshold; By the role of the software module in the business scenario processing flow, it is judged whether the software module is a core module for processing a specific business scenario, if there is a response delay or a functionally abnormal core module, the influence of the condition on the processing efficiency of the corresponding business scenario is evaluated, the influence evaluation adopts an efficiency decay model to calculate the decay degree of the processing efficiency, the processing efficiency includes the speed and quality of processing the business scenario, and the efficiency description in the scene adaptation capability is adjusted according to the decay degree, the efficiency description includes specific indicators of processing speed and processing quality; The data interaction state between the hardware components is checked to identify whether there is an interaction link with data transmission interruption or transmission error, the data transmission interruption is determined by the difference between the data interaction frequency and the normal interaction frequency reaching a set frequency threshold, and the transmission error is determined by the difference between the data interaction success rate and the normal success rate reaching a set proportion threshold; By the importance of the interaction link in the business scenario data flow process, it is judged whether the interaction link is a key path for the data flow of a specific business scenario, if the key path has a data interaction problem, the influence of the data interaction problem on the continuity of the specific business scenario is judged, the influence judgment adopts a continuity interruption risk evaluation model to calculate the interruption risk level, and a restriction condition description is added in the scene adaptation capability according to the interruption risk level, the restriction condition description includes the prerequisite and risk prompt of scene execution; According to the analysis results of the hardware components, the software modules and the data interaction state, the scene adaptation capability description in the capability dimension information is corrected, the correction content includes the range adjustment of the adaptation scene, the efficiency description update or the addition of the restriction condition, the range adjustment of the adaptation scene includes the addition of the adaptation scene or the removal of the non-adaptation scene, the efficiency description update includes the adjustment of the processing speed and the processing quality indicators, and the addition of the restriction condition includes the supplement of the prerequisite and the risk prompt of the scene execution; The corrected scene adaptation capability description is updated to the service node dynamic capability portrait, and a correction log is generated, the correction log includes the correction time, the correction content, the correction basis and the correction executor. 8.The operation flow optimization method for operation and maintenance technical service of claim 4, wherein, The resource supply efficiency of the candidate service node in the scene adaptation capability is analyzed, and the candidate service node whose resource supply efficiency meets the operation and maintenance task execution rhythm demand is determined as an adaptation candidate node, including: The resource use rhythm description is extracted from the resource demand attribute of the operation and maintenance task, the resource use rhythm description includes the start time of resource use, the preparation time, the duration and the use intensity change trend, the start time of resource use includes the time point of starting to use the resource and the preparation time, the duration includes the shortest duration and the longest duration of resource use, and the use intensity change trend includes the rising stage, the stable stage and the falling stage of resource use intensity; The data interaction state between the hardware components is checked to identify whether there is an interaction link with data transmission interruption or transmission error, the data transmission interruption is determined by the difference between the data interaction frequency and the normal interaction frequency reaching a set frequency threshold, and the transmission error is determined by the difference between the data interaction success rate and the normal success rate reaching a set proportion threshold; By the importance of the interaction link in the business scenario data flow process, it is judged whether the interaction link is a key path for the data flow of a specific business scenario, if the key path has a data interaction problem, the influence of the data interaction problem on the continuity of the specific business scenario is judged, the influence judgment adopts a continuity interruption risk evaluation model to calculate the interruption risk level, and a restriction condition description is added in the scene adaptation capability according to the interruption risk level, the restriction condition description includes the prerequisite and risk prompt of scene execution; According to the analysis results of the hardware components, the software modules and the data interaction state, the scene adaptation capability description in the capability dimension information is corrected, the correction content includes the range adjustment of the adaptation scene, the efficiency description update or the addition of the restriction condition, the range adjustment of the adaptation scene includes the addition of the adaptation scene or the removal of the non-adaptation scene, the efficiency description update includes the adjustment of the processing speed and the processing quality indicators, and the addition of the restriction condition includes the supplement of the prerequisite and the risk prompt of the scene execution; The corrected scene adaptation capability description is updated to the service node dynamic capability portrait, and a correction log is generated, the correction log includes the correction time, the correction content, the correction basis and the correction executor. Extract resource supply efficiency information in the corresponding scene from the dynamic capability profile of the candidate service node, the resource supply efficiency information contains the response speed, the continuous supply stability and the supply intensity adjustment capability of the resource supply, the response speed contains the preparation time and the start time of the resource supply, the continuous supply stability contains the fluctuation range and the interruption frequency of the resource supply, and the supply intensity adjustment capability contains the adjustment range and the adjustment speed of the resource supply intensity; Compare the start time in the resource use rhythm description with the response speed in the resource supply efficiency information, and determine whether the response speed of the resource supply can meet the start time demand of the resource use through time comparison algorithm, that is, whether the resource supply can be ready before the start time, and the judgment standard of the preparation is that the sum of the preparation time and the start time of the resource supply is less than the advance preparation time threshold of the resource use; Compare the continuous duration in the resource use rhythm description with the continuous supply stability in the resource supply efficiency information, and determine whether the resource supply can remain stable without interruption or large fluctuation within the continuous duration of the resource use through duration comparison and stability analysis, and the judgment standard of stability is that the interruption frequency of the resource supply is lower than the preset interruption frequency threshold and the fluctuation amplitude is less than or equal to the allowed fluctuation amplitude threshold; Compare the use intensity change trend in the resource use rhythm description with the supply intensity adjustment capability in the resource supply efficiency information, and determine whether the resource supply intensity can be adjusted to follow the change of the use intensity through trend comparison and adjustment capability analysis, so that the change trends are consistent, and the judgment standard of the consistent change trend is that the adjustment range of the resource supply intensity covers the change range of the use intensity and the adjustment rate of the resource supply intensity is greater than or equal to the change rate threshold of the use intensity; If the response speed, the continuous supply stability and the supply intensity adjustment capability of the resource supply all meet the resource use rhythm demand, establish the association between the candidate service node identifier and the adaptation identifier, and mark the candidate service node as an adaptation candidate node; If any one or more does not meet the demand, record the specific aspects that do not meet the demand, which include that the response speed cannot meet the start time, the continuous supply stability cannot cover the continuous duration, or the supply intensity adjustment capability cannot match the use intensity change, and record the specific data difference that does not meet the demand, which includes the difference between the actual value and the required value; For the service node determined as the adaptation candidate node, record the specific performance of the resource supply efficiency of the service node that meets the demand, which includes the specific data of the response speed that meets the start time, the specific data of the continuous supply stability that covers the continuous duration, and the specific data of the supply intensity adjustment capability that matches the use intensity change; Sort all adaptation candidate nodes to form a preliminary list of adaptation candidate nodes for each operation and maintenance task, which contains the service node identifier and the efficiency performance description that meets the demand, and the efficiency performance description that meets the demand contains complete qualified data and analysis results. 9.The operation flow optimization method for operation and maintenance technical service of claim 5, wherein, The operation and maintenance task with adjustable execution order is adjusted in reference to the resource supply description of the service node and the resource use period of other operation and maintenance tasks, and the resource use period of the operation and maintenance task is adjusted, including: From the operation and maintenance task execution association relationship in the resource supply conflict adaptation combination, the basis for adjusting the execution order is extracted, the basis for adjusting the execution order includes the order flexibility description and the cooperative execution requirement description in the execution association relationship, the order flexibility description includes whether the execution order is allowed to be adjusted, and the cooperative execution requirement description includes whether it is mandatory synchronous execution; The current resource use period of the operation and maintenance task with adjustable execution order is determined, the current resource use period is obtained by querying the resource demand attribute of the operation and maintenance task, and the current resource use period includes the start time period and the end time period of resource use; The resource use period of other operation and maintenance tasks on the same service node is analyzed, the resource use period of other operation and maintenance tasks is obtained by traversing the resource demand attribute of all operation and maintenance tasks associated with the same service node, and the time period overlapping with the current resource use period is found, the specific time range of the overlap is determined through a time interval comparison algorithm, and the specific time range includes the start time and the end time of the overlapping time period; In combination with the execution association relationship of the operation and maintenance task, the earliest start time period and the latest end time period of the operation and maintenance task with adjustable execution order are determined, the earliest start time period is determined by the completion time of the preceding task in the execution association relationship, and the latest end time period is determined by the start time of the subsequent task in the execution association relationship; Between the earliest start time period and the latest end time period, a time range without overlap with the resource use period of other operation and maintenance tasks is selected, a suitable time range is selected from the available time range as the adjusted resource use period, and the adjusted resource use period includes a new start time period and an end time period; If there are multiple non-overlapping time ranges, a time range with a time span similar to the original resource use period is preferentially selected, and the judgment standard of the time span similar to the original time span is that the difference between the new time span and the original time span accounts for less than a preset proportion threshold of the original time span; The start time period and the end time period of the adjusted resource use period are determined, the resource use period description in the resource demand attribute of the operation and maintenance task is updated, the resource use period data before and after the update is recorded to form an update log, the update log includes the update time, the operation and maintenance task identifier, the original period data and the new period data; The related cooperative execution operation and maintenance task is notified, and the adjusted resource use period is informed, the notification mode adopts a data pushing mechanism, and the adjusted resource use period data is pushed to the associated data node of the related cooperative execution operation and maintenance task; The adjustment content of the resource use period is recorded, the adjustment content includes the period before the adjustment, the period after the adjustment and the adjustment basis, and the adjustment basis includes the execution association relationship analysis result and the other operation and maintenance task period analysis result, the execution association relationship analysis result includes the basis data of the adjustable order, and the other operation and maintenance task period analysis result includes the overlapping time period data and the available time range data.

10. An operation flow optimization system applied to operation and maintenance technical service, characterized in that, The application comprises a processor and a computer readable storage medium, the computer readable storage medium stores machine executable instructions, and the machine executable instructions are executed by the processor to realize the operation and maintenance process optimization method applied to operation and maintenance technical service in any one of claims 1-9.

Images