Marketing management information platform-based intelligent operation and maintenance monitoring method

Abstract

The invention discloses an intelligent operation and maintenance monitoring method based on a marketing management information platform, particularly relates to the field of intelligent monitoring, and comprises a data acquisition module, a data processing module, an index generation module, an index grading module, an intelligent decision module and an automatic optimization module. According to the method, a stability-efficiency-cost three-dimensional evaluation model is constructed, index quantification is realized by adopting a dynamic weighting algorithm, and the false alarm rate is reduced compared with a traditional static threshold value alarm mode; unified acquisition of heterogeneous data is realized through a distributed sensor network, and compared with a traditional architecture, quick response and positioning are realized; the resource bottleneck during the marketing activity period can be predicted in advance by combining the intelligent decision-making module with historical data pattern recognition, and active prevention is realized; and the quantitative evaluation of the severity of the fault is realized through the index grading module, a multi-level automatic response strategy from slight abnormity to major accidents is supported, and the operation and maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and more specifically, to an intelligent operation and maintenance monitoring method based on a marketing management information platform. Background Art

[0002] With the acceleration of enterprise digital transformation, marketing management information platforms have become the core infrastructure supporting the operations of modern enterprises. Current mainstream systems generally adopt microservice architecture and process more than 1 billion transaction data per day, which poses severe challenges to traditional operation and maintenance models.

[0003] Traditional technologies use sensor networks to collect hardware performance and business metrics in real time. Log analysis relies on tools to parse error logs, identify system anomalies, calculate single quantitative metrics, and display alerts through visual dashboards. Alerts are triggered by predefined threshold rules, and notifications to operations and maintenance personnel are sent through a work order system. The operation and maintenance data platform integrates log, performance, and topology data, and cross-system data lineage tracing relies on manual configuration. This presents the following problems:

[0004] Insufficient business elasticity: Static alarm thresholds cannot adapt to sudden increases in marketing activity traffic, resulting in a high false alarm rate.

[0005] Inefficient fault location: Network, hardware, and application data belong to independent systems, and manually correlating logs to locate the root cause takes a long time on average.

[0006] Lack of predictive capabilities: The existing system can only perform post-event analysis and cannot predict resource bottlenecks during marketing activities.

[0007] High operation and maintenance costs: Sensor deployment and repetitive work order processing account for a large proportion of the budget, and dedicated operation and maintenance personnel spend a large proportion of their time handling inefficient alarms.

[0008] Therefore, a method of dynamic adaptation, rapid positioning, early prediction and low-cost monitoring is needed to solve the above problems. Summary of the Invention

[0009] In view of this, the present invention proposes an intelligent operation and maintenance monitoring method based on a marketing management information platform. This solution deploys a distributed sensor network to collect multi-dimensional operation data of the marketing management information platform in real time, and extracts three core indicators of operation and maintenance stability, operation and maintenance efficiency, and operation and maintenance cost through a data processing module. Then, an operation and maintenance stability index, efficiency index, and cost index that can be dynamically weighted and calculated are generated. Based on the index classification, an intelligent decision-making module is triggered to realize fault warning and root cause location. Finally, a "monitoring-evaluation-optimization" closed loop is formed through the automatic optimization module, thereby achieving the collaborative goals of improving platform stability, optimizing operation and maintenance efficiency, and reducing costs.

[0010] The purpose of the present invention can be achieved by the following technical solution: an intelligent operation and maintenance monitoring method based on a marketing management information platform, specifically comprising the following steps:

[0011] S1. Deploy a distributed sensor network to collect multi-dimensional data from the marketing management information platform in real time through the data acquisition module;

[0012] S2. Processing the collected data through the data processing module to obtain stability-related indicators, operation and maintenance efficiency-related indicators, and operation and maintenance cost-related indicators;

[0013] S3. Generate an operation and maintenance stability index, an operation and maintenance efficiency index, and an operation and maintenance cost index through an index generation module based on the obtained relevant indicators;

[0014] S4. Use the index grading module to classify the operation and maintenance stability index, operation and maintenance efficiency index, and operation and maintenance cost index into different levels;

[0015] S5. Use the intelligent decision-making module to warn of potential failures in advance based on index grading, locate the root cause of the problem, and take appropriate measures;

[0016] S6. Continuously optimize the operation and maintenance monitoring methods through the automatic optimization module to improve the stability and operation and maintenance efficiency of the platform and reduce operation and maintenance costs.

[0017] Technical effects and advantages of the present invention:

[0018] 1. The present invention constructs a three-dimensional evaluation model of "stability-efficiency-cost" and adopts a dynamic weighting algorithm to quantify indicators, which reduces the false alarm rate compared with the traditional static threshold alarm method.

[0019] 2. The present invention realizes the unified collection of heterogeneous data such as marketing system performance parameters, business indicators, resource consumption, etc. through a distributed sensor network, breaking through the data island problem of traditional single monitoring tools and achieving rapid response and positioning compared to traditional architectures.

[0020] 3. The present invention uses an intelligent decision-making module combined with historical data pattern recognition to predict resource bottlenecks during marketing activities in advance, achieving a shift from "passive repair" to "active prevention."

[0021] 4. The present invention implements quantitative assessment of fault severity through an index grading module, supports multi-level automatic response strategies from minor anomalies to major accidents, and reduces operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Attachment Figure 1 This is a system block diagram of the present invention.

[0023] Attachment Figure 2 Flowchart of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0025] As attached Figure 1 The intelligent operation and maintenance monitoring method based on the marketing management information platform shown includes a data acquisition module, a data processing module, an index generation module, an index grading module, an intelligent decision-making module and an automatic optimization module.

[0026] like Figure 2 As shown, the specific implementation steps of the present invention include the following steps:

[0027] S1. Deploy a distributed sensor network to collect multi-dimensional data of the marketing management information platform in real time through the data acquisition module.

[0028] It should be noted that the multidimensional data includes relevant data used to calculate system availability, business continuity rate and resource stability rate, fault handling efficiency, resource utilization and manpower efficiency, hardware cost, software cost, manpower cost and energy consumption cost.

[0029] S2. The collected data is processed by the data processing module to obtain stability-related indicators, operation and maintenance efficiency-related indicators, and operation and maintenance cost-related indicators.

[0030] It should be noted that the stability-related indicators include system availability, business continuity rate and resource stability rate.

[0031] It should be explained that system availability = (1-service unavailable time / total service time)*100%, and the total service time is calculated based on natural months / years.

[0032] It should be explained that business continuity rate = 1-[transaction failure rate * α + proportion of transactions affected by failures * β], where transaction failure rate = 100% * failed transactions / total transactions, proportion of transactions affected by failures = 100% * failed transactions during the failure period / total transactions in the period, α ranges from 0.65 to 0.7, and β ranges from 0.3 to 0.35.

[0033] It should be explained that resource stability rate = 1-[hardware failure rate * 0.5 + network packet loss rate * 0.5], where hardware failure rate = 100% * number of failed servers / total number of servers, network packet loss rate = 100% * number of probe packet losses / total number of probe packets.

[0034] It should be noted that the operation and maintenance efficiency-related indicators include fault handling efficiency, resource utilization and manpower efficiency.

[0035] It should be explained that fault handling efficiency is the weighted sum of the standardized value of average repair time and the standardized value of automation processing rate, where average repair time = total fault repair time / number of faults, with a weight between 0.55-0.6; automation processing rate = 100% * number of automatically repaired faults / total number of faults, with a weight between 0.4-0.45.

[0036] It should be explained that the lower the deviation / idle rate, the better. The inverse needs to be taken and then normalized. The resource utilization is the weighted sum of the normalized value of the inverse of the peak deviation and the normalized value of the inverse of the limit rate, where the deviation = |actual resource peak utilization - predicted peak utilization| / predicted peak utilization, with a weight between 0.65-0.7; the idle rate = (1-actual average utilization / total resource capacity) * 100%, with a weight between 0.3-0.35.

[0037] It should be explained that the lower the repetition rate, the better, and the inverse should be taken. Human efficiency is the weighted sum of the standardized value of the number of systems per capita and the standardized value of the inverse of the repetition rate, where the number of systems per capita = the total number of managed systems / the number of people in the operation and maintenance team, with a weight between 0.55-0.6; the repetition rate = 100% * the number of repeated work orders / the total number of work orders, with a weight between 0.4-0.45.

[0038] It should be noted that the operation and maintenance cost-related indicators include hardware costs, software costs, labor costs and energy costs.

[0039] It should be specifically explained that the hardware cost calculation is specifically as follows:

[0040] Annual depreciation = total equipment purchase price * (1-residual value rate) / expected useful life, where the residual value rate is the proportion of the equipment's residual value after it is scrapped, and the useful life is the length of time a general-purpose server / network equipment is in use;

[0041] Annual maintenance fee = total equipment purchase price * maintenance rate;

[0042] Total hardware cost = depreciation + maintenance fee.

[0043] The software cost calculation is specifically as follows:

[0044] Annual license fee = ∑(unit price of each software license * number of licenses);

[0045] Open source support fee = open source software equivalent commercial license fee * (1-savings rate) + additional labor costs, where the savings rate is the percentage of license fees saved compared to commercial software, and the additional labor costs are the additional manpower input for open source software operations and maintenance.

[0046] Total software cost = commercial license fee + open source support fee.

[0047] The labor cost calculation is specifically as follows:

[0048] Annual internal cost = ∑(operation and maintenance personnel annual salary + benefits + training fees), where annual salary includes basic salary and performance, and benefits include social security and provident fund;

[0049] Annual outsourcing fee = fixed service fee + volume-based billing;

[0050] Total labor cost = internal cost + outsourcing fee.

[0051] The energy consumption cost calculation is specifically as follows:

[0052] Annual electricity cost = total server power * 24 hours * 365 days * electricity price, where total server power = single server power consumption * number of servers;

[0053] Cooling cost = server electricity cost * cooling ratio;

[0054] Total energy cost = server electricity cost + cooling cost.

[0055] It needs to be explained that the standardization formula is specifically:

[0056]

[0057] Where Z is the normalized value, X is the original data value, μ is the arithmetic mean of the data set, and σ is the standard deviation of the data set.

[0058] The mean calculation is as follows:

[0059]

[0060] The standard deviation is calculated as follows:

[0061]

[0062] Where n is the number of original data values, and Xi is any original data value.

[0063] S3. Generate an operation and maintenance stability index, an operation and maintenance efficiency index, and an operation and maintenance cost index through an index generation module based on the obtained relevant indicators.

[0064] It should be noted that the operation and maintenance stability index is specifically:

[0065] D=υ1*X+υ2*Y+υ3*Z;

[0066] D is the operation and maintenance stability index, which is the core quantitative indicator for evaluating the health status of the marketing management information platform.

[0067] It should be explained that X is the standardized value of system availability. System availability refers to the accessibility of the platform's basic services. It is measured by the proportion of service uptime and reflects the reliability of the infrastructure. System availability is the bottom line indicator of stability, but a single dimension cannot cover the business impact. A high weight ensures that basic services do not go down, but it needs to be balanced with other dimensions. The value range of weight υ1 is between 0.35 and 0.4.

[0068] Y is the standardized value of the business continuity rate. The business continuity rate is the fault resistance capability of key business links, such as transaction success rate and peak performance, which is directly related to user benefits. The core value of the marketing platform lies in business smoothness. Failures during the promotion period may lead to significant losses, so it is given the highest weight. The value range of weight υ2 is between 0.4-0.45.

[0069] Z is the standardized value of the resource stability rate. The resource stability rate represents the health status of hardware / network resources, including CPU failure rate and packet loss rate, and is the support layer for availability. Resource problems are often exposed through availability or business indicators. As a front-end warning layer, its weight is slightly lower but indispensable. The value range of weight υ3 is between 0.25 and 0.3.

[0070] It should be noted that the operation and maintenance efficiency index is specifically:

[0071] W=G*ω1+F*ω2+L*ω3;

[0072] W is the operation and maintenance efficiency index, which is a comprehensive indicator to quantify the overall effectiveness of the operation and maintenance team.

[0073] It should be explained that G is the standardized value of fault handling efficiency. Fault handling efficiency measures the ability of the operation and maintenance team to quickly restore services. Core indicators include mean repair time and automated processing rate, which directly affect system availability and user experience. Fault recovery delays can lead to business losses. Fault handling efficiency is directly related to business continuity rate. Fault response speed reflects the core value of operation and maintenance. Therefore, the value range of weight ω1 is between 0.4 and 0.45.

[0074] F is the standardized value of resource utilization. Resource utilization reflects the effective use of hardware resources such as CPU and memory, and includes peak deviation and idle rate indicators. Resource overload can cause failures, while idleness causes cost waste. It is necessary to balance efficiency and stability. Resource utilization optimization can reduce operating costs and affect the long-term stability of the system. Therefore, the value range of the weight ω2 is between 0.3 and 0.35.

[0075] L is the standardized value of human efficiency. Human efficiency evaluates the output efficiency of the operation and maintenance team, including the number of operation and maintenance systems per capita and the proportion of duplicate work orders, reflecting the organization's management level and knowledge accumulation ability. A highly efficient human efficiency team can reduce repetitive work, but this requires process standardization and long-term accumulation. Its weight ω3 ranges from 0.2 to 0.25.

[0076] It should be noted that the operation and maintenance cost index is specifically:

[0077] C=J*δ1+R*δ2+P*δ3+N*δ4;

[0078] Where C is the operation and maintenance cost index, which achieves systematic evaluation and dynamic control of operation and maintenance cost efficiency by quantifying and integrating multi-dimensional cost data.

[0079] It should be explained that J is the standardized value of hardware cost. Hardware cost includes depreciation and maintenance fees of servers and network equipment. Hardware cost is the basic supporting cost. Equipment depreciation and maintenance are fixed costs and cannot be optimized in the short term. In addition, the hardware failure rate directly affects the business continuity rate and requires high-weight monitoring. Therefore, the hardware cost weight δ1 is the highest, and the value range is between 0.35 and 0.4.

[0080] R is the standardized value of software cost. Software cost is an alternative cost, including annual license fees for commercial software and open source technical support fees. The weight is compressible under the cloud-native architecture, and the unit cost is significantly reduced after large-scale deployment. Therefore, the software cost weight δ2 is the lowest, and the value range is between 0.15-0.2.

[0081] P is the standardized value of labor costs, which include internal team salaries and outsourcing service fees. It is a performance-determining cost. Tool automation can reduce labor requirements. The weight of finance and government affairs needs to be increased to meet the weight of security operations and maintenance. The value range of δ3 is between 0.2 and 0.25.

[0082] N is the standardized value of energy consumption cost. Energy consumption cost includes electricity consumption and cooling system energy consumption. It is a continuous operating cost with long-term consumption properties. The electricity and cooling systems operate 24 hours a day and are highly technically sensitive. Green electricity technology can significantly reduce the weight. The value range of weight δ4 is between 0.2-0.25.

[0083] S4. The operation and maintenance stability index, operation and maintenance efficiency index and operation and maintenance cost index are divided into different levels through the index grading module.

[0084] It should be noted that the operation and maintenance stability index is divided into the following levels:

[0085] When D ≥ d1, the grade is excellent;

[0086] When d2≤D<d1, the grade is normal;

[0087] When D<d2, the level is high risk.

[0088] The value range of d1 is between 89-90, and the value range of d2 is between 69-70

[0089] It should be noted that the operation and maintenance efficiency index is divided into the following levels:

[0090] When W ≥ w1, the grade is excellent;

[0091] When w2≤W<w1, the level is normal;

[0092] When W<w2, the level is high-risk.

[0093] The value range of w1 is between 84 and 85, and the value range of w2 is between 69 and 70.

[0094] It should be noted that the operation and maintenance cost index is divided into the following levels:

[0095] When C≤c1, the grade is excellent, indicating efficient cost control;

[0096] When c1<C≤c2, the grade is normal, indicating that the cost meets the industry standard;

[0097] When c2<C≤c3, the level is warning, indicating that some modules are overspent;

[0098] When C>c3, the level is high-risk, indicating that systemic costs are out of control.

[0099] The value range of c1 is between 85 and 86, the value range of c2 is between 100 and 101, and the value range of c3 is between 115 and 116.

[0100] S5. Based on the index classification, the intelligent decision-making module can provide early warning of potential failures, locate the root cause of the problem and take corresponding measures.

[0101] It should be noted that the corresponding measures taken according to the level of operation and maintenance stability index are as follows:

[0102] When the level is excellent, a hardware health scan is performed weekly, using tools to detect disk bad sectors and check server fan speed and temperature thresholds; a network link stress test is performed monthly, using tools to simulate peak traffic.

[0103] Based on the model, resource demand is predicted for the next three months. When the predicted CPU usage exceeds the threshold, a cloud resource expansion application is submitted in advance. Database tablespace monitoring is also used, and cleanup is triggered when the threshold is exceeded.

[0104] When the level is normal, performance optimization is performed. For services with pause times greater than the threshold, a garbage collector is used and corresponding parameters are set.

[0105] Identify the top ten most time-consuming queries in the slow query log, use tools to analyze the execution plan, and then add composite indexes.

[0106] For memory utilization that is continuously greater than the threshold, a taint mark is added and the device is migrated.

[0107] For areas where access latency is greater than the threshold, increase the edge node cache duration.

[0108] When the risk level is high, pre-configured circuit breaker rules are immediately activated to shut down non-core functions.

[0109] Switch all resolution weights from the primary data center to the disaster recovery center, and verify the synchronization status;

[0110] By locating the timeout nodes in the microservice call chain, the thread pool capacity is expanded for services whose response time exceeds the threshold.

[0111] When data inconsistency is detected, the tool triggers the incremental data synchronization task and configures the expiration time.

[0112] It should be noted that the corresponding measures taken according to the level of operation and maintenance efficiency index are as follows:

[0113] When the rating is excellent, duplicate tickets are automatically marked and associated with the root cause. For example, tickets related to insufficient disk space are prompted to optimize the log cleanup strategy.

[0114] Visualize the distribution of team technical capabilities and push targeted training content. For example, automatically recommend courses when a team member is inefficient in troubleshooting.

[0115] Generate comprehensive reports daily covering key development / test / operations metrics, highlighting cross-team blocking issues, such as release delays linked to insufficient test environments;

[0116] Automated acceptance pipeline: After operations and maintenance propose infrastructure requirements, developers must complete performance testing and acceptance in the pipeline before going online.

[0117] When the level is normal, predictive capacity expansion is performed. Based on model analysis of business cycles, such as e-commerce promotion seasons, computing nodes are expanded in advance and the connection pool is preheated.

[0118] Isolate mixed loads, deploy online services and processing tasks to different resource pools, and limit the maximum resource usage of processing tasks;

[0119] Intelligent recycling: For VMs whose CPU utilization is below a certain value for several consecutive days, migration notifications are automatically sent. If there is no feedback after 24 hours, the VMs are shut down and archived.

[0120] Tiered cold data storage automatically migrates logs that have not been accessed for six months to object storage, reducing storage costs.

[0121] When the level is high-risk, deploy an intelligent diagnostic robot that automatically identifies the fault type and matches the repair script based on the historical work order training model;

[0122] Establish a knowledge base linkage mechanism. When automated repair fails, similar case solutions are automatically pushed to the on-duty personnel's terminal, requiring them to provide a new solution within 15 minutes.

[0123] Monitor critical paths, implant probes in transaction links, track core indicators of the database connection pool in real time, and automatically trigger backup channel switching when thresholds are exceeded;

[0124] A hierarchical response mechanism is implemented, where the fault level response team includes architects, operations and maintenance, and development personnel, with different level limits and response times.

[0125] It should be noted that the corresponding measures taken according to the level of operation and maintenance cost index are as follows:

[0126] When the rating is excellent, an automated audit pipeline is established to generate resource utilization / idle rate heat maps every week;

[0127] Implement hybrid cloud elastic scheduling, automatically migrate non-core businesses to instances, and save cloud costs;

[0128] Memory tiered storage automatically downgrades cold data to persistent memory, reducing memory procurement costs.

[0129] When the level is normal, a hardware health model is introduced to predict hard drive failures using vibration and temperature sensors, allowing hard drive replacement three months in advance to reduce sudden failure rates.

[0130] Implement a plan to reuse old equipment, retiring servers and converting them into edge computing nodes to extend their service life;

[0131] Build a knowledge graph system to convert troubleshooting experience into searchable cases and reduce duplicate work orders;

[0132] Implement the T-shaped talent plan, so that core operation and maintenance personnel can master at least three types of cross-domain skills and increase the number of operation and maintenance systems per capita.

[0133] When the level reaches the warning level, a liquid cooling modification kit is deployed to complete the high-density cabinet modification within 48 hours;

[0134] Enable dynamic voltage regulation to reduce CPU voltage during off-peak hours to save power;

[0135] Establishing a temporary cost committee to freeze overspending project budgets and review procurement requests daily;

[0136] Initiate supplier renegotiations and leverage historical purchasing data to secure tiered discounts.

[0137] When the risk level is high, split the application into microservices and restructure the monolithic application to reduce operation and maintenance complexity.

[0138] Implement chaos engineering to identify redundant resources through fault injection and reduce annual hardware purchases;

[0139] Use computing power futures contracts to lock in cloud computing resource prices for the next six months and avoid the risk of price increases during peak seasons;

[0140] Establish an equipment sharing pool and jointly purchase high-end network equipment with peer companies to share costs.

[0141] S6. Continuously optimize the operation and maintenance monitoring methods through the automatic optimization module to improve the stability and operation and maintenance efficiency of the platform and reduce operation and maintenance costs.

[0142] The implementation scheme of the automatic optimization module is specifically as follows:

[0143] Data feedback closed-loop mechanism: Establish a closed-loop verification system of indicators, decisions, and results, and input the execution results of the S5 decision module, such as fault repair time and resource allocation accuracy, back into the indicator calculation model;

[0144] Use the A / B testing framework to compare the actual effects of different optimization strategies and retain the optimal algorithm path.

[0145] Dynamic weight tuning technology: Based on an algorithm, the weight parameters used in the S3 index calculation are periodically optimized, and a representative parameter combination is selected through uniform design to establish a prediction model.

[0146] A reinforcement learning mechanism is introduced to automatically adjust the weights of stability, efficiency, and cost according to changes in business scenarios.

[0147] Multi-source data fusion and optimization: Connecting to the big data platform enables correlation analysis between operation and maintenance data and business data such as marketing activity flow and transaction peaks;

[0148] Use online analytical processing technology to cross-mine high-dimensional static data such as historical fault libraries and low-dimensional dynamic data real-time monitoring streams.

[0149] Self-learning capability development: Build a fault prediction model library, annotate historical fault features through supervised learning, and discover potential abnormal patterns through unsupervised learning;

[0150] Regularly clean up redundant monitoring strategies automatically and retain core monitoring points based on data similarity assessment.

[0151] Resource scheduling optimization: Combined with aggressive strategy mechanisms, fault-tolerant deployment is implemented for non-critical path components;

[0152] Through containerized deployment, elastic scaling of computing resources is achieved, and instance size is automatically adjusted according to predicted load.

[0153] Secondly: The drawings of the embodiments disclosed in the present invention only involve structures related to the embodiments disclosed in the present invention. Other structures may refer to conventional designs. The same embodiment and different embodiments of the present invention may be combined with each other without conflict.

[0154] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent operation and maintenance monitoring method based on a marketing management information platform, characterized in that: include: S1. Deploy a distributed sensor network to collect multi-dimensional data from the marketing management information platform in real time through the data acquisition module; S2. Processing the collected data through the data processing module to obtain stability-related indicators, operation and maintenance efficiency-related indicators, and operation and maintenance cost-related indicators; S3. Generate an operation and maintenance stability index, an operation and maintenance efficiency index, and an operation and maintenance cost index through an index generation module based on the obtained relevant indicators; S4. Use the index grading module to classify the operation and maintenance stability index, operation and maintenance efficiency index, and operation and maintenance cost index into different levels; S5. Use the intelligent decision-making module to warn of potential failures in advance based on index grading, locate the root cause of the problem, and take appropriate measures; S6. Continuously optimize the operation and maintenance monitoring methods through the automatic optimization module to improve the stability and operation and maintenance efficiency of the platform and reduce operation and maintenance costs.

2. The intelligent operation and maintenance monitoring method based on the marketing management information platform according to claim 1 is characterized by: The stability-related indicators include system availability, business continuity rate and resource stability rate; The operation and maintenance efficiency related indicators include fault handling efficiency, resource utilization and manpower efficiency; The operation and maintenance cost related indicators include hardware cost, software cost, labor cost and energy consumption cost.

3. The intelligent operation and maintenance monitoring method based on the marketing management information platform according to claim 1 is characterized by: The operation and maintenance stability index is specifically: D=υ1*X+υ2*Y+υ3*Z; Where D is the operation and maintenance stability index, X is the standardized value of system availability, υ1 is the weight of system availability; Y is the standardized value of business continuity rate, υ2 is the weight of business continuity rate; Z is the standardized value of resource stability rate, υ3 is the weight of resource stability rate; The specific normalization formula is: Where Z is the standardized value, X is the original data value, μ is the arithmetic mean of the data set, and σ is the standard deviation of the data set; The mean calculation is as follows: The standard deviation is calculated as follows: Where n is the number of original data values, and Xi is any original data value.

4. The intelligent operation and maintenance monitoring method based on the marketing management information platform according to claim 1 is characterized by: The operation and maintenance efficiency index is specifically: W=G*ω1+F*ω2+L*ω3; Where W is the operation and maintenance efficiency index, G is the standardized value of fault handling efficiency, and ω1 is the fault handling efficiency weight; F is the standardized value of resource utilization, and ω2 is the resource utilization weight; L is the standardized value of human efficiency, and ω3 is the human efficiency weight.

5. The intelligent operation and maintenance monitoring method based on the marketing management information platform according to claim 1 is characterized in that: The operation and maintenance cost index is specifically: C=J*δ1+R*δ2+P*δ3+N*δ4; Where C is the operation and maintenance cost index, J is the standardized value of hardware cost, and δ1 is the hardware cost weight; R is the standardized value of software cost, and δ2 is the software cost weight; P is the standardized value of labor cost, and δ3 is the labor cost weight; N is the standardized value of energy consumption cost, and δ4 is the energy consumption cost weight.

6. The intelligent operation and maintenance monitoring method based on the marketing management information platform according to claim 1 is characterized by: The operation and maintenance stability index is divided into the following levels: When D ≥ d1, the grade is excellent; When d2≤D<d1, the grade is normal; When D < d1, the level is high risk; The operation and maintenance efficiency index is divided into the following levels: When W ≥ w1, the grade is excellent; When w2≤W<w1, the level is normal; When W < w2, the level is high-risk; The operation and maintenance cost index is divided into the following levels: When C≤c1, the grade is excellent; When c1<C≤c2, the grade is normal; When c2<C≤c3, the level is warning; When C>c3, the level is high risk.

Images