CN119762084B

CN119762084B - Intelligent product after-sale management method and system

Info

Publication number: CN119762084B
Application number: CN202510258429.9A
Authority: CN
Inventors: 胡远雄; 李明; 苏志雄; 马雪梅
Original assignee: Charisma Technology Co ltd
Current assignee: Charisma Technology Co ltd
Priority date: 2025-03-06
Filing date: 2025-03-06
Publication date: 2025-06-20
Anticipated expiration: 2045-03-06
Also published as: CN119762084A

Abstract

The invention relates to the technical field of product management, in particular to an intelligent after-sale product management method and system. The method comprises the steps of obtaining an after-sales intelligent product problem record set, wherein the after-sales intelligent product problem record set comprises at least one or more after-sales intelligent product problem records, quantifying customer demand on the after-sales intelligent product problem records to generate after-sales intelligent product demand emergency degree data, comparing the after-sales intelligent product demand emergency degree data with a preset standard demand emergency degree threshold to generate priority response demand data and delay response demand data, and performing fault handling association analysis on the after-sales intelligent product problem record set through the priority response demand data and the delay response demand data to generate a product fault handling automatic solution. The invention improves the efficiency and the efficiency of after-sales management through demand quantification, fault association analysis, feedback diagnosis and trend evaluation.

Description

Intelligent product after-sale management method and system

Technical Field

The invention relates to the technical field of product management, in particular to an intelligent after-sale product management method and system.

Background

Along with the rapid development of modern technology, intelligent products are widely applied to various industries. From early single-function electronic devices to today's multi-function smart devices, the complexity of the product and the diversity of user needs has increased significantly. Traditional after-sales management methods, such as manual repair, telephone support and paper recording, are increasingly unable to meet the user's expectations for efficient service due to inefficiency, slow response and susceptibility to errors. To optimize after-market management processes, businesses begin introducing informationized technologies, such as customer relationship management systems (CRM) and enterprise resource planning systems (ERP), which enable partial automation of processes by digitally storing and managing customer information. However, this approach still is based on static data processing and cannot dynamically respond to the large real-time data demands of intelligent products. In recent years, with the maturation of internet of things (IoT), artificial Intelligence (AI), and big data analysis technologies, intelligent after-market management has entered a new stage. However, at present, the traditional after-sales treatment mainly uses manual analysis, so that missed diagnosis or misjudgment is easy to occur, and meanwhile, in the existing after-sales flow, failure treatment often lacks an effective feedback and diagnosis mechanism, so that the repeated maintenance efficiency is low, and further, the after-sales management efficiency and the time efficiency are low.

Disclosure of Invention

Based on the foregoing, there is a need for an intelligent after-market management method and system for solving at least one of the above-mentioned problems.

To achieve the above object, an intelligent after-market product management method includes the steps of:

step S1, acquiring an after-sales problem record set of an intelligent product, wherein the after-sales problem record set of the intelligent product comprises at least one or more after-sales problem records of the intelligent product;

Step S2, comparing the after-sale demand emergency degree data of the intelligent product with a preset standard demand emergency degree threshold value to generate priority response demand data and delay response demand data, and performing fault processing association analysis on an after-sale problem record set of the intelligent product through the priority response demand data and the delay response demand data to generate a product fault processing automation solution;

Step S3, performing functional damage evaluation on the intelligent product based on the after-sale failure processing result to generate product functional damage evaluation data, performing intervention fault diagnosis on the after-sale failure processing result according to the product functional damage evaluation data to generate intervention fault diagnosis data, performing second remote processing feedback result acquisition on the intervention fault diagnosis data to obtain a second after-sale failure processing result, and performing after-sale maintenance scheduling on the second after-sale failure result to generate after-sale maintenance service scheduling data;

And S4, carrying out the after-sales trend analysis on the first after-sales failure processing result, the second after-sales failure processing result and the after-sales maintenance service scheduling data to generate the after-sales trend data, and carrying out the after-sales quality assessment visualization on the after-sales quality trend data to execute the intelligent after-sales management optimization operation.

According to the invention, the after-sale problem record set of the intelligent product is obtained and the customer demand is quantified, so that the type and the emergency degree of the after-sale problem can be comprehensively mastered, the problem priority division is realized, the reasonable allocation of resources is ensured, and the after-sale response efficiency is improved. The priority response and delay response demand data are generated through intelligent analysis, fault processing association analysis is carried out, the problem key points can be accurately identified, an automatic solution can be quickly formulated, human intervention is reduced, and the success rate and the efficiency of primary remote processing are improved. And performing functional damage evaluation and intervention fault diagnosis based on after-sales failure results, further refining problem positioning, feeding back an optimized solution through multiple rounds of remote processing, and reasonably scheduling maintenance resources, so that maintenance success rate and repeated problem solving efficiency are improved. By comprehensively analyzing a plurality of processing results and scheduling data, the method generates staged after-sale trend data and visually displays the staged after-sale trend data, is favorable for finding potential rules and trends of after-sale problems, provides data support and decision basis for after-sale management optimization, and improves service quality and customer satisfaction. And each step forms a full-flow closed-loop management from acquisition, analysis and diagnosis of after-sales problem records to trend evaluation, so that the efficiency, the accuracy and the service quality of after-sales treatment are obviously improved, and the continuous optimization of after-sales management of the intelligent power-assisted products is realized. Therefore, the invention improves the efficiency and the efficiency of after-sales management through demand quantification, fault association analysis, feedback diagnosis and trend evaluation.

Preferably, step S1 comprises the steps of:

step S11, acquiring an after-sales problem record set of the intelligent product, wherein the after-sales problem record set of the intelligent product comprises at least one or more after-sales problem records of the intelligent product;

step S12, performing record type division on the after-sales problem record set of the intelligent product to generate a text type record and a voice type record;

S13, performing user demand speech speed analysis on the after-sales demand speech recognition data to generate user after-sales demand speech speed data;

And S14, carrying out word-aid extraction on the basis of the after-sales demand semantic recognition data to obtain after-sales demand word-aid data, and carrying out customer demand quantification on the after-sales problem records of the intelligent product through the after-sales demand word-aid data and the user after-sales demand word-speed data to generate the after-sales demand emergency degree data of the intelligent product.

The invention processes the data in different forms (such as voice recognition and text semantic recognition) respectively by distinguishing the text and the voice type records, fully utilizes the information in the after-sales problem records and improves the comprehensiveness and the accuracy of the data processing. The method comprises the steps of carrying out user demand speech speed analysis on the voice record, identifying the urgency expressed by the user, and combining the extraction of the mood assisted words in the text, further deducing the emotional state and the demand urgency of the user and providing key references for after-sales service. Through the combination analysis of the speech speed and the language gas assisted words, the user demands are quantized, the after-sales demand emergency degree data are generated, scientific basis can be provided for subsequent priority processing, and the after-sales service efficiency is improved. Based on the generated emergency degree data, the after-sales service team can quickly respond to the high-priority demands, optimize resource allocation, realize more accurate and personalized service, and improve customer satisfaction. The after-sales problems are systematically recorded and analyzed, so that enterprises can find potential product defects or service bottlenecks, and data support is provided for product improvement and service flow optimization.

Preferably, the customer demand level quantification of the after-sales problem record for the intelligent product by the after-sales demand word-aid data and the user after-sales demand word-speed data includes:

Performing language characteristic analysis on the after-sales requirement language and word assisting data to obtain customer language characteristic data;

Carrying out emotion fluctuation analysis according to the user speech speed characteristic data and the client speech gas characteristic data so as to obtain client emotion fluctuation data;

And carrying out comprehensive demand urgency analysis on the after-sales problem priority quantized data based on the time sequence after-sales records so as to obtain the after-sales demand urgency data of the intelligent product.

According to the invention, emotion information (such as impatience, pleasure, anxiety and the like) in the language of the client is extracted through the language-qi word-assisting data, so that depth information is provided for evaluating the emotion of the client. And by utilizing the feature extraction of the user speed change, the urgency or the speed smoothness degree in the user expression is captured, and the urgency judgment of the user requirement is further perfected. Based on emotion fluctuation analysis of user speech speed characteristic and mood characteristic data, the variation amplitude and trend of the emotion of the user can be quantified, and a scientific basis is provided for priority quantification of after-sales problems. The mood swings data provides more dimensional information to the after-market service personnel to help more accurately determine the urgency of the user's needs. And (3) rapidly prioritizing after-sales problems by using customer emotion fluctuation data, so as to provide a basis for efficient processing. Through time sequence after-sale record collection, the problem urgency degree is comprehensively analyzed by combining priority quantized data, deviation caused by single dimension evaluation is avoided, and accuracy of demand urgency degree judgment is improved. Through intelligent priority evaluation, after-sales teams can quickly respond to the high-urgency problem, optimize resource allocation, shorten problem solving time, and accordingly improve customer satisfaction. Dynamic updating and adjusting of priorities may ensure flexibility and effectiveness of the after-market service process.

Preferably, step S2 comprises the steps of:

step S21, comparing the after-sales demand emergency degree data of the intelligent product with a preset standard demand emergency degree threshold, and when the after-sales demand emergency degree data of the intelligent product is greater than or equal to the preset standard demand emergency degree threshold, performing priority response on corresponding after-sales problem records of the intelligent product to generate priority response demand data;

Step S22, when the after-sales demand emergency degree data of the intelligent product is smaller than a preset standard demand emergency degree threshold value, carrying out delayed response on the corresponding after-sales problem record of the intelligent product to generate delayed response demand data;

Step S23, constructing a product knowledge graph, importing product fault prediction data into the product knowledge graph for fault processing association analysis, and generating a product fault processing automatic solution;

and step S24, carrying out first remote processing feedback data acquisition on the automatic product fault processing solution, repeating data acquisition until the remote processing feedback data is received when the corresponding remote processing feedback data is not acquired, carrying out feedback result division on the remote processing feedback data, and eliminating a first after-sale successful processing result to obtain a first after-sale failure processing result.

The invention distinguishes between high priority issues and low priority issues by comparing the after-market demand urgency to a standard threshold. The high-priority questions are responded in time (priority response demand data are generated), so that the key questions are ensured to be processed quickly, and the customer satisfaction is improved. The low priority problem is delayed to respond (generating delay response demand data), resources are reasonably arranged, and operation cost is reduced. The priority response demand data and the delay response demand data are analyzed, and potential product failure problems can be predicted, so that measures are taken in advance, and further expansion of the problems is avoided. The prospective and preventive performance of after-sales service are improved, and customer complaints and potential risks are reduced. By constructing a product knowledge graph, the fault prediction data is related to the prior knowledge, so that the fault cause is rapidly positioned, and the problem diagnosis time is shortened. Based on the association analysis of the knowledge graph, an automatic fault processing solution is generated, so that the manual participation is reduced, and the processing efficiency and accuracy are improved. By collecting and classifying the first far-end processing feedback data (successful processing and failed processing), a closed-loop feedback mechanism is formed. Further analyzing and optimizing the failure processing result to ensure that the success rate of subsequent processing is gradually improved. When feedback data is not acquired, the system repeatedly acquires the feedback data, so that the integrity and the reliability of the data are ensured, and high-quality data support is provided for subsequent decisions.

Preferably, predicting product failure for the intelligent after-market problem record set by prioritizing the response demand data and delaying the response demand data includes:

The after-sales problem time series data is obtained by carrying out after-sales problem time series analysis according to the priority response demand data and the delay response demand data;

The after-sales problem demand frequency distribution data is utilized to classify the demand characteristics of the priority response demand data and the delay response demand data so as to obtain priority demand characteristic data and delay demand characteristic data;

And carrying out fault probability prediction calculation on the intelligent after-sale problem record set according to the potential product fault mode data so as to obtain product fault prediction data.

According to the method, through time sequence analysis and demand frequency distribution extraction, time regularity and frequency characteristics of after-sales problems can be accurately captured, so that the prediction accuracy of the occurrence trend of the after-sales problems is improved. The priority response demands and the delay response demands are classified separately, and demand characteristic classification is carried out by combining frequency distribution data, so that a problem classification result is more targeted, and a high-quality data basis is provided for subsequent problem clustering and fault analysis. Through problem record feature cluster analysis and potential fault pattern matching, potential rules in the problem records can be effectively identified, and hidden fault patterns are mined, so that deep analysis of complex after-sales problems is realized. The method and the system have the advantages that the fault probability prediction calculation is carried out based on the potential fault mode data, a quantitative risk assessment result can be provided for a team after sale, the method and the system help to make countermeasures in advance, and uncontrollability of fault occurrence is reduced. The demand characteristic data and the potential fault prediction data with definite priorities are provided, so that after-sales teams can quickly respond according to the priority demand characteristics, meanwhile, resources are reasonably allocated to solve delay demands, and after-sales problem solving efficiency is improved. By analyzing the potential failure mode and predicting the failure probability, feedback data can be provided for product design and manufacturing links, so that common failure points are improved, and the reliability and user experience of the product are improved.

Preferably, step S23 includes the steps of:

Step S231, acquiring a product basic information data set and a historical intelligent after-sale problem record set, extracting common fault modes and corresponding solutions in the historical intelligent after-sale problem record set, and generating a historical fault resolution data set;

step S232, carrying out semantic enhancement processing on the initial product knowledge graph to generate a product knowledge graph, analyzing core elements in the product fault prediction data, including potential fault types, occurrence probability, influence ranges and time ranges, and generating a fault prediction element data set;

Step S233, carrying out path analysis on the dynamic knowledge graph of the product, extracting functional modules, components and historical processing methods related to potential faults, thereby constructing a fault processing association model;

Step S234, carrying out scheme step by step and logic verification on the initial product fault handling automation solution, thereby generating the product fault handling automation solution.

According to the invention, through the integration of the basic information data set and the historical fault solving data set of the product, the initial product knowledge graph is constructed by utilizing the graph structure construction technology, and comprehensive and systematic knowledge support is provided for product problem analysis and solution. The semantic enhancement processing enables the knowledge graph to be more accurate in expressing product information and fault relations, and high-quality semantic association data is provided for subsequent analysis. The fault prediction element data set analyzes core elements (such as types, probabilities, influence ranges and time ranges) of potential faults, and establishes association relation between fault prediction data and a knowledge graph through similarity analysis, so that a dynamic knowledge graph is generated, and knowledge interpretation of a prediction result is realized. The real-time updating of the dynamic knowledge graph enhances the timeliness and accuracy of knowledge and adapts to the change of the product state and the fault mode. By analyzing the paths of the dynamic knowledge graphs, functional modules, components and historical processing methods related to potential faults are extracted, a fault processing association model is constructed, and a data-driven logic basis is provided for analysis and solution of complex faults. And the fault processing association model is utilized for carrying out association mining, an initial automatic solution is generated, and the efficient and accurate automatic solution is finally output through solution step by step and logic verification, so that the requirement of manual intervention is greatly reduced. And carrying out associated node weight updating on the semantic enhanced knowledge graph through fusion of the fault prediction data and the dynamic knowledge graph, so as to form a dynamic optimization closed-loop mechanism of the knowledge graph, and continuously enhancing the fault processing capability.

Preferably, step S3 comprises the steps of:

Step S31, performing functional damage evaluation on the intelligent product based on the after-sale failure processing result to generate product functional damage evaluation data;

step S32, performing intervention fault diagnosis on after-sale failure processing results according to product function damage evaluation data to generate intervention fault diagnosis data, performing second remote processing feedback data acquisition through the intervention fault diagnosis data, and repeating data acquisition until the remote processing feedback data is received when the corresponding remote processing feedback data is not acquired;

And step S33, dividing the feedback result of the remote processing feedback data, eliminating the second after-sale successful processing result to obtain a second after-sale failure processing result, and carrying out after-sale maintenance scheduling on the second after-sale failure result to generate after-sale maintenance service scheduling data.

According to the invention, the functional problems of the intelligent product can be accurately identified and quantified by carrying out damage evaluation on the product functions based on the after-sale failure processing result, so that clear basic data is provided for subsequent fault diagnosis and maintenance. The functional damage assessment provides a systematic perspective for further fault analysis, reducing false positives and omissions on product problems. By analyzing the functional damage evaluation data, the method is beneficial to accurately diagnosing the problem sources, thereby reducing unnecessary fault investigation and optimizing maintenance flow. The collection of the remote processing feedback data provides a basis for dynamically adjusting the after-sales strategy, and ensures that each fault diagnosis can be optimized based on the latest data. The repeated data acquisition mechanism ensures that the integrity and accuracy of the information can be ensured no matter whether the feedback is timely or not. The feedback result division can accurately distinguish successful from failed after-sales processing, and interference of error data on after-sales decision is avoided. Only the failure result is reserved by eliminating the successful processing result, so that the follow-up processing is focused on the problems in practice. The generation of the after-sales repair service schedule can automatically schedule repair resources based on the second after-sales failure processing result, so that the repair process is more efficient and accords with the priority and severity of the actual failure of the product.

Preferably, performing the intervention fault diagnosis on the after-sales failure processing result according to the product function damage evaluation data includes:

Marking the damage module of the intelligent product according to the product function damage evaluation data to obtain damage module marking data of the intelligent product; performing module fault intervention analysis on the after-sale failure processing result based on the intelligent product damage module marking data to generate intervention analysis result data;

performing fault chain expansion on the intervention analysis result data by utilizing the product knowledge graph to generate fault diagnosis expansion data, performing fault mode priority ranking on the fault diagnosis expansion data, identifying an optimal processing path to generate fault diagnosis path data, performing simulation verification on the fault diagnosis path data, and generating optimized fault diagnosis data;

and integrating the optimized fault diagnosis data into a product knowledge graph to perform fault expansion node mapping, so as to generate intervention fault diagnosis data.

According to the invention, through analyzing the functional damage evaluation data of the intelligent product, the damaged specific module in the product is accurately identified and marked, so that a clear direction and focus are provided for subsequent fault diagnosis, and the blind investigation of the problem is reduced. Accurate damage module marking helps to improve failure diagnosis efficiency and avoids unnecessary time waste on extraneous modules. The fault module can be rapidly positioned by accurately performing intervention analysis on the after-sale failure processing result based on the damage module marking data, and the problem cause can be further confirmed by careful analysis, so that the pertinence and the effectiveness of fault processing are ensured. The product knowledge graph is used for carrying out fault chain expansion on the intervention analysis result, and potential association between fault modules can be revealed, so that the influence range and the fault source are identified, a comprehensive view angle is provided for fault diagnosis, and neglect of single module faults to other modules is reduced. The fault chain expansion is beneficial to finding hidden problems, improving the depth and breadth of fault diagnosis and avoiding the condition of missed diagnosis or misdiagnosis. The priority ordering of the fault modes and the identification of the optimal processing paths can order different fault modes according to the severity degree and the occurrence probability of the faults, so that after-sales teams can be helped to process the faults with high risk and high priority preferentially, and the key faults can be solved as soon as possible. After the optimal processing path is identified, a clear and direct fault solving route can be provided for a maintenance team, redundant steps are reduced, and the repair efficiency is improved.

Preferably, step S4 comprises the steps of:

step S41, carrying out after-sales trend analysis on the first after-sales failure processing result, the second after-sales failure processing result and the after-sales maintenance service scheduling data to generate after-sales trend data;

And S42, performing data visualization on the intelligent product after-sale quality evaluation data so as to generate an intelligent product after-sale quality evaluation report to execute intelligent product after-sale management optimization operation.

According to the method, the first after-sale failure processing result, the second after-sale failure processing result and the after-sale maintenance service scheduling data are comprehensively analyzed, and the generated staged after-sale trend data can comprehensively show trend change and distribution rules of after-sale problems. The after-sales quality assessment is based on the trend data to carry out deep analysis, intelligent after-sales quality assessment data of the product is generated, core problems affecting the after-sales service quality are accurately revealed, and data support is provided for optimization. The after-sales service performance including key indexes such as trend graphs, fault rate distribution, service response time and success rate is presented in an intuitive manner by visualizing the intelligent product after-sales quality evaluation data to generate an evaluation report. The data visualization presentation mode is easy to understand, not only provides the management layer with the capability of quickly observing the problem, but also facilitates the communication and coordination among cross departments. The intelligent after-sales quality assessment report provides a clear direction for after-sales management optimization, and supports the formulation of improved strategies, such as enhancing reliability tests of key modules, optimizing service flows, or enhancing resource allocation. The staged trend analysis helps the management layer predict potential problems, improves the response speed to after-sales demands, deploys resources in advance, and avoids the concentrated outbreak of problems.

In this specification, there is provided an after-sales management system for performing the above-mentioned after-sales management method for an intelligent product, the after-sales management system comprising:

The system comprises an emergency degree analysis module, a customer demand degree analysis module and an emergency degree analysis module, wherein the emergency degree analysis module is used for acquiring an intelligent after-sales problem record set, and the intelligent after-sales problem record set comprises at least one or more intelligent after-sales problem records;

The automatic processing module is used for comparing the after-sale demand emergency degree data of the intelligent product with a preset standard demand emergency degree threshold value to generate priority response demand data and delay response demand data, and performing fault processing association analysis on the after-sale problem record set of the intelligent product through the priority response demand data and the delay response demand data to generate a product fault processing automatic solution;

The system comprises an after-sale failure processing module, an intervention processing module, an after-sale failure processing module, an after-sale maintenance scheduling module and a maintenance scheduling module, wherein the after-sale failure processing module is used for performing functional damage evaluation on an intelligent product based on the after-sale failure processing result to generate product functional damage evaluation data;

The after-sales quality evaluation module is used for carrying out after-sales trend analysis on the first after-sales failure processing result, the second after-sales failure processing result and the after-sales maintenance service scheduling data to generate staged after-sales trend data, and carrying out after-sales quality evaluation visualization on the staged after-sales quality trend data so as to execute intelligent after-sales management optimization operation of the product.

The invention has the beneficial effects that the emergency degree analysis module can clearly distinguish the emergency degree of different problems by quantifying the customer demand degree on the after-sale problem records, so that reasonable priority is allocated to each problem, and the mechanism ensures that the key problem can be processed in the shortest time and the customer dissatisfaction caused by processing delay is avoided. By comparing the critical problems with the standard requirement emergency threshold, the critical problems can be ensured to be responded preferentially, and the response time to the problems with lower priority is reduced, so that the service efficiency is improved. The automatic processing module automatically generates a product fault processing automatic solution and carries out remote processing feedback for the first time, so that manual intervention can be reduced, human error risks are reduced, and meanwhile, the fault processing accuracy is improved. The automatic solution can improve the consistency and efficiency of the processing, shorten the processing period, ensure that the problem can be effectively solved when the problem is processed for the first time at the far end, and reduce the requirement of subsequent manual intervention. The intervention processing module can accurately position the fault cause through functional damage assessment based on the first after-sale failure processing result and further intervention fault diagnosis, and the process can provide a more accurate repairing scheme by combining historical data and fault chain expansion, reduce unnecessary reworking and improve the overall maintenance efficiency. Through the remote processing feedback collection of many rounds, the system can track the fault repair progress in real time, ensures that the repair scheme is verified and improved in a plurality of stages. After the after-sales quality evaluation module analyzes the after-sales trend data, a detailed after-sales quality evaluation report can be generated, and the after-sales service quality is visually displayed, so that an enterprise management layer is helped to clearly know the problem, and the after-sales service flow is further optimized. By analyzing the staged after-sale quality trend data, the service bottleneck or the area with reduced quality can be quickly found, timely adjustment is made on the strategy for enterprises, and the trust and satisfaction degree of clients are enhanced. Therefore, the invention improves the efficiency and the efficiency of after-sales management through demand quantification, fault association analysis, feedback diagnosis and trend evaluation.

Drawings

FIG. 1 is a schematic flow chart of steps of an intelligent after-sales management method;

FIG. 2 is a flowchart illustrating the detailed implementation of step S2 in FIG. 1;

FIG. 3 is a flowchart illustrating the detailed implementation of step S3 in FIG. 1;

The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.

Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments, with the term "and/or" as used herein including any and all combinations of one or more of the associated items listed.

To achieve the above object, referring to fig. 1 to 3, an intelligent after-market management method for products includes the following steps:

In the embodiment of the present invention, as described with reference to fig. 1, the step flow chart of an after-sale management method of an intelligent product of the present invention is shown, and in this example, the after-sale management method of an intelligent product includes the following steps:

In the embodiment of the invention, after-sales problem records related to intelligent products are collected through an enterprise after-sales service system and a customer feedback platform (such as online customer service, mail, telephone records and the like). And screening out data directly related to product faults, use problems, maintenance requests and the like, and ensuring the integrity and accuracy of the after-sales problem record set. And carrying out standardized processing on the collected records, including deduplication, format unification and semantic cleaning, and generating a structured after-sales problem record set. The text content of the after-sales problem records is analyzed by using Natural Language Processing (NLP) technology, and key fields (such as 'immediate', 'serious', 'unavailable', and the like) reflecting the emergency degree of the customer demand are extracted. And calculating the occurrence frequency and the context semantic relation of the keywords, and generating a preliminary customer demand measurement index. A plurality of evaluation dimensions affecting the degree of urgency of the demand are defined, such as question type weights, product functionality loss, performance issues, appearance issues, and the like. Customer level weight-VIP customer, ordinary customer, etc. Problem scope-single equipment problem or batch equipment problem. A preliminary demand quantification model is formed based on these dimensions scoring each record. Calculating a composite score using a weighted model: wherein As an index of the degree of urgency,In order to be the speech rate data,Is the occurrence frequency of the Chinese Qi and the auxiliary words,AndThe weight coefficients (adjusted according to actual application) respectively generate intelligent after-sales demand emergency degree data according to the calculation results, and the after-sales problem records are classified according to comprehensive scores (such as high priority, medium priority and low priority). The emergency degree data is bound with the original problem record and stored as an intelligent product after-sales demand emergency degree data set, so that subsequent processing and decision support are facilitated.

in the embodiment of the invention, the emergency classification standard is defined, for example, a high priority threshold is used for recording more than 90 minutes, a medium priority threshold is used for recording between 60 and 90 minutes, and a low priority threshold is used for recording less than 60 minutes. The after-sales demand emergency degree data are compared with a threshold value one by one, namely, if the emergency degree score is more than or equal to a high priority threshold value, the after-sales demand emergency degree data are marked as 'priority response demand data', and otherwise, the after-sales demand emergency degree data are marked as 'delay response demand data'. The classification result is stored in layers into two groups of data sets, wherein the priority response demand data set comprises all problem records needing quick response and immediate processing, and the delay response demand data set comprises problem records capable of being processed in a delayed mode for subsequent batch analysis and optimization. A rule base between the product fault type and the processing scheme is established, wherein the rule base comprises fault reasons (hardware, software, operation problems and the like) and processing steps (remote debugging, part replacement, field maintenance and the like). Using data mining methods (e.g., association rule analysis, decision trees, etc.), the association patterns of the problem and historical solutions are identified in combination with the feature data of the after-market problem records. Based on the priority response demand data and the delay response demand data, generating a product fault handling automation solution according to the following logic, wherein the product fault handling automation solution comprises priority matching of historical solutions in a rule base, prediction of the optimal solution by a machine learning model (such as random forest and XGBoost) if no direct matching item exists, and generation of the fault handling automation solution comprising fault classification, cause analysis and proposal processing flow. Based on the generated fault handling automation solution, remote processing operations are performed such as pushing a software patch or configuration adjustment instruction to the client device, remotely debugging the fault module, logging the processing log, requesting the client to provide more diagnostic information (uploading the device log, taking a picture of the problem, etc.). The feedback data after the remote processing is collected mainly comprises the steps of recording the time of successfully solving the problem, executing the scheme and recording the reasons of failure (such as unsuccessful instruction execution, problem exceeding the remote capability range and the like). And storing the failure processing result as a first after-sale failure processing result for further analysis.

In the embodiment of the invention, after-sale failure processing result data comprising product model, failure description, failure occurrence time, processing measures, results and the like is collected. And extracting a functional module list and corresponding performance indexes of the intelligent product. Based on the fault description and the processing result, the affected functional modules are identified and the degree of damage is marked. The degree of functional impairment (e.g., mild damage, moderate damage, complete failure) of each damaged module is assessed using historical data and machine learning models (e.g., classification models or regression models). And integrating the evaluation results to generate product function damage evaluation data, wherein the product function damage evaluation data comprises damage levels and damage reasons of all modules. And positioning a specific damage module according to the function damage evaluation data, extracting relevant fault information, and generating damage module marking data. And carrying out module fault intervention analysis on the marked data, and outputting detailed fault reasons and associated modules. And matching the damaged module with the failure cause by using the product knowledge graph to generate failure chain extension data. And sequencing the fault modes according to the occurrence probability and the influence range, and screening the optimal processing path. And performing simulation verification on the optimal fault diagnosis path, and testing the effectiveness and reliability of the path. Optimizing the diagnostic path and generating final interventional fault diagnostic data. And establishing a data acquisition flow, and monitoring a remote processing result, wherein the remote processing result comprises maintenance operation records, user feedback and performance test data. If not, repeating the acquisition process until the complete data is obtained. Classifying the acquired data, eliminating successful processing results, and screening out second failure processing results. And carrying out deep analysis on the result of the second failure processing, and identifying the type and the area of the failure needing to be processed preferentially. And (3) sequencing the priority of the faults according to the emergency degree and the influence range of the faults, and allocating resources for maintenance tasks by using a scheduling optimization algorithm (such as linear programming, genetic algorithm and the like) to determine maintenance time and execution team. After-market repair service dispatch data including dispatch plans including repair team assignments, task priorities, predicted completion times, etc. are output.

In the embodiment of the invention, the first after-sale failure processing result, the second after-sale failure processing result and the after-sale maintenance service scheduling data are integrated into a staged after-sale data set. And counting the frequency and reason distribution of failed maintenance according to time periods (such as week, month and quarter), and generating time trend data. Based on fault classification (such as hardware, software and interface problems), the frequency and the duty ratio of each type of faults are counted, and fault type trend data are generated. And counting the distribution condition of maintenance failure according to the region or market partition, and generating regional trend data. The future failure to repair trend is predicted using linear regression, moving average, etc. Such as ARIMA or Prophet models, for analysis of long-term trends and periodic fluctuations. And integrating analysis results to generate staged after-market trend data, including time trend, fault type trend and regional trend. Based on the staged after-sales trend data, various evaluation indexes are calculated to form intelligent after-sales quality evaluation data of the product, wherein the various evaluation indexes comprise the proportion of successful maintenance, the average time from fault report to completion of repair, and the repeated occurrence proportion of after-sales service quality and similar problems after repair. The staged after-market quality trend data is visualized for after-market quality assessment using a Power BI, tableau, or the like visualization tool, and the visualization chart is integrated into an intelligent product after-market quality assessment report.

Preferably, step S1 comprises the steps of:

In the embodiment of the invention, the historical after-sales service records are extracted from the enterprise after-sales system, and the historical after-sales service records comprise text records and voice call records fed back by clients. Ensure that the data covers a variety of intelligent product categories and problem types, such as equipment failure, functional anomalies, user guidance, and the like. An intelligent after-sales problem record set is constructed, wherein the record set needs to contain a time stamp, a product model number, a problem description and customer basic information of each record. The data in the intelligent product after-sales problem record set is automatically classified into a text type record and a voice type record according to the content format. Quick classification is accomplished using a format detection algorithm, e.g., based on file extensions (e.g.,. Txt or. Mp 3), file size, header parsing, etc. For the voice type record, a voice recognition algorithm (such as an ASR model based on deep learning) is adopted to extract text content in voice, and after-sales demand voice recognition data is generated. In the recognition process, the problems of multi-language support and background noise interference in the voice are required to be processed, and the accuracy of text transcription is ensured. And extracting the time length of the voice fragments and the number of transcribed words by utilizing the voice recognition result, calculating the speech speed index of each record, and generating after-sales demand speech speed data of the user. And carrying out cluster analysis on the speech speed data to distinguish normal speech speed, faster speech speed and rapid speech speed so as to reflect the emergency degree of the client. For text type records, semantic analysis is performed by adopting a Natural Language Processing (NLP) technology, key problem description and demand intention in customer feedback are identified, and after-sales demand semantic identification data are generated. Specific keywords, emotional trends (e.g., positive, negative, neutral), and problem categories (e.g., hardware faults, software problems, operational consultations, etc.) are extracted. Further word analysis is carried out on the after-sales demand semantic recognition data, words reflecting the customer's word (such as ' urgent ', ' immediate ', ' quick ', etc.) are extracted, and after-sales demand word data is generated. And accurately extracting the relevant information of the language and the words by using a predefined language and words assisting dictionary and a context dependent analysis method. Combining the speech speed data with the speech gas word-assisting data, and calculating the emergency degree of the client demand through a weighting algorithm: wherein As an index of the degree of urgency,In order to be the speech rate data,Is the occurrence frequency of the Chinese Qi and the auxiliary words,AndAnd respectively generating intelligent after-sales demand emergency degree data for the weight coefficients (adjusted according to actual application) according to the calculation results, and grading the data according to the emergency degree (such as ordinary, urgent and emergency) so as to guide the priority ranking of the after-sales services.

In the embodiment of the invention, the frequency and distribution of the word assisting are counted by utilizing the after-sales requirement word assisting data. The term-assisting feature vector is constructed, and feature weights are extracted through TF-IDF (word frequency-inverse document frequency), for example. The mood-analysis model (e.g., BERT-based mood classification model) is used to combine mood-assisted words with context, analyze mood tendencies (e.g., urgency, anger, satisfaction, etc.) of the customer, and generate customer mood feature data including mood category, mood intensity, and mood type. The speed change trend and fluctuation amplitude are extracted by calculating the speed change rate (such as sentence-by-sentence or segment-by-segment speed difference value). Classifying the speech speed change characteristics into three categories of 'stable', 'slight fluctuation', 'severe fluctuation', generating user speech speed characteristic data, and reflecting anxiety or calm states in customer expressions. And analyzing the emotion fluctuation mode of the user by combining the user speech speed characteristic data and the client speech gas characteristic data and utilizing machine learning models such as decision trees, random forests and the like. And outputting the intensity scores of the emotion fluctuation of the clients (such as 0-100), wherein the higher the scores are, the larger the fluctuation is represented, and generating the emotion fluctuation data of the clients, including the fluctuation types (such as stable and fluctuation), the fluctuation intensity and the trend. Combining emotion fluctuation data with after-sales demand classification data, and primarily quantifying according to customer problem urgency rules: Wherein, the method comprises the steps of, The score is quantized for the priority level,As the intensity of the mood swings,For the question category weight,AndTo adjust the parameters. And outputting the preliminary priority scores of each after-sales problem, and generating after-sales problem priority quantized data. And (3) sorting the problem records according to a time sequence based on the multiple after-sales records of the clients to form time sequence after-sales records, and marking the occurrence time and the progress of the problems. And extracting the change trend of the customer demand in the record by using a time sequence analysis technology, and predicting the subsequent emergency. Based on the time sequence after-sales records, the preliminary priority quantification data is combined with the time sequence information, whether the problem has the necessity of priority improvement is analyzed, the intelligent after-sales demand emergency degree data is generated, and the intelligent after-sales demand emergency degree data is classified according to the emergency degree grade (such as ordinary, urgent and urgent).

As an example of the present invention, referring to fig. 2, the step S2 in this example includes:

In the embodiment of the invention, the after-sale demand emergency degree data of the intelligent product and the preset standard demand emergency degree threshold value are obtained. The after-market demand urgency data is compared to a standard demand urgency threshold. When the emergency degree data is greater than or equal to the threshold value, marking the corresponding after-sales problem record as a priority response requirement, and generating priority response requirement data for quick processing of the after-sales problem. When the urgency data is less than the threshold, the corresponding after-market issue record is marked as a delayed response demand. And summarizing the priority response demand data and the delay response demand data, analyzing all after-sales problem records, and applying a fault prediction algorithm (such as time sequence analysis or machine learning model) to generate product fault prediction data. And summarizing life cycle data of the product, wherein the life cycle data comprise link information such as design, production, sales, use, after-sales and the like. And defining relation nodes among product characteristics, common fault types, fault reasons and processing modes. And (5) establishing a dynamic knowledge graph by using a graph database (such as Neo4 j). And importing the product fault prediction data into a knowledge graph, and analyzing the relevance of the faults and the processing method. A rule reasoning algorithm (such as an OWL framework based on logic reasoning) is applied to automatically generate an automatic solution for product fault treatment. And executing remote processing operation on the product fault processing automation solution, and monitoring feedback data in real time through an API interface. If the feedback data is not received, a retry mechanism is set, and the acquisition is circulated until the remote processing feedback data is received. And analyzing the result of the received feedback data, and recording the result as a first after-sale successful processing result if the processing is successful. If the processing fails, after the successful record is removed, the remaining data is finished into a first after-sale failure processing result.

In the embodiment of the invention, the time sequence modeling is performed on the priority response demand data and the delay response demand data based on the time dimension. And selecting a proper time sequence analysis method (such as ARIMA model, LSTM model and the like) to generate after-sales problem time sequence data. And analyzing the time sequence trend and the periodicity of the after-sales problems, and identifying the peak and valley time periods of the demand fluctuation. Counting the occurrence frequency of the demand in the time series of the after-sales problems, and extracting frequency distribution characteristics according to time windows (such as days, weeks or months). The frequency distribution curve or histogram is used for data visualization to determine the distribution characteristics of the priority demand and the delay demand. And classifying the characteristics of the priority response requirement and the delay response requirement by using a classification algorithm (such as KNN, decision tree or SVM). And extracting key characteristics of the requirements (such as problem type, occurrence area, emergency degree, time consumption of processing and the like) and generating priority requirement characteristic data and delay requirement characteristic data. And carrying out cluster analysis on the problem records based on the feature data, and finding out feature similarity of the problem records by adopting K-means, DBSCAN or hierarchical clustering algorithm. The clustering dimension comprises the occurrence frequency, the equipment model, the geographic distribution and the like of the problems, and generates the problem record characteristic clustering data. A potential failure mode database (containing historical failure modes and problem records) is constructed. Based on the clustering result and the fault pattern database, a pattern matching algorithm (such as a nearest neighbor algorithm or semantic similarity matching) is utilized to identify potential fault patterns related to the problem record clustering data. And (3) performing correlation analysis and fault probability calculation on the potential product fault modes and the problem record sets by using a Bayesian probability model or a machine learning model (such as random forests and XGBoost). The probability value of each potential failure mode is output, and potential failures with high risks are identified.

Preferably, step S23 includes the steps of:

In the embodiment of the invention, basic information related to the product, such as model, configuration, production batch, key technical parameters and the like, is collected. The method comprises the steps of obtaining intelligent fault problems recorded in past product after-sales service and solving the problems to form a historical fault solving data set. And analyzing the fault mode in the historical after-sales problem record, extracting a corresponding solution, and generating a historical fault resolution data set. By adopting a graph structure construction technology, various attributes, fault modes, solutions and the like of a product are associated into nodes and edges of a graph by integrating a product basic information data set and a historical fault resolution data set, so that an initial product knowledge graph is constructed. The initial product knowledge graph is semantically enhanced, and the knowledge graph is enabled to have depth and accuracy by adding more associated information, context understanding and reasoning rules and the like, and finally the product knowledge graph is generated. And analyzing the product fault prediction data, extracting key factors such as potential fault types, occurrence probability, influence range, time range and the like, and generating a fault prediction element data set. And matching each element in the fault prediction element data set with nodes (such as a functional module, a component, a history processing method and the like) and edges (such as association relations) in the product knowledge graph, establishing the association relation between the fault prediction data and the prior knowledge through similarity analysis, and generating the product dynamic knowledge graph. And carrying out path analysis on the dynamic knowledge graph of the product, and extracting functional modules, components and historical fault processing methods thereof related to potential faults, wherein the paths can help to understand the root cause and the solution of the faults in depth. And constructing a fault processing association model by analyzing the association relation between the fault mode and the solution, and guiding logic of fault processing. And excavating the fault prediction data by using the constructed fault processing association model to generate an initial product fault processing automatic solution. And carrying out step-by-step processing on the initial automatic solution for product fault processing, and determining the operation requirement and processing mode of each step. The automated solution is verified, so that each step can be effectively executed in practical application, the processing logic is correct, and the complete and effective product fault processing automated solution is finally generated.

As an example of the present invention, referring to fig. 3, the step S3 in this example includes:

In the embodiment of the invention, the past after-sale failure processing records are collected and analyzed, wherein the records comprise the type of failure, the maintained or replaced parts, the processing time, the cost and the like. And constructing a function damage evaluation model based on the after-sale failure processing result data. The model comprehensively considers factors such as the service condition, the fault frequency, the damage degree of the components, the repair history and the like of the product, and evaluates the function damage degree of the product. And outputting the functional damage condition of the product through the evaluation model, and generating product functional damage evaluation data. The evaluation data will help identify faulty modules and subsequent processing steps that require significant attention. And carrying out interventional fault diagnosis on each product with problems according to the product functional damage evaluation data. The diagnosis is based on the damaged functional module and historical fault information, analyzes the source of the fault, and generates interventional fault diagnostic data. Based on the intervention fault diagnosis data, the remote technical support team performs a second remote processing operation. After processing, remote feedback data is collected, and the data comprises whether fault processing is successful or not, effectiveness of a processing mode and the like. If sufficient remote feedback data is not available in the first acquisition, a repeated process of data acquisition is required. The system will continue to monitor until valid remote processing feedback data is received. The collected remote processing feedback data is classified, the second after-sale successful processing result is removed, and the second after-sale failure processing result is reserved, so that the step is used for identifying products which fail to solve the problem. And for the second after-sale failure processing result, scheduling after-sale maintenance service according to the failure type, the influence range and the functional damage evaluation of the product. The scheduling will assign the relevant product to the appropriate repair site or technician and determine the priority, schedule, etc. of repair. Based on the scheduling scheme, after-sales repair service scheduling data are generated, and each fault processing link can be ensured to be carried out effectively on time, so that the efficiency and quality of after-sales service are improved.

In the embodiment of the invention, the damage condition of each module is extracted from the product function damage evaluation data. The evaluation data contains information such as the number of times of failure of each module, failure probability, influence range (if it affects the functions of other modules), and the like. The evaluation data is processed using algorithms (e.g., machine learning models such as decision trees, random forests, etc.), marking each module as to whether there is damage, and assigning each damaged module a priority (e.g., based on the module's centrality, damage severity, and frequency of occurrence). The tag data for each module is integrated into "intelligent product damage module tag data" that includes the fault status and priority of each module (e.g., module 1, damage, priority 1). The failure mode analysis is performed on the damaged module by analyzing historical failure data (such as failure type, repair method, etc.). For example, if the module 1 frequently fails, its source (e.g., power management chip failure, voltage instability, etc.) may be analyzed. A preset processing scheme is provided for each failure mode, the efficiency and cost of different processing schemes are analyzed, and a preliminary intervention analysis result is generated for each failure mode (for example, the repair method of the power failure of the module 1 is to replace a power management chip). And outputting intervention analysis result data, and recording diagnosis and suggested treatment schemes of each fault mode. According to the knowledge graph of the product, the nodes (module type, fault cause, history repairing method and the like) in the graph are utilized to expand the fault chain. First, the failure mode in the interventional analysis result data is matched to the relevant node in the map (e.g., a power failure is associated with a "battery module" or "power chip" node). The extended fault chain may contain multiple levels of information, for example, the fault of a certain module is not just "power management chip damaged", but also is related to "battery voltage instability" or "software configuration problem", and diagnostic extension data (for example, power failure chain: battery damage- > power management chip fault- > power supply module failure) containing all potential fault sources and repair paths is generated. A prioritization algorithm based on failure mode severity and scope of impact is employed. For example, if a failure of the power management chip would result in paralysis of the entire device, but only insufficient battery power would affect only the device endurance, the power management chip failure may be prioritized as high. Based on the prioritized results, fault diagnosis path data is generated, the path including a number of steps, such as first checking the power module, then checking the battery, and finally checking the system settings. Each step is ordered according to importance, ensuring fault priority handling of high priority. Simulation tools (e.g., fault tree analysis, monte Carlo simulation, etc.) are used to validate the fault diagnosis path. And evaluating the success rate and time cost of each processing scheme by simulating the influence of different processing paths on equipment fault recovery. For example, a repair path for a battery under-run and a power management chip failure is simulated, and which path is observed to restore the device to normal operation faster. And optimizing a fault diagnosis path according to the simulation result. If the simulation shows that some paths are more efficient than others, the diagnostic strategy can be adjusted to generate a final optimized version of the fault diagnosis data. The optimized fault diagnosis data are integrated into the existing product knowledge graph, so that when similar faults occur again, the system can quickly reference the historical data and give the best repair suggestion. For example, when a power management chip fails, the system can refer to the historical repair record, quickly recommend replacement of the chip or check the battery module. A map is created in the product knowledge graph for each new failsafe expansion node. For example, after a diagnosis path of "power management chip fault" is newly added, the fault path is marked as a new fault expansion node in the knowledge graph, and forms a new relationship with other related nodes (such as a battery, a power supply module and the like). And through node mapping, the expansion and upgrading of the product knowledge graph are ensured. Finally, the optimized diagnosis data are integrated into a product knowledge graph to form final 'intervention fault diagnosis data', and the data are used for future fault diagnosis, intelligent processing path recommendation and fault pattern recognition.

Preferably, step S4 comprises the steps of:

In the embodiment of the invention, the first after-sale failure processing result, the second after-sale failure processing result and the after-sale maintenance service scheduling data are collected and integrated. Data collection comes from a number of sources, such as customer feedback, service center records, maintenance history, and the like. The data is partitioned by time dimension (e.g., monthly, quarterly, etc.), and the after-market failure processing results and maintenance service schedules for each time period are aggregated. Data for each time period is analyzed using data mining and analysis methods (e.g., moving average, time series analysis, trend line analysis, etc.). The objective of the analysis is to identify trends in after-market failure types, frequency changes, repair response times, etc. that are common at a certain stage. For example, during a quarter, the battery failure rate increases, the user feedback processing speed is slower, or a module (such as a display screen) frequently fails. Based on the analysis results, staged after market trend data is generated. The data will show the fluctuating trends of after-market problems, such as failure type frequency changes, increased periods of customer complaints, after-market service response times, etc. Based on the after-market trend data, a series of after-market quality assessment indicators are set, which may include the frequency of failure occurrences over time, the average time from receipt of a failure report to completion of repair, customer satisfaction score for after-market treatment, typically through customer questionnaires or return visits, and the proportion of customer complaints processed. And calculating the numerical value of each evaluation index by using the staged after-sale trend data. Weights may be set (e.g., failure rate 40%, repair efficiency 30%, user satisfaction 30%), and the final after-market quality score calculated by weighted average. And summarizing and calculating all indexes, and finally generating after-sales quality evaluation data. For example, an after-market quality score of 85 points over a period of time indicates good after-market quality at that stage. Based on the after-market quality assessment data, an appropriate visualization method is selected. For example, the line graph is used to show after-market trend data, such as the trend of failure rate over time. The histogram is used to show the frequency of occurrence of different types of faults, or the distribution of after-market service response times. Pie charts are used to show the duty cycle of each type of fault in the total fault. The radar graph is used for displaying the comprehensive performance of a plurality of evaluation indexes and highlighting the advantages and disadvantages of different dimensions. A data label is added to the visual chart to identify the specific value and time point for each data point (e.g., 15% failure rate for a battery for a month). At the same time, notes may be added indicating the cause of the abnormal fluctuations (e.g., increased display screen failure over a period of time, particularly due to quality issues with new model lots). Reports are automatically generated using visualization tools (e.g., matplotlib and Seaborn in Power BI, tableau, python, etc.) and exported in PDF or Excel format for further analysis and sharing.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent after-sales management method for products is characterized by comprising the following steps:

Step S2, comparing the after-sale demand emergency degree data of the intelligent product with a preset standard demand emergency degree threshold value to generate priority response demand data and delay response demand data, performing fault processing association analysis on an after-sale problem record set of the intelligent product through the priority response demand data and the delay response demand data to generate a product fault processing automation solution, performing first remote processing feedback result acquisition on the product fault processing automation solution to obtain a first after-sale failure processing result, wherein the step S2 comprises the following steps:

Step S23, constructing a product knowledge graph, importing product fault prediction data into the product knowledge graph for fault processing association analysis to generate a product fault processing automatic solution, wherein the step S23 comprises the following steps:

step S234, carrying out scheme step by step and logic verification on the initial product fault handling automation solution so as to generate the product fault handling automation solution;

step S24, carrying out first remote processing feedback data acquisition on the automatic solution of product fault processing, and repeating data acquisition until the remote processing feedback data is received when corresponding remote processing feedback data is not acquired;

2. The intelligent after-market product management method according to claim 1, wherein step S1 comprises the steps of:

3. The method of claim 2, wherein quantifying customer demand on the intelligent after-market problem record by the after-market demand word-aid data and the user after-market demand word-speed data comprises:

4. The intelligent after-market product management method of claim 1, wherein predicting product failure for the intelligent after-market product issue record set by prioritizing response demand data and delaying response demand data comprises:

5. The intelligent after-market product management method according to claim 1, wherein step S3 comprises the steps of:

6. The intelligent after-market product management method according to claim 5, wherein performing an intervention fault diagnosis on the after-market failure processing result according to the product function damage evaluation data comprises:

7. The intelligent after-market product management method according to claim 1, wherein step S4 comprises the steps of:

8. An intelligent after-market product management system for performing the intelligent after-market product management method of claim 1, the intelligent after-market product management system comprising: