Motor vehicle maintenance management service system

Abstract

The invention discloses a motor vehicle maintenance management service system, and relates to the technical field of motor vehicle maintenance management, and the system comprises a client information management module which builds data for storing client multi-information, supports multi-format import and export, intelligently recognizes and updates, and guarantees the information to be timely and accurate; the vehicle fault diagnosis module is provided with various interfaces and expert systems, analyzes and diagnoses faults based on big data, and gives reasons, suggestions and cost estimation; the maintenance work order management module supports work order full-process operation, automatically distributes personnel and updates the progress in real time for clients to inquire; the maintenance accessory management module is provided with an inventory database and can intelligently match accessories and provide position query navigation; and the maintenance personnel management module is used for storing personnel information and evaluating performance. According to the invention, management efficiency and accuracy can be improved, accessory management and cost are optimized, personnel quality and service quality are improved, customer experience and satisfaction are enhanced, powerful support can be provided for enterprise decision making, and intelligentization of the motor vehicle maintenance industry is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of motor vehicle maintenance management, and in particular to a motor vehicle maintenance management service system. Background Art

[0002] With the continuous growth of the number of motor vehicles, the scale of the motor vehicle maintenance industry continues to expand, but the traditional maintenance management service model faces many problems.

[0003] In the traditional way, maintenance companies rely on manual recording and management of customer information, which is inefficient and prone to errors. Customer information is scattered in paper documents or simple spreadsheets, making it difficult to find data, update data in a timely manner, and difficult to achieve information sharing and effective use. For example, when a customer's vehicle annual inspection is due or there is a new maintenance record, the maintenance company cannot obtain and notify the customer in time, affecting the customer experience and the timeliness of vehicle maintenance.

[0004] Vehicle fault diagnosis mainly relies on the experience of maintenance personnel and simple diagnostic tools, which lacks systematicity and accuracy. For complex faults, it often takes multiple attempts to determine the problem, which prolongs the maintenance time and increases the waiting cost for customers. Moreover, the experience of different maintenance personnel varies greatly, and the diagnostic results lack consistency and scientificity.

[0005] The maintenance work order management process is cumbersome. From work order creation and allocation to maintenance progress tracking, most of the information is transmitted and recorded manually, which is prone to problems such as work order loss and untimely progress updates. Customers cannot understand the vehicle maintenance status in real time, which increases their anxiety and is not conducive to the maintenance company's reasonable arrangement of work and allocation of resources.

[0006] In terms of maintenance parts management, inventory management is unscientific and there is a lack of accurate demand forecasting and inventory early warning mechanisms. Parts are often overstocked or out of stock. Overstocking leads to capital occupation and waste, while out of stock delays maintenance time, affecting customer satisfaction and corporate reputation.

[0007] The maintenance personnel management lacks an effective performance evaluation system, and the salary adjustment and promotion rewards lack objective basis, which makes it difficult to stimulate the maintenance personnel's work enthusiasm and motivation to improve their skills. At the same time, it is difficult for enterprises to formulate targeted training plans based on market demand and the current status of maintenance personnel's skills, which is not conducive to improving the overall quality of maintenance personnel.

[0008] In terms of customer service evaluation, traditional evaluation methods have limited collection channels, mostly paper questionnaires or oral inquiries. Data collection is incomplete and untimely, and the analysis methods are single. It is impossible to deeply explore customer needs and opinions, making it difficult for companies to effectively improve service quality based on this.

[0009] In addition, the traditional model has weak data statistics and analysis functions, and enterprises cannot fully utilize maintenance business data for decision support, making it difficult to adapt to market changes and competitive demands. Therefore, it is urgent to develop an efficient and intelligent motor vehicle maintenance management service system to improve the management level and service quality of maintenance enterprises and meet the growing demand of the motor vehicle maintenance market. Summary of the invention

[0010] The present invention proposes a motor vehicle maintenance management service system to solve the problems mentioned in the above-mentioned prior art.

[0011] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a motor vehicle maintenance management service system, comprising: Customer information management module: build a customer information database to store basic customer information, support CSV and Excel data format import and export, use MySQL relational database management system, optimize database table structure design, establish customer basic information table, vehicle information table, and association relationship table; use database index technology to index the frame number and contact information fields; By connecting with the vehicle management department database or a third-party vehicle information platform, the annual inspection information, insurance information, and repair and maintenance records of the customer's vehicle can be obtained in real time and automatically updated to the customer information database. The connection process uses API interface technology, and the data transmission process is encrypted. The encryption algorithm uses the AES-256-bit encryption algorithm. The data comparison algorithm is used to compare the new data obtained with the data in the existing database to identify the changed parts; Vehicle fault diagnosis module: equipped with a variety of vehicle fault diagnosis interfaces, supporting OBD-II and CAN bus diagnosis protocols. The diagnostic interface hardware uses high-performance chips, including Texas Instruments' automotive communication chips; Built-in vehicle fault diagnosis system, the system is built based on big data analysis and machine learning algorithms. The number of algorithm model training samples is not less than 500,000 groups, providing fault cause analysis, maintenance suggestions and maintenance cost estimates. The convolutional neural network deep learning algorithm is used to extract features and recognize patterns of vehicle fault data. The random gradient descent algorithm is used to optimize the model parameters during training, and the learning rate is dynamically adjusted. The initial learning rate is 0.01. When the loss function decreases by less than 0.001 for 5 consecutive iterations, the learning rate is reduced to 0.5 times the original value. The L2 regularization term is introduced to prevent overfitting, and the regularization coefficient is 0.001. The Bayesian network algorithm is used to perform probabilistic reasoning on the cause of the fault, and the maintenance suggestions are determined in combination with historical maintenance cases and expert experience. The maintenance cost estimate adopts a multivariate linear regression model, which is calculated based on factors such as fault type, parts price, and labor time. The regression model determination coefficient R² is not less than 0.8.

[0012] Maintenance work order management module: supports the creation, editing, query and printing of maintenance work orders. The work order management module is built using the Web development technology of the SpringBoot framework. The front-end page adopts a responsive design to adapt to a variety of terminal devices. The query function uses SQL optimization technology. It has the function of automated management of maintenance work order process, automatically assigns maintenance personnel according to maintenance projects, and maintenance personnel can receive work order information in real time through mobile terminals. The work order allocation algorithm adopts the principle of skill matching and workload balance. First, according to the skills required for the maintenance project, personnel with corresponding skills are screened out from the maintenance personnel skill database, and then the current workload of each person is calculated. The maintenance personnel with the smallest workload and the highest skill matching degree are selected for allocation, and the progress information is pushed to customers in a timely manner using Firebase Cloud Messaging message push technology; Repair parts management module: Establish a repair parts inventory database to record the name, specification, brand, inventory quantity, purchase price, sales price, and supplier information of the parts. The inventory database uses Redis real-time database technology to store inventory quantity data, and uses time series analysis algorithms to analyze historical sales data to predict the demand trend of parts. The inventory warning threshold is determined by calculating the average consumption speed, procurement cycle, and safety inventory coefficient of the parts. The calculation formula is: warning threshold = average consumption speed × procurement cycle × safety inventory coefficient. The safety inventory coefficient is between 1.2 and 1.5 according to the importance of the parts and the stability of supply. It has the function of intelligent matching of accessories, automatically matching the required repair accessories according to the vehicle fault diagnosis results and repair items, and providing accessories compatibility detection; Maintenance personnel management module: Establish maintenance personnel information database, store basic information of maintenance personnel, support the entry, modification and query of maintenance personnel information. The database adopts a layered architecture design, which is divided into data access layer, business logic layer and presentation layer. Memcached caching technology is used to cache frequently queried data; It has the function of evaluating the performance of maintenance personnel. Based on the multi-dimensional data of maintenance work order completion, customer satisfaction evaluation, and maintenance quality inspection results, the weighted average algorithm is used to calculate the performance score of maintenance personnel. The performance evaluation formula is: Performance score = work order completion rate × weight 1 + customer satisfaction score × weight 2 + maintenance quality score × weight 3. Weight 1, weight 2, and weight 3 are between 0.3 and 0.5 according to the management needs of the enterprise, and weight 1 + weight 2 + weight 3 = 1. Work order completion rate = number of completed work orders ÷ total number of work orders; customer satisfaction score is calculated by multiplying customer evaluation score by customer evaluation weight; maintenance quality score is calculated by multiplying maintenance quality inspection result by maintenance quality weight; Customer service evaluation module: Build a customer service evaluation platform. After the maintenance service is completed, customers can evaluate the maintenance service through multiple channels such as mobile phone text messages, WeChat public accounts, and system APPs. The evaluation content includes maintenance quality, maintenance speed, service attitude, and reasonableness of charges. The evaluation data is collected in real time and stored in the database. The evaluation platform is developed using a microservice architecture. Each evaluation channel runs as an independent microservice. Data asynchronous transmission and collection are achieved through the Kafka message queue. Data storage uses the Hadoop HDFS distributed file system. The Apriori association rule mining algorithm is used to analyze the relationship between maintenance items and customer satisfaction, and the key factors affecting customer satisfaction are discovered to provide a basis for service improvement. Data statistics and analysis module: It can perform statistical analysis on various types of data in the system. The statistical analysis function is implemented based on the Spark big data processing framework, using Spark's distributed computing capabilities to quickly process massive data. The pivot table function builds a multidimensional data set. Provide decision support functions based on data analysis results, predict peaks and troughs of maintenance business according to vehicle brand and model and maintenance frequency, and provide reference for maintenance resource allocation; ARIMA time series prediction model is used to predict maintenance business volume in the future based on historical maintenance record data; inventory optimization decisions are made by calculating the turnover rate of accessories, inventory turnover rate = sales cost ÷ average inventory balance, and sorting accessories according to turnover rate. Accessories with turnover rate lower than the threshold of 0.5 are listed as slow-moving accessories, and it is recommended to reduce inventory; accessories with turnover rate higher than the threshold of 2 are listed as popular accessories, and it is recommended to increase inventory.

[0013] Furthermore, when entering customer information, the customer information management module uses OCR optical character recognition technology to automatically identify the information on the vehicle driving license photo. The OCR technology is trained using an OCR model based on a convolutional neural network. The training samples include no less than 100,000 vehicle driving license photos of different types and resolutions. The identified information is compared and verified with the vehicle management department database through a data verification algorithm.

[0014] Furthermore, the vehicle fault diagnosis module performs real-time remote diagnosis during vehicle driving and transmits vehicle diagnostic data to the system server through the on-board mobile network module. The on-board mobile network module adopts an industrial-grade communication module, and the network signal strength monitoring algorithm monitors the network signal quality in real time. When the signal strength is lower than the threshold, the network mode is automatically switched or the transmission parameters are adjusted. The vehicle diagnostic data transmission adopts the GZIP compression algorithm. After the server receives the data, it quickly processes the data through asynchronous processing technology.

[0015] Furthermore, the maintenance work order management module supports the electronic signature function of the maintenance work order. Maintenance personnel and customers confirm the signature through a touch screen or electronic handwriting board. The electronic signature is implemented using digital certificate technology. The signature process uses the SHA-256 hash algorithm to hash the signature content to generate a unique hash value, which is stored together with the digital certificate. When verifying the signature, the hash value is calculated again and compared with the stored hash value. At the same time, the signature is encrypted using public key encryption technology.

[0016] Furthermore, the repair parts management module is connected to the supplier management system to realize the online generation, sending and tracking of purchase orders. Suppliers can update the order processing progress and delivery information in real time. The information synchronization delay does not exceed 1 minute, and the connection adopts the standardized electronic data interchange EDI protocol.

[0017] Furthermore, the maintenance personnel management module has the function of skill training management for maintenance personnel. It formulates personalized training plans based on the skill shortcomings of maintenance personnel and market demand. The skill assessment adopts a skill testing system, including theoretical knowledge tests and practical operation tests. The test questions cover all areas of automobile maintenance, and the skill shortcomings of maintenance personnel are analyzed based on the test results. The training plan is formulated based on the skill demand matrix, and the required skills are determined according to the popular car models and new technology development trends in the market, and personalized training plans are generated based on the current skill status of maintenance personnel.

[0018] Furthermore, the customer service evaluation module automatically converts the customer evaluation into text evaluation through the voice input function when the customer makes an evaluation. The voice recognition function adopts Baidu's voice recognition API or iFlytek's voice recognition engine. The voice recognition model is trained with a large number of voice samples. When voice is input, the voice signal is first preprocessed, including noise reduction and filtering operations. The converted text evaluation is semantically understood and corrected through natural language processing technology.

[0019] Furthermore, the data statistics and analysis module supports exporting statistical analysis results as documents in PDF and Word formats, which facilitates users to share data and generate reports. The document export function is implemented using open source document processing libraries, including Apache POI for Word and iText for PDF. During the export process, the statistical analysis results are filled into the corresponding document template according to the format and template selected by the user, generating a document with standardized format and complete content. The document format complies with industry standards.

[0020] Compared with the prior art, the present invention has the following beneficial effects: The system integrates customer information management, realizes centralized information storage and intelligent update, and improves information query and utilization efficiency. Vehicle fault diagnosis is more accurate and faster, reducing maintenance time. The maintenance work order process is automated, improving the speed and accuracy of work order processing, tracking maintenance progress in real time, and enhancing customer satisfaction.

[0021] Accurately predict the demand for repair parts, reasonably control inventory levels, avoid backlogs or out-of-stocks, and reduce inventory costs. Intelligent matching of parts ensures accuracy, improves repair efficiency, and reduces additional costs caused by mismatched parts.

[0022] A scientific performance evaluation system motivates maintenance personnel to improve their skills, and personalized training plans are highly targeted, which helps to improve the overall maintenance service level. High-quality maintenance services attract more customers and enhance the company's market competitiveness.

[0023] The multi-channel customer service evaluation is collected comprehensively and timely, and the data analysis is in-depth. Enterprises can improve services based on feedback. Customers can understand the progress of repairs in real time and participate in service evaluation through convenient methods such as voice input evaluation, which improves customer experience.

[0024] Powerful data statistics and analysis functions provide a basis for corporate decision-making, optimize resource allocation, predict market trends, help companies formulate scientific development strategies, adapt to market changes, and promote the intelligent and efficient development of the motor vehicle maintenance industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic block diagram of a motor vehicle maintenance management service system proposed by the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0027] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0028] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0029] Reference Figure 1 :A motor vehicle maintenance management service system, comprising: Customer information management module: Build a customer information database to store basic customer information, including name, contact information, vehicle brand and model, frame number, purchase time, mileage, etc. The data storage capacity is not less than 1 million records, and supports multiple data formats such as CSV, Excel, etc. The data import and export speed is not less than 10MB / s. Adopt a relational database management system (such as MySQL), optimize the database table structure design, establish a customer basic information table, vehicle information table, and relationship table, etc., to ensure the efficiency of data storage and data integrity. Use database indexing technology to index frequently queried fields (such as frame number and contact information) to improve query speed, and the index query efficiency is improved by no less than 50%.

[0030] It has the function of intelligently identifying and updating customer information. By connecting with the database of the vehicle management department or a third-party vehicle information platform, it can obtain the annual inspection information, insurance information, repair and maintenance records of the customer's vehicle in real time, and automatically update it to the customer information database. The update frequency is not less than once a week to ensure the timeliness and integrity of customer information. The connection process uses a safe and reliable API interface technology, and the data transmission process is encrypted. The encryption algorithm uses the AES-256-bit encryption algorithm to ensure data transmission security. Using the data comparison algorithm, the new data obtained is compared with the data in the existing database to identify the changed parts. The update algorithm has an accuracy rate of more than 99%, which effectively avoids duplicate data and erroneous updates.

[0031] Vehicle fault diagnosis module: equipped with a variety of vehicle fault diagnosis interfaces, supporting common diagnostic protocols such as OBD-II and CAN bus, and can communicate with the on-board diagnostic systems of more than 95% of the models on the market. The communication baud rate is adaptively adjusted in the range of 10kbps-1Mbps. The diagnostic interface hardware uses high-performance chips (such as Texas Instruments' automotive communication chips), with strong signal processing and anti-interference capabilities, and a signal processing accuracy of more than 99%. At the software level, the data of different protocols are parsed and converted through protocol parsing algorithms, and the protocol parsing success rate is not less than 98%.

[0032] Built-in vehicle fault diagnosis expert system, the system is built based on big data analysis and machine learning algorithms, the number of algorithm model training samples is not less than 500,000 groups, covering common faults and maintenance cases of various models, and can automatically diagnose vehicle faults based on the data obtained from the vehicle diagnostic interface, with a fault diagnosis accuracy of not less than 90%, and provide fault cause analysis, maintenance suggestions and maintenance cost estimates, with an estimated error range of ±10%. The expert system uses deep learning algorithms (such as convolutional neural networks) to extract features and pattern recognition from vehicle fault data. During the training process, the random gradient descent algorithm is used to optimize the model parameters, and the learning rate is dynamically adjusted. The initial learning rate is 0.01. When the loss function decreases by less than 0.001 for 5 consecutive iterations, the learning rate is reduced to 0.5 times the original. Regularization terms (such as L2 regularization) are introduced to prevent overfitting, and the regularization coefficient is 0.001. The Bayesian network algorithm is used to perform probabilistic reasoning on the cause of the fault, and the maintenance suggestions are determined in combination with historical maintenance cases and expert experience. The matching degree of maintenance suggestions is not less than 85%. The maintenance cost estimate is calculated using a multivariate linear regression model based on factors such as fault type, parts price, and labor costs. The coefficient of determination (R²) of the regression model is not less than 0.8.

[0033] Maintenance work order management module: supports the creation, editing, query and printing of maintenance work orders. The maintenance work order contains vehicle information, customer information, fault description, maintenance items, maintenance personnel, estimated maintenance time, estimated maintenance costs, etc. The work order creation time does not exceed 1 minute, and the query response time does not exceed 5 seconds. The system uses Web development technology (such as SpringBoot framework) to build the work order management module. The front-end page adopts responsive design and is compatible with a variety of terminal devices (such as computers, tablets, and mobile phones). The work order creation process uses a form validation algorithm to ensure the accuracy and completeness of the input data, and the verification accuracy rate reaches 100%. The query function uses SQL optimization technology (such as index optimization and query statement optimization) to improve query efficiency. After optimization, the query speed is increased by no less than 30%.

[0034] It has the function of automated management of maintenance work order processes, automatically assigns maintenance personnel according to maintenance projects, and maintenance personnel can receive work order information in real time through mobile terminals. The maintenance progress can be updated in real time and fed back to customers. Customers can query the work order status through the system, and the query result update delay does not exceed 10 seconds. The work order allocation algorithm adopts the principle of skill matching and workload balance. First, according to the skills required for the maintenance project, personnel with corresponding skills are screened out from the maintenance personnel skill database, and then the current workload of each person (the number of unfinished work orders and the estimated working hours) is calculated. The maintenance personnel with the smallest workload and the highest skill matching degree are selected for allocation, and the allocation accuracy is not less than 95%. The maintenance progress update is triggered in real time by the maintenance personnel's operation on the mobile terminal (such as starting maintenance, completing maintenance steps, pausing maintenance, etc.), and the progress information is pushed to the customer in a timely manner using message push technology (such as Firebase Cloud Messaging), and the message push success rate is not less than 98%.

[0035] Repair parts management module: Establish a repair parts inventory database to record the name, specification, brand, inventory quantity, purchase price, sales price, supplier information, etc. of the parts. The inventory data is updated in real time, and the data update delay does not exceed 1 second. The purchase plan can be automatically generated based on the historical sales data and the inventory warning threshold. The accuracy of the purchase plan generation is not less than 95%. The inventory database uses real-time database technology (such as Redis) to store key data such as inventory quantity to ensure the real-time and high-performance reading and writing of data. The time series analysis algorithm is used to analyze historical sales data and predict the demand trend of parts. The prediction accuracy is not less than 85%. The inventory warning threshold is determined by calculating the average consumption speed of the parts, the procurement cycle, the safety stock coefficient and other factors. The calculation formula is: Warning threshold = average consumption speed × procurement cycle × safety stock coefficient (the safety stock coefficient is between 1.2-1.5 according to the importance of the parts and the stability of supply).

[0036] It has the function of intelligent matching of accessories. It can automatically match the required repair accessories according to the vehicle fault diagnosis results and maintenance items, provide accessories compatibility detection, ensure that the selected accessories are compatible with the vehicle, and the accuracy of accessories matching is not less than 98%. At the same time, it provides the function of querying and navigating the location of accessories in the warehouse, so that maintenance personnel can quickly find accessories. The navigation path planning time does not exceed 5 seconds. The accessories matching algorithm is based on information such as vehicle model, fault code, and maintenance items. It performs fuzzy query and matching in the accessories knowledge base. The knowledge base covers no less than 100,000 types of vehicle models and 500,000 types of accessories information. The matching algorithm uses the cosine similarity algorithm to calculate the matching degree and select the accessories with the highest matching degree. The compatibility detection is judged by querying the accessories compatibility database (including the adaptation relationship between vehicle models and accessories, compatibility information between different accessories, etc.), and the accuracy of database information is not less than 99%. The query and navigation of the location of accessories in the warehouse use indoor positioning technology (such as Bluetooth positioning or Wi-Fi positioning), with a positioning accuracy of 1-3 meters. The navigation algorithm uses the A* algorithm or Dijkstra algorithm to plan the shortest path.

[0037] Maintenance personnel management module: Establish maintenance personnel information database to store basic information of maintenance personnel, including name, age, gender, qualification certificate, training record, vehicle models and projects that they are good at repairing, etc. Support the entry, modification and query of maintenance personnel information, and the query response time shall not exceed 3 seconds. The database adopts a layered architecture design, which is divided into data access layer, business logic layer and presentation layer. By optimizing the interaction and data transmission between each layer, the query performance is improved. Utilize cache technology (such as Memcached) to cache frequently queried data, with a cache hit rate of not less than 80%, effectively reducing the number of database queries and improving the query speed.

[0038] It has the function of performance evaluation of maintenance personnel. Based on the completion of maintenance work orders, customer satisfaction evaluation, maintenance quality inspection results and other multi-dimensional data, the weighted average algorithm is used to calculate the performance score of maintenance personnel. The performance evaluation cycle is once a month, and the evaluation results can be used as the basis for salary adjustment and promotion rewards for maintenance personnel. The performance evaluation formula is: Performance score = work order completion rate × weight 1 + customer satisfaction score × weight 2 + maintenance quality score × weight 3 (weight 1, weight 2, weight 3 are between 0.3-0.5 according to the management needs of the enterprise, and weight 1 + weight 2 + weight 3 = 1). Work order completion rate = number of completed work orders ÷ total number of work orders; customer satisfaction score is calculated by multiplying the customer evaluation score (full score 10 points) by the customer evaluation weight (determined by the number and importance of the evaluated customers); the maintenance quality score is calculated by multiplying the maintenance quality inspection result (qualified is 1, unqualified is 0) by the maintenance quality weight. The maintenance quality inspection adopts the sampling method, and the sampling ratio is not less than 10%.

[0039] Customer service evaluation module: Build a customer service evaluation platform. After the maintenance service is completed, customers can evaluate the maintenance service through various channels such as mobile phone text messages, WeChat public accounts, system apps, etc. The evaluation content includes maintenance quality, maintenance speed, service attitude, reasonableness of charges, etc. The evaluation data is collected and stored in the database in real time, and the data storage accuracy rate reaches 100%. The evaluation platform is developed using a microservice architecture. Each evaluation channel runs as an independent microservice, and data is asynchronously transmitted and collected through message queues (such as Kafka) to ensure data stability and efficiency. Data storage uses a distributed file system (such as Hadoop HDFS), which has high reliability and scalability and can store massive evaluation data.

[0040] It has the function of analyzing customer service evaluation data, uses data mining technology to analyze customer evaluation data, generates customer satisfaction reports, maintenance service quality analysis charts, etc., and provides data support for enterprises to improve service quality. The data analysis time does not exceed 1 hour, and the visualization of the analysis results is clear and intuitive. The data analysis uses natural language processing technology (such as sentiment analysis algorithm) to analyze the sentiment tendency of customer evaluation texts, and divides the evaluation into positive, negative and neutral. The accuracy of sentiment analysis is not less than 85%. The association rule mining algorithm (such as Apriori algorithm) is used to analyze the correlation between maintenance items and customer satisfaction, and the key factors affecting customer satisfaction are mined to provide a basis for service improvement. The visualization display uses visualization libraries such as Echarts to generate a variety of charts such as bar charts, line charts, and pie charts. The chart loading speed does not exceed 3 seconds, ensuring that users can quickly and intuitively view the analysis results.

[0041] Data statistics and analysis module: It can perform statistical analysis on various data in the system, including vehicle maintenance records, maintenance parts sales data, customer information change trends, etc. It provides a variety of visual analysis tools such as pivot tables, bar charts, line charts, pie charts, etc. It can customize the statistical analysis dimensions and time range according to user needs, and the data statistical analysis response time does not exceed 10 seconds. The statistical analysis function is implemented based on a big data processing framework (such as Spark), and uses Spark's distributed computing capabilities to quickly process massive data. The pivot table function enables users to flexibly summarize, filter and analyze data by building a multidimensional data set, and the pivot table calculation speed is increased by no less than 40%. The visualization tool is implemented using technologies such as WebGL to improve the performance of chart rendering and ensure smooth display even with large amounts of data.

[0042] Based on the results of data analysis, decision support functions are provided, such as predicting the peak and valley of maintenance business according to the brand and model of the vehicle and the maintenance frequency, providing a reference for the allocation of maintenance resources; analyzing popular and slow-selling parts according to the sales of repair parts, optimizing the inventory structure, and the effectiveness of decision support suggestions is not less than 80%. The maintenance business forecast adopts a time series forecasting model (such as the ARIMA model) to predict the future maintenance business volume based on historical maintenance record data, with a forecast error within ±15%. Inventory optimization decisions are made by calculating the turnover rate of parts, inventory turnover rate = sales cost ÷ average inventory balance, sorting parts according to turnover rate, and listing parts with a turnover rate below a certain threshold (such as 0.5) as slow-selling parts, and recommending to reduce inventory; listing parts with a turnover rate above a certain threshold (such as 2) as popular parts, and recommending to increase inventory. The optimized inventory structure can reduce inventory costs by not less than 10%.

[0043] In the present invention, when entering customer information, the customer information management module uses optical character recognition (OCR) technology to automatically identify the information on the vehicle driving license photo, with a recognition accuracy of no less than 95%, reducing the workload of manual entry. OCR technology uses a deep learning algorithm (such as an OCR model based on a convolutional neural network) for training, and the training samples include no less than 100,000 vehicle driving license photos of different types and resolutions. After multiple iterations of optimization, the model can accurately identify the text information on the driving license, including license plate number, vehicle type, owner, address, nature of use, brand model, vehicle identification code, engine number, registration date, date of issuance and other fields. The identified information is compared and verified with the vehicle management department database through a data verification algorithm, and the verification accuracy rate is more than 99%, ensuring the accuracy of the input information.

[0044] In the present invention, the vehicle fault diagnosis module can perform real-time remote diagnosis during vehicle driving, and transmit vehicle diagnostic data to the system server through the on-board mobile network module (such as 4G / 5G module). The remote diagnosis delay does not exceed 5 seconds, and potential faults are discovered in time and the owner is reminded. The on-board mobile network module adopts an industrial-grade communication module with high stability and low latency characteristics. The network signal strength monitoring algorithm monitors the network signal quality in real time. When the signal strength is lower than a certain threshold, it automatically switches the network mode or adjusts the transmission parameters to ensure stable data transmission. Vehicle diagnostic data transmission adopts data compression technology (such as GZIP compression algorithm), and the compression ratio is not less than 30%, which reduces data transmission traffic and improves transmission efficiency. After receiving the data, the server side quickly processes the data through asynchronous processing technology (such as message queues and thread pools) to ensure the timeliness of system response, and the asynchronous processing efficiency is improved by not less than 50%.

[0045] In the present invention, the maintenance work order management module supports the electronic signature function of the maintenance work order. Maintenance personnel and customers can confirm the signature through a touch screen or an electronic handwriting board. The electronic signature is legally binding and complies with the requirements of relevant laws and regulations. The signature verification accuracy rate reaches 100%. The electronic signature is implemented using digital certificate technology. The digital certificate is issued by an authoritative certification body. The certificate is valid for 1 year. The certificate update process is automatic and does not affect the normal operation of the system. The signature process uses a hash algorithm (such as SHA-256) to perform a hash operation on the signature content to generate a unique hash value. The hash value is stored together with the digital certificate. When verifying the signature, the hash value is calculated again and compared with the stored hash value to ensure that the signature has not been tampered with. At the same time, the signature is encrypted using public key encryption technology to ensure the security and non-repudiation of the signature.

[0046] In the present invention, the maintenance parts management module is connected to the supplier management system to realize the online generation, sending and tracking of purchase orders. Suppliers can update the order processing progress and delivery information in real time, and the information synchronization delay does not exceed 1 minute, thereby improving procurement efficiency and transparency. The connection adopts a standardized electronic data interchange (EDI) protocol to ensure the consistency and compatibility of the data format. The purchase order generation process is implemented through smart contract technology to ensure the accuracy and non-tamperability of the order terms, and the order generation time does not exceed 10 seconds. The order tracking function uses blockchain technology to record every status change of the order, including order placement, review, delivery, receipt, etc. The blockchain nodes are distributed among maintenance companies, suppliers and other related parties to ensure the transparency and traceability of information, and the response time for querying the order history status does not exceed 5 seconds.

[0047] In the present invention, the maintenance personnel management module has the function of skill training management for maintenance personnel. According to the skill shortcomings of maintenance personnel and market demand, a personalized training plan is formulated. The accuracy of training plan execution tracking is not less than 95%, thereby improving the overall skill level of maintenance personnel. Skill assessment adopts a skill testing system, including theoretical knowledge test and practical operation test. The test questions cover all areas of automobile maintenance, and the skill shortcomings of maintenance personnel are analyzed according to the test results. The training plan is formulated based on the skill demand matrix. The required skills are determined according to the popular models and new technology development trends in the market, and a personalized training plan is generated in combination with the current status of maintenance personnel's skills. The execution tracking of the training plan is realized through the training management system, which records the courses, time, test scores and other information of maintenance personnel participating in the training, and uses reminder functions (such as SMS reminders, system push) to ensure that maintenance personnel attend the training on time, with a reminder accuracy of more than 98%.

[0048] In the present invention, the customer service evaluation module can automatically convert the customer evaluation into text evaluation through the voice input function, and the voice recognition accuracy is not less than 90%, which is convenient for customer operation. The voice recognition function adopts a voice recognition engine (such as Baidu voice recognition API or iFlytek voice recognition engine), and the voice recognition model is trained with a large number of voice samples, including voice data with different accents, speaking speeds, and environmental noises, and the number of training samples is not less than 1 million. During voice input, the voice signal is first preprocessed, including noise reduction, filtering and other operations to improve the accuracy of voice recognition. The converted text evaluation is semantically understood and corrected through natural language processing technology to ensure the accuracy of the evaluation content, and the accuracy of semantic understanding is not less than 80%.

[0049] In the present invention, the data statistics and analysis module supports exporting the statistical analysis results into documents in PDF, Word and other formats, which is convenient for users to share data and generate reports, and the document export time does not exceed 1 minute. The document export function is implemented using open source document processing libraries (such as Apache POI for Word and iText for PDF), and the export efficiency is improved by optimizing the data format conversion algorithm and document generation process. During the export process, according to the format and template selected by the user, the statistical analysis results are filled into the corresponding document template to generate a document with standardized format and complete content. The document format meets industry standards to ensure the compatibility of the document on different platforms and software.

[0050] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A motor vehicle maintenance management service system, characterized in that: include: Customer information management module: build a customer information database to store basic customer information, support CSV and Excel data format import and export, use MySQL relational database management system, optimize database table structure design, establish customer basic information table, vehicle information table, and association relationship table; Use database indexing technology to create indexes for vehicle frame number and contact information fields; By connecting with the vehicle management department database or a third-party vehicle information platform, the annual inspection information, insurance information, and repair and maintenance records of the customer's vehicle can be obtained in real time and automatically updated to the customer information database. The connection process uses API interface technology, and the data transmission process is encrypted. The encryption algorithm uses the AES-256-bit encryption algorithm. The data comparison algorithm is used to compare the new data obtained with the data in the existing database to identify the changed parts; Vehicle fault diagnosis module: equipped with a variety of vehicle fault diagnosis interfaces, supporting OBD-II and CAN bus diagnosis protocols. The diagnostic interface hardware uses high-performance chips, including Texas Instruments' automotive communication chips; Built-in vehicle fault diagnosis system, which is built based on big data analysis and machine learning algorithms. The number of algorithm model training samples is not less than 500,000 groups. It provides fault cause analysis, maintenance suggestions and maintenance cost estimates. It uses convolutional neural network deep learning algorithm to extract features and recognize patterns of vehicle fault data. During the training process, the random gradient descent algorithm is used to optimize the model parameters. The learning rate is dynamically adjusted. The initial learning rate is 0.

01. When the loss function decreases by less than 0.001 for 5 consecutive iterations, the learning rate is reduced to 0.5 times the original value. The L2 regularization term is introduced to prevent overfitting, and the regularization coefficient is 0.

001. The Bayesian network algorithm is used to perform probabilistic reasoning on the cause of the fault, and the maintenance suggestions are determined by combining historical maintenance cases and expert experience. The maintenance cost estimate uses a multivariate linear regression model, which is calculated based on factors such as fault type, parts price, and labor time. The regression model determination coefficient R² is not less than 0.

8. Maintenance work order management module: supports the creation, editing, query and printing of maintenance work orders. The work order management module is built using the Web development technology of the SpringBoot framework. The front-end page adopts a responsive design to adapt to a variety of terminal devices. The query function uses SQL optimization technology. It has the function of automated management of maintenance work order process, automatically assigns maintenance personnel according to maintenance projects, and maintenance personnel can receive work order information in real time through mobile terminals. The work order allocation algorithm adopts the principle of skill matching and workload balance. First, according to the skills required for the maintenance project, personnel with corresponding skills are screened out from the maintenance personnel skill database, and then the current workload of each person is calculated. The maintenance personnel with the smallest workload and the highest skill matching degree are selected for allocation, and the progress information is pushed to customers in a timely manner using Firebase Cloud Messaging message push technology; Repair parts management module: Establish a repair parts inventory database to record the name, specification, model, brand, inventory quantity, purchase price, sales price, and supplier information of the parts. The inventory database uses Redis real-time database technology to store inventory quantity data, and uses time series analysis algorithms to analyze historical sales data and predict the demand trend of parts; The inventory warning threshold is determined by calculating the average consumption rate, procurement cycle, and safety inventory coefficient of the accessories. The calculation formula is: Warning threshold = average consumption rate × procurement cycle × safety inventory coefficient. The safety inventory coefficient is between 1.2 and 1.5 according to the importance of the accessories and the stability of supply. It has the function of intelligent matching of accessories, automatically matching the required repair accessories according to the vehicle fault diagnosis results and repair items, and providing accessories compatibility detection; Maintenance personnel management module: Establish maintenance personnel information database, store basic information of maintenance personnel, support the entry, modification and query of maintenance personnel information. The database adopts a layered architecture design, which is divided into data access layer, business logic layer and presentation layer. Memcached caching technology is used to cache frequently queried data; It has the function of evaluating the performance of maintenance personnel. Based on the multi-dimensional data of maintenance work order completion, customer satisfaction evaluation, and maintenance quality inspection results, the weighted average algorithm is used to calculate the performance score of maintenance personnel. The performance evaluation formula is: Performance score = work order completion rate × weight 1 + customer satisfaction score × weight 2 + maintenance quality score × weight 3. Weight 1, weight 2, and weight 3 are between 0.3 and 0.5 according to the management needs of the enterprise, and weight 1 + weight 2 + weight 3 = 1. Work order completion rate = number of completed work orders ÷ total number of work orders. The customer satisfaction score is calculated by multiplying the customer evaluation score by the customer evaluation weight; the maintenance quality score is calculated by multiplying the maintenance quality inspection result by the maintenance quality weight; Customer service evaluation module: Build a customer service evaluation platform. After the maintenance service is completed, customers can evaluate the maintenance service through multiple channels such as mobile phone text messages, WeChat public accounts, and system APPs. The evaluation content includes maintenance quality, maintenance speed, service attitude, and reasonableness of charges. The evaluation data is collected in real time and stored in the database. The evaluation platform is developed using a microservice architecture. Each evaluation channel runs as an independent microservice. Data asynchronous transmission and collection are achieved through the Kafka message queue. Data storage uses the Hadoop HDFS distributed file system. The Apriori association rule mining algorithm is used to analyze the relationship between maintenance items and customer satisfaction, and the key factors affecting customer satisfaction are discovered to provide a basis for service improvement. Data statistics and analysis module: It can perform statistical analysis on various types of data in the system. The statistical analysis function is implemented based on the Spark big data processing framework, using Spark's distributed computing capabilities to quickly process massive data. The pivot table function builds a multidimensional data set. Provide decision support functions based on data analysis results, predict peaks and troughs of maintenance business according to vehicle brand and model and maintenance frequency, and provide reference for maintenance resource allocation; ARIMA time series prediction model is used to predict maintenance business volume in the future based on historical maintenance record data; inventory optimization decisions are made by calculating the turnover rate of accessories, inventory turnover rate = sales cost ÷ average inventory balance, and sorting accessories according to turnover rate. Accessories with turnover rate lower than the threshold of 0.5 are listed as slow-moving accessories, and it is recommended to reduce inventory; accessories with turnover rate higher than the threshold of 2 are listed as popular accessories, and it is recommended to increase inventory.

2. The motor vehicle maintenance management service system according to claim 1, characterized in that: When entering customer information, the customer information management module uses OCR optical character recognition technology to automatically identify the information on the vehicle driving license photo. The OCR technology is trained using an OCR model based on a convolutional neural network. The training samples include no less than 100,000 vehicle driving license photos of different types and resolutions. The identified information is compared and verified with the vehicle management department database through a data verification algorithm.

3. The motor vehicle maintenance management service system according to claim 1, characterized in that: The vehicle fault diagnosis module performs real-time remote diagnosis during vehicle driving and transmits vehicle diagnostic data to the system server through the on-board mobile network module. The on-board mobile network module adopts an industrial-grade communication module, and the network signal strength monitoring algorithm monitors the network signal quality in real time. When the signal strength is lower than the threshold, the network mode is automatically switched or the transmission parameters are adjusted. The vehicle diagnostic data transmission adopts the GZIP compression algorithm. After the server receives the data, it quickly processes the data through asynchronous processing technology.

4. The vehicle maintenance management service system according to claim 1, characterized in that: The maintenance work order management module supports the electronic signature function of the maintenance work order. Maintenance personnel and customers confirm their signatures through a touch screen or electronic handwriting pad. The electronic signature is implemented using digital certificate technology. The signature process uses the SHA-256 hash algorithm to hash the signature content to generate a unique hash value, which is stored together with the digital certificate. When verifying the signature, the hash value is calculated again and compared with the stored hash value. At the same time, the signature is encrypted using public key encryption technology.

5. The vehicle maintenance management service system according to claim 1, characterized in that: The repair parts management module is connected to the supplier management system to realize the online generation, sending and tracking of purchase orders. Suppliers can update the order processing progress and delivery information in real time. The information synchronization delay does not exceed 1 minute. The connection adopts the standardized electronic data interchange EDI protocol.

6. The vehicle maintenance management service system according to claim 1, characterized in that: The maintenance personnel management module has the maintenance personnel skill training management function. According to the maintenance personnel's skill shortcomings and market demand, a personalized training plan is formulated. The skill assessment adopts a skill test system, including theoretical knowledge test and practical operation test. The test questions cover all areas of automobile maintenance, and the maintenance personnel's skill shortcomings are analyzed according to the test results. The training plan is formulated based on the skill demand matrix. The required skills are determined according to the popular models and new technology development trends in the market, and personalized training plans are generated based on the current skill status of maintenance personnel.

7. The vehicle maintenance management service system according to claim 1, characterized in that: When a customer makes a comment, the customer service evaluation module automatically converts it into text evaluation through the voice input function. The voice recognition function uses Baidu's voice recognition API or iFlytek's voice recognition engine. The voice recognition model is trained with a large number of voice samples. When voice is input, the voice signal is preprocessed, including noise reduction and filtering operations. The converted text evaluation is semantically understood and corrected through natural language processing technology.

8. The vehicle maintenance management service system according to claim 1, characterized in that: The data statistics and analysis module supports exporting statistical analysis results to documents in PDF and Word formats, which facilitates users to share data and generate reports. The document export function is implemented using open source document processing libraries, including Apache POI for Word and iText for PDF. During the export process, the statistical analysis results are filled into the corresponding document template according to the format and template selected by the user, generating documents with standardized format and complete content. The document format complies with industry standards.

Images