Intelligent operation system for digital automobile maintenance and repair market

Abstract

The invention discloses a digital automobile maintenance and repair market intelligent operation system, particularly relates to the field of automobile maintenance and repair, and comprises a database, a vehicle information acquisition module, a data preprocessing module, a repair auxiliary diagnosis module, a work order intelligent scheduling module and a user information terminal. The maintenance auxiliary diagnosis module constructs a fault diagnosis model based on a vehicle data processing text and a historical maintenance case library, the accuracy and efficiency of fault diagnosis are improved, a maintenance work order is generated according to a diagnosis result and an association rule table, accurate selection of accessories needed by maintenance is ensured, and the maintenance time and cost are reduced; according to the intelligent work order scheduling module, the objective function and the genetic algorithm are constructed, related variables are comprehensively considered, reasonable distribution of maintenance work orders is achieved, resource utilization of maintenance enterprises is effectively balanced, work order waiting time and total maintenance time are shortened, and overall operation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile maintenance and repair, and more specifically, to an intelligent operation system for the digital automobile maintenance and repair market. Background Art

[0002] The intelligent operation system for the digital automobile maintenance and repair market is a technical solution proposed for the core pain points such as low efficiency, data silos, and insufficient service standardization in the automobile after-sales service market. With the continuous growth of the automobile ownership and the increase in the penetration rate of new energy vehicle models, the traditional maintenance mode relying on manual experience has been difficult to meet the users' needs for precision, transparency, and preventive maintenance. Therefore, building an intelligent operation system based on the new generation of information technology plays a key role in improving service efficiency, reducing operation and maintenance costs, and optimizing the user experience.

[0003] In the prior art, there are also some solutions related to the intelligent operation of the digital automobile maintenance and repair market. For example, a vehicle fault repair system with the Chinese patent publication number CN111079952A, which is characterized by including a central server for data processing and transmission. The input end of the central server is connected with a repair registration module for registering maintenance work task information. The output end of the central processor is connected with a repair module for processing vehicle repair information. The input end of the repair module is connected with a repairman information module for recording the information of the vehicle repairman. The output end of the repair module is connected with a repair plan module for the specific repair plan proposed by the repairman according to the vehicle fault situation. Through the direct communication between the vehicle owner and the repairman, the repairman is responsible for the vehicle repair situation. The transparent vehicle repair process is shown to the vehicle owner by using the repair plan module, so that the vehicle owner can fully understand the repairman, the repair plan, and the fault cause, and provide high-quality vehicle fault repair services.

[0004] However, when it is actually used, there are still some disadvantages. For example, the vehicle information is obtained incompletely and untimely, it is difficult to achieve precise management of the vehicle's entire life cycle. The maintenance business data is scattered, lacking effective integration and analysis, and unable to provide strong support for decision-making. The fault diagnosis method relying on manual experience has low diagnosis efficiency and is prone to misdiagnosis. The vehicle sensor data and the historical maintenance case library are not fully utilized, and it is difficult to quickly and accurately judge the fault type. The maintenance work order scheduling lacks scientific planning, and the resource status of the maintenance enterprise is not fully considered, resulting in idle or overused workstations, uneven workloads of technicians, and chaotic spare parts inventory management, increasing the operation cost. Summary of the Invention

[0005] To overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an intelligent operation system for the digital automotive maintenance and repair market, and through the following solutions, to solve the problems raised in the above-mentioned background art.

[0006] To achieve the above object, the present invention provides the following technical solutions: An intelligent operation system for the digital automotive maintenance and repair market, including a database, a vehicle information acquisition module, a data preprocessing module, a maintenance auxiliary diagnosis module, a work order intelligent scheduling module, and a user information terminal;

[0007] The database: is used to store all data texts obtained by the intelligent operation system for the digital automotive maintenance and repair market;

[0008] The vehicle information acquisition module: includes a vehicle sensor interface and a business data collection unit, and is used to acquire vehicle data texts and maintenance business data texts;

[0009] The data preprocessing module: cleans the numerical data in the acquired vehicle data texts and maintenance business data texts, removes noise data, duplicate data, and abnormal data, and performs data standardization processing to obtain vehicle data processing texts and maintenance business data processing texts;

[0010] The maintenance auxiliary diagnosis module: constructs a fault diagnosis model based on the vehicle data processing texts and the historical maintenance case library, assists maintenance personnel in diagnosing the fault type based on the fault diagnosis model, and selects the accessories required for maintenance to generate a maintenance work order text based on the diagnosis result and the association rule table;

[0011] The work order intelligent scheduling module: is used to obtain a scheduling variable text according to the maintenance work order text, import the scheduling variable text into the objective function construction text to obtain an objective function, construct a genetic algorithm text based on the objective function and the scheduling variable, and iteratively obtain a work order allocation plan based on the genetic algorithm text;

[0012] The user information terminal: includes an owner client and a maintenance enterprise terminal, and is used for information interaction between the owner and the maintenance enterprise.

[0013] The technical effects and advantages of the present invention:

[0014] 1. The present invention comprehensively collects vehicle data and maintenance business data through the vehicle information acquisition module and stores them in the database, realizing centralized management and sharing of information;

[0015] 2. The maintenance auxiliary diagnosis module of the present invention constructs a fault diagnosis model based on the vehicle data processing texts and the historical maintenance case library, and uses technologies such as data annotation, feature selection, and feature encoding to improve the accuracy and efficiency of fault diagnosis, generates a maintenance work order according to the diagnosis result and the association rule table, ensures the accurate selection of accessories required for maintenance, and reduces maintenance time and costs;

[0016] 3. The work order intelligent scheduling module of the present invention constructs an objective function and a genetic algorithm, comprehensively considers variables related to maintenance enterprises, technicians, workstations, and spare parts, realizes the reasonable allocation of maintenance work orders, effectively balances the resource utilization of maintenance enterprises, reduces the work order waiting time and the total maintenance time, and improves the overall operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall structure of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] As shown in the attached Figure 1 The digital intelligent operation system for the automotive maintenance and repair market includes a database, a vehicle information acquisition module, a data preprocessing module, a maintenance auxiliary diagnosis module, a work order intelligent scheduling module, and a user information terminal;

[0020] The database: is used to store all data texts obtained by the digital intelligent operation system for the automotive maintenance and repair market;

[0021] The vehicle information acquisition module: includes a vehicle sensor interface and a business data collection unit, and is used to acquire vehicle data texts and maintenance business data texts;

[0022] The vehicle sensor interface is used to connect to the vehicle internal network through the on-vehicle diagnostic system interface of the vehicle, use the on-vehicle diagnostic system reading device to acquire sensor data, connect to the on-vehicle diagnostic system interface of the vehicle through a fault diagnostic instrument to read the fault codes stored in the vehicle electronic control unit, and summarize the data to obtain vehicle data texts;

[0023] The business data collection unit is used to acquire work order information, customer information, and spare parts inventory data from the business system of the maintenance enterprise, acquire reservation information from the vehicle owner client, and summarize the acquired information to obtain maintenance business data texts;

[0024] It should be specifically noted that the sensor data includes engine data, driving system data, and electrical data;

[0025] The engine data includes: engine speed, intake air volume, throttle opening, coolant temperature, oil pressure, and fuel injection volume. These data can directly reflect the operating state of the engine. For example, unstable engine speed may imply ignition system failure or mechanical component wear; abnormal increase in coolant temperature may be due to cooling system failure.

[0026] The driving system data includes: vehicle speed, tire pressure, four-wheel alignment parameters, and suspension system sensor data. Vehicle speed data is used to judge the driving conditions of the vehicle. Abnormal tire pressure may lead to driving safety problems. Abnormal four-wheel alignment parameters will affect vehicle handling and tire wear. Body height sensor data can assist in diagnosing suspension system failures.

[0027] The electrical data includes: battery voltage, generator output voltage, and current consumption of headlights, air conditioners, and radios. Too low battery voltage may be due to battery aging or charging system failure. Abnormal current in electrical components may indicate internal short circuit or poor contact of the component.

[0028] It should be further noted that when the vehicle electronic control system detects a fault, it will generate fault codes. These codes follow specific standard protocols, such as the fault codes in the OBD-II standard. Each fault code corresponds to a specific fault type. For example, P0101 indicates air flow sensor failure, and P0300 indicates engine misfire failure. The fault codes also include relevant information such as the time and frequency of the fault occurrence and the vehicle operating conditions when the fault occurred;

[0029] The data preprocessing module: cleans the numerical data in the obtained vehicle data text and maintenance business data text, removes noise data, duplicate data, and abnormal data, and performs data standardization processing to obtain the vehicle data processing text and maintenance business data processing text;

[0030] The maintenance assistance diagnosis module: constructs a fault diagnosis model based on the vehicle data processing text and the historical maintenance case library, assists maintenance personnel in diagnosing the fault type based on the fault diagnosis model, and selects the accessories required for maintenance to generate a maintenance work order text based on the diagnosis result and the association rule table;

[0031] Specifically, the historical maintenance case library specifically refers to extracting historical maintenance case data from the business management system of the maintenance enterprise. These cases include vehicle models, vehicle identification numbers, production dates, fault descriptions, diagnostic tools used and test data, final fault reasons, and maintenance measures. Through the data interface, the maintenance case library is synchronously updated monthly to ensure the timeliness of the cases;

[0032] The specific method for constructing the fault diagnosis model is as follows:

[0033] A1: Data annotation: For vehicle sensor data, according to the fault causes in the maintenance case library, the corresponding sensor data segments are annotated as the corresponding fault types. For example, if the fault cause in a certain maintenance case is "engine coolant leakage", then the coolant temperature sensor data, liquid level sensor data, etc. during a period before and after the occurrence of this case are annotated as the "engine coolant leakage" fault type; for the historical maintenance case library, standardize the annotation format of fault types, and classify similar fault descriptions into the same standard fault type to facilitate classification processing during subsequent model training;

[0034] A2: Feature selection: Utilize the professional knowledge in the field of automotive maintenance to exclude features that are obviously irrelevant to the fault type. For example, vehicle interior color, audio brand, etc. are irrelevant to engine fault types and can be directly removed from the feature set. For features with a change rate less than 5% in all data samples, such as the tire specifications being the same in all maintenance cases of a certain vehicle model, or the measured value volatility of the feature value being less than 5% under normal conditions, they are excluded because these features contribute less to distinguishing different fault types. For numerical features, calculate the Pearson correlation coefficient between each feature and the fault type, and discard features with an absolute value less than 0.2. For example, calculate the Pearson correlation coefficient between the engine coolant temperature and the "engine coolant leakage" fault type. If the absolute value of the correlation coefficient is less than 0.2, then discard this feature;

[0035] It should be further noted that when calculating the Pearson correlation coefficient, the fault type needs to be encoded as 0 - 1. For example, 0 represents no "engine coolant leakage" fault, and 1 represents having this fault;

[0036] A3: Feature encoding: Encode the non - numerical features selected in the feature selection step according to their frequencies of occurrence in a certain fault type. For example, in the engine coolant leakage fault, the frequency of engine jitter is 0.3, then encode engine jitter as 0.3;

[0037] A4: Model construction: Determine the number of input layer nodes according to the number of features determined in the feature selection step. Suppose that after screening and processing, there are 40 features including sensor data features and maintenance case features, then the number of input layer nodes is 40. Set 2 - 3 hidden layers, with 50 nodes in each layer. The hidden layers are fully connected, and the ReLU activation function is used, whose expression is f(x) = max(0, x) to introduce non - linear factors and enhance the model's expressive ability. The number of output layer nodes is equal to the number of fault types. Suppose there are 20 standard fault types and their sub - categories in total, then the number of output layer nodes is 20. The Softmax activation function is used to convert the model output into the probability distribution of each fault type, and its expression is:

[0038] where z is the input vector, K is the number of categories, and σ(z)j represents the probability of the j-th type of fault;

[0039] A5: Fault judgment: After obtaining the final predicted probability of the fault type, select the fault type with the highest probability as the diagnostic result;

[0040] Specifically, the association rule table refers to the list of maintenance parts associated with the fault type, and customers can select the required maintenance parts according to their own needs;

[0041] It should be further noted that the association rule table is based on the mapping relationship table between the fault type and the maintenance parts. The table clearly shows the necessary maintenance parts, optional maintenance parts corresponding to each fault type, and the combination relationship between them. When generating a maintenance work order, according to the diagnosed fault type, the system automatically filters out the required accessory information based on the association rule table, and the customer can select and add it to the maintenance work order text;

[0042] The maintenance work order text specifically includes: accessory information required for maintenance, estimated market for maintenance, and fault type;

[0043] The work order intelligent scheduling module: is used to obtain the scheduling variable text according to the maintenance work order text, import the scheduling variable text into the objective function construction text to obtain the objective function, construct the genetic algorithm text based on the objective function and the scheduling variables, and iteratively obtain the work order allocation plan based on the genetic algorithm text;

[0044] Specifically, the scheduling variable text specifically includes variables related to maintenance enterprises, variables related to technicians, variables related to workstations, variables related to accessories, variables related to maintenance work orders, decision variables, and work order waiting time variables;

[0045] The variables related to maintenance enterprises include:

[0046] Set of maintenance enterprises: M = {m1, m2,..., m n}

[0047] Definition: The set containing all maintenance enterprises in the system, and each element m i represents a maintenance enterprise;

[0048] Obtaining method: In the system initialization stage, the administrator enters the information of all cooperative maintenance enterprises into the system, and the system automatically generates this set.

[0049] Number of workstations of maintenance enterprise m i : |S i |

[0050] Definition: Represents the total number of workstations owned by maintenance enterprise m.

[0051] Obtaining method: When a maintenance enterprise enters the system, it provides the system with information about the number of its workstations, and the system records and stores it.

[0052] Maintenance enterprise m i Workstation s in ij Available quantity:

[0053] Definition: The j-th workstation of maintenance enterprise m i is currently in an idle state and the available quantity, which is represented as 0 or 1, where 0 means unavailable and 1 means available.

[0054] Obtaining method: The status of the workstation is monitored in real time through sensors installed on the workstation. When there is no vehicle for maintenance work on the workstation, the sensor feeds back a signal to the system, and the system updates the value to 1; conversely, when there is a vehicle being repaired at this workstation, it is updated to 0. At the same time, before each scheduling work order, the system reads the latest workstation status information to obtain this value.

[0055] Maintenance enterprise m i Workstation s in ij Current occupancy quantity:

[0056] Definition: The number of the j-th workstation of maintenance enterprise m i that is currently being occupied for maintenance work, which is represented as 0 or 1, where 0 means unoccupied and 1 means occupied.

[0057] Obtaining method: Also relying on the workstation sensor, when it is detected that there is a vehicle being repaired at the workstation, the system will update it to 1, and after the vehicle leaves the workstation after the repair is completed, it is updated to 0. When the system processes the work order scheduling, it obtains this value in real time to judge the workstation occupancy situation.

[0058] The technician-related variables include:

[0059] Technician set: T = {t1, t2,..., t p}

[0060] Definition: The set consisting of all technicians in the system, and each element t k represents a technician.

[0061] Obtaining method: The maintenance enterprise enters the information of its technicians into the system, and the system generates a technician set accordingly.

[0062] Skill level vector of technician t k :

[0063] Definition: This vector represents technician t kThe degree of mastery of different types of maintenance skills. kn Indicates technician k The mastery level of the nth type of maintenance skill ranges from 0 to 1, where 0 indicates no mastery at all and 1 indicates proficiency.

[0064] How to obtain: After the maintenance company conducts a skill assessment on the technician, the assessment results are entered into the system. The system generates the technician's skill level vector based on the entered information. At the same time, as the technician continues to participate in training and actual maintenance work, his skill level may change. The maintenance company needs to regularly update the technician's skill assessment information, and the system will update the skill level vector synchronously.

[0065] Technician k Current workload: w tk

[0066] Definition: indicates technician k The number of maintenance work orders currently assigned.

[0067] How to obtain: When the system assigns a work order to a technician, the system will assign the technician's corresponding w tk The value increases by 1; when the technician completes a work order maintenance task and confirms it in the system, the system will tk The value is reduced by 1. When the system schedules a new work order, it reads the value in real time to evaluate the technician's workload.

[0068] The station-related variables include:

[0069] Workstation collection:

[0070] Definition: A set of workstations of all maintenance enterprises in the system, each element s ij Representative maintenance company i The j-th workstation.

[0071] Acquisition method: It is generated based on the summary of the workstation information entered by each maintenance enterprise. When the system is initialized, all the workstation information of each maintenance enterprise is integrated into this collection.

[0072] The accessory related variables include:

[0073] Accessory set: P = {p1, p2, ..., p r}

[0074] Definition: The set of all car parts involved in the system, each element p r Represents an accessory.

[0075] How to obtain: The parts supplier enters the parts information it supplies into the system, and the system generates a parts collection based on the information.

[0076] Maintenance Companyi Spare part p r Inventory quantity: q pli

[0077] Definition: Maintenance enterprise m i The quantity of spare part p in the current inventory r Quantity

[0078] Obtaining method: The maintenance enterprise updates the inventory quantity of spare parts in real time through the spare part management module of the system. When new spare parts are purchased and put into storage, the inventory quantity of the corresponding spare parts is increased; when spare parts are used for maintenance, the inventory quantity is decreased.

[0079] The relevant variables of the maintenance work order include:

[0080] Set of maintenance work orders: O, which is represented by ∣O∣ in the objective function and some constraint conditions to indicate the number of work orders;

[0081] Definition: The set of all pending maintenance work orders in the system.

[0082] Obtaining method: The car owner submits a maintenance appointment request through the system client, and the system generates a maintenance work order according to the request information and adds it to this set.

[0083] Maintenance work order o j Set of required spare parts: P oj

[0084] Definition: The set composed of the spare parts required for maintenance work order o j Set of required spare parts

[0085] Obtaining method: After the maintenance technician views the vehicle fault information corresponding to the work order, combined with maintenance experience and maintenance manuals, selects or enters the spare part information required for this work order in the system, and the system generates set P accordingly oj Set

[0086] Maintenance work order o j Required maintenance duration: d oj

[0087] Definition: The estimated time required to complete the vehicle maintenance task corresponding to maintenance work order o j Estimated time required

[0088] Obtaining method: The system calculates the average maintenance duration of similar faults in the historical maintenance data, calculates the maintenance efficiency of the current maintenance enterprise and the actual vehicle fault situation based on the past maintenance duration data of this enterprise, and estimates the maintenance duration d through a weighted average algorithm oj . At the same time, the maintenance technician can also appropriately adjust the estimated duration in the system according to actual experience.

[0089] Maintenance work order o jFault type: type oj

[0090] Definition: Represents a maintenance work order o j The type to which the vehicle fault corresponding to the work order belongs.

[0091] Obtaining method: The maintenance technician obtains the fault code through the vehicle fault diagnosis equipment, combines the fault code library and the actual detection situation, selects or enters the fault type information in the system, and the system records it as type oj .

[0092] The decision variables include:

[0093] Work order o j Whether it is assigned to the work station s of the maintenance enterprise m i On: ij On:

[0094] Definition: The value is 0 or 1. 0 means the work order o j Is not assigned to the work station s of the maintenance enterprise m i On, 1 means it is assigned to this work station. ij On, 1 means it is assigned to this work station.

[0095] Obtaining method: This variable is assigned by the intelligent scheduling algorithm when calculating the work order assignment plan according to various constraint conditions and objective functions. In each iteration of the algorithm, by calculating and comparing different assignment combinations, the value of this variable is determined to achieve the optimal work order assignment plan.

[0096] Work order o j Whether it is assigned to the technician t k :

[0097] Definition: The value is 0 or 1. 0 means the work order o j Is not assigned to the technician t k , 1 means it is assigned to this technician.

[0098] Obtaining method: Similarly, the intelligent scheduling algorithm assigns a value to this variable during the calculation process based on constraint conditions such as the matching between the technician's skill level and the work order fault type and the technician's workload to achieve a reasonable assignment of technician work orders.

[0099] Work order o j Whether it is assigned to the maintenance enterprise m i :

[0100] Definition: The value is 0 or 1. 0 means the work order o j Is not assigned to the maintenance enterprise m i , 1 means it is assigned to this enterprise.

[0101] Obtaining method: The intelligent scheduling algorithm comprehensively considers factors such as the availability of workstations in the maintenance enterprise and the inventory of spare parts, and assigns a value to this variable during the calculation to determine the matching relationship between the work order and the maintenance enterprise.

[0102] The work order waiting time variable includes:

[0103] Work order o j Waiting time: w oj

[0104] Definition: The time elapsed from when work order o j enters the system and waits for the start of maintenance to the actual start of maintenance.

[0105] Obtaining method: The system records the current time t1 when the work order enters the waiting queue, and records the time t2 when the work order is assigned to a maintenance workstation and starts maintenance. Then w oj = t2 - t1. The system obtains the waiting time of each work order through real-time recording and calculation, which is used for the calculation of the objective function to optimize the work order scheduling plan and reduce the total waiting time.

[0106] The text for constructing the objective function includes: constraint conditions and the objective function;

[0107] The constraint conditions include:

[0108] Workstation constraint: Each maintenance work order can only be assigned to an available workstation of one maintenance enterprise, that is where is a decision variable, indicating whether work order o j is assigned to workstation s i of maintenance enterprise m ij , with a value of 0 or 1. At the same time, ensure that the workstation is not overloaded after assignment.

[0109] Technician constraint: The technician assigned to the maintenance work order must have the corresponding skills. where is a decision variable, indicating whether work order o j is assigned to technician t k , with a value of 0 or 1, and the workload of the technician cannot exceed its maximum workload.

[0111] Spare part constraint: The spare parts required for the maintenance work order must be in stock in the maintenance enterprise. where is a decision variable, indicating whether work order o j is assigned to maintenance enterprise m i , with a value of 0 or 1.

[0112] The specific objective function refers to: where w oj is the waiting time of work order o j and d oj为 is the required repair duration of maintenance work order o j This function aims to minimize the total waiting time and total repair time of all maintenance work orders.

[0113] The construction method of the genetic algorithm text is specifically as follows:

[0114] B1: Encoding: Using integer encoding, encode the information of the maintenance enterprise, work station, and technician assigned to each work order into a chromosome. For example, the chromosome [m2, s 23 , t5, m1, s 12 , t3, …] represents the assignment scheme of different work orders.

[0115] B2: Initialize the population: Randomly generate 100 initial chromosomes to form the initial population.

[0116] B3: Fitness calculation: Calculate the fitness of each chromosome according to the objective function. The smaller the fitness value, the better the assignment scheme. For example, for the chromosome [m2, s 23 , t5, m1, s 12 , t3, …], calculate the total waiting time and total repair time of the corresponding work orders to obtain the fitness value.

[0117] B4: Selection operation: Use the roulette wheel selection method to select according to the fitness ratio of the chromosomes. Calculate the ratio of the fitness of each chromosome in the total fitness of the population. The higher the fitness of the chromosome, the greater the probability of being selected.

[0118] B5: Crossover operation: Randomly select two chromosomes and perform crossover according to a certain crossover probability (such as 0.8), exchange some gene segments to generate new chromosomes. For example, for two chromosomes [m2, s 23 , t5, m1, s 12 , t3] and [m3, s 31 , t7, m4, s 42 , t6], after crossover at a certain crossover point, generate new chromosomes [m2, s 23 , t7, m4, s 42 , t3] and [m3, s 31 , t5, m1, s 12 , t6].

[0119] B6: Mutation operation: Mutate the genes in the chromosome with a mutation probability of 0.01 to change the assignment information of a certain work order to increase the diversity of the population. For example, for the chromosome [m2, s 23 , t5, m1, s12 in [t3], a certain gene m2 mutates to m4, obtaining a new chromosome [m4, s 23 , t5, m1, s 12 , t3].

[0120] B7: Iterative optimization: Repeat the above selection, crossover, and mutation operations, continuously iterate the population until the minimum fitness change value is less than 0.01, and decode the chromosome with the minimum fitness to obtain the optimal work order allocation plan.

[0121] It should be further noted that when an emergency occurs, such as when some workstations of a certain maintenance enterprise become unavailable due to equipment failure, the system immediately updates the workstation status information of the relevant maintenance enterprise, that is and the values of, and then recalculate the waiting time and repair time of the affected work orders, and use these work orders as new inputs to restart the genetic algorithm for optimization calculation, adjust the work order allocation plan to ensure the continuity and timeliness of maintenance services;

[0122] The user information terminal: includes an owner client and a maintenance enterprise client for information interaction between the owner and the maintenance enterprise;

[0123] Specifically, the owner client is used to help query vehicle information, reserve maintenance services, view the repair progress and cost details;

[0124] The maintenance enterprise client is used to receive reservation work orders, assign technician tasks, view vehicle repair diagnosis reports, and manage parts inventory;

[0125] Secondly: In the attached drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved. For other structures, reference can be made to the general design. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0126] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. The intelligent operation system of digital automobile maintenance and repair market is characterized by: include: Database, vehicle information acquisition module, data preprocessing module, maintenance auxiliary diagnosis module, work order intelligent scheduling module and user information terminal; The database is used to store all data texts acquired by the intelligent operation system of the digital automobile maintenance and repair market; The vehicle information acquisition module includes a vehicle sensor interface and a service data acquisition unit, and is used to acquire vehicle data text and maintenance service data text; The data preprocessing module cleans the numerical data in the acquired vehicle data text and maintenance business data text, removes noise data, duplicate data and abnormal data, and performs data standardization processing to obtain vehicle data processing text and maintenance business data processing text; The maintenance auxiliary diagnosis module: constructs a fault diagnosis model based on the vehicle data processing text and the historical maintenance case library, assists maintenance personnel in diagnosing the fault type based on the fault diagnosis model, and selects the parts required for maintenance based on the diagnosis results and the association rule table to generate a maintenance work order text; The work order intelligent scheduling module is used to obtain the scheduling variable text according to the maintenance work order text, import the scheduling variable text into the target function construction text to obtain the target function, construct the genetic algorithm text based on the target function and the scheduling variable, and iterate the genetic algorithm text to obtain the work order allocation plan; The user information terminal includes a car owner client and a maintenance enterprise client, which are used for information exchange between the car owner and the maintenance enterprise.

2. The digital automobile maintenance and repair market intelligent operation system according to claim 1 is characterized by: The vehicle sensor interface is used to connect to the vehicle internal network through the vehicle's on-board diagnostic system interface, obtain sensor data using the on-board diagnostic system reading device, read the fault code stored in the vehicle electronic control unit through the fault diagnosis instrument connected to the vehicle's on-board diagnostic system interface, and summarize the data to obtain vehicle data text; The business data collection unit is used to obtain work order information, customer information, and spare parts inventory data from the business system of the maintenance enterprise, obtain appointment information from the car owner client, and summarize the obtained information to obtain the maintenance business data text; The sensor data includes engine data, driving system data and electrical data; The engine data includes: engine speed, air intake, throttle opening, coolant temperature, oil pressure, and fuel injection amount; The driving system data includes: vehicle speed, tire pressure, four-wheel alignment parameters, and suspension system sensor data; The electrical data include: battery voltage, generator output voltage, and current consumption of lights, air conditioner, and radio.

3. The digital automobile maintenance and repair market intelligent operation system according to claim 1 is characterized by: The historical maintenance case database specifically refers to extracting historical maintenance case data from the business management system of the maintenance enterprise, which includes vehicle model, frame number, production date, fault description, diagnostic tools used and test data, final fault cause and maintenance measures; The method for constructing the fault diagnosis model is specifically as follows: A1: Data labeling: For vehicle sensor data, the corresponding sensor data segments are labeled as corresponding fault types according to the fault causes in the maintenance case library. For the historical maintenance case library, the labeling format of the fault type is standardized and similar fault descriptions are classified into the same standard fault type. A2: Feature selection: Using professional knowledge in the field of automobile maintenance, exclude features that are obviously unrelated to the fault type. For features whose value change rate in all data samples is less than 5%, exclude them. For numerical features, calculate the Pearson correlation coefficient between each feature and the fault type, and discard features with an absolute value less than 0.2; A3: Feature encoding: Encode the non-numerical features selected in the feature selection step according to their frequency of occurrence in a certain fault type; A4: Model construction: According to the number of features determined in the feature selection step, determine the number of input layer nodes, set 2-3 hidden layers, each with 50 nodes, use full connection between hidden layers, use ReLU activation function, the expression is f(x) = max(0,x), the number of output layer nodes is equal to the number of fault types, use Softmax activation function, convert the model output into the probability distribution of each fault type, the expression is: Where z is the input vector, K is the number of categories, and σ(z)j represents the probability of the jth type of failure; A5: Fault judgment: After obtaining the final predicted probability of the fault type, select the fault type with the highest probability as the diagnosis result; The association rule table refers to a list of repair parts associated with fault types, and customers can select the required repair parts according to their own needs; The maintenance work order text specifically includes: information on spare parts required for maintenance, an estimated market for maintenance, and the type of fault.

4. The digital automobile maintenance and repair market intelligent operation system according to claim 1 is characterized by: The scheduling variable text specifically includes maintenance enterprise related variables, technician related variables, workstation related variables, parts related variables, maintenance work order related variables, decision variables and work order waiting time variables.

5. The digital automobile maintenance and repair market intelligent operation system according to claim 4 is characterized by: The maintenance enterprise related variables include: Maintenance enterprise set: M = {m1, m2, ..., m n }; Maintenance Company i Number of workstations: |S i |; Maintenance Company i Middle stations ij Available quantity: Maintenance Company i Middle stations ij Current occupancy: The technician-related variables include: Technician set: T = {t1, t2, ..., t p }; Technician k The skill level vector is: Technician k Current workload: w tk； The workstation-related variables include: a workstation set; The accessory related variables include: Accessory set: P = {p1, p2, ..., p r }; Maintenance Company i Accessories r Stock quantity: q pli ; The maintenance work order related variables include: Maintenance work order set: O, where |O| is used to represent the number of work orders in the objective function and some constraints; Maintenance work order j Required accessories: P oj ; Maintenance work order j Required maintenance time: d oj ; Maintenance work order j Fault type: type oj ; The decision variables include: Work Order j Whether to allocate to maintenance enterprise m i Workstations ij superior: Work Order j Is it assigned to a technician? k : Work Order j Whether it is assigned to the maintenance enterprise m i : The work order waiting time variables include: Work Order j Waiting time: w oj .

6. The digital automobile maintenance and repair market intelligent operation system according to claim 1 is characterized by: The objective function construction text includes: constraint conditions and objective function; The constraints include: Workstation constraint: Each maintenance work order can only be assigned to an available workstation in a maintenance enterprise, that is, in is a decision variable, representing work order o j Whether to allocate to maintenance enterprise m i Workstations ij The value is 0 or 1. Ensure that the assigned workstations are not overloaded; Technician constraints: The technicians assigned to the maintenance work order must have the corresponding skills. in is a decision variable, representing work order o j Is it assigned to a technician? k , the value is 0 or 1, and the technician workload cannot exceed its maximum load, Parts constraint: The parts required for the maintenance work order must be in stock at the maintenance company. in is a decision variable, representing work order o j Whether to allocate to maintenance enterprise m i , the value is 0 or 1; The objective function specifically refers to: where w oj For work order j The waiting time, d oj为 Maintenance work order j The required maintenance time is the function that aims to minimize the total waiting time and total maintenance time of all maintenance work orders.

7. The digital automobile maintenance and repair market intelligent operation system according to claim 1 is characterized by: The construction method of the genetic algorithm text is as follows: B1: Coding: Integer coding is used to encode the maintenance company, workstation and technician information assigned to each work order into a chromosome; B2: Initialize the population: randomly generate 100 initial chromosomes to form the initial population; B3: Fitness calculation: Calculate the fitness of each chromosome according to the objective function. The smaller the fitness value, the better the allocation scheme. B4: Selection operation: Roulette selection method is used to select according to the fitness ratio of chromosomes, and the ratio of the fitness of each chromosome in the total fitness of the population is calculated. The higher the fitness of the chromosome, the greater the probability of being selected; B5: Crossover operation: Randomly select two chromosomes, perform crossover according to a certain crossover probability (such as 0.8), exchange some gene fragments, and generate new chromosomes; B6: Mutation operation: mutate the genes in the chromosome with a mutation probability of 0.01, change the allocation information of a certain work order to increase the diversity of the population; B7: Iterative optimization: Repeat the above selection, crossover, and mutation operations, and continuously iterate the population until the minimum fitness change value is less than 0.

01. Decode the chromosome with the smallest fitness to obtain the optimal work order allocation plan.

Images