Automobile maintenance quality evaluation system and method based on big data

Abstract

The invention discloses an automobile maintenance quality evaluation system and method based on big data, and the system comprises a multi-source data collection module, an evaluation standard generation module and a maintenance quality evaluation module, and the method comprises the steps: 1, obtaining data; 2, generating an evaluation index; step 3, obtaining an evaluation result; step 4, comprehensive evaluation; according to the invention, the multi-source data acquisition module is used for acquiring multi-source data related to a maintenance project, and the maintenance quality is evaluated by taking performance actual measurement data as a main part and feedback data and operation data as auxiliary parts, so that the problem of lack of reliability of a result caused by single evaluation data is solved; the maintenance quality evaluation module fuses various performance scores according to correlation, and then integrates the fused scores with scores based on feedback data and operation data to reflect the maintenance quality; and the evaluation standard generation module generates a dynamic evaluation standard based on the vehicle maintenance record and the historical operation data analysis result, so that the actual effect of the maintenance can be reflected more accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile maintenance, and in particular to an automobile maintenance quality assessment system and method based on big data. Background Art

[0002] With the continued growth in the number of vehicles on the road, the automotive repair industry is expanding. The quality of automotive repairs not only impacts vehicle safety, reliability, and service life, but also directly impacts consumers' experience and economic benefits. Therefore, scientifically and accurately evaluating automotive repair quality is crucial. Existing automotive repair quality assessment techniques primarily test various performance indicators of repaired vehicles and compare and analyze the test data with standard data to assess the quality of each repair. While this approach can, to a certain extent, determine whether a repair project meets basic technical requirements, it still suffers from the following shortcomings: 1. Relying solely on post-repair test data, the assessment fails to reflect the actual performance of the repair results during subsequent vehicle use, making it difficult to ensure the long-term effectiveness of the repair results, resulting in a lack of reliability in the repair quality assessment. 2. Existing techniques assess automotive repair quality using multiple performance evaluation results separately. These separate evaluations fail to comprehensively reflect the repair quality, making it difficult for users to intuitively understand the repair quality assessment results. 3. Existing techniques use a unified, fixed evaluation standard that fails to consider the vehicle's service life and maintenance history. For older vehicles with multiple repairs, evaluating the repair quality using the standards of a new vehicle can result in a low evaluation result that fails to truly reflect the actual effectiveness of the repair. Summary of the Invention

[0003] The purpose of the present invention is to provide a system and method for evaluating automobile maintenance quality based on big data to solve the problems raised in the above background technology.

[0004] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: a big data-based automobile maintenance quality assessment system, comprising a multi-source data acquisition module, an assessment standard generation module, and a maintenance quality assessment module, wherein the multi-source data acquisition module establishes a data connection with the assessment standard generation module, and the maintenance quality assessment module establishes data connections with the multi-source data acquisition module and the assessment standard generation module, respectively.

[0005] Preferably, the multi-source data acquisition module includes a data acquisition unit, a maintenance item identification unit, a data screening unit and a data preprocessing unit. The data acquisition unit is used to collect multi-source data, the maintenance item identification unit is used to identify the specific content of the maintenance item, the data screening unit is used to screen data related to the vehicle system corresponding to the maintenance item, and the data preprocessing unit is used to preprocess the screened data.

[0006] Preferably, the data acquisition unit includes a maintenance work order acquisition subunit, a measured data acquisition subunit, a historical data acquisition subunit, a feedback data acquisition subunit and a vehicle operation data acquisition subunit; the maintenance item identification unit includes a work order information parsing subunit and a vehicle system association subunit; the maintenance work order acquisition subunit is used to acquire maintenance work order data; the measured data acquisition subunit is used to acquire vehicle performance measured data; the historical data acquisition subunit is used to acquire vehicle historical maintenance records and operation data; the feedback data acquisition subunit is used to acquire user feedback data; the vehicle operation data acquisition subunit is used to acquire vehicle operation data within a period of time after maintenance; the work order information parsing subunit is used to extract the specific content of the maintenance item from the maintenance work order; and the vehicle system association subunit is used to associate the vehicle with the corresponding system according to the maintenance item.

[0007] Preferably, the data screening unit includes a measured data screening subunit, a historical data screening subunit, a feedback data screening subunit and a vehicle operation data screening subunit; the data preprocessing unit includes a data cleaning subunit, a data conversion subunit and a missing value processing subunit; the measured data screening subunit is used to obtain data related to maintenance items from the measured vehicle performance data; the historical data screening subunit is used to obtain data related to maintenance items from the vehicle's historical maintenance records and operation data; the feedback data screening subunit is used to obtain data related to maintenance items from user feedback data; the vehicle operation data screening subunit is used to obtain data related to maintenance items from the operation data within a period of time after the vehicle is repaired; the data cleaning subunit is used to remove noise data and duplicate data from the screened data; the data conversion subunit is used to convert data in different formats into a unified format that can be analyzed; and the missing value processing subunit uses interpolation to process missing data.

[0008] Preferably, the evaluation criteria generation module includes a historical data analysis unit and a performance index prediction unit. The historical data analysis unit includes a maintenance record analysis subunit and an operation data analysis subunit. The historical data analysis unit is used to analyze the screened vehicle historical maintenance records and operation data. The performance index prediction unit is used to predict the vehicle's performance index based on the analysis results and use the performance index as the evaluation criteria. The maintenance record analysis subunit is used to analyze the screened vehicle historical maintenance records. The operation data analysis subunit is used to analyze the screened vehicle historical operation data.

[0009] Preferably, the maintenance quality assessment module includes an initial assessment unit, a correlation analysis unit, an assessment result fusion unit, a feedback data assessment unit, a vehicle operation data assessment unit and a comprehensive assessment unit. The initial assessment unit is used to compare the measured performance data of the vehicle after maintenance with the assessment standard to obtain a performance score. The correlation analysis unit adopts a correlation analysis algorithm to analyze the correlation between each performance, and divides the performance into strong correlation, weak correlation and no correlation through a preset correlation strength judgment threshold. The assessment result fusion unit is used to fuse the performance scores according to the performance correlation. The feedback data assessment unit adopts natural language processing to convert the screened user feedback data into a score for the maintenance quality. The vehicle operation data assessment unit adopts a reliability scoring algorithm to convert the screened vehicle operation data into a score for the maintenance quality. The comprehensive assessment unit is used to comprehensively integrate the fused performance score, the score based on the feedback data and the score based on the vehicle operation data to obtain the final maintenance quality score.

[0010] Preferably, the evaluation method adopted by the initial evaluation unit is as follows: the lower the actual measured data is than the evaluation standard, the higher the performance score is; if the actual measured data is lower than the evaluation standard, the score is 100 points; if the actual measured data is higher than the evaluation standard, the score is calculated using a scoring formula; the higher the actual measured data is than the evaluation standard, the higher the performance score is; if the actual measured data is higher than the evaluation standard, the score is 100 points; if the actual measured data is lower than the evaluation standard, the score is calculated using a scoring formula, and the scoring formula is as follows:

[0011]

[0012] Among them, S i Score the i-th performance, X i 实测 is the measured data of the performance item i, X i 预测 is the performance index of the i-th performance.

[0013] Preferably, the evaluation result fusion unit divides the performance into different groups according to the correlation relationship, and groups the unrelated performance separately. The formula used for the intra-group fusion is as follows:

[0014] For strongly associated groups:

[0015]

[0016] Among them, S 强关联组 is the performance score after intra-group fusion of the strong correlation group, ω i is the maximum absolute value of the correlation coefficient between the i-th indicator and other indicators in the group, S i Score the i-th performance.

[0017] For weakly associated groups:

[0018]

[0019] Among them, S 弱关联组 is the performance score after fusion within the weakly associated group, I is the number of indicators within the weakly associated group, S i Score the i-th performance;

[0020]

[0021] Among them, S 融合 is the performance score after all fusion, m is the number of association groups, σ g It is the group weight, which is set according to the importance of the indicators in the group.

[0022] A method for evaluating automobile maintenance quality based on big data includes the following steps: first, data acquisition; second, generating evaluation indicators; third, obtaining evaluation results; and fourth, comprehensive evaluation.

[0023] In the above step 1, the vehicle performance measurement data, vehicle historical maintenance records, vehicle historical operation data, user feedback data and vehicle operation data related to the maintenance project are obtained through the multi-source data acquisition module;

[0024] In the above step 2, the evaluation criteria generation module analyzes the vehicle's historical maintenance records and vehicle historical operating data obtained in step 1, and then predicts the vehicle's performance indicators based on the analysis results, and uses the performance indicators as the evaluation criteria;

[0025] In the above step 3, the maintenance quality assessment module compares and analyzes the measured vehicle performance data obtained in step 1 with the assessment criteria obtained in step 2 to obtain a performance score, then fuses the performance scores based on performance correlation to obtain a fused performance score, and converts the user feedback data obtained in step 1 into a score for maintenance quality to obtain a score based on the user feedback data, and converts the vehicle operation data obtained in step 1 into a score for maintenance quality to obtain a score based on the vehicle operation data;

[0026] In the above step 4, the maintenance quality assessment module integrates the fused performance score obtained in step 3, the score based on user feedback data, and the score based on vehicle operation data to obtain a comprehensive maintenance quality score.

[0027] Preferably, in step 4, the maintenance quality assessment module uses a comprehensive assessment unit to integrate the scores, and the specific formula is as follows:

[0028] S 最终 =α·S 融合 +β·S 反馈 +γ·S 运行

[0029] Among them, S 最终 S is the comprehensive score of maintenance quality. 融合 is the performance score after fusion, α is its weight coefficient, S 反馈 is the rating based on user feedback data, β is its weight coefficient, S 运行 is the score based on the vehicle operation data, γ is its weight coefficient, and α+β+γ=1.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention utilizes a multi-source data acquisition module to acquire multi-source data related to maintenance projects, mainly based on measured performance data, supplemented by feedback data and operation data, to evaluate maintenance quality, thereby solving the problem of lack of reliability of results due to single evaluation data; the maintenance quality evaluation module fuses various performance scores according to correlation, and then combines the fused scores with the scores based on feedback data and operation data to reflect the maintenance quality; the evaluation standard generation module generates dynamic evaluation standards based on the analysis results of vehicle maintenance records and historical operation data, which can more accurately reflect the actual effect of this maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a system structure block diagram of the present invention;

[0032] Figure 2 This is a structural block diagram of the multi-source data acquisition module of the present invention;

[0033] Figure 3 This is a structural block diagram of the data acquisition unit of the present invention;

[0034] Figure 4 This is a structural block diagram of the maintenance item identification unit of the present invention;

[0035] Figure 5 This is a structural block diagram of the data screening unit of the present invention;

[0036] Figure 6 This is a structural block diagram of the data preprocessing unit of the present invention;

[0037] Figure 7 Generate a module structure diagram for the evaluation criteria of the present invention;

[0038] Figure 8 This is a structural block diagram of the maintenance quality assessment module of the present invention;

[0039] Figure 9 Flow chart of the method of the present invention.

[0040] In the figure: 1. Multi-source data acquisition module; 11. Data acquisition unit; 111. Maintenance work order acquisition subunit; 112. Measured data acquisition subunit; 113. Historical data acquisition subunit; 114. Feedback data acquisition subunit; 115. Vehicle operation data acquisition subunit; 12. Maintenance item identification unit; 121. Work order information parsing subunit; 122. Vehicle system association subunit; 13. Data screening unit; 131. Measured data screening subunit; 132. Historical data screening subunit; 133. Feedback data screening subunit; 134. Vehicle operation data screening subunit; 14. Data preprocessing unit; 141. Data cleaning subunit; 142. Data conversion subunit; 143. Missing value processing subunit; 2. Evaluation standard generation module; 21. Historical data analysis unit; 211. Maintenance record analysis subunit; 212. Operation data analysis subunit; 22. Performance indicator prediction unit; 3. Maintenance quality assessment module; 31. Initial assessment unit; 32. Correlation analysis unit; 33. Assessment result fusion unit; 34. Feedback data evaluation unit; 35. Vehicle operation data evaluation unit; 36. Comprehensive evaluation unit. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] Please see the attached Figure 1 -Attached Figure 8The present invention provides an embodiment: a car maintenance quality assessment system based on big data, including a multi-source data acquisition module 1, an assessment standard generation module 2 and a maintenance quality assessment module 3, the multi-source data acquisition module 1 establishes a data connection with the assessment standard generation module 2, and the maintenance quality assessment module 3 establishes a data connection with the multi-source data acquisition module 1 and the assessment standard generation module 2 respectively; the multi-source data acquisition module 1 includes a data acquisition unit 11, a maintenance project identification unit 12, a data screening unit 13 and a data preprocessing unit 14, the data acquisition unit 11 is used to collect multi-source data, the maintenance project identification unit 12 is used to identify the specific content of the maintenance project, and the data screening unit 13 is used to screen the data related to the maintenance project. The data preprocessing unit 14 is used to preprocess the filtered data; the data acquisition unit 11 includes a maintenance work order acquisition subunit 111, a measured data acquisition subunit 112, a historical data acquisition subunit 113, a feedback data acquisition subunit 114 and a vehicle operation data acquisition subunit 115, and the maintenance project identification unit 12 includes a work order information parsing subunit 121 and a vehicle system association subunit 122. The maintenance work order acquisition subunit 111 is used to obtain maintenance work order data, the measured data acquisition subunit 112 is used to obtain vehicle performance measured data, the historical data acquisition subunit 113 is used to collect vehicle historical maintenance records and operation data, and the feedback data acquisition subunit The unit 114 is used to obtain user feedback data, the vehicle operation data acquisition subunit 115 is used to obtain the operation data of the vehicle within a period of time after maintenance, the work order information parsing subunit 121 is used to extract the specific content of the maintenance project from the maintenance work order, and the vehicle system association subunit 122 is used to associate the vehicle with the corresponding system according to the maintenance project; the data screening unit 13 includes a measured data screening subunit 131, a historical data screening subunit 132, a feedback data screening subunit 133 and a vehicle operation data screening subunit 134, the data preprocessing unit 14 includes a data cleaning subunit 141, a data conversion subunit 142 and a missing value processing subunit 143, the measured data screening subunit 131 is used to Data related to maintenance items are obtained from measured vehicle performance data. The historical data screening subunit 132 is used to obtain data related to maintenance items from historical vehicle maintenance records and operating data. The feedback data screening subunit 133 is used to obtain data related to maintenance items from user feedback data. The vehicle operating data screening subunit 134 is used to obtain data related to maintenance items from operating data within a period of time after vehicle maintenance. The data cleaning subunit 141 is used to remove noise data and duplicate data from the screened data. The data conversion subunit 142 is used to convert data in different formats into a unified format that can be analyzed. The missing value processing subunit 143 uses interpolation to process missing data.The evaluation standard generation module 2 includes a historical data analysis unit 21 and a performance index prediction unit 22. The historical data analysis unit 21 includes a maintenance record analysis subunit 211 and an operation data analysis subunit 212. The historical data analysis unit 21 is used to analyze the selected vehicle historical maintenance records and operation data. The performance index prediction unit 22 is used to predict the vehicle's performance index based on the analysis results and use the performance index as the evaluation standard. The maintenance record analysis subunit 211 is used to analyze the selected vehicle historical maintenance records. The operation data analysis subunit 212 is used to analyze the selected vehicle historical operation data; the maintenance quality evaluation module 3 includes an initial evaluation unit 31, a correlation analysis unit 32, an evaluation result fusion unit 33, a feedback data evaluation unit 34, a vehicle operation data evaluation unit 35 and a comprehensive evaluation unit 36. The initial evaluation unit 31 is used to compare the actual performance data of the vehicle after maintenance with the evaluation standard to obtain a performance score. The correlation analysis unit 32 uses a correlation analysis algorithm to analyze the correlation between each performance and judges the performance by a preset correlation strength. The performance is divided into strong correlation, weak correlation and no correlation according to the threshold. The evaluation result fusion unit 33 is used to fuse the performance scores according to the performance correlation. The feedback data evaluation unit 34 uses natural language processing to convert the screened user feedback data into a score for the maintenance quality. The vehicle operation data evaluation unit 35 uses a reliability scoring algorithm to convert the screened vehicle operation data into a score for the maintenance quality. The comprehensive evaluation unit 36 ​​is used to comprehensively integrate the fused performance score, the score based on the feedback data and the score based on the vehicle operation data to obtain the final maintenance quality score. The evaluation method adopted by the initial evaluation unit 31 is specifically as follows: the lower the measured data is from the evaluation standard, the higher the performance score is. If the measured data is lower than the evaluation standard, the score is 100 points. If the measured data is higher than the evaluation standard, the score is calculated using the scoring formula. The higher the measured data is from the evaluation standard, the higher the performance score is. If the measured data is higher than the evaluation standard, the score is 100 points. If the measured data is lower than the evaluation standard, the score is calculated using the scoring formula. The scoring formula is as follows:

[0043]

[0044] Among them, S i Score the i-th performance, X i 实测 is the measured data of the performance item i, X i 预测 is the performance indicator of the i-th performance; the evaluation result fusion unit 33 divides the performance into different groups according to the correlation relationship, and the unrelated performance is grouped separately. The formula used for group fusion is as follows:

[0045] For strongly associated groups:

[0046]

[0047] Among them, S 强关联组 is the performance score after intra-group fusion of the strong correlation group, ω i is the maximum absolute value of the correlation coefficient between the i-th indicator and other indicators in the group, S i Score the i-th performance.

[0048] For weakly associated groups:

[0049]

[0050] Among them, S 弱关联组 is the performance score after fusion within the weakly associated group, I is the number of indicators within the weakly associated group, S i Score the i-th performance;

[0051]

[0052] Among them, S 融合 is the performance score after all fusion, m is the number of association groups, σ g It is the group weight, which is set according to the importance of the indicators in the group.

[0053] See also Figure 9 The present invention provides an embodiment: a method for evaluating automobile maintenance quality based on big data, comprising the steps of: first, acquiring data; second, generating evaluation indicators; third, obtaining evaluation results; and fourth, comprehensive evaluation.

[0054] In the above step 1, the multi-source data acquisition module 1 is used to obtain vehicle performance measurement data, vehicle historical maintenance records, vehicle historical operation data, user feedback data and vehicle operation data related to the maintenance project;

[0055] In the above step 2, the evaluation standard generation module 2 analyzes the vehicle's historical maintenance records and vehicle historical operation data obtained in step 1, and then predicts the vehicle's performance index based on the analysis results, and uses the performance index as the evaluation standard;

[0056] In the above step 3, the maintenance quality assessment module 3 compares and analyzes the measured vehicle performance data obtained in step 1 with the assessment criteria obtained in step 2 to obtain a performance score, then fuses the performance scores based on performance correlation to obtain a fused performance score, converts the user feedback data obtained in step 1 into a score for maintenance quality, and obtains a score based on the user feedback data, and converts the vehicle operation data obtained in step 1 into a score for maintenance quality, and obtains a score based on the vehicle operation data;

[0057] In step 4 above, the maintenance quality assessment module 3 integrates the fused performance score obtained in step 3, the score based on user feedback data, and the score based on vehicle operation data to obtain a comprehensive maintenance quality score. The maintenance quality assessment module 3 uses the comprehensive assessment unit 36 ​​to integrate the scores. The specific formula is as follows:

[0058] S 最终 =α·S 融合 +β·S 反馈 +γ·S 运行

[0059] Among them, S 最终 S is the comprehensive score of maintenance quality. 融合 is the performance score after fusion, α is its weight coefficient, S 反馈 is the rating based on user feedback data, β is its weight coefficient, S 运行 is the score based on the vehicle operation data, γ is its weight coefficient, and α+β+γ=1.

[0060] Based on the above, the advantage of the present invention is that when the present invention is used, data is collected through the data collection unit 11 in the multi-source data collection module 1, the maintenance work order acquisition subunit 111 is used to obtain maintenance work order data, the measured data collection subunit 112 obtains vehicle performance measured data, the historical data collection subunit 113 collects vehicle historical maintenance records and operation data, the feedback data collection subunit 114 obtains user feedback data, and the vehicle operation data collection subunit 115 obtains vehicle operation data within a period of time after maintenance, for example, vehicle mileage, fuel consumption, fault code and other operation data are obtained from the on-board diagnostic system, and maintenance data are obtained from the maintenance enterprise management system. The system collects data such as repair work orders, maintenance personnel operation records, etc., and collects feedback data such as customer satisfaction with maintenance services and problems encountered during use from the customer evaluation platform. Then, the maintenance project identification unit 12 is used to identify the current maintenance project, and the work order information parsing subunit 121 is used to extract the specific content of the maintenance project from the maintenance work order, such as engine maintenance, brake system maintenance, etc. The vehicle system association subunit 122 is associated with the corresponding system of the vehicle according to the maintenance project. For example, if the engine is repaired, it is associated with the entire power system. Then, the data screening unit 13 is used to screen the collected data according to the corresponding system of the vehicle, and the measured data screening subunit 131 is used to screen the vehicle performance. The data related to the maintenance project can be obtained from the measured data. For example, if the maintenance is done on the engine, the measured data such as engine speed, power, fuel consumption, etc. are filtered. The historical data filtering subunit 132 obtains data related to the maintenance project from the vehicle's historical maintenance records and operation data, such as the engine's previous maintenance records, fault data, etc. The feedback data filtering subunit 133 obtains data related to the maintenance project from user feedback data. The vehicle operation data filtering subunit 134 obtains data related to the maintenance project from the operation data of the vehicle within a period of time after maintenance. Finally, the filtered data is preprocessed by the data preprocessing unit 14 and data cleaning is performed. Subunit 141 removes noise and duplicate data from the filtered data, data conversion subunit 142 converts data in different formats into a unified format that can be analyzed, and missing value processing subunit 143 uses interpolation to process missing data; evaluation standard generation module 2 uses historical data analysis unit 21 to analyze the filtered vehicle historical maintenance records and operating data, maintenance record analysis subunit 211 analyzes the filtered vehicle historical maintenance records, and operating data analysis subunit 212 analyzes the filtered vehicle historical operating data. Performance index prediction unit 22 predicts the vehicle's performance index based on the analysis results and uses the performance index as the evaluation standard;The maintenance quality assessment module 3 uses the initial assessment unit 31 to compare the measured performance data of the vehicle after maintenance with the assessment standard to obtain a performance score. The lower the measured data is from the assessment standard, the higher the performance score is. If the measured data is lower than the assessment standard, the score is 100 points. If the measured data is higher than the assessment standard, the score is calculated using a scoring formula. For example, for fuel consumption, the higher the measured data is from the assessment standard, the higher the performance score is. If the measured data is higher than the assessment standard, the score is 100 points. If the measured data is lower than the assessment standard, the score is calculated using a scoring formula. For example, for speed, the correlation analysis unit 32 analyzes the correlation between various performances and divides the performance into strong correlation, weak correlation and no correlation by presetting the correlation strength judgment threshold. For example, emissions, power and fuel consumption, these performance scores are correlated. The assessment result fusion unit 33 fuses the performance scores according to the performance correlation to obtain a fused performance score. The fused performance score takes into account the correlation between the performance scores and is therefore more accurate. For example, if the fuel consumption score is low, the power score is medium, and the emission score is medium, then the fused performance score is The score will be relatively low. If the power rating is considered alone, an accurate evaluation result cannot be obtained. The feedback data evaluation unit 34 uses natural language processing to convert the filtered user feedback data into a repair quality score, obtaining a score based on the user feedback data. For example, after the engine repair, the user reported that the vehicle was shaking severely. The score based on the user feedback data can reflect the actual effect of the repair. The vehicle operation data evaluation unit 35 uses a reliability scoring algorithm to convert the filtered vehicle operation data into a repair quality score, obtaining a score based on the vehicle operation data. The reliability scoring algorithm evaluates the reliability of the repair by analyzing the operation of the repaired vehicle over a certain period of time. For example, if the engine repair passes the inspection score, but after a period of subsequent use, it malfunctions or has unstable speed, the reliability of the repair can be reflected by the reliability score. The comprehensive evaluation unit 36 ​​combines the integrated performance score, the feedback data score, and the vehicle operation data score to obtain a comprehensive repair quality score.

[0061] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A big data-based automobile maintenance quality assessment system, comprising a multi-source data acquisition module (1), an assessment standard generation module (2) and a maintenance quality assessment module (3), characterized in that: The multi-source data acquisition module (1) establishes a data connection with the evaluation standard generation module (2), and the maintenance quality evaluation module (3) establishes data connections with the multi-source data acquisition module (1) and the evaluation standard generation module (2).

2. The automobile maintenance quality assessment system based on big data according to claim 1, characterized in that: The multi-source data acquisition module (1) comprises a data acquisition unit (11), a maintenance item identification unit (12), a data screening unit (13) and a data pre-processing unit (14); the data acquisition unit (11) is used to acquire multi-source data; the maintenance item identification unit (12) is used to identify the specific content of the maintenance item; the data screening unit (13) is used to screen data related to the vehicle system corresponding to the maintenance item; and the data pre-processing unit (14) is used to pre-process the screened data.

3. The automobile maintenance quality assessment system based on big data according to claim 2 is characterized by: The data acquisition unit (11) includes a maintenance work order acquisition subunit (111), a measured data acquisition subunit (112), a historical data acquisition subunit (113), a feedback data acquisition subunit (114), and a vehicle operation data acquisition subunit (115); the maintenance item identification unit (12) includes a work order information parsing subunit (121) and a vehicle system association subunit (122); the maintenance work order acquisition subunit (111) is used to acquire maintenance work order data; the measured data acquisition subunit (112) is used to acquire vehicle performance measured data; the historical data acquisition subunit (113) is used to acquire vehicle historical maintenance records and operation data; the feedback data acquisition subunit (114) is used to acquire user feedback data; the vehicle operation data acquisition subunit (115) is used to acquire vehicle operation data within a period of time after maintenance; the work order information parsing subunit (121) is used to extract specific content of a maintenance item from a maintenance work order; and the vehicle system association subunit (122) is used to associate the vehicle with the corresponding system according to the maintenance item.

4. The automobile maintenance quality assessment system based on big data according to claim 2, characterized in that: The data screening unit (13) includes a measured data screening subunit (131), a historical data screening subunit (132), a feedback data screening subunit (133) and a vehicle operation data screening subunit (134); the data preprocessing unit (14) includes a data cleaning subunit (141), a data conversion subunit (142) and a missing value processing subunit (143); the measured data screening subunit (131) is used to obtain data related to maintenance items from the measured vehicle performance data; the historical data screening subunit (132) is used to obtain data related to maintenance items from the vehicle's historical maintenance records and operation data. The invention relates to a method for obtaining data related to maintenance items from user feedback data, a feedback data screening subunit (133) is used to obtain data related to maintenance items from user feedback data, a vehicle operation data screening subunit (134) is used to obtain data related to maintenance items from operation data within a period of time after vehicle maintenance, a data cleaning subunit (141) is used to remove noise data and duplicate data from the screened data, a data conversion subunit (142) is used to convert data in different formats into a unified format that can be analyzed, and a missing value processing subunit (143) uses an interpolation method to process missing data.

5. The automobile maintenance quality assessment system based on big data according to claim 1 is characterized by: The evaluation standard generation module (2) includes a historical data analysis unit (21) and a performance index prediction unit (22). The historical data analysis unit (21) includes a maintenance record analysis subunit (211) and an operation data analysis subunit (212). The historical data analysis unit (21) is used to analyze the selected vehicle historical maintenance records and operation data. The performance index prediction unit (22) is used to predict the vehicle's performance index based on the analysis result and use the performance index as the evaluation standard. The maintenance record analysis subunit (211) is used to analyze the selected vehicle historical maintenance records. The operation data analysis subunit (212) is used to analyze the selected vehicle historical operation data.

6. The automobile maintenance quality assessment system based on big data according to claim 1, characterized in that: The maintenance quality evaluation module (3) includes an initial evaluation unit (31), a correlation analysis unit (32), an evaluation result fusion unit (33), a feedback data evaluation unit (34), a vehicle operation data evaluation unit (35) and a comprehensive evaluation unit (36). The initial evaluation unit (31) is used to compare the measured performance data of the vehicle after maintenance with the evaluation standard to obtain a performance score. The correlation analysis unit (32) uses a correlation analysis algorithm to analyze the correlation between each performance and divides the performance into strong correlation, weak correlation and no correlation by presetting a correlation strength judgment threshold. The evaluation result fusion unit (33) is used to fuse the performance scores according to the performance correlation. The feedback data evaluation unit (34) uses natural language processing to convert the screened user feedback data into a score for maintenance quality. The vehicle operation data evaluation unit (35) uses a reliability scoring algorithm to convert the screened vehicle operation data into a score for maintenance quality. The comprehensive evaluation unit (36) is used to comprehensively combine the fused performance score, the score based on the feedback data and the score based on the vehicle operation data to obtain a final maintenance quality score.

7. The automobile maintenance quality assessment system based on big data according to claim 6, characterized in that: The evaluation method adopted by the initial evaluation unit (31) is specifically as follows: the lower the actual measured data is than the evaluation standard, the higher the performance score is; if the actual measured data is lower than the evaluation standard, the score is 100 points; if the actual measured data is higher than the evaluation standard, the score is calculated using a scoring formula; the higher the actual measured data is than the evaluation standard, the higher the performance score is; if the actual measured data is higher than the evaluation standard, the score is 100 points; if the actual measured data is lower than the evaluation standard, the score is calculated using a scoring formula, and the scoring formula is as follows: Among them, S i Score the i-th performance, X i 实测 is the measured data of the performance item i, X i 预测 is the performance indicator of the i-th performance.

8. The automobile maintenance quality assessment system based on big data according to claim 6 is characterized by: The evaluation result fusion unit (33) divides the performance into different groups according to the correlation relationship, and the unrelated performance is grouped separately. The formula used for the fusion within the group is as follows: For strongly associated groups: Among them, S 强关联组 is the performance score after intra-group fusion of the strong correlation group, ω i is the maximum absolute value of the correlation coefficient between the i-th indicator and other indicators in the group, S i Score the i-th performance. For weakly associated groups: Among them, S 弱关联组 is the performance score after fusion within the weakly associated group, I is the number of indicators within the weakly associated group, S i Score the i-th performance; Among them, S 融合 is the performance score after all fusion, m is the number of association groups, σ g It is the group weight, which is set according to the importance of the indicators in the group.

9. A method for evaluating automobile repair quality based on big data, comprising the following steps: first, acquiring data; second, generating evaluation indicators; third, obtaining evaluation results; and fourth, comprehensive evaluation; characterized in that: In the above step 1, the vehicle performance measurement data, vehicle historical maintenance records, vehicle historical operation data, user feedback data and vehicle operation data related to the maintenance project are obtained through the multi-source data acquisition module (1); In the above step 2, the evaluation standard generation module (2) analyzes the vehicle historical maintenance records and vehicle historical operation data obtained in step 1, and then predicts the vehicle performance index based on the analysis results, and uses the performance index as the evaluation standard; In the above step 3, the maintenance quality evaluation module (3) compares and analyzes the measured vehicle performance data obtained in step 1 with the evaluation criteria obtained in step 2 to obtain a performance score, and then fuses the performance scores according to the performance correlation to obtain a fused performance score, and converts the user feedback data obtained in step 1 into a score for maintenance quality to obtain a score based on the user feedback data, and converts the vehicle operation data obtained in step 1 into a score for maintenance quality to obtain a score based on the vehicle operation data; In the above step 4, the maintenance quality assessment module (3) integrates the fused performance score obtained in step 3, the score based on user feedback data and the score based on vehicle operation data to obtain a comprehensive maintenance quality score.

10. The automobile maintenance quality assessment method based on big data according to claim 9, characterized in that: In step 4, the maintenance quality assessment module (3) uses the comprehensive assessment unit (36) to synthesize the scores. The specific formula is as follows: S 最终 =α·S 融合 +β·S 反馈 +γ·S 运行 Among them, S 最终 S is the comprehensive score of maintenance quality. 融合 is the performance score after fusion, α is its weight coefficient, S 反馈 is the rating based on user feedback data, β is its weight coefficient, S 运行 is the score based on the vehicle operation data, γ is its weight coefficient, and α+β+γ=1.

Images