Multi-dimensional business scene track geometric detection data quality evaluation method and device

Abstract

The invention provides a multi-dimensional business scene track geometric detection data quality evaluation method and device. The method comprises the following steps: determining a detection normativity index value, a detection integrity index value, a detection timeliness index value, an accessibility index value, a detection validity index value and a detection accuracy index value of track geometric detection data; determining weights and sequences of the index values by adopting an analytic hierarchy process; determining the quality grade of the track geometric detection data according to the weight and the sequence of the index values; a systematic framework is provided for quality evaluation of comprehensive detection data of the high-speed railway, track geometric detection data quality evaluation is efficiently and accurately carried out, accuracy and reliability of track geometric detection data quality evaluation are ensured, data support is provided for operation management and maintenance decision making of the high-speed railway, and the track geometric detection data quality evaluation method is suitable for popularization and application. Therefore, the track maintenance strategy is optimized, and the maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of track detection technology, and in particular to a method and device for evaluating the quality of track geometry detection data in a multi-dimensional business scenario. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] The geometric state of railway tracks directly affects the running safety of trains and the service life of tracks. Its quality has a significant impact on the safety, stability and comfort of train travel. By regularly conducting track infrastructure inspections, track geometry inspection data including parameters such as gauge, level, height, and twist can be obtained.

[0004] With the increase in railway transportation speed and train load, the analysis and research on track geometry detection data has become more in-depth and refined. In the existing technology, high-quality track geometry detection data is the prerequisite for extracting useful information, making accurate predictions and formulating effective maintenance strategies. Data quality is directly related to the accuracy of track geometry status assessment and the scientific nature of track maintenance decisions. At present, data quality evaluation, as a key link to ensure data accuracy and reliability, has received widespread attention in many fields. However, different fields have different focuses on data objects, and data quality evaluation is field-specific, resulting in the lack of a set of universally applicable data quality evaluation methods in the field of track detection technology, and the quality evaluation results are neither efficient nor accurate. Summary of the invention

[0005] The embodiment of the present invention provides a method for evaluating the quality of track geometry detection data in a multi-dimensional business scenario, which is used to form a set of generally applicable quality evaluation methods in the field of track detection technology, and improve the efficiency and accuracy of the quality evaluation of track geometry detection data; the method for evaluating the quality of track geometry detection data in a multi-dimensional business scenario includes:

[0006] Match the file name of the track geometry detection data with a pre-set business dictionary table, and determine the matching result as the detection standardization index value; the business dictionary table records the codes corresponding to different business information;

[0007] Extract the start and end mileages of the actual inspection from the file header information of the track geometry inspection data, and calculate the actual inspection mileage length based on the start and end mileages of the actual inspection; calculate the mileage coverage rate of the inspection plan based on the actual inspection mileage length and the inspection plan mileage length; determine the mileage coverage rate of the inspection plan as the inspection integrity index value;

[0008] Clustering the collection time of track geometry detection data, and determining the clustering result as the detection timeliness index value;

[0009] Extracting key metadata from the file header information of the track geometry detection data according to the predefined byte position and length; determining the accuracy of the extracted key metadata as the accessibility index value of the track geometry detection data;

[0010] According to the actual detection mileage length and the mileage length of invalid data in the track geometry detection data, the proportion of valid start and end mileage in the track geometry detection data is determined; the proportion of valid start and end mileage is determined as the detection effectiveness index value;

[0011] Determine the matching degree between the track geometry detection data and the reference data, determine the detection mileage deviation value according to the matching degree, and determine the detection mileage deviation value as the detection accuracy index value;

[0012] The analytic hierarchy process is used to determine the weights and rankings of the test standardization index value, the test integrity index value, the test timeliness index value, the accessibility index value, the test effectiveness index value, and the test accuracy index value;

[0013] The quality level of track geometry inspection data is determined based on the weights and rankings of the inspection standardization index value, inspection integrity index value, inspection timeliness index value, accessibility index value, inspection effectiveness index value and inspection accuracy index value.

[0014] The embodiment of the present invention further provides a multi-dimensional business scenario track geometry detection data quality evaluation device, which is used to form a set of generally applicable quality evaluation methods in the field of track detection technology, and improve the efficiency and accuracy of track geometry detection data quality evaluation; the multi-dimensional business scenario track geometry detection data quality evaluation device includes:

[0015] The detection standardization index determination module is used to match the file name of the track geometry detection data with a pre-set business dictionary table, and determine the matching result as the detection standardization index value; the business dictionary table records the codes corresponding to different business information;

[0016] The detection integrity index determination module is used to extract the start and end mileages of the actual detection from the file header information of the track geometry detection data, calculate the actual detection mileage length according to the start and end mileages of the actual detection; calculate the mileage coverage rate of the detection plan according to the actual detection mileage length and the detection plan mileage length; determine the mileage coverage rate of the detection plan as the detection integrity index value;

[0017] A detection timeliness index determination module is used to cluster the collection time of track geometry detection data and determine the clustering result as the detection timeliness index value;

[0018] An accessibility index determination module is used to extract key metadata from the file header information of the track geometry detection data according to a predefined byte position and length; and determine the accuracy of the extracted key metadata as the accessibility index value of the track geometry detection data;

[0019] The detection effectiveness index determination module is used to determine the proportion of valid start and end mileages in the track geometry detection data according to the actual detection mileage length and the mileage length of invalid data in the track geometry detection data; and determine the proportion of valid start and end mileages as the detection effectiveness index value;

[0020] A detection accuracy index determination module is used to determine the matching degree between the track geometry detection data and the reference data, determine the detection mileage deviation value according to the matching degree, and determine the detection mileage deviation value as the detection accuracy index value;

[0021] A hierarchical analysis module is used to determine the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value and the detection accuracy index value by using the hierarchical analysis method;

[0022] The quality analysis module is used to determine the quality level of track geometry inspection data according to the weights and rankings of inspection standardization index value, inspection integrity index value, inspection timeliness index value, accessibility index value, inspection effectiveness index value and inspection accuracy index value.

[0023] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for evaluating the quality of track geometry detection data in multi-dimensional business scenarios when executing the computer program.

[0024] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned method for evaluating the quality of track geometry detection data in multi-dimensional business scenarios.

[0025] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the above-mentioned method for evaluating the quality of track geometry detection data in multi-dimensional business scenarios.

[0026] The multi-dimensional business scenario track geometry detection data quality evaluation method and device of the embodiment of the present invention determines the detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value, and detection accuracy index value of the track geometry detection data; uses the hierarchical analysis method to determine the weights and rankings of these index values; determines the quality level of the track geometry detection data according to the weights and rankings of these index values; the embodiment of the present invention will provide a systematic framework for the quality evaluation of high-speed railway comprehensive detection data, efficiently and accurately perform track geometry detection data quality evaluation, ensure the accuracy and reliability of track geometry detection data quality evaluation, and provide data support for the operation management and maintenance decision-making of high-speed railways, so as to optimize track maintenance strategies and reduce maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0028] Figure 1 This is an example diagram of a method for evaluating quality of track geometry detection data in a multi-dimensional business scenario in an embodiment of the present invention;

[0029] Figure 2 A specific example diagram of the clustering evaluation result of track geometry detection data upload timeliness in an embodiment of the present invention;

[0030] Figure 3 A heat map of evaluation indicators of a single file in a detection business scenario in an embodiment of the present invention;

[0031] Figure 4 A hierarchical structure model diagram in an embodiment of the present invention;

[0032] Figure 5 This is an example diagram of a single track geometry detection file data quality evaluation result in a detection business scenario in an embodiment of the present invention;

[0033] Figure 6 This is a structural example diagram of a device for evaluating quality of track geometry detection data in a multi-dimensional business scenario according to an embodiment of the present invention;

[0034] Figure 7 It is a specific example diagram of the structure of the multi-dimensional business scenario track geometry detection data quality evaluation device in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] 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.

[0036] The inventors' research found that track geometry detection data is crucial to ensuring the safety and efficiency of railway transportation; however, different fields have different focuses on data objects, and data quality evaluation is field-specific, resulting in the lack of a universally applicable data quality evaluation method.

[0037] In view of the problems existing in the prior art, the present invention proposes that according to the characteristics of track geometry detection data and the data flow process, in the data collection stage, the evaluation indicators are mainly concentrated on the standardization, timeliness, accessibility and integrity of the data, while in the data analysis and processing stage, the evaluation indicators are expanded to accuracy and effectiveness; by calculating the values ​​of various indicators and assigning corresponding weights according to different application scenarios, a comprehensive evaluation result of the data quality is finally obtained.

[0038] Figure 1 : is an example diagram of a method for evaluating the quality of track geometry detection data in a multi-dimensional business scenario according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0039] Step 101, matching the file name of the track geometry detection data with a preset business dictionary table, and determining the matching result as the detection standardization index value; the business dictionary table records the codes corresponding to different business information;

[0040] Step 102: extract the start and end mileages of the actual detection from the file header information of the track geometry detection data, calculate the actual detection mileage length according to the start and end mileages of the actual detection; calculate the mileage coverage rate of the detection plan according to the actual detection mileage length and the detection plan mileage length; determine the mileage coverage rate of the detection plan as the detection integrity index value;

[0041] Step 103, clustering the collection time of the track geometry detection data, and determining the clustering result as the detection timeliness index value;

[0042] Step 104: extract key metadata from the file header information of the track geometry detection data according to the predefined byte position and length; determine the accuracy of the extracted key metadata as the accessibility index value of the track geometry detection data;

[0043] Step 105: Determine the proportion of valid start and end mileages in the track geometry detection data according to the actual detection mileage length and the mileage length of invalid data in the track geometry detection data; determine the proportion of valid start and end mileages as the detection effectiveness index value;

[0044] Step 106, determining the matching degree between the track geometry detection data and the reference data, determining the detection mileage deviation value according to the matching degree, and determining the detection mileage deviation value as the detection accuracy index value;

[0045] Step 107: using the analytic hierarchy process to determine the weights and rankings of the test standardization index value, the test integrity index value, the test timeliness index value, the accessibility index value, the test effectiveness index value, and the test accuracy index value;

[0046] Step 108: Determine the quality level of the track geometry detection data according to the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value, and the detection accuracy index value.

[0047] In specific implementation, in step 101, the file name of the track geometry detection data is matched with a pre-set business dictionary table, and the matching result is determined as the detection normative index value; the business dictionary table records the codes corresponding to different business information. In an embodiment, matching the file name of the track geometry detection data with a pre-set business dictionary table, and determining the matching result as the detection normative index value may include:

[0048] Extract key business information from the file name of track geometry inspection data according to the predefined regular expression; the regular expression defines the naming format of track geometry inspection data; the key business information includes line code, line code, starting station, ending station, inspection date and inspection time;

[0049] Match key business information with the business dictionary table;

[0050] If the key business information matches the corresponding code, the detection normative indicator value is determined to be yes, otherwise the detection normative indicator value is determined to be no.

[0051] In the embodiment, the file name of the track geometry detection data must follow a strict naming format, and the file name contains key business information such as line code, line code, starting station, ending station, detection date and detection time. The specific naming format can be, for example: line (line code line type-starting station pinyin-end station pinyin)-date-time-[detection direction]-(starting mileage-end mileage).geo.

[0052] In order to efficiently extract these key business information from the file name of the track geometry detection data, the embodiment of the present invention proposes a precise regular expression:

[0053] ^([a-ZA-2]{4,8})-([a-zA-Z]+)-([a-2A-2]+)-( \\d{8})-(\\d{6}).*.geo$

[0054] Among them, the ^([a-ZA-2]{4,8}) part is used to match the line code part at the beginning of the file name, ^ means matching the beginning of the string, [a-ZA-2]{4,8} means matching a string consisting of 4 to 8 letters or numbers; the front part of the string (except the last digit) matches the line information, and the last digit of the string matches the line information, that is, whether the track is up or down, etc.; ([a-zA-Z]+)-([a-zA-Z]+) is used to match the names of the starting and ending stations, + means matching the string of the starting and ending station names consisting of multiple uppercase or lowercase letters; (\\d{8}) is used to match the inspection date consisting of 8 digits; (\\d{6}) is used to match the inspection time consisting of 6 digits;

[0055] Through this regular expression, key business information such as line code, line code, starting station, ending station, inspection date and inspection time can be effectively extracted from the file name of track geometry inspection data.

[0056] After extracting the test date, the test date can also be checked for compliance. If the test date does not conform to the specified format, the test compliance index value is judged as "no"; if the check passes, the subsequent code conversion work will continue. The key business information such as the parsed line code, line code, starting station, ending station, test date and test time will be matched with the pre-set business dictionary table and converted into the corresponding code. If the corresponding code can be successfully matched, the test compliance index value is "yes"; if the corresponding code cannot be matched, the test compliance index value is judged as "no".

[0057] In step 102, the start and end mileages of the actual inspection are extracted from the file header information of the track geometry inspection data, and the actual inspection mileage length is calculated based on the start and end mileages of the actual inspection; the mileage coverage of the inspection plan is calculated based on the actual inspection mileage length and the inspection plan mileage length; the mileage coverage of the inspection plan is determined as the inspection integrity index value.

[0058] In the embodiment, before calculating the mileage coverage of the inspection plan according to the actual inspection mileage length and the inspection plan mileage length, the following steps may also be included:

[0059] The inspection plan data is structured; the structured inspection plan data is decomposed to extract the inspection plan mileage length.

[0060] The inspection integrity index value mainly measures whether the line mileage actually inspected completely covers the line mileage required by the inspection plan; the integrity evaluation of the inspection file is achieved by calculating the mileage coverage of the inspection plan. For example, the inspection plan data is first structured according to "one car, one line, one route per day", and then the inspection plan data is decomposed into the corresponding minimum calculation units to obtain key information such as the inspection date, inspection vehicle number, route, route, starting mileage, ending mileage, and inspection plan mileage length; then the actual inspection start and end mileages are extracted from the file header information of the track geometry inspection data, and the actual inspection mileage length is calculated; then the mileage coverage of the inspection plan is calculated based on the inspection plan mileage length and the actual inspection mileage length, and the mileage coverage of the inspection plan is determined as the inspection integrity index value. The calculation formula can be as follows:

[0061]

[0062] Among them, the C value represents the detection integrity index value. The closer the value is to 100%, the higher the integrity of the track geometry detection data is, that is, the actual detected line mileage is closer to or equal to the line mileage required by the detection plan.

[0063] In step 103, the collection time of the track geometry detection data is clustered, and the clustering result is determined as the detection timeliness index value.

[0064] In an embodiment, clustering the acquisition time of track geometry detection data may include:

[0065] The collection time of track geometry detection data is clustered by the K-means clustering method. The collection time of track geometry detection data is divided into multiple data clusters, and the clustering goal is to minimize the sum of square errors within each data cluster.

[0066] For example, the sum of the distances from the data points in each data cluster to the cluster center can be minimized according to the following formula to achieve the clustering goal:

[0067]

[0068] Among them, SSE is the sum of the distances from the data points in each data cluster to the cluster center; K is the number of data clusters, C i is the data point set of the i-th data cluster, x is the data point of C i The detection timeliness index value, μ i is the center of the ith data cluster.

[0069] In an embodiment, clustering the acquisition time of track geometry detection data may include:

[0070] The number of data clusters is determined based on the predefined silhouette coefficient index value; the silhouette coefficient index value is used to measure the compactness and separation of the clustering results.

[0071] For example, to determine the optimal number of data clusters, the silhouette coefficient index is used to measure the compactness and separation of the time-sensitive clustering results. For example, the number of data clusters is determined to be 7 according to the silhouette coefficient index.

[0072] For example, K-means clustering is performed on the acquisition time of track geometry detection data. Figure 2 FIG. 1 is a specific example diagram of the clustering evaluation result of the track geometry detection data upload time efficiency in an embodiment of the present invention; Figure 2 As shown, data points of different colors represent timeliness evaluations determined according to the detection timeliness index values; from the clustering results, it can be seen that the upload time within 24 hours is fast upload, and Table 1 is a clustering evaluation result table of track geometry detection data upload timeliness in an embodiment of the present invention;

[0073] Table 1 Clustering evaluation results of track geometry detection data upload timeliness

[0074]

[0075] As shown in Table 1, [24,40] hours are fast uploads, [41,72] hours are qualified uploads that meet the data collection requirements, [73,86] hours are timed uploads, [87,108] hours are slow uploads, [109,144] hours are slow uploads, and [145,+∞] hours are serious timed uploads.

[0076] In step 104, key metadata is extracted from the file header information of the track geometry detection data according to the predefined byte position and length; the accuracy of the extracted key metadata is determined as the accessibility index value of the track geometry detection data.

[0077] The accessibility index value of track geometry detection data focuses on the readability of the detection file header information. The file header information of track geometry detection data contains rich key metadata. Through parsing, key metadata can be extracted from the file header information, including file version number, increase and decrease mileage mark, channel data record length, sampling interval, unit, detection date, detection time, area name and detection code, etc. If these key metadata can be accurately identified and extracted according to the predefined byte position and length, the accessibility index value of the track geometry detection data is "yes"; conversely, if the key metadata cannot be correctly extracted or there is an identification obstacle, it means that the file header information is unreadable, and the accessibility index value of the track geometry detection data is "no".

[0078] In step 105, the proportion of valid start and end mileages in the track geometry detection data is determined based on the actual detection mileage length and the mileage length of invalid data in the track geometry detection data; the proportion of valid start and end mileages is determined as the detection effectiveness index value.

[0079] The effectiveness evaluation of track geometry detection data depends on the accurate identification and marking of invalid data. The invalid data identification algorithm using high-pass filtering and moving window technology can automatically identify and mark invalid data caused by factors such as interference from sunlight, rain and snow, local burrs, widened turnouts, single-sided track gauge straightening, over-phasing, and low speed. By comparing with the records of corresponding lines and lines in the long-short chain database, ensure that the start and end mileage of the invalid data correctly reflects the long-short chain information, and then calculate the accurate invalid detection mileage. Summarize the identification results of various types of invalid data in the track geometry detection data, superimpose their start and end mileage distances, and finally obtain the proportion of valid start and end mileage in the track geometry detection data. The calculation formula can be, for example, as follows:

[0080]

[0081] Among them, the E value represents the detection effectiveness index value. The closer the value is to 100%, the higher the effectiveness of the track geometry detection data, that is, the lower the proportion of invalid data in the track geometry detection data.

[0082] In step 106, the matching degree between the track geometry detection data and the reference data is determined, and the detection mileage deviation value is determined according to the matching degree, and the detection mileage deviation value is determined as the detection accuracy index value.

[0083] In an embodiment, determining the degree of matching between the track geometry detection data and the reference data may include:

[0084] Eliminate invalid data from the track geometry detection data; sample the track geometry detection data after the invalid data has been eliminated, slide the sampled data points, and calculate the correlation coefficient between each data point and the benchmark data; determine the degree of match between the track geometry detection data and the benchmark data based on the correlation coefficient.

[0085] In an embodiment, the correlation coefficient between each data point and the reference data may be calculated according to the following formula:

[0086]

[0088] Among them, r is the correlation coefficient; x i and i are the corresponding data points in the test data and the benchmark data respectively; and are the average values ​​of the detection data and the benchmark data, respectively.

[0089] The accuracy evaluation of track geometry detection data mainly focuses on the accuracy of the mileage field in the detection data; by accurately dividing the track geometry detection data, eliminating invalid data per kilometer, counting the number of data points per kilometer and applying the interquartile range (IQR) for screening, further eliminating kilometers with very few data points.

[0090] On this basis, sampling is carried out every kilometer, data points in specific sections are selected for sliding, and the correlation coefficient between the track geometry detection data and the benchmark data is calculated to determine the degree of match between the track geometry detection data and the benchmark data.

[0091] By sorting the mileage deviation values ​​and selecting the mileage deviation values ​​corresponding to the 95% and 99% units, the accuracy of the data is quantified, ensuring the clarity and systematization of the evaluation results.

[0092] In the embodiment, before using the hierarchical analysis method to determine the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value, and the detection accuracy index value, the following may also be included:

[0093] The test standardization index value, test integrity index value, test timeliness index value, accessibility index value, test effectiveness index value and test accuracy index value are normalized.

[0094] For example, normalizing the detection timeliness index value may include:

[0095] The gradient assignment method is used to normalize the detection timeliness index value;

[0096] For example, the Z-Score method is used to standardize the detection mileage deviation value.

[0097] In the embodiment, due to the richness of evaluation indicators and the diversity of data measurement, in order to solve the problem that different evaluation indicator measurements affect the weight effect, a data normalization method is used to assign and unify each evaluation indicator, so that the evaluation score range of each evaluation indicator is unified in the range of 0-1.

[0098] For example:

[0099] ① For the detection timeliness index value, a gradient of 0.15 can be used for assignment. For example, the sample with the best timeliness and label 6 is assigned a value of 1, the sample with label 5 is assigned a value of 0.85, and so on;

[0100] ② For the detection timeliness index value, due to the extremely large value of the detection mileage deviation value, Z-

[0101] The Score method standardizes the detection mileage deviation value. Z-Score standardization can convert data of different magnitudes into a uniformly measured Z-Score value, which represents the difference between the data point and the overall data statistic. It is suitable for normalizing data with a relatively scattered distribution. The specific formula of the Z-Score method is as follows:

[0102] z=(x-δ) / σ

[0103] Among them, z is the Z-Score value, x is the detection mileage deviation value that needs to be standardized, δ represents the overall mean of the data, and σ represents the standard deviation of the data. After obtaining the z value, map the z value to the range of 0-1.

[0104] ③ For Boolean indicators such as detection normative index value and accessibility index value, "yes" is assigned a value of 1.

[0105] If "No", the value is 0;

[0106] ④For percentage data such as detection effectiveness index values, simply convert them into decimals.

[0107] Figure 3 : is a heat map of single file evaluation indexes in the detection business scenario in the embodiment of the present invention; after normalizing the detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value and detection accuracy index value, the heat map of each index is as follows: Figure 3 shown.

[0108] In step 107, the analytic hierarchy process is used to determine the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value, and the detection accuracy index value.

[0109] In an embodiment, the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value, and the detection accuracy index value are determined by using the hierarchical analysis method, which may include:

[0110] Using the hierarchical analysis method, according to the business scenarios, a judgment matrix of detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value and detection accuracy index value is constructed;

[0111] Adopting the hierarchical analysis method, the values ​​of the test standardization index, the test integrity index, the test timeliness index, the accessibility index, the test effectiveness index and the test accuracy index are ranked according to the business scenarios;

[0112] Calculate the eigenphasor and maximum eigenvalue of the judgment matrix, and determine the weights of the sorted detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value, and detection accuracy index value based on the eigenphasor and maximum eigenvalue of the judgment matrix.

[0113] The Analytic Hierarchy Process (AHP) is a method that can reasonably combine qualitative decision-making with quantitative indicators. First, the data quality evaluation process is divided into different levels according to different factors, and then the importance of the final layer compared to the higher level is compared. The indicators between the levels are given corresponding weights and rankings.

[0114] (1) Establishing a hierarchical model

[0115] Figure 4 is a hierarchical structure model diagram in an embodiment of the present invention, such as Figure 4 As shown, the hierarchical model includes a target layer, a criterion layer, a sub-criterion layer, and an object layer; the target layer includes the data quality evaluation results; the criterion layer includes E1 detection timeliness index value, E2 detection standardization index value, E3 accessibility index value, E4 detection integrity index value, E5 detection accuracy index value, and E6 detection effectiveness index value; the sub-criterion layer includes E21 naming standardization, E22 type standardization, E51 mileage 95% quantile, and E52 mileage 99% quantile; the object layer includes data quality evaluation objects.

[0116] In the process of data quality evaluation, the object layer is the data file to be evaluated, namely the track geometry detection data; the criterion layer and sub-criterion layer are the values ​​of various indicators that affect the evaluation results, and the sub-criterion layer is the subclass of the criterion layer; the target layer is the evaluation result of the final data quality.

[0117] (2) Constructing the judgment matrix

[0118] One of the key points of the hierarchical analysis method is pairwise comparison, which grades the relative importance of elements between levels, usually using a 1-9 scale. In this example, a simple scaling method is used, that is, for the six major indicators in data quality evaluation, the scale between every two indicators is set to 1 / n, where n is the number of indicators belonging to each higher level in this layer.

[0119] Table 2 is a table of the judgment scale and meaning of the hierarchical analysis method in the embodiment of the present invention. As shown in Table 2, the simple scale is more suitable for hierarchical analysis in multiple business scenarios. In different business scenarios, the importance of each indicator is different, so it is only necessary to sort the importance of the indicators in different business scenarios. Therefore, the simple scale has certain advantages when used in multiple business scenarios. First, it reduces multiple judgment links; second, it can reduce the bias of indicators to a certain extent, causing some indicators to be given too much attention; third, it ensures that the consistency verification conditions are met and reduces the complexity of calculation.

[0120] Table 2. Analytical hierarchy process judgment scale and meaning table

[0121]

[0122] For the criterion layer belonging to the target layer, the number of criterion layer indicator gradients is different in different business scenarios. When two indicators are equally important, the number of indicator gradients is reduced by 1. In most business scenarios, there are no two indicators of equal importance. The number of indicator gradients is 6. Then the criterion layer judgment matrix R 1 As shown below:

[0123]

[0124] For the indicators belonging to the criterion layer in the sub-criterion layer, the number of indicators is 2, then the sub-criterion layer judgment matrix R 2 As shown below:

[0125]

[0126] (3) Hierarchical single sorting in various business scenarios

[0127] Through the ranking method, the importance of the six indicators is ranked according to different businesses. For example, in the three business scenarios: detection task execution, detection business management, and data analysis application:

[0128] The ranking results of the detection task execution criterion layer are: E4 detection integrity index value > E5 detection accuracy index value > E1 detection timeliness index value > E2 detection standardization index value > E3 accessibility index value > E6 detection effectiveness index value, and the ranking results of the sub-criterion layer are: E22 type standardization > E21 naming standardization, E51 mileage 95% quantile > E52 mileage 99% quantile.

[0129] From the perspective of detection business management, the ranking results at the criterion layer are: E4 detection integrity index value > E1 detection timeliness index value > E2 detection standardization index value > E3 accessibility index value > E6 detection effectiveness index value > E5 detection accuracy index value; the ranking results at the sub-criterion layer are: E22 type standardization > E21 naming standardization, E51 mileage 95% quantile > E52 mileage 99% quantile.

[0130] The ranking results of the criterion layer from the data analysis application perspective are: E3 accessibility index value = E2 detection standardization index value > E4 detection integrity index value > E5 detection accuracy index value > E6 detection effectiveness index value > E1 detection timeliness index value, and the ranking results of the sub-criterion layer are: E22 type standardization > E21 naming standardization, E51 mileage 95% quantile > E52 mileage 99% quantile.

[0131] (4) Calculate weight

[0132] By calculating the eigenvector and the maximum eigenvalue of the judgment matrix, the weights of each ranking after the indicators of the criterion layer are sorted can be obtained. Table 3 is a criterion layer weight table in an embodiment of the present invention, as shown in Table 3.

[0133] Table 3 Criteria layer weight table

[0134]

[0135] At the same time, the ranking weights of the indicators in the sub-criteria layer are obtained after sorting. Table 4 is a weight table of the sub-criteria layer in an embodiment of the present invention, as shown in Table 4:

[0136] Table 4 Sub-criteria layer weight table

[0137]

[0138] In step 108, the quality level of the track geometry detection data is determined according to the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value, and the detection accuracy index value.

[0139] For example, in different scenarios, the quality results of track geometry detection data are calculated based on the weights of the criterion layer and the sub-criteria layer:

[0140]

[0141] Where R is the quality level evaluation result of track geometry detection data, W i is the weight corresponding to each indicator value, P i is the value of each indicator; if the indicator value has sub-item P ki , then through the weight W of the sub-item ki With sub-item Pki Obtained P by weighted multiplication k .

[0142] Figure 5 This is an example diagram of the data quality evaluation result of a single track geometry detection file in the detection business scenario of the embodiment of the present invention. The analysis result of the track geometry detection data quality in the detection business scenario is as Figure 5 shown, where the quality index corresponds to different quality levels. For example, the quality index [0.9, 1) represents high-quality track geometry detection data; the quality index [0.7, 0.9) represents qualified track geometry detection data; the quality index [0.5, 0.7) represents unqualified track geometry detection data; the quality index [0, 0.5) represents invalid track geometry detection data.

[0143] As described above, a method for evaluating the quality of track geometry detection data in a multi-dimensional business scenario proposed in the embodiment of the present invention has the following beneficial effects:

[0144] (1) Defined the quality evaluation indicators for track geometry detection data, and proposed methods such as the aging classification method based on K-means, the standardization judgment method based on regular expressions, the calculation method of the accuracy index value based on a sliding window, and the multi-index normalization method based on Z-score. These index values can not only comprehensively evaluate the quality of track geometry detection data and provide a quantitative tool, but also contribute to scientific decision-making and algorithm optimization in different business scenarios.

[0145] (2) For the first time, a data quality analysis method for track geometry detection data was proposed, filling the research gap in the current field and providing a new idea for the data quality analysis of other professional detections.

[0146] (3) It was sorted out from different business scenarios such as detection business execution, detection business management, and detection analysis application. For the situations in different scenarios, the quality of track geometry detection data was analyzed by combining the analytic hierarchy process, which not only improves the accuracy and reliability of data quality analysis, but also helps to optimize the track maintenance strategy, focus on the quality status of detection data in different scenarios, reduce the maintenance cost, and improve the transportation efficiency.

[0147] In summary, the embodiment of the present invention proposes for the first time a complete multi-dimensional business scenario track geometry detection data quality evaluation method, clarifies the track geometry detection data quality evaluation indicators, designs a set of efficient quality evaluation algorithm processes, and promotes the standardization and scientific process of data quality management in related fields. Through the embodiment of the present invention, a systematic framework will be provided for the quality evaluation of high-speed railway comprehensive detection data, and the quality evaluation of track geometry detection data will be carried out efficiently and accurately to ensure the accuracy and reliability of the quality evaluation of track geometry detection data, and provide data support for the operation management and maintenance decision-making of high-speed railways, so as to optimize the track maintenance strategy and reduce maintenance costs.

[0148] The present invention also provides a device for evaluating the quality of track geometry detection data in a multi-dimensional business scenario, as described in the following embodiments. Since the principle of solving the problem by the device is similar to the method for evaluating the quality of track geometry detection data in a multi-dimensional business scenario, the implementation of the device can refer to the implementation of the method for evaluating the quality of track geometry detection data in a multi-dimensional business scenario, and the repeated parts will not be repeated.

[0149] Figure 6 FIG. 1 is a structural example diagram of a device for evaluating quality of track geometry detection data in a multi-dimensional business scenario according to an embodiment of the present invention. Figure 6 As shown, the device comprises:

[0150] The detection standardization index value determination module 601 is used to match the file name of the track geometry detection data with a preset business dictionary table, and determine the matching result as the detection standardization index value; the business dictionary table records the codes corresponding to different business information;

[0151] The detection integrity index value determination module 602 is used to extract the start and end mileages of the actual detection from the file header information of the track geometry detection data, calculate the actual detection mileage length according to the start and end mileages of the actual detection; calculate the mileage coverage rate of the detection plan according to the actual detection mileage length and the detection plan mileage length; and determine the mileage coverage rate of the detection plan as the detection integrity index value;

[0152] The detection timeliness index value determination module 603 is used to cluster the collection time of the track geometry detection data and determine the clustering result as the detection timeliness index value;

[0153] The accessibility index value determination module 604 is used to extract key metadata from the file header information of the track geometry detection data according to the predefined byte position and length; and determine the accuracy of the extracted key metadata as the accessibility index value of the track geometry detection data;

[0154] The detection validity index value determination module 605 is used to determine the proportion of valid start and end mileages in the track geometry detection data according to the actual detection mileage length and the mileage length of invalid data in the track geometry detection data; and determine the proportion of valid start and end mileages as the detection validity index value;

[0155] A detection accuracy index value determination module 606 is used to determine the matching degree between the track geometry detection data and the reference data, determine the detection mileage deviation value according to the matching degree, and determine the detection mileage deviation value as the detection accuracy index value;

[0156] A hierarchy analysis module 607 is used to determine the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value, and the detection accuracy index value by using a hierarchy analysis method;

[0157] The quality analysis module 608 is used to determine the quality level of the track geometry detection data according to the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value and the detection accuracy index value.

[0158] In one embodiment, the detection standardization index determination module 601 is specifically used to extract key business information from the file name of the track geometry detection data according to a predefined regular expression; the regular expression defines the naming format of the track geometry detection data; the key business information includes the line code, the line code, the starting station, the ending station, the detection date and the detection time; the key business information is matched with the business dictionary table; if the key business information matches the corresponding code, the detection standardization index value is determined to be yes, otherwise the detection standardization index value is determined to be no.

[0159] In one embodiment, the inspection integrity index determination module 602 is also used to structure the inspection plan data before calculating the mileage coverage of the inspection plan based on the actual inspection mileage length and the inspection plan mileage length; decompose the structured inspection plan data to extract the inspection plan mileage length.

[0160] In one embodiment, the detection timeliness index determination module 603 is specifically used to cluster the collection time of the track geometry detection data by using the K-means clustering method; wherein the collection time of the track geometry detection data is divided into multiple data clusters, and the clustering goal is to minimize the sum of square errors in each data cluster; for example, the sum of distances from data points in each data cluster to the cluster center can be minimized according to the following formula to achieve the clustering goal:

[0161]

[0162] Among them, SSE is the sum of the distances from the data points in each data cluster to the cluster center; K is the number of data clusters, C i is the data point set of the i-th data cluster, x is the data point of C i The detection timeliness index value, μ i is the center of the i-th data cluster;

[0163] In one embodiment, the detection timeliness index determination module 603 is specifically used to determine the number of data clusters according to a predefined silhouette coefficient index value; the silhouette coefficient index value is used to measure the compactness and separation of the clustering results.

[0164] In one embodiment, the detection accuracy index determination module 606 is specifically used to remove invalid data from the track geometry detection data; sample the track geometry detection data after the invalid data is removed, slide the sampled data points, and calculate the correlation coefficient between each data point and the reference data; determine the matching degree between the track geometry detection data and the reference data according to the correlation coefficient; for example, the correlation coefficient between each data point and the reference data can be calculated according to the following formula:

[0165]

[0167] Among them, r is the correlation coefficient; x i and i are the corresponding data points in the test data and the benchmark data respectively; and are the average values ​​of the detection data and the benchmark data, respectively.

[0168] Figure 7 FIG. 1 is a specific example diagram of the structure of the multi-dimensional business scenario track geometry detection data quality evaluation device in an embodiment of the present invention. Figure 7 As shown, in one embodiment, Figure 6 The device for evaluating quality of track geometry detection data in a multi-dimensional business scenario in the embodiment of the present invention may further include a normalization module 701 .

[0169] In this example, the normalization module 701 is specifically used to normalize the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value and the detection accuracy index value before the hierarchical analysis module 607 adopts the hierarchical analysis method to determine the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value and the detection accuracy index value; for example, the normalization of the detection timeliness index value may include: adopting a gradient assignment method to normalize the detection timeliness index value; for another example, adopting a Z-Score method to standardize the detection mileage deviation value.

[0170] In one embodiment, the hierarchical analysis module 607 is specifically used to adopt the hierarchical analysis method to construct a judgment matrix of detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value and detection accuracy index value according to the business scenario; adopt the hierarchical analysis method to sort the detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value and detection accuracy index value according to the business scenario; calculate the eigenvector and maximum eigenvalue of the judgment matrix, and determine the weights of the sorted detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value and detection accuracy index value according to the eigenvector and maximum eigenvalue of the judgment matrix.

[0171] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for evaluating the quality of track geometry detection data in multi-dimensional business scenarios when executing the computer program.

[0172] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned method for evaluating the quality of track geometry detection data in multi-dimensional business scenarios.

[0173] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the above-mentioned method for evaluating the quality of track geometry detection data in multi-dimensional business scenarios.

[0174] The multi-dimensional business scenario track geometry detection data quality evaluation method and device of the embodiment of the present invention determines the detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value, and detection accuracy index value of the track geometry detection data; uses the hierarchical analysis method to determine the weights and rankings of these index values; determines the quality level of the track geometry detection data according to the weights and rankings of these index values; the embodiment of the present invention will provide a systematic framework for the quality evaluation of high-speed railway comprehensive detection data, efficiently and accurately perform track geometry detection data quality evaluation, ensure the accuracy and reliability of track geometry detection data quality evaluation, and provide data support for the operation management and maintenance decision-making of high-speed railways, so as to optimize track maintenance strategies and reduce maintenance costs.

[0175] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0176] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0177] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0178] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0179] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for evaluating the quality of track geometry detection data in multi-dimensional business scenarios, characterized in that: include: Match the file name of the track geometry detection data with a pre-set business dictionary table, and determine the matching result as the detection standardization index value; The business dictionary table records the codes corresponding to different business information; Extract the start and end mileages of the actual inspection from the file header information of the track geometry inspection data, and calculate the actual inspection mileage length based on the start and end mileages of the actual inspection; Calculate the mileage coverage of the inspection plan based on the actual inspection mileage length and the inspection plan mileage length; Determine the mileage coverage of the inspection plan as the inspection completeness index value; Clustering the collection time of track geometry detection data, and determining the clustering result as the detection timeliness index value; Extract key metadata from the file header information of track geometry detection data according to predefined byte positions and lengths; Determine the accuracy of the extracted key metadata as the accessibility index value of the track geometry detection data; According to the actual detection mileage length and the mileage length of invalid data in the track geometry detection data, determine the proportion of valid start and end mileages in the track geometry detection data; The proportion of effective start and end mileage is determined as the detection effectiveness index value; Determine the matching degree between the track geometry detection data and the reference data, determine the detection mileage deviation value according to the matching degree, and determine the detection mileage deviation value as the detection accuracy index value; The analytic hierarchy process is used to determine the weights and rankings of the test standardization index value, the test integrity index value, the test timeliness index value, the accessibility index value, the test effectiveness index value, and the test accuracy index value; The quality level of track geometry inspection data is determined based on the weights and rankings of the inspection standardization index value, inspection integrity index value, inspection timeliness index value, accessibility index value, inspection effectiveness index value and inspection accuracy index value.

2. The method according to claim 1, characterized in that The file name of the track geometry detection data is matched with the pre-set business dictionary table, and the matching result is determined as the detection standardization index value, including: Extract key business information from the file name of track geometry inspection data according to the predefined regular expression; the regular expression defines the naming format of track geometry inspection data; the key business information includes line code, line code, starting station, ending station, inspection date and inspection time; Match key business information with the business dictionary table; If the key business information matches the corresponding code, the detection normative indicator value is determined to be yes, otherwise the detection normative indicator value is determined to be no.

3. The method according to claim 1, characterized in that Based on the actual inspection mileage length and the inspection plan mileage length, before calculating the mileage coverage of the inspection plan, it also includes: Structuring the inspection plan data; Decompose the structured inspection plan data and extract the inspection plan mileage length.

4. The method according to claim 1, characterized in that Clustering the acquisition time of track geometry detection data, including: The collection time of track geometry detection data is clustered by the K-means clustering method. The collection time of track geometry detection data is divided into multiple data clusters, and the clustering goal is to minimize the sum of square errors within each data cluster.

5. The method according to claim 4, characterized in that Clustering the acquisition time of track geometry detection data, including: According to the following formula, the sum of the distances from the data points in each data cluster to the cluster center is minimized to achieve the clustering goal: Among them, SSE is the sum of the distances from the data points in each data cluster to the cluster center; K is the number of data clusters, C i is the data point set of the i-th data cluster, x is the data point of C i The detection timeliness index value, μ i is the center of the ith data cluster.

6. The method according to claim 4, characterized in that Clustering the acquisition time of track geometry detection data, including: The number of data clusters is determined based on the predefined silhouette coefficient index value; the silhouette coefficient index value is used to measure the compactness and separation of the clustering results.

7. The method according to claim 1, characterized in that Determine the degree of match between track geometry inspection data and reference data, including: Eliminate invalid data from track geometry detection data; Sampling the track geometry detection data after removing invalid data, sliding the sampled data points, and calculating the correlation coefficient between each data point and the benchmark data; The matching degree between the track geometry detection data and the reference data is determined based on the correlation coefficient.

8. The method according to claim 7, characterized in that The correlation coefficient between each data point and the benchmark data is calculated according to the following formula: Where r is the correlation coefficient; x i and i are the corresponding data points in the test data and the benchmark data respectively; and are the average values ​​of the detection data and the benchmark data, respectively.

9. The method according to claim 1, characterized in that Before using the hierarchical analysis method to determine the weights and rankings of the test standardization index value, the test integrity index value, the test timeliness index value, the accessibility index value, the test effectiveness index value and the test accuracy index value, it also includes: The test standardization index value, test integrity index value, test timeliness index value, accessibility index value, test effectiveness index value and test accuracy index value are normalized.

10. The method according to claim 9, characterized in that Normalize the detection timeliness index values, including: The gradient assignment method is used to normalize the detection timeliness index value.

11. The method according to claim 9, characterized in that Normalize the detection accuracy index values, including: The Z-Score method is used to standardize the detection mileage deviation value.

12. The method according to claim 1, characterized in that The analytic hierarchy process is used to determine the weights and rankings of the test standardization index value, the test integrity index value, the test timeliness index value, the accessibility index value, the test effectiveness index value, and the test accuracy index value, including: Using the hierarchical analysis method, according to the business scenarios, a judgment matrix of detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value and detection accuracy index value is constructed; Adopting the hierarchical analysis method, the values ​​of the test standardization index, the test integrity index, the test timeliness index, the accessibility index, the test effectiveness index and the test accuracy index are ranked according to the business scenarios; Calculate the eigenphasor and maximum eigenvalue of the judgment matrix, and determine the weights of the sorted detection standardization index value, detection integrity index value, detection timeliness index value, accessibility index value, detection effectiveness index value, and detection accuracy index value based on the eigenphasor and maximum eigenvalue of the judgment matrix.

13. A device for evaluating the quality of track geometry detection data in a multi-dimensional business scenario, characterized in that: include: A detection normative index value determination module is used to match the file name of the track geometry detection data with a preset business dictionary table, and determine the matching result as the detection normative index value; The business dictionary table records the codes corresponding to different business information; The detection integrity index value determination module is used to extract the start and end mileages of the actual detection from the file header information of the track geometry detection data, and calculate the actual detection mileage length according to the start and end mileages of the actual detection; Calculate the mileage coverage of the inspection plan based on the actual inspection mileage length and the inspection plan mileage length; Determine the mileage coverage of the inspection plan as the inspection completeness index value; A detection timeliness index value determination module is used to cluster the collection time of track geometry detection data and determine the clustering result as the detection timeliness index value; An accessibility index value determination module is used to extract key metadata from the file header information of the track geometry detection data according to a predefined byte position and length; and determine the accuracy of the extracted key metadata as the accessibility index value of the track geometry detection data; A detection effectiveness index value determination module is used to determine the proportion of valid start and end mileages in the track geometry detection data according to the actual detection mileage length and the mileage length of invalid data in the track geometry detection data; and determine the proportion of valid start and end mileages as the detection effectiveness index value; A detection accuracy index value determination module is used to determine the matching degree between the track geometry detection data and the reference data, determine the detection mileage deviation value according to the matching degree, and determine the detection mileage deviation value as the detection accuracy index value; A hierarchical analysis module is used to determine the weights and rankings of the detection standardization index value, the detection integrity index value, the detection timeliness index value, the accessibility index value, the detection effectiveness index value and the detection accuracy index value by using the hierarchical analysis method; The quality analysis module is used to determine the quality level of track geometry inspection data according to the weights and rankings of inspection standardization index value, inspection integrity index value, inspection timeliness index value, accessibility index value, inspection effectiveness index value and inspection accuracy index value.

14. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 12 is implemented.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

16. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

Images