CN115271565A - DEA-based method, device and equipment for evaluating highway pavement maintenance measures - Google Patents
DEA-based method, device and equipment for evaluating highway pavement maintenance measures Download PDFInfo
- Publication number
- CN115271565A CN115271565A CN202211199571.3A CN202211199571A CN115271565A CN 115271565 A CN115271565 A CN 115271565A CN 202211199571 A CN202211199571 A CN 202211199571A CN 115271565 A CN115271565 A CN 115271565A
- Authority
- CN
- China
- Prior art keywords
- maintenance
- disease
- pavement
- dea
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention discloses a DEA-based method, a DEA-based device and DEA-based equipment for evaluating highway pavement maintenance measures, wherein the DEA-based device comprises the following steps: acquiring historical pavement quality conditions and corresponding historical pavement maintenance conditions of a highway to be researched, and extracting historical disease data, maintenance measures and maintenance cost of each damaged road section and pavement performance indexes before and after maintenance; carrying out noise filtering by adopting a K nearest neighbor algorithm; calculating unit maintenance cost and pavement performance index change quantity of each damaged road section; and (3) taking each disease maintenance and repair measure as a label, taking the unit maintenance cost and the pavement performance index change quantity as input and output respectively, establishing a DEA evaluation model of the pavement maintenance measure of the expressway to be researched, evaluating the benefit of each disease maintenance and repair measure, and giving a maintenance decision suggestion. The method can integrate economy and benefit, objectively evaluate various maintenance measures and help the road management and maintenance department to decide which diseases are mainly repaired under the condition of limited funds.
Description
Technical Field
The invention relates to the technical field of traffic, in particular to a DEA-based method, a DEA-based device and DEA-based equipment for evaluating highway pavement maintenance measures.
Background
The rapid development of social economy makes expressways play an increasingly important role in national construction and development. But the highway is also subject to wear and damage due to heavy tasks after greatly improving the comprehensive transportation efficiency and the back of the transportation structure. With the gradual increase of service life and vehicle-passing mileage, various diseases can occur on the road surface of many highways, especially under heavy-load traffic conditions. The service life of the highway is seriously influenced by the pavement diseases, the service performance of the pavement is greatly reduced, and at the moment, engineering projects can be developed by a highway management department to repair the highway. Limited by resources and construction funds, the highway management and maintenance departments usually only adopt various maintenance and repair measures to specifically repair pavement diseases. In order to fully utilize limited resources and funds and enable the overall road surface condition of the expressway to be relatively optimal, various maintenance measures must be evaluated, and measures with the maximum benefits under the condition of equal money amount are preferably adopted during construction. For this purpose, in recent decades, there have been common evaluation methods such as expert empirical models and economic-benefit models.
However, the conventional evaluation methods have the following problems:
the classical expert experience model is too dependent on the personal view of experts, and the problems of strong subjectivity, inconsistent evaluation results of multiple times and the like exist. Meanwhile, the recruitment expert also needs to spend money, which is a small burden under the condition of limited money.
The common economic-benefit evaluation model can analyze the benefit conditions of various maintenance measures under the constraint of economic conditions, and is a better evaluation method capable of integrating the economic efficiency and the benefit. However, in practical application, the influence of random errors and noise is caused, the influence of the data needs to be eliminated, and the accuracy of the evaluation model is improved.
Evaluation models based on various machine learning algorithms which are common in recent years have certain application scenes, but the models are too dark boxes, and the models cannot be artificially evaluated. In addition, the accuracy of the model has a great relationship with the training data amount, and the evaluation effect is not ideal in a small sample set scene.
In conclusion, the single evaluation method for evaluating the highway maintenance measures is difficult to adapt to a real scene with a small sample set and a large amount of noise data, and accurate effect evaluation cannot be objectively and reasonably carried out on the maintenance measures.
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides a method, a device and equipment for evaluating highway pavement maintenance measures based on DEA (data envelope analysis), which can provide maintenance measure priority under the condition of limited fund so as to obtain the best benefit.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
a method for evaluating highway pavement maintenance measures based on data envelope analysis comprises the following steps:
step 1, acquiring historical pavement quality conditions and corresponding historical pavement maintenance conditions of a highway to be researched, and extracting historical disease data of each highway section, maintenance measures and maintenance cost of each damaged highway section and pavement performance indexes before and after maintenance from the historical pavement quality conditions and the corresponding historical pavement maintenance conditions; forming a data sample by each historical disease, the corresponding maintenance measure and the pavement performance indexes before and after maintenance;
step 2, performing noise filtering processing on all data samples by adopting a K nearest neighbor algorithm according to a disease data set formed by all disease data and a maintenance and maintenance measure data set formed by corresponding maintenance and maintenance measures;
step 3, counting all the filtered data samples, and calculating the unit maintenance cost and the pavement performance index change quantity of the maintenance measures corresponding to each disease based on all the data samples of each disease;
step 4, taking the maintenance measures corresponding to each disease as labels, taking the unit maintenance cost and the pavement performance index variation as input and output respectively, and establishing a DEA evaluation model of the pavement maintenance measures of the expressway to be researched;
and 5, evaluating the benefit index of each disease maintenance measure of the expressway to be researched based on the DEA evaluation model, and providing a maintenance decision suggestion.
Further, the road surface performance indexes comprise a road surface damage condition index PCI, a road surface running quality index RQI and a road surface rutting depth index RDI.
Further, the historical road surface maintenance condition comprises historical major repair, middle repair and minor repair records; and the historical major repair, medium repair and minor repair records comprise maintenance measures and maintenance costs of all the diseases of the expressway to be researched and the number of the starting point pile and the number of the stopping point pile of all the road sections included by the diseases.
Further, the method for performing noise filtering on all data samples by adopting the K nearest neighbor algorithm comprises the following steps:
averagely dividing a disease data set into two subsets, wherein the number of each disease in the two subsets is the same;
clustering the first disease subset by adopting a K nearest neighbor algorithm and based on an Euclidean distance to obtain an initial disease classification model;
classifying the disease data in the second disease subset using an initial disease classification model; if the classification output label of a certain disease data is not consistent with the actual disease label, the disease data is discarded from the second disease subset; simultaneously discarding the data sample corresponding to the disease data;
clustering the current second disease subset by adopting a K nearest neighbor algorithm and based on the Euclidean distance to obtain an improved disease classification model;
classifying the disease data in the first subset of diseases using an improved disease classification model; if the classification output label of a certain disease data is not consistent with the actual disease label, the disease data is discarded from the first disease subset; simultaneously discarding the data sample corresponding to the disease data;
and performing noise filtering treatment on the maintenance and repair measure data set and the data sample by adopting a filtering method the same as that of the disease data set.
Further, the step 5 of evaluating and obtaining the benefit indexes of each disease maintenance measure comprises the following steps: technical benefit, scale benefit, comprehensive benefit, relaxation variable, redundancy variable and scale reward coefficient; the technical benefit reflects an efficiency situation brought by technical factors, the scale benefit reflects an efficiency situation brought by input scale, the comprehensive benefit reflects a comprehensive efficiency situation brought by technical factors and input scale, the redundancy variable reflects how much input is reduced to achieve the target efficiency, and the relaxation variable reflects how much output is increased to achieve the target efficiency; the consideration scale factor reflects how many units of input need to be increased for each unit of output increase.
Further, effectiveness evaluation is carried out on the DEA evaluation model on the basis of each benefit index, and the effectiveness evaluation results are 3, namely DEA strong effectiveness, DEA weak effectiveness and non-DEA effectiveness; the method for obtaining the DEA effectiveness evaluation result from the benefits comprises the following steps: if the comprehensive benefit is 1 and the redundancy variable and the relaxation variable are 0, DEA is strong and effective; if the comprehensive benefit is 1 and the redundancy variable or the relaxation variable is more than 0, DEA is weak and effective; if the comprehensive benefit is less than 1, the DEA is not effective.
Further, on the basis of each benefit index, scale reward analysis, investment redundancy analysis and output insufficiency analysis are further carried out on various disease maintenance measures.
A DEA-based expressway pavement maintenance measure evaluation device comprises:
a sample acquisition module to: obtaining historical pavement quality conditions and corresponding historical pavement maintenance conditions of an expressway to be researched, and extracting historical disease data of each section of the expressway, maintenance measures and maintenance cost of each damaged section and pavement performance indexes before and after maintenance from the historical pavement quality conditions and the corresponding historical pavement maintenance conditions; forming a data sample by each historical disease, corresponding maintenance measures and pavement performance indexes before and after maintenance;
a noise filtering processing module for: performing noise filtering processing on all data samples by adopting a K neighbor algorithm according to a disease data set formed by all disease data and a maintenance and repair measure data set formed by corresponding maintenance and repair measures;
a data processing module to: counting all the filtered data samples, and calculating the unit maintenance cost and the pavement performance index change amount of maintenance and repair measures corresponding to each disease based on all the data samples of each disease;
a DEA evaluation model construction module for: taking maintenance and repair measures corresponding to each disease as labels, taking unit maintenance cost and pavement performance index variation as input and output respectively, and establishing a DEA evaluation model of the pavement maintenance measures of the expressway to be researched;
the index evaluation and suggestion module is used for: and evaluating the benefit index of each disease maintenance and repair measure of the expressway to be researched based on the DEA evaluation model, and providing a maintenance decision suggestion.
An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor implements the method for evaluating a measure of maintenance of a highway pavement based on DEA according to any one of the above technical solutions.
Advantageous effects
The method has the technical effects that various maintenance measures adopted on the expressway can be objectively evaluated by integrating the economy and the benefit, scientific and reasonable references are provided for the highway management and maintenance department, and the highway management and maintenance department is helped to make a decision on mainly repairing diseases under the condition of limited fund subsequently, so that a better overall maintenance effect is achieved.
Drawings
FIG. 1 is a schematic flow chart of a method according to an embodiment of the present application.
Detailed Description
The following describes embodiments of the present invention in detail, which are developed based on the technical solutions of the present invention, and give detailed implementation manners and specific operation procedures to further explain the technical solutions of the present invention.
The embodiment provides a method for evaluating a pavement maintenance measure of an expressway based on DEA, which is shown in fig. 1 and comprises the following steps:
step 1, acquiring historical pavement quality conditions and corresponding historical pavement maintenance conditions of a highway to be researched, and extracting historical disease data of each highway section, maintenance measures and maintenance cost of each damaged highway section and pavement performance indexes before and after maintenance from the historical pavement quality conditions and the corresponding historical pavement maintenance conditions; and forming a data sample by each historical disease, the corresponding maintenance measure and the pavement performance indexes before and after maintenance, wherein all the data samples form a sample set.
The historical pavement quality condition of the expressway comprises historical pavement disease data and historical pavement condition data of the expressway to be researched. The historical pavement disease data comprises historical existence conditions of various pavement diseases on each road section of the researched expressway; the historical road surface condition data includes historical values of road surface performance indexes PCI, RQI, RDI of respective sections of the studied expressway. And the extracted disease data comprises the type and area/length of the pavement diseases.
In the example, the Hunan section of the Jinggang and Australian expressway is selected as a research object, and the obtained historical pavement quality conditions are as follows:
1.1 Presence of various road surface defects on each section of the south-Hunan section of the Kyoto, australian expressway in Beijing harbor in 2019 and 2020. For example: in 2019, strip-shaped repair with the influence area of 2.4029 square meters exists on the descending lane road section with the starting point pile number of 1310.0 and the end point pile number of 1311.0; in 2020, there were transverse cracks of length 4.2185 m, block repairs of area 0.0706 m and strip repairs of area of influence 52.9963 m on this route.
1.2 Values of road surface performance indexes PCI, RQI, RDI of each road section of the kano-australia highway hunan section in 2019 and 2020. For example: in 2019, in a downlink lane section with a starting point pile number of 1310.0 and an end point pile number of 1311.0, the road surface damage condition index PCI value is 98.1273, the road surface running quality index RQI value is 94.8290, and the road surface rutting depth index RDI value is 96.0768; in 2020, the road surface damage condition index PCI value of the road section is 95.9387, the road surface running quality index RQI value is 94.9219, and the road surface rutting depth index RDI value is 96.1125.
The historical road surface maintenance condition described in this embodiment includes historical major repair, medium repair, and minor repair records. In a specific example, the historical major repair, intermediate repair and minor repair records include the starting point stake number and the ending point stake number of each damaged section of the expressway, maintenance and repair measures implemented on each damaged section and costs thereof, and information of branch companies responsible for maintenance and repair work, and if the record is major repair and intermediate repair, the record also includes engineering project information.
In the example, the Hunan section of the Jinggang Australian expressway is selected as a research object, and the obtained historical pavement maintenance conditions comprise:
2.1 No major repairs were made in 2019 and 2020 on the Hunan section of the expressway in Kyoto harbor.
2.2 Record of the mid-school record of the Hunan section of the Kyoto, hong Kong, australia highway in 2019 and 2020. For example: in 2019, a project of 'repairing in the road surface in 2019 (temporary growth)' is developed by a Changsha division, and for a road section with a starting point pile number of 1444.3 and a stopping point pile number of 1444.77, a maintenance measure of 'additionally paving an AC-13 cover surface' is adopted, so that 234775.30 yuan is consumed.
2.3 Minor repair record of the south Hunan section of the Jinggang Australian expressway in 2019 and 2020. For example: in 2019, the Changsha division company takes corresponding maintenance measures for the disease of the 'pot hole' on the road section with the starting point pile number of 1473.15 and the stopping point pile number of 1474, and costs 200 yuan.
And 2, performing noise filtering treatment on the sample set by adopting a K neighbor algorithm according to a disease data set formed by all disease data and a maintenance and maintenance measure data set formed by corresponding maintenance and maintenance measures.
In the manually recorded disease data, whether the disease data is of a type or a length, obvious errors can occur. The K nearest neighbor algorithm is utilized to filter outliers, the inaccuracy of the front surface of the envelope analysis caused by error data can be effectively avoided, and therefore the accuracy of the model is guaranteed. The method comprises the following specific steps:
firstly, according to the specific information of the diseases of each road section of the Hunan section of the Kyowa expressway in Kyowa, hong Kong and Australia in 2019 and 2020, half of the cases are extracted from each disease category and used as initial data points. And calling a KNeighborsClassifier algorithm in the neighbor module from a sklern toolkit of python, taking the other half of road segment disease data as the input of the algorithm, taking the nearest neighbor number K as 20, defining the distance as the Euclidean distance, classifying the road segments according to the categories, and discarding the value if the distance does not accord with the actual disease label. Then, the screened data is used as an initial data point, the operation is repeated for the disease data of the first half road section, and finally the total screened disease data is obtained.
Similarly, noise filtering treatment can be performed according to specific information of maintenance and repair measures of each disease adopted in 2019 and 2020 on the Hunan section of the expressway in Kyoto, hong Kong and Australia.
And discarding the data sample corresponding to the outlier data while discarding the outlier disease data and maintenance and repair measure data from the corresponding data set.
And 3, performing statistical treatment on all the filtered data samples, and calculating the unit maintenance cost and the pavement performance index variation of the maintenance measures corresponding to each disease based on all the data samples of each disease.
After screening and removing error data, counting and processing the data samples to obtain the unit maintenance cost (repair cost per unit length or unit area) of each data sample and the road surface performance index improvement condition before and after maintenance of each road section.
In this example, the respective input amount and repair mileage of 16 types of repair measures for the diseases on the south-Hunan section of the expressway in Kyoto, australia and Hongkong are obtained through statistics, and based on the calculation, the repair cost per kilometer of each repair measure can be obtained. Meanwhile, according to the PCI, RQI and RDI values of each section of the Hunan section of the Kyowa expressway in Kyowa of Kyowa, hong Kong and Australia in 2019 and 2020, the road surface improvement condition after repairing each damaged section can be obtained, namely the difference between the PCI, RQI and RDI after adopting the maintenance and repair measures respectively relative to the difference before maintenance and repair.
And 4, establishing a DEA evaluation model of the pavement maintenance measures of the expressway to be researched by taking the maintenance and maintenance measures corresponding to each disease as labels and taking the unit maintenance cost and the pavement performance index change quantity as input and output respectively.
The DEA evaluation method is an efficiency evaluation method which has objectivity and is not doped with any subjective factor, and is suitable for the situation that an evaluation object has a plurality of decision units such as a plurality of inputs or a plurality of outputs. Therefore, in the embodiment, the maintenance effect of various road surface diseases is evaluated by selecting the DEA evaluation model. The method comprises the following specific steps:
and (3) establishing a DEA evaluation model of the highway pavement maintenance measures by taking the maintenance measures of different diseases as labels, taking the repair cost of each kilometer of the maintenance measures as input and taking the final improvement condition of the pavement condition as output. Through the statistics and processing of the data, maintenance and repair data of 16 diseases in 2019-2020 on the Hunan section of the expressway in Kyoto and Australia are obtained, and the average improvement condition of performance indexes of each road surface after each disease is repaired by adopting maintenance and repair measures is based on the data sample population. For example: the maintenance cost of the crack is 2871.0694 yuan per kilometer, and after the crack is repaired, the PCI value is increased by 0.6186, the RQI value is increased by 0.4863, and the RDI value is increased by 0.1182. And (3) taking the 'crack' as a label, taking 2871.0694 yuan per kilometer as an input index, taking the index of 0.6186 increase of a PCI value, 0.4863 increase of an RQI value and 0.1182 increase of an RDI value as an index of multi-item output, and so on, continuously inputting corresponding data of other 15 diseases, and establishing a DEA evaluation model.
And 5, evaluating the benefit index of each disease maintenance and repair measure of the expressway to be researched based on the DEA evaluation model, and giving a maintenance decision suggestion.
After data is input into a DEA evaluation model, the model can give respective technical benefits, scale benefits, comprehensive benefits, relaxation variables, redundancy variables and scale reward coefficients of 16 maintenance and repair measures. Wherein, the technical benefit reflects the efficiency condition brought by technical factors, the value of which is 1 indicates that the invested capital is used efficiently, otherwise indicates that the implementation technology of the maintenance and repair measures needs to be promoted; the scale benefit reflects the efficiency condition brought by the input scale, and if the value is 1, the optimal input scale is reached, otherwise, the input can be continuously increased; the comprehensive benefit reflects the comprehensive efficiency condition of the elements, and the value of the comprehensive benefit is the product of the technical benefit and the scale benefit and is not more than 1; the redundancy variable reflects the target efficiency achieved when reducing the investment; the relaxation variable reflects how much output is increased to achieve the target efficiency; the scale reward factor reflects how many units of investment need to be increased to increase the yield of one unit, and a value of 0 indicates that the scale profit is unchanged, less than 1 indicates that the scale profit is increased progressively, and more than 1 indicates that the scale profit is decreased progressively.
According to the indexes, the benefit condition of maintenance and repair measures can be analyzed to obtain the final DEA effectiveness evaluation result. There are 3 results for the DEA efficacy evaluation, DEA strong, weak and non-DEA effective, respectively. If the comprehensive benefit is 1 and the redundancy variable and the relaxation variable are 0, DEA is strong and effective; if the comprehensive benefit is 1 and one of the redundancy variable and the relaxation variable is more than 0, DEA is weak and effective; if the comprehensive benefit is less than 1, the DEA is not effective. For example: for repairing the disease of 'crack', the technical benefit is 0.022, the scale benefit is 0.538, the comprehensive benefit is 0.154, the redundancy variable is 0, and the relaxation variable is 0.074. On the contrary, for repairing the disease of the pit slot, the technical benefit is 1, the scale benefit is 1, the comprehensive benefit is 1, the redundancy variable is 0, and the relaxation variable is 0.
And secondly, according to the indexes, performing scale reward analysis, investment redundancy analysis and output insufficiency analysis, and giving corresponding decision suggestions.
The reward analysis is based on the reward factor. For example: for repairing the disease of 'crack', the scale reward coefficient is 0.672; for repairing the disease of 'block crack', the scale reward coefficient is 1.024. This indicates that the curing investment for the "crack" disease can be preferentially increased relative to the "massive crack".
The invested redundancy analysis is based on redundancy variables and invested redundancy rates. The input redundancy rate is a ratio of a redundancy variable to an input amount, and a larger value means a larger input ratio to be reduced. The result can provide a target for improving the repair efficiency of various diseases. For example: for repairing the disease of 'settlement', the redundancy variable of the repair cost per kilometer of the input index is 0, and the input redundancy rate is also 0. This means that it is not necessary to improve the efficiency of the sink repair from the viewpoint of reducing the investment.
The underyield analysis is based on relaxation variables and underyield rates. The underyield is a ratio of the relaxation variable to the throughput, and a larger value means more "underyield", and if null, it means that the original "underyield" data is 0. The result can provide a target for improving the repair efficiency of various diseases. For example: for repairing the crack, the relaxation variable of the output index PCI is 0.105, the relaxation variable of the output index RQI and RDI is 0, and correspondingly, the output deficiency rate of the output index PCI is 0.170, and the output deficiency rate of the output index RQI and RDI is 0. This indicates that for improving crack repair, the PCI value of the road surface after crack repair needs to be increased by 0.105 to reach the optimum input-output state.
The above embodiments are preferred embodiments of the present application, and those skilled in the art can make various changes or modifications without departing from the general concept of the present application, and such changes or modifications should fall within the scope of the claims of the present application.
Claims (9)
1. A DEA-based method for evaluating highway pavement maintenance measures is characterized by comprising the following steps:
step 1, acquiring historical pavement quality conditions and corresponding historical pavement maintenance conditions of a highway to be researched, and extracting historical disease data of each highway section, maintenance measures and maintenance cost of each damaged highway section and pavement performance indexes before and after maintenance from the historical pavement quality conditions and the corresponding historical pavement maintenance conditions; forming a data sample by each historical disease, the corresponding maintenance measure and the pavement performance indexes before and after maintenance;
step 2, performing noise filtering processing on all data samples by adopting a K neighbor algorithm according to a disease data set formed by all disease data and a maintenance and repair measure data set formed by corresponding maintenance and repair measures;
step 3, counting all the filtered data samples, and calculating the unit maintenance cost and the pavement performance index variation of the maintenance measures corresponding to each disease based on all the data samples of each disease;
step 4, establishing a DEA evaluation model of the pavement maintenance measures of the expressway to be researched by taking the maintenance and maintenance measures corresponding to each disease as a label and taking the unit maintenance cost and the pavement performance index variation as input and output respectively;
and 5, evaluating the benefit index of each disease maintenance and repair measure of the expressway to be researched based on the DEA evaluation model, and giving a maintenance decision suggestion.
2. The method of evaluating a measure for maintaining a pavement of a highway according to claim 1, wherein the pavement performance indicators include a pavement damage status index PCI, a pavement quality of travel index RQI, and a pavement rutting depth index RDI.
3. The method of evaluating a maintenance measure for a highway pavement according to claim 1, wherein the historical road repair situation includes historical major repair, intermediate repair and minor repair records; and the historical major repair, medium repair and minor repair records comprise maintenance measures and maintenance costs of all the diseases of the expressway to be researched and the starting point pile number and the stop point pile number of all the road sections included by the diseases.
4. The method for evaluating the maintenance measures for the highway pavement according to claim 1, wherein the method for filtering the noise of all the data samples by adopting the K-nearest neighbor algorithm comprises the following steps:
averagely dividing a disease data set into two subsets, wherein the number of each disease in the two subsets is the same;
clustering the first disease subset by adopting a K nearest neighbor algorithm and based on an Euclidean distance to obtain an initial disease classification model;
classifying the disease data in the second disease subset by using an initial disease classification model; if the classification output label of a certain disease data is not consistent with the actual disease label, the disease data is discarded from the second disease subset; meanwhile, discarding the data sample corresponding to the disease data;
clustering the current second disease subset by adopting a K nearest neighbor algorithm and based on the Euclidean distance to obtain an improved disease classification model;
classifying the disease data in the first subset of diseases using an improved disease classification model; if the classification output label of certain disease data is not consistent with the actual disease label, the disease data is discarded from the first disease subset; meanwhile, discarding the data sample corresponding to the disease data;
and performing noise filtering treatment on the maintenance and repair measure data set and the data sample by adopting a filtering method the same as that of the disease data set.
5. The method for evaluating pavement maintenance measures for expressway according to claim 1, wherein the step 5 of evaluating the benefit index of each disease maintenance measure comprises: technical benefits, scale benefits, comprehensive benefits, relaxation variables, redundancy variables, and scale reward coefficients; the technical benefit reflects an efficiency situation brought by technical factors, the scale benefit reflects an efficiency situation brought by input scale, the comprehensive benefit reflects a comprehensive efficiency situation brought by technical factors and input scale, the redundancy variable reflects how much input is reduced to achieve the target efficiency, and the relaxation variable reflects how much output is increased to achieve the target efficiency; the consideration scale factor reflects how many units of input need to be increased for each unit of output increase.
6. The method for evaluating measures for maintaining a pavement of an expressway according to claim 5, wherein the effectiveness evaluation of the DEA evaluation model is performed on the basis of each benefit index, and the effectiveness evaluation results include 3 types, namely DEA strong effective, DEA weak effective and non-DEA effective; the method for obtaining the DEA effectiveness evaluation result from the benefits comprises the following steps: if the comprehensive benefit is 1 and the redundancy variable and the relaxation variable are 0, DEA is strong and effective; if the comprehensive benefit is 1 and the redundancy variable or the relaxation variable is more than 0, DEA is weak and effective; if the combined benefit is less than 1, then non-DEA is effective.
7. The method of claim 5, wherein a scale consideration analysis, a investment redundancy analysis, and an output deficiency analysis are further performed on various disease maintenance measures on the basis of each benefit index.
8. The utility model provides a highway road surface maintenance measure evaluation device based on DEA which characterized in that includes:
a sample acquisition module to: obtaining historical pavement quality conditions and corresponding historical pavement maintenance conditions of an expressway to be researched, and extracting historical disease data of each section of the expressway, maintenance measures and maintenance cost of each damaged section and pavement performance indexes before and after maintenance from the historical pavement quality conditions and the corresponding historical pavement maintenance conditions; forming a data sample by each historical disease, the corresponding maintenance measure and the pavement performance indexes before and after maintenance;
a noise filtering processing module for: performing noise filtering processing on all data samples by adopting a K neighbor algorithm according to a disease data set formed by all disease data and a maintenance and repair measure data set formed by corresponding maintenance and repair measures;
a data processing module to: counting all the filtered data samples, and calculating the unit maintenance cost and the pavement performance index change amount of maintenance and repair measures corresponding to each disease based on all the data samples of each disease;
a DEA evaluation model construction module for: taking maintenance measures corresponding to each disease as a label, taking unit maintenance cost and pavement performance index variation as input and output respectively, and establishing a DEA evaluation model of the pavement maintenance measures of the expressway to be researched;
the index evaluation and suggestion module is used for: and evaluating the benefit index of each disease maintenance and repair measure of the expressway to be researched based on the DEA evaluation model, and providing a maintenance decision suggestion.
9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program, wherein the computer program, when executed by the processor, causes the processor to implement the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211199571.3A CN115271565B (en) | 2022-09-29 | 2022-09-29 | DEA-based method, device and equipment for evaluating highway pavement maintenance measures |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211199571.3A CN115271565B (en) | 2022-09-29 | 2022-09-29 | DEA-based method, device and equipment for evaluating highway pavement maintenance measures |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115271565A true CN115271565A (en) | 2022-11-01 |
| CN115271565B CN115271565B (en) | 2023-01-24 |
Family
ID=83757485
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211199571.3A Active CN115271565B (en) | 2022-09-29 | 2022-09-29 | DEA-based method, device and equipment for evaluating highway pavement maintenance measures |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115271565B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116151522A (en) * | 2023-04-24 | 2023-05-23 | 中南大学 | DEA-based expressway pavement maintenance auxiliary decision-making method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103745128A (en) * | 2014-01-26 | 2014-04-23 | 东南大学 | Maintenance measure decision method based on road disease causes |
| CN106600503A (en) * | 2016-11-04 | 2017-04-26 | 华南理工大学 | Data envelopment analysis-based expressway network preventive maintenance efficiency evaluation method |
| CN107609757A (en) * | 2017-08-28 | 2018-01-19 | 东南大学 | A kind of method for evaluating maintenance of surface long-term benefit |
| CN109829492A (en) * | 2019-01-22 | 2019-05-31 | 长安大学 | A kind of classification method for unbalanced data |
| CN111062583A (en) * | 2019-11-28 | 2020-04-24 | 武汉理工大学 | Asphalt pavement historical maintenance benefit quantitative evaluation method based on principal component analysis method |
| CN111160728A (en) * | 2019-12-09 | 2020-05-15 | 华南农业大学 | Road and bridge maintenance decision optimization method and device |
| CN112686396A (en) * | 2020-12-15 | 2021-04-20 | 中公高科养护科技股份有限公司 | Method, medium and system for selecting pavement maintenance property based on disease number |
| CN114722947A (en) * | 2022-04-12 | 2022-07-08 | 北京邮电大学 | Power dispatching monitoring data anomaly detection method based on neighbor search clustering |
| CN114998852A (en) * | 2021-08-05 | 2022-09-02 | 浙江杉工智能科技有限公司 | Intelligent detection method for road pavement diseases based on deep learning |
-
2022
- 2022-09-29 CN CN202211199571.3A patent/CN115271565B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103745128A (en) * | 2014-01-26 | 2014-04-23 | 东南大学 | Maintenance measure decision method based on road disease causes |
| CN106600503A (en) * | 2016-11-04 | 2017-04-26 | 华南理工大学 | Data envelopment analysis-based expressway network preventive maintenance efficiency evaluation method |
| CN107609757A (en) * | 2017-08-28 | 2018-01-19 | 东南大学 | A kind of method for evaluating maintenance of surface long-term benefit |
| CN109829492A (en) * | 2019-01-22 | 2019-05-31 | 长安大学 | A kind of classification method for unbalanced data |
| CN111062583A (en) * | 2019-11-28 | 2020-04-24 | 武汉理工大学 | Asphalt pavement historical maintenance benefit quantitative evaluation method based on principal component analysis method |
| CN111160728A (en) * | 2019-12-09 | 2020-05-15 | 华南农业大学 | Road and bridge maintenance decision optimization method and device |
| CN112686396A (en) * | 2020-12-15 | 2021-04-20 | 中公高科养护科技股份有限公司 | Method, medium and system for selecting pavement maintenance property based on disease number |
| CN114998852A (en) * | 2021-08-05 | 2022-09-02 | 浙江杉工智能科技有限公司 | Intelligent detection method for road pavement diseases based on deep learning |
| CN114722947A (en) * | 2022-04-12 | 2022-07-08 | 北京邮电大学 | Power dispatching monitoring data anomaly detection method based on neighbor search clustering |
Non-Patent Citations (2)
| Title |
|---|
| 谢振文 等: "基于空-地协同的公路边坡施工监测与策略调控探究", 《公路工程》 * |
| 马艳玲: ""基于DEA的广东省国省公路路面养护效果评价及有效的路面养护策略研究"", 《优秀硕士论文集》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116151522A (en) * | 2023-04-24 | 2023-05-23 | 中南大学 | DEA-based expressway pavement maintenance auxiliary decision-making method and system |
| CN116151522B (en) * | 2023-04-24 | 2023-07-14 | 中南大学 | DEA-based expressway pavement maintenance auxiliary decision-making method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115271565B (en) | 2023-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114596700B (en) | Real-time traffic estimation method for expressway section based on portal data | |
| CN105374209B (en) | A kind of urban area road network running status characteristics information extraction method | |
| CN108665093B (en) | Deep learning-based expressway traffic accident severity prediction method | |
| CN114049765B (en) | Urban road network traffic flow OD estimation method based on automatic vehicle number plate identification data | |
| CN110807919A (en) | Urban road network traffic operation situation evaluation method based on vehicle passing data | |
| CN115271565B (en) | DEA-based method, device and equipment for evaluating highway pavement maintenance measures | |
| CN114048436A (en) | Construction method and construction device for forecasting enterprise financial data model | |
| CN101794345A (en) | Data processing method and device | |
| CN110738523A (en) | maintenance order quantity prediction method and device | |
| CN111784022A (en) | Short-time adjacent fog prediction method based on combination of Wrapper method and SVM method | |
| CN111640304A (en) | Automatic quantitative extraction method for traffic jam propagation characteristics of continuous flow traffic facility | |
| CN111786385A (en) | Power grid operation and maintenance scheme planning method, system and equipment | |
| CN104751369A (en) | Improved line loss analysis method for K-means clustering | |
| CN116468160A (en) | Aluminum alloy die casting quality prediction method based on production big data | |
| CN112116810A (en) | Whole road network segment travel time estimation method based on urban road checkpoint data | |
| CN108710967B (en) | Expressway traffic accident severity prediction method based on data fusion and support vector machine | |
| CN112562311B (en) | Method and device for obtaining working condition weight factor based on GIS big data | |
| CN111145535B (en) | Travel time reliability distribution prediction method under complex scene | |
| CN112668967A (en) | Cold-chain logistics tracking and tracing processing method based on big data analysis and big data cloud platform | |
| CN107590747A (en) | Power grid asset turnover rate computational methods based on the analysis of comprehensive energy big data | |
| CN113362597B (en) | Traffic sequence data anomaly detection method and system based on non-parametric modeling | |
| CN111222678B (en) | Road surface technical condition prediction method | |
| CN116151522B (en) | DEA-based expressway pavement maintenance auxiliary decision-making method and system | |
| MATSUMOTO et al. | Modeling an optimal track maintenance schedule in consideration of timing of grinding and tamping | |
| CN114013285B (en) | Actual driving range evaluation method for electric automobile |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |