Low-efficiency industrial land data identification method

Abstract

The invention discloses a low-efficiency industrial land data identification method, and relates to the technical field of land resource management, and the method comprises the steps: collecting the business income, tax, power consumption, employees, land area, building area and other basic information of an industrial land through on-site treading, and calculating the land average evaluation index of each land block. A natural discontinuity point grading method is adopted to carry out grading assignment on each index data, and an analytic hierarchy process is adopted to determine the evaluation weight of each index. And calculating the efficiency comprehensive score of each land parcel in combination with the grading assignment and the evaluation weight, and judging whether the land parcel is a low-efficiency industrial land or not through a preset score threshold value. And finally, performing coupling verification on a model judgment result and a preliminary result of field exploration of a professional, and establishing a final low-efficiency industrial land database after taking a union set. According to the method, the quantitative model and field verification are integrated, and effective support is provided for fine management and efficiency improvement of land resources.

Description

Technical Field

[0001] This invention relates to the field of land resource management technology, specifically a method for identifying inefficient industrial land data. Background Technology

[0002] With the advancement of my country's industrialization, urbanization, and modernization, "seeking efficiency from existing land" has become a key focus of urban construction. A series of policy documents issued by national ministries and commissions on the redevelopment of inefficient land have made the redevelopment of inefficient industrial land an important tool for promoting the economical and intensive use of land. The accurate identification of inefficient industrial land has become the prerequisite and key to carrying out this work.

[0003] Currently, the identification of inefficient industrial land mainly relies on manual surveys, experience-based judgments, and single-indicator assessments, lacking a systematic and scientific identification method. While commonly used assessment indicators in existing technologies, such as tax revenue per unit area and operating income per unit area, can reflect some land use efficiency, they often overlook the synergistic effects and weight differences among multiple indicators, resulting in one-sided and highly subjective assessment results that fail to comprehensively and objectively reflect the actual utilization efficiency of industrial land.

[0004] Existing methods for classifying inefficient industrial land use often employ equal-interval or equal-frequency classification, failing to fully consider the inherent patterns in data distribution and easily leading to problems such as unreasonable classification and low identification accuracy.

[0005] Therefore, there is an urgent need for a method to identify inefficient industrial land that can integrate multi-source data, scientifically classify and reasonably assign weights, and has verifiability, so as to improve the identification accuracy and decision support capabilities, and provide technical support for the optimal allocation of land resources and industrial upgrading. Summary of the Invention

[0006] The purpose of this invention is to provide a method for identifying inefficient industrial land use data, so as to solve the problems raised in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for identifying inefficient industrial land use data, the method comprising the following steps: Step S100: Conduct on-site investigations and visits to industrial zones, communicate face-to-face with enterprises and management departments to obtain first-hand information, collect basic information on industrial land, conduct investigations on inefficient industrial land, establish a preliminary database of inefficient industrial land, and calculate the per-land assessment index for each plot. In step S100, the basic information on industrial land includes the enterprise's operating revenue R, tax revenue T, electricity consumption E, number of employees P, land area A, and building area S for each industrial plot. In the preliminary screening, professionals investigate inefficient industrial land and collect information on land for industries that are eliminated or prohibited by national industrial policies and land that does not meet safety and environmental protection requirements, forming a preliminary set of inefficient industrial land. This not only provides the numbers in the table but also gives a more intuitive understanding of the land use situation. Calculate the following evaluation indicators per unit area: Revenue per unit area I1 = R / A, where I1 represents revenue per unit area of ​​land; Tax revenue per unit area I2 = T / A, where I2 represents tax contribution per unit area of ​​land; Electricity consumption per unit area I3 = E / A, where I3 represents electricity consumption per unit area of ​​land; Employers per unit area I4 = P / A, where I4 represents the number of employees per unit area of ​​land; Building area per unit area I5 = S / A, where I5 ​​represents building area per unit area of ​​land.

[0008] Step S200: Use a grading algorithm to assign grading values ​​to the obtained local average evaluation indicators. Divide the continuous values ​​of each indicator into different levels with significant differences according to their inherent distribution patterns. This makes the performance of different plots on the same indicator clear and comparable, paving the way for subsequent comprehensive evaluation. The hierarchical algorithm used in step S200 is the natural discontinuity algorithm, specifically as follows: For the five land area assessment indicators, each is classified using the natural discontinuity algorithm. A dataset X={x1,x2,...,x} containing m industrial land parcels is set up for indicator I. m},x1,x2,...,x m The values ​​of industrial plots X on the per-land evaluation index I represent the values ​​of the 1st, 2nd, ..., mth industrial plots. The dataset X is divided into k consecutive levels, where k is an integer between 3 and 10. The natural discontinuity algorithm automatically finds inherent, natural breakpoints and grouping patterns in the data to ensure that plots within each level are as similar as possible. Calculate the total sum of squared deviations (SDAM) of dataset X, specifically as follows: ; In the formula, Let x be the arithmetic mean of dataset X. i This represents the value of the i-th industrial plot in the per-land assessment index I; For the current classification, the SDAM is used to measure the dispersion of the internal data, and the within-group sum of squared deviations (SDCM) is calculated as follows: ; In the formula, m j Let x be the number of industrial plots included in the j-th level. ji Let be the average assessment index value per unit area for the i-th plot in the j-th grade. This is the arithmetic mean of all indicator values ​​within the j-th level; Calculate the variance-goodness-of-fit (GVF) of the current grading scheme. The closer the GVF is to 1, the smaller the within-group variance (SDCM) after grouping is compared to the total variance (SDAM), indicating better grouping, greater similarity within groups, and greater differences between groups. Specifically: ; In the formula, the variance fit goodness-of-fit GVF traverses all grading schemes from k to 10, selects the optimal grading number k0 that maximizes the GVF value, and divides the value of the current industrial land parcel's per-unit evaluation index into k0 grading schemes based on the optimal grading number k0, with corresponding grading values ​​assigned from 1 to k0.

[0009] Step S300: Use a weighted calculation method to determine the evaluation weights of the evaluation indicators for each region, ensuring that the final evaluation results meet the actual development needs and value orientations of the local area; The weighting method used in step S300 is the Analytic Hierarchy Process (AHP), which uses matrix operations to qualitatively compare the importance of the land area assessment indicators. Specifically: Construct the judgment matrix A=[a ij ] n×n In the formula, n represents the number of average assessment indicators for the current land parcel, and a ij The importance scale value of the i-th land area assessment indicator relative to the j-th land area assessment indicator is set by professionals. The judgment matrix A is column-normalized to obtain matrix A'. The sum of each row element of matrix A' is calculated to obtain the row sum vector V. The i-th component of the row sum vector V... Normalize V to obtain the weight vector W, where the i-th component of the weight vector W is... The W i The evaluation weight of the i-th average evaluation index for the current land parcel; In step S300, after obtaining the weight vector, a large number of pairwise comparisons may occur, which may lead to logical contradictions. These contradictions are identified and quantified through consistency checks, specifically as follows: Calculate the consistency index CI: CI = (λ) max -n) / (n-1), in the formula, λ max It is the largest eigenvalue of the judgment matrix; Calculate the consistency ratio CR: CR = CI / RI, where RI is the average random consistency index corresponding to the matrix order n; When CR≤a, the consistency of the judgment matrix is ​​acceptable. When CR>a, it indicates that there is a significant logical inconsistency in the current judgment, and the consistency of the judgment matrix is ​​unacceptable. The system issues a warning, prompting experts to re-examine and adjust their judgment matrix until it passes the consistency test, thus ensuring the scientific nature and logical rigor of the weighting results. Here, 'a' is the consistency threshold.

[0010] Step S400: Combine the results obtained from the first two steps and calculate the comprehensive efficiency score of the current land parcel based on the graded assignment and the evaluation weight of the local evaluation indicators, so as to objectively reflect the comprehensive efficiency level of its land use. In step S400, the overall performance score of the current land parcel is calculated as follows: ; In the formula, Score represents the overall performance score of the current industrial site, and W... i G represents the weight of the i-th evaluation indicator. i This represents the grade assignment of the current industrial land parcel on the i-th evaluation indicator.

[0011] Step S500: Determine inefficient industrial land use based on the comprehensive performance score; The specific steps for determining inefficient industrial land in step S500 are as follows: A performance score threshold b is set by professionals. When the score is greater than or equal to b, the land is considered to be inefficiently used.

[0012] Step S600: Couple and verify the industrial plots identified as inefficient industrial land with the results of the site survey, establish an inefficient industrial land database, and form a mechanism of human-machine combination and mutual verification.

[0013] The coupling verification in step S600 specifically involves: Let M be the set of inefficient industrial land identified in step S500, and F be the set of inefficient industrial land discovered in step S100 through field visits. Take the union of set M and set F to obtain the inefficient industrial land database D: D=M∪F.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. An inefficient industrial land identification model based on multivariate data analysis is established, which is highly operable and replicable. This invention is based on enterprise operating income, tax revenue, electricity consumption, number of employees, land use and building area data. The selected data types are closely related to the enterprise's operating status and construction status, making it highly replicable and easy to promote.

[0015] 2. By coupling and verifying the model recognition results with the field survey results, and mutually confirming and supplementing each other, the final database has both theoretical scientific validity and practical conformity, providing highly reliable data support for land management.

[0016] 3. High efficiency in comprehensive evaluation: This invention integrates complex multi-indicator information into a quantifiable comprehensive performance score, making the performance evaluation and ranking of a large number of industrial lands simple and efficient, and improving the efficiency of identification work. Attached Figure Description

[0017] Figure 1 This is a schematic diagram illustrating the steps of applying the present invention to a method for identifying inefficient industrial land data; Figure 2 This is a schematic diagram of the structure of the present invention applied to a method for identifying inefficient industrial land data. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example: Figures 1-2 As shown, the present invention provides a technical solution. Given an industrial zone containing multiple industrial plots, this method is used to systematically evaluate and identify inefficient industrial land. The method includes the following steps: Step S100: Conduct on-site investigations and visits to industrial zones, communicate face-to-face with enterprises and management departments to obtain first-hand information, collect basic information on industrial land, conduct investigations on inefficient industrial land, establish a preliminary database of inefficient industrial land, and calculate the per-land assessment index for each plot. In step S100, the basic information on industrial land includes the enterprise's operating revenue R, tax revenue T, electricity consumption E, number of employees P, land area A, and building area S for each industrial plot. In the preliminary screening, professionals investigate inefficient industrial land and collect information on land for industries that are eliminated or prohibited by national industrial policies and land that does not meet safety and environmental protection requirements, forming a preliminary set of inefficient industrial land. This not only provides the numbers in the table but also gives a more intuitive understanding of the land use situation. Calculate the following evaluation indicators per unit area: Revenue per unit area I1 = R / A, where I1 represents revenue per unit area of ​​land; Tax revenue per unit area I2 = T / A, where I2 represents tax contribution per unit area of ​​land; Electricity consumption per unit area I3 = E / A, where I3 represents electricity consumption per unit area of ​​land; Employers per unit area I4 = P / A, where I4 represents the number of employees per unit area of ​​land; Building area per unit area I5 = S / A, where I5 ​​represents building area per unit area of ​​land.

[0020] Example 1: Professional staff conducted a site visit to the industrial zone and collected basic information on 5 industrial plots (numbered G1 to G5), including enterprise operating income (R), enterprise tax revenue (T), enterprise electricity consumption (E), number of employees (P), enterprise land area (A), and enterprise building area (S). The preset data is as follows: Plot G1: R=5 million yuan, T=500,000 yuan, E=100,000 degrees, P=100 people, A=100 mu, S=80 mu; Plot G2: R=3 million yuan, T=300,000 yuan, E=80,000 degrees, P=80 people, A=120 mu, S=90 mu; Plot G3: R=8 million yuan, T=800,000 yuan, E=150,000 degrees, P=150 people, A=150 mu, S=120 mu; Plot G4: R=2 million yuan, T=200,000 yuan, E=50,000 degrees, P=60 people, A=80 mu, S=60 mu; Plot G5: R=6 million yuan, T=600,000 yuan, E=120,000 degrees, P=120 people, A=130 mu, S=100 mu; Based on the above data, calculate the per-plot evaluation index for each plot: G1: I1 = 50,000 yuan / mu, I2 = 5,000 yuan / mu, I3 = 1,000 kWh / mu, I4 = 1 person / mu, I5 = 0.8; G2: I1 = 25,000 yuan / mu, I2 = 2,500 yuan / mu, I3 = 6700 kWh / mu, I4 = 670 people / mu, I5 = 7500 yuan / mu; G3: I1 = 53,300 yuan / mu, I2 = 5,300 yuan / mu, I3 = 1,000 kWh / mu, I4 = 1.0 person / mu, I5 = 0.8; G4: I1 = 25,000 yuan / mu, I2 = 2,500 yuan / mu, I3 = 6,250 yuan / mu, I4 = 7,500 people / mu, I5 = 7,500 people / mu; G5: I1 = 46,200 yuan / mu, I2 = 4,600 yuan / mu, I3 = 9,200 yuan / mu, I4 = 9,200 people / mu, I5 = 7,700 yuan / mu; Meanwhile, through site visits, professionals have preliminarily identified potentially inefficient plots of land, forming a preliminary set of inefficient industrial land F (G2, G4).

[0021] Step S200: Use a grading algorithm to assign grading values ​​to the obtained local average evaluation indicators. Divide the continuous values ​​of each indicator into different levels with significant differences according to their inherent distribution patterns. This makes the performance of different plots on the same indicator clear and comparable, paving the way for subsequent comprehensive evaluation. The hierarchical algorithm used in step S200 is the natural discontinuity algorithm, specifically as follows: For the five land area assessment indicators, each is classified using the natural discontinuity algorithm. A dataset X={x1,x2,...,x} containing m industrial land parcels is set up for indicator I. m},x1,x2,...,x m The values ​​of industrial plots X on the per-land evaluation index I represent the values ​​of the 1st, 2nd, ..., mth industrial plots. The dataset X is divided into k consecutive levels, where k is an integer between 3 and 10. The natural discontinuity algorithm automatically finds inherent, natural breakpoints and grouping patterns in the data to ensure that plots within each level are as similar as possible. Calculate the total sum of squared deviations (SDAM) of dataset X, specifically as follows: ; In the formula, Let x be the arithmetic mean of dataset X. i This represents the value of the i-th industrial plot in the per-land assessment index I; For the current classification, the SDAM is used to measure the dispersion of the internal data, and the within-group sum of squared deviations (SDCM) is calculated as follows: ; In the formula, m j Let x be the number of industrial plots included in the j-th level. ji Let be the average assessment index value per unit area for the i-th plot in the j-th grade. This is the arithmetic mean of all indicator values ​​within the j-th level; Calculate the variance-goodness-of-fit (GVF) of the current grading scheme. The closer the GVF is to 1, the smaller the within-group variance (SDCM) after grouping is compared to the total variance (SDAM), indicating better grouping, greater similarity within groups, and greater differences between groups. Specifically: ; In the formula, the variance fit goodness-of-fit GVF traverses all grading schemes from k to 10, selects the optimal grading number k0 that maximizes the GVF value, and divides the value of the current industrial land parcel's per-unit evaluation index into k0 grading schemes based on the optimal grading number k0, with corresponding grading values ​​assigned from 1 to k0.

[0022] Example 2: The Natural Discontinuity (GVF) algorithm is used to classify each land area evaluation indicator. Taking the land area revenue I1 as an example, the dataset X={5, 2.5, 5.33, 2.5, 4.62} is composed of the I1 values ​​of 5 plots. The algorithm iterates through the classification schemes from k=3 to 10 and calculates the variance goodness of fit for each scheme. It is set that the GVF is maximized when k0=3, so I1 is divided into 3 levels: values ​​≤2.5 are assigned a value of 1, values ​​between 2.5 and 4.62 are assigned a value of 2, and values ​​≥4.62 are assigned a value of 3. The same classification is applied to other indicators. I1 hierarchical assignment: G1=3, G2=1, G3=3, G4=1, G5=3; I2 hierarchical assignment: Dataset {0.5, 0.25, 0.53, 0.25, 0.46}, k0=3, G1=3, G2=1, G3=3, G4=1, G5=2; I3 hierarchical assignment: Dataset {0.1, 0.067, 0.1, 0.0625, 0.092}, k0=2, G1=2, G2=1, G3=2, G4=1, G5=2; I4 hierarchical assignment: Dataset {1.0, 0.67, 1.0, 0.75, 0.92}, k0=2, G1=2, G2=1, G3=2, G4=1, G5=2; I5: Dataset {0.8, 0.75, 0.8, 0.75, 0.77}, k0=3, G1=3, G2=2, G 3= 3, G4=2, G5=2.

[0023] Step S300: Use a weighted calculation method to determine the evaluation weights of the evaluation indicators for each region, ensuring that the final evaluation results meet the actual development needs and value orientations of the local area; The weighting method used in step S300 is the Analytic Hierarchy Process (AHP), which uses matrix operations to qualitatively compare the importance of the land area assessment indicators. Specifically: Construct the judgment matrix A=[a ij ] n×n In the formula, n represents the number of average assessment indicators for the current land parcel, and a ij The importance scale value of the i-th land area assessment indicator relative to the j-th land area assessment indicator is set by professionals. The judgment matrix A is column-normalized to obtain matrix A'. The sum of each row element of matrix A' is calculated to obtain the row sum vector V. The i-th component of the row sum vector V... Normalize V to obtain the weight vector W, where the i-th component of the weight vector W is... The W iThe evaluation weight of the i-th average evaluation index for the current land parcel; Example 3: The weights of the evaluation indicators for each region are determined using the analytic hierarchy process (AHP). The threshold a = 0.1. The relative importance scale of indicators I1 to I5 is represented by matrix A. Column normalization of A yields matrix A'. The row sum vector V is calculated and then normalized to obtain the weight vector W. The preset weights are W1 = 0.35 (I1), W2 = 0.25 (I2), W3 = 0.15 (I3), W4 = 0.15 (I4), and W5 = 0.10 (I5).

[0024] In step S300, after obtaining the weight vector, a large number of pairwise comparisons may occur, which may lead to logical contradictions. These contradictions are identified and quantified through consistency checks, specifically as follows: Calculate the consistency index CI: CI = (λ) max -n) / (n-1), in the formula, λ max It is the largest eigenvalue of the judgment matrix; Calculate the consistency ratio CR: CR = CI / RI, where RI is the average random consistency index corresponding to the matrix order n; When CR≤a, the consistency of the judgment matrix is ​​acceptable. When CR>a, it indicates that there is a significant logical inconsistency in the current judgment, and the consistency of the judgment matrix is ​​unacceptable. The system issues a warning, prompting experts to re-examine and adjust their judgment matrix until it passes the consistency test, thus ensuring the scientific nature and logical rigor of the weighting results. Here, 'a' is the consistency threshold.

[0025] Example 4: Perform a consistency check, calculate the consistency index CI and the consistency ratio CR, and set CR=0.05, which is less than the threshold a=0.1, so the consistency is acceptable and the weights are effective.

[0026] Step S400: Combine the results obtained from the first two steps and calculate the comprehensive efficiency score of the current land parcel based on the graded assignment and the evaluation weight of the local evaluation indicators, so as to objectively reflect the comprehensive efficiency level of its land use. In step S400, the overall performance score of the current land parcel is calculated as follows: ; In the formula, Score represents the overall performance score of the current industrial site, and W... i G represents the weight of the i-th evaluation indicator. i This represents the grade assignment of the current industrial land parcel on the i-th evaluation indicator.

[0027] Example 5: Based on the graded assignment and weights, the comprehensive performance score for each plot is calculated, resulting in G1: Score = 2.7, G2: Score = 1.1, G3: Score = 2.7, G4: Score = 1.1, and G5: Score = 2.35. Step S500: Determine inefficient industrial land use based on the comprehensive performance score; The specific steps for determining inefficient industrial land in step S500 are as follows: A performance score threshold b is set by professionals. When the score is greater than or equal to b, the land is considered to be inefficiently used.

[0028] Example 6: Set the efficiency score threshold b=2.5. The scores of plots G2, G4 and G5 are all below 2.5, and they are judged as inefficient industrial land, forming set M (G2, G4, G5).

[0029] Step S600: Couple and verify the industrial plots identified as inefficient industrial land with the results of the site survey, establish an inefficient industrial land database, and form a mechanism of human-machine combination and mutual verification.

[0030] The coupling verification in step S600 specifically involves: Let M be the set of inefficient industrial land identified in step S500, and F be the set of inefficient industrial land discovered in step S100 through field visits. Take the union of set M and set F to obtain the inefficient industrial land database D: D=M∪F.

[0031] Example 7: The union of the identified set of inefficient industrial land M (G2, G4, G5) and the preliminary set of inefficient land obtained from field surveys F (G2, G4) is taken to obtain the final database of inefficient industrial land D. The union D contains G2, G4, and G5.

[0032] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for identifying inefficient industrial land use data, characterized in that: The method includes the following steps: Step S100: Conduct site visits to industrial zones, collect basic information on industrial land, conduct preliminary screening of inefficient industrial land, establish a preliminary database of inefficient industrial land, and calculate the per-plot evaluation index for each plot. Step S200: Assign values ​​to the obtained regional average evaluation indicators using a tiered algorithm; Step S300: Determine the evaluation weights of the evaluation indicators for each region using a weighting calculation method; Step S400: Calculate the comprehensive performance score of the current land parcel based on the graded assignment and the evaluation weights of the evaluation indicators for each region; Step S500: Determine inefficient industrial land use based on the comprehensive performance score; Step S600: Couple and verify the industrial plots identified as inefficient industrial land with the results of the site survey to establish an inefficient industrial land database.

2. The method for identifying inefficient industrial land use data according to claim 1, characterized in that, In step S100, the basic information of industrial land includes the enterprise's operating income R, enterprise tax T, enterprise electricity consumption E, number of employees P, enterprise land area A, and enterprise building area S for each industrial plot. In the initial screening, professionals investigate inefficient industrial land, collect information on land for industries that are to be phased out or prohibited by national industrial policies, and land that does not meet safety and environmental protection requirements, and form a preliminary set of inefficient industrial land. Calculate the following evaluation indicators per unit area: Revenue per unit area I1 = R / A, where I1 represents revenue per unit area of ​​land; Tax revenue per unit area I2 = T / A, where I2 represents tax contribution per unit area of ​​land; Electricity consumption per unit area I3 = E / A, where I3 represents electricity consumption per unit area of ​​land; Employers per unit area I4 = P / A, where I4 represents the number of employees per unit area of ​​land; Building area per unit area I5 = S / A, where I5 ​​represents building area per unit area of ​​land.

3. The method for identifying inefficient industrial land use data according to claim 1, characterized in that, The hierarchical algorithm used in step S200 is the natural discontinuity algorithm, specifically as follows: For the five land area assessment indicators, each is classified using the natural discontinuity algorithm. A dataset X={x1,x2,...,x} containing m industrial land parcels is set up on indicator I. m },x1,x2,...,x m This represents the values ​​of the 1st, 2nd, ..., mth industrial plots on the per-land assessment index I. The dataset X is divided into k consecutive levels, where k is an integer between 3 and 10. The total sum of squared deviations (SDAM) of dataset X is calculated as follows: ； In the formula, Let x be the arithmetic mean of dataset X. i This represents the value of the i-th industrial plot in the per-land assessment index I; For the current classification, the sum of squared within-group deviations (SDCM) is calculated as follows: ； In the formula, m j Let x be the number of industrial plots included in the j-th level. ji Let be the average assessment index value per unit area for the i-th plot in the j-th grade. This is the arithmetic mean of all indicator values ​​within the j-th level; The variance goodness of fit (GVF) of the current grading scheme is calculated as follows: ； In the formula, the variance fit goodness-of-fit GVF traverses all grading schemes from k to 10, selects the optimal grading number k0 that maximizes the GVF value, and divides the value of the current industrial land parcel's per-unit evaluation index into k0 grading schemes based on the optimal grading number k0, with corresponding grading values ​​assigned from 1 to k0.

4. The method for identifying inefficient industrial land use data according to claim 1, characterized in that, The weight calculation method used in step S300 is the Analytic Hierarchy Process (AHP), specifically: Construct the judgment matrix A=[a ij ] n×n In the formula, n represents the number of average assessment indicators for the current land parcel, and a ij The importance scale value of the i-th land area assessment indicator relative to the j-th land area assessment indicator is set by professionals. The judgment matrix A is column-normalized to obtain matrix A'. The sum of each row element of matrix A' is calculated to obtain the row sum vector V. The i-th component of the row sum vector V... Normalize V to obtain the weight vector W, where the i-th component of the weight vector W is... The W i This represents the evaluation weight of the i-th average evaluation indicator for the current land parcel.

5. The method for identifying inefficient industrial land use data according to claim 4, characterized in that, In step S300, after obtaining the weight vector, a consistency check is performed, specifically as follows: Calculate the consistency index CI: CI = (λ) max -n) / (n-1), in the formula, λ max It is the largest eigenvalue of the judgment matrix; Calculate the consistency ratio CR: CR = CI / RI, where RI is the average random consistency index corresponding to the matrix order n; When CR≤a, the consistency of the judgment matrix is ​​acceptable; when CR>a, the consistency of the judgment matrix is ​​unacceptable, and the weights are manually evaluated by professionals. Here, a is the consistency threshold.

6. The method for identifying inefficient industrial land use data according to claim 1, characterized in that, In step S400, the overall performance score of the current land parcel is calculated, specifically as follows: ； In the formula, Score represents the overall performance score of the current industrial site, and W... i G represents the weight of the i-th evaluation indicator. i This indicates the grade assignment of the current industrial land parcel on the i-th evaluation indicator.

7. The method for identifying inefficient industrial land use data according to claim 1, characterized in that, The specific steps for determining inefficient industrial land in step S500 are as follows: A performance score threshold b is set by professionals. When the score is greater than or equal to b, the land is considered to be inefficiently used.

8. The method for identifying inefficient industrial land use data according to claim 1, characterized in that, The coupling verification in step S600 specifically involves: Let M be the set of inefficient industrial land identified in step S500, and F be the set of inefficient industrial land discovered in step S100 through field visits. Take the union of set M and set F to obtain the inefficient industrial land database D: D=M∪F.

Images