A performance evaluation method based on improved analytic hierarchy process

Abstract

The application provides a performance evaluation method based on improved analytic hierarchy process, and relates to the technical field of project performance evaluation.The method comprises the following steps: constructing a performance evaluation index system comprising n-level index layers; comparing each level index of the performance evaluation index system to obtain a comparison result, and constructing multiple judgment matrices according to the comparison result; integrating and processing the multiple judgment matrices to form a comprehensive judgment matrix; calculating a consistency ratio through the characteristic vector and the maximum eigenvalue of the comprehensive judgment matrix; judging the consistency ratio to obtain a judgment result; and determining a final performance evaluation index system according to the judgment result and the weight value of each index at each level.The application ensures the rationality and consistency of the performance evaluation index system by constructing the comprehensive judgment matrix and calculating the consistency ratio, and improves the accuracy and reliability of the performance evaluation.

Description

Technical Field

[0001] This invention relates to the field of project performance evaluation technology, and in particular to a performance evaluation method based on an improved analytic hierarchy process. Background Technology

[0002] As a decision-making method, the Analytic Hierarchy Process (AHP) can treat the research object as a system and make decisions according to the thinking methods of decomposition, comparison, judgment, and synthesis. It combines qualitative and quantitative methods and has become the preferred choice for determining the weight of evaluation indicators.

[0003] However, the traditional 1-9 scale method falls short when dealing with a large number of evaluation indicators. When there are numerous evaluation factors with complex relationships, it is difficult to accurately and consistently provide a comparison of the importance of each pair of factors. This makes the construction of the judgment matrix susceptible to subjective influence, reducing the objectivity and accuracy of the evaluation results. When constructing the judgment matrix, it is required that the matrix meet the consistency condition, that is, the judgment results should be logically self-consistent. However, in practice, due to the subjectivity and complexity of judgments, it is difficult for the judgment matrix to achieve complete consistency, which affects the rationality and scientific nature of the weight allocation.

[0004] The Analytic Hierarchy Process (AHP) involves numerous matrix operations, including finding eigenvectors, calculating the largest eigenvalue, and performing consistency checks. These calculations are relatively cumbersome, especially when dealing with large-scale evaluation systems, increasing computational and time costs and limiting the method's practical application efficiency. AHP is also sensitive to changes in input data; even small changes in the judgment matrix can alter the final weight allocation, reducing the stability and reliability of the evaluation results. Although AHP attempts to combine qualitative and quantitative information, accurately quantifying qualitative judgments into specific numerical values ​​remains a challenge in practice. This can lead to information loss or distortion during the evaluation process, affecting the accuracy of the evaluation results. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a performance evaluation method based on an improved analytic hierarchy process, which improves the scientificity and accuracy of performance evaluation.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0007] Firstly, a performance evaluation method based on an improved analytic hierarchy process (AHP), the method comprising:

[0008] Construct a performance evaluation indicator system that includes n levels of indicators;

[0009] The indicators at each level of the performance evaluation indicator system are compared to obtain the comparison results. Based on the comparison results, multiple judgment matrices are constructed.

[0010] Multiple judgment matrices are integrated to form a comprehensive judgment matrix;

[0011] The consistency ratio is calculated by comprehensively judging the eigenvectors and the largest eigenvalue of the matrix.

[0012] The consistency ratio is evaluated to obtain the evaluation result;

[0013] Based on the judgment results and the weight values ​​of each indicator at each level, the final performance evaluation indicator system is determined.

[0014] Furthermore, n≥2 for the n-level indicator layer, and the indicator names for each level of the indicator layer have been determined.

[0015] Furthermore, the indicators at each level of the performance evaluation indicator system are compared to obtain comparison results. Based on these results, multiple judgment matrices are constructed, including:

[0016] In each level of the performance evaluation indicator system, two indicators are selected to form a pair for comparison;

[0017] For each pair of selected indicators, a corresponding scale value is given according to the preset comparison standard to obtain the comparison result of each pair of indicators, including the indicator pair being compared and the scale value information.

[0018] Based on the comparison results of each pair of indicators, multiple judgment matrices are constructed.

[0019] Furthermore, the multiple judgment matrices are integrated to form a comprehensive judgment matrix, including:

[0020] Each judgment matrix is ​​standardized to ensure that the comparison scale values ​​are within a uniform range, resulting in a standardized set of judgment matrices.

[0021] Calculate the position of each comparison scale value in the standardized matrix;

[0022] Calculate the trust weight based on the standardized location;

[0023] For each pair of indicators, a weighted average is calculated using the confidence weight and the original comparison scale value to obtain the final comparison scale value between each pair of indicators;

[0024] The final comparison scale values ​​are imported into the matrix in the order of the index pairs to construct a comprehensive judgment matrix.

[0025] Furthermore, the formula for calculating the final comparison scale value between each pair of indicators is as follows:

[0026]

[0027] Among them, W fThis represents the final comparison scale value between each pair of indicators; m represents the number of comparison scale values; w i R represents the trust weight of the i-th degree; i Represents the trust weight of the i-th degree; α represents the adjustment coefficient; R j This represents the trust weight of the j-th element; i and j represent index variables.

[0028] Furthermore, the formula for calculating the consistency ratio is as follows:

[0029]

[0030] Where CR represents the consistency ratio; λ max represents the largest eigenvalue; n represents the order of the judgment matrix; RI represents the average random consistency index.

[0031] Furthermore, the consistency ratio is judged to obtain the judgment result, including:

[0032] Determine if the consistency ratio is less than 0.1. If the consistency ratio is less than 0.1, the judgment matrix passes the consistency test. If the consistency ratio is greater than or equal to 0.1, adjust the judgment matrix and repeat the operation of constructing the comprehensive judgment matrix and calculating the consistency ratio until the consistency test is passed and the judgment result is obtained.

[0033] Secondly, a performance evaluation system based on an improved analytic hierarchy process includes:

[0034] The acquisition module is used to construct a performance evaluation indicator system containing n levels of indicators; compare the indicators at each level of the performance evaluation indicator system to obtain comparison results; construct multiple judgment matrices based on the comparison results; and integrate the multiple judgment matrices to form a comprehensive judgment matrix.

[0035] The processing module is used to calculate the consistency ratio by comprehensively judging the eigenvectors and the largest eigenvalue of the matrix; judge the consistency ratio to obtain the judgment result; and determine the final performance evaluation index system based on the judgment result and the weight values ​​of each indicator at each level.

[0036] Thirdly, a computing device includes:

[0037] One or more processors;

[0038] A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0039] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0040] The above-described solution of the present invention has at least the following beneficial effects:

[0041] By optimizing the comparison process, such as by using more refined scaling or introducing fuzzy evaluation, the comparison of indicators at each level of the performance evaluation indicator system becomes more precise, reducing the subjectivity and arbitrariness of expert judgment and thus improving the accuracy and reliability of the judgment matrix. After constructing multiple judgment matrices, a comprehensive judgment matrix is ​​formed through scientific integration methods, such as weighted averaging and fuzzy comprehensive evaluation. This process helps eliminate inconsistencies in individual judgment matrices, making the final evaluation results more logically consistent and reasonable. The improved analytic hierarchy process may employ more efficient algorithms or simplify calculation steps to reduce computational complexity and time costs. This makes the method more feasible and convenient in practical applications, especially when dealing with large-scale evaluation systems, thereby improving evaluation efficiency.

[0042] This means that minor changes in the judgment matrix will not lead to changes in the final weight allocation, thus improving the stability and reliability of the evaluation results. The improved method employs more advanced quantification techniques to more accurately convert expert qualitative judgments into specific numerical values. This helps reduce information loss or distortion, making the evaluation results more comprehensive and accurate. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating a performance evaluation method based on an improved analytic hierarchy process, provided by an embodiment of the present invention.

[0044] Figure 2 This is a schematic diagram of a performance evaluation system based on an improved analytic hierarchy process, provided by an embodiment of the present invention. Detailed Implementation

[0045] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0046] like Figure 1 As shown, an embodiment of the present invention proposes a performance evaluation method based on an improved analytic hierarchy process (AHP), the method comprising the following steps:

[0047] Step 11: Construct a performance evaluation indicator system containing n levels of indicators;

[0048] Step 12: Compare the indicators at each level of the performance evaluation indicator system to obtain the comparison results, and construct multiple judgment matrices based on the comparison results;

[0049] Step 13: Integrate multiple judgment matrices to form a comprehensive judgment matrix;

[0050] Step 14: Calculate the consistency ratio by comprehensively judging the eigenvectors and the largest eigenvalue of the matrix;

[0051] Step 15: Judge the consistency ratio to obtain the judgment result;

[0052] Step 16: Based on the judgment results and the weight values ​​of each indicator at each level, determine the final performance evaluation indicator system.

[0053] In this embodiment of the invention, by meticulously comparing indicators at each level, the relative importance between indicators can be captured more accurately, thereby constructing a judgment matrix that is closer to reality. This refined processing helps reduce the bias of subjective judgment and improves the objectivity and accuracy of the evaluation system. The integration of multiple judgment matrices can gather evaluation information from different experts or from different perspectives, forming a more comprehensive and integrated judgment matrix. This integration helps eliminate the one-sidedness that may exist in a single judgment matrix, making the final evaluation result more comprehensive and reliable.

[0054] By calculating the eigenvectors and maximum eigenvalue of the comprehensive judgment matrix, and then deriving the consistency ratio, the rationality of the evaluation system can be scientifically verified. The judgment of the consistency ratio can directly guide the adjustment and optimization of the evaluation system. A high consistency ratio indicates that the evaluation system is relatively reasonable; a low consistency ratio suggests that further adjustments and improvements are needed. This clear judgment helps improve the practicality and effectiveness of the evaluation system. Based on the judgment results and the weight values ​​of each level and indicator, the final performance evaluation indicator system can be reasonably determined.

[0055] In a preferred embodiment of the present invention, step 11 above, which constructs a performance evaluation indicator system comprising an n-level indicator layer; wherein n≥2 for the n-level indicator layer and the indicator names for each level of the indicator layer are already determined, may include:

[0056] Based on the characteristics of the evaluation object and the evaluation objectives, determine the specific content of each indicator level. Use appropriate methods (such as expert consultation, analytic hierarchy process, etc.) to determine the weights of each indicator level. The weights should reflect the importance and relative contribution of each indicator in the performance evaluation. Develop specific evaluation standards and methods for each indicator. Collect data and information related to the evaluation indicators. Based on the evaluation indicator system and evaluation standards, conduct performance evaluation of the evaluation object. Based on the feedback results, make necessary adjustments and improvements to the performance evaluation indicator system.

[0057] Taking a company's performance evaluation indicator system as an example, suppose the company wants to build a performance evaluation indicator system with a three-level indicator layer.

[0058] Determine the indicator level and name:

[0059] Primary Indicator: Corporate Performance (Overall Evaluation)

[0060] Secondary indicators: financial condition, market performance, internal management, and innovation capability.

[0061] Level 3 indicators (taking financial status as an example): operating revenue growth rate, total profit growth rate, debt-to-equity ratio, and cash flow ratio.

[0062] Construct a framework with three levels of indicators, clearly defining the logical relationships between each level. For example, corporate performance (level one indicator) consists of financial condition, market performance, internal management, and innovation capability (level two indicators); while financial condition (level two indicator) consists of specific indicators such as revenue growth rate, total profit growth rate, debt-to-equity ratio, and cash flow ratio (level three indicators). Use expert consultation to determine the weight of each level of indicator. For example, financial condition might account for 30% of the total weight of corporate performance, while revenue growth rate might account for 20% of the weight of financial condition. Develop specific evaluation criteria and methods for each level of indicator. For example, revenue growth rate can be evaluated by comparing it with the industry average or historical levels; the debt-to-equity ratio can be evaluated using a reasonable range. Collect data and information related to each level of indicator, such as financial statements, market research reports, and internal management documents. Process and analyze the data to ensure its accuracy and reliability. For example, calculate the actual values ​​of revenue growth rate and total profit growth rate and compare them with the industry average or historical levels. Evaluate the company's performance based on the evaluation indicator system and evaluation criteria. Calculate the scores for each level of indicators and the overall evaluation score to generate a performance evaluation report. The report should include the scores for each level of indicators, the overall evaluation results, existing problems, and suggestions for improvement.

[0063] In a preferred embodiment of the present invention, step 12 above, which compares the indicators at each level of the performance evaluation indicator system to obtain comparison results, and constructs multiple judgment matrices based on the comparison results, may include:

[0064] Step 122: In each level of the performance evaluation indicator system, select two indicators to form a pair for comparison;

[0065] Step 123: For each pair of selected indicators, according to the preset comparison standard, give the corresponding scale value to obtain the comparison result of each pair of indicators, including the indicator pair being compared and the scale value information.

[0066] Step 124: Construct multiple judgment matrices based on the comparison results of each pair of indicators.

[0067] In this embodiment of the invention, a hierarchical structure of the performance evaluation indicator system is clearly defined, including primary indicators, secondary indicators, and specific indicators at each level. Within each level, two indicators are selected to form a pair for comparison. This pair is determined through expert consultation to ensure the representativeness and comparability of the selected indicator pairs. For each pair of selected indicators, a corresponding scale value is given according to a preset comparison standard (e.g., 1 indicates that the two indicators are equally important, 3 indicates that the former indicator is slightly more important than the latter, 5 indicates that the former indicator is much more important than the latter, etc.). This scale value reflects the relative importance between the two indicators. Based on the comparison results of each pair of indicators, multiple judgment matrices are constructed. Each judgment matrix is ​​a square matrix, where the rows and columns represent the compared indicators, and the elements in the matrix are the corresponding scale values. An expert consultation method is used, inviting multiple experts to conduct pairwise comparisons of the indicators at each level of the indicator system and provide scale values. Through multiple rounds of iteration and feedback, the judgment matrices can be gradually adjusted and improved.

[0068] In practical applications, the expert consultation method is adopted, inviting three experts to provide scale values ​​for pairwise comparisons of indicators at each level of the indicator system, thus forming three judgment matrices.

[0069] The judgment matrix given by expert 1

[0070]

[0071] The judgment matrix given by expert 2

[0072]

[0073]

[0074] The judgment matrix given by expert 3

[0075]

[0076] By selecting indicators for comparison and providing scale values ​​based on pre-defined comparison standards, the influence of subjective assumptions and biases on evaluation results can be reduced. The constructed judgment matrix can objectively reflect the relative importance of each indicator, thereby improving the accuracy and credibility of performance evaluation. Comparing each level of the performance evaluation indicator system ensures that the evaluation covers all key areas and indicators, avoiding omissions or overemphasis on certain aspects. By constructing multiple judgment matrices, the position and role of each indicator in the entire performance evaluation indicator system can be comprehensively reflected. Inviting experts to participate in discussions and consultations during the comparison of indicators and the construction of judgment matrices helps promote consensus and communication among all parties. Through joint participation in the evaluation process, team members' understanding and acceptance of performance evaluation can be enhanced, improving the acceptance and effectiveness of the evaluation.

[0077] In a preferred embodiment of the present invention, step 13 above, which integrates multiple judgment matrices to form a comprehensive judgment matrix, may include:

[0078] Step 131: Standardize each judgment matrix to make the comparison scale values ​​within a uniform range, thus obtaining a set of standardized judgment matrices.

[0079] Step 132: Calculate the position of each comparison scale value in each judgment matrix after standardization;

[0080] Step 133: Calculate the trust weight based on the standardized position;

[0081] Step 134: For each pair of indicators, a weighted average is calculated using the trust level weight and the original comparison scale value to obtain the final comparison scale value between each pair of indicators; the formula for calculating the final comparison scale value between each pair of indicators is as follows: Among them, W f This represents the final comparison scale value between each pair of indicators; m represents the number of comparison scale values; w i R represents the trust weight of the i-th degree; i Represents the trust weight of the i-th degree; α represents the adjustment coefficient; R j This represents the trust weight of the j-th element; i and j represent index variables.

[0082] Step 135: Import the final comparison scale values ​​into the matrix according to the order of the index pairs to construct the comprehensive judgment matrix.

[0083] In this embodiment of the invention, the comparison scale values ​​are standardized to ensure they are within a uniform range. The standardization formula is: Among them, w i W is the original comparison scale value given by the i-th expert for a certain pair of indicators. iIt is the standardized value of the comparison scale given by the i-th expert for a certain pair of indicators, where m is the number of scale values; 1≤i≤m; w m It is the m-th value in the standardized comparison scale values.

[0084] In the standardized judgment matrix, the position of each comparison scale value can be represented by its standardized value. P = W i ×(max(W1,W2,…,W m )-min(W1,W2,…,W m ))+min(W1,W2,…,W m ); where P represents the comparison scale after position adjustment. This formula makes the position value the same as the standardized value (all W). i (All values ​​are between 0 and 1) Based on the standardized position, calculate the trust weight for each expert. The formula is: Among them, R i W is the trust weight of the i-th expert; ij This is the comparison scale value given by the i-th expert for the j-th pair of indicators. For each pair of indicators, a weighted average is calculated using the confidence weight and the original comparison scale value to obtain the final comparison scale value between each pair of indicators.

[0085] Determine the number of comparison scale values, m. Obtain the trust weight w for each expert. i And the original comparison scale values ​​they provided for each pair of indicators. Set an adjustment factor α, which is a parameter used to adjust for the influence of expert weights. Calculate the average of all expert trust weights. For each expert i, calculate its adjusted weight. For each pair of indicators, calculate the sum of the weighted comparison scale values ​​given by all experts, i.e. Calculate the sum of the adjusted weights, i.e. Divide the weighted sum by the adjusted sum of weights to obtain the final comparison scale value W for each pair of indicators. f .

[0086] Suppose there are 3 experts (m=3) comparing the indicators "overall construction scale" and "R&D investment level", and the original comparison scale values ​​and confidence weights given are as follows:

[0087] Expert 1: R1 = 3, w1 = 0.4;

[0088] Expert 2: R² = 5, w² = 0.3;

[0089] Expert 3: R3 = 2, w3 = 0.3.

[0090] Set the adjustment factor α = 0.5.

[0091] Average weight = 33 + 5 + 2 = 3.33 (more decimal places should be retained in actual calculations).

[0092] The adjusted weight for Expert 1 is 3 + 0.5 × (3 - 3.33) = 2.835.

[0093] The adjusted weight for Expert 2 is 5 + 0.5 × (5 - 3.33) = 5.835.

[0094] The adjusted weight for expert 3 is 2 + 0.5 × (2 - 3.33) = 1.335.

[0095] Weighted sum = 0.4 × 2.835 + 0.3 × 5.835 + 0.3 × 1.335 = 3.003;

[0096] Denominator = 2.835 + 5.835 + 1.335 = 10.005;

[0097] Calculate the final comparison scale value (Retain two decimal places). Import the final comparison scale values ​​into the matrix according to the order of the index pairs to form a comprehensive judgment matrix.

[0098] The comprehensive judgment matrix obtained by combining the three judgment matrices

[0099]

[0100] By integrating multiple judgment matrices, opinions from different sources or experts can be comprehensively considered, thus avoiding the bias and subjectivity that may arise from a single judgment matrix. Standardization and the introduction of trust weights allow information from each judgment matrix to be integrated into the comprehensive judgment matrix in a more reasonable way, improving the scientific rigor and accuracy of decision-making. By clarifying the standardization process for each judgment matrix, the calculation method for trust weights, and the formula for calculating the final comparison scale value, the entire decision-making process becomes more transparent and traceable. This helps decision-makers understand how the comprehensive judgment matrix is ​​formed and the degree of influence of each judgment matrix on the final result. It can handle multiple judgment matrices without being limited by the number, size, or form of the matrices, exhibiting strong flexibility and applicability. By adjusting parameters such as adjustment coefficients, differentiated processing of information from different judgment matrices can be achieved to meet the needs of different decision-making scenarios. Developing clear steps and formulas for calculating the comprehensive judgment matrix promotes the standardization and normalization of the decision-making process. This helps reduce arbitrariness and uncertainty in the decision-making process and improves the consistency and comparability of decision results. When facing complex decision-making problems involving multiple indicators, multiple experts, or multiple data sources, this method can provide a systematic and comprehensive solution. By integrating opinions and information from all parties to form a comprehensive judgment matrix, decision-makers can be provided with a comprehensive and objective basis for decision-making, which helps to improve the quality and efficiency of decision-making.

[0101] In a preferred embodiment of the present invention, step 14 above calculates the consistency ratio by comprehensively judging the eigenvectors and the largest eigenvalue of the matrix; the formula for calculating the consistency ratio is:

[0102]

[0103] Where CR represents the consistency ratio; λ max represents the largest eigenvalue; n represents the order of the judgment matrix; RI represents the average random consistency index.

[0104] In this embodiment of the invention, a characteristic polynomial is used to solve for the eigenvectors and corresponding eigenvalues ​​of the comprehensive judgment matrix. The characteristic polynomial P(λ) = det(A - λI); where det represents the determinant; A represents the comprehensive judgment matrix; λ represents the variable of the characteristic polynomial; and I represents the identity matrix. By solving the characteristic polynomial, all eigenvalues ​​λ1, λ2, ..., λI of the matrix are obtained. n By solving the linear equation system (A-λ) n I) Find the corresponding eigenvector v by setting v = 0. This system of equations has multiple solutions, which together form a system with respect to the eigenvalue λ. n The corresponding eigenspace. Find the maximum value among all eigenvalues, which is the maximum eigenvalue λ. max Using the formula To calculate the consistency ratio, we use n, where n is the order of the judgment matrix (i.e., the number of rows or columns), and RI is the average random consistency index, which is a constant related to the matrix order and can be obtained by looking up a table.

[0105] Suppose we have a 9th-order comprehensive judgment matrix A, and through linear algebra calculations, we have obtained its eigenvalues ​​λ1, λ2, ..., λ9. Among these eigenvalues, we find the maximum value, denoted as λ. max At the same time, it also obtained the relationship with λ. max The corresponding eigenvectors (in practical applications, these values ​​should be obtained through specific calculations). The largest eigenvalue λ is obtained. max Then, the formula can be used. To calculate the consistency index, we have the matrix order n = 9, and through calculation, we obtained CI = 0.2783 (this is an assumed result; in practical applications, it should be calculated using specific methods). The formula is used... To calculate the consistency ratio, for a 9th-order matrix, the average random consistency index RI obtained from the table is 1.46 (this is a standard value; in practical applications, relevant tables should be consulted for confirmation). Substituting CI = 0.2783 and RI = 1.46 into the formula, we get CR = 1.46 * 0.2783 ≈ 0.1906.

[0106] The consistency ratio (CR) is an important indicator for measuring the consistency of a judgment matrix. Calculating the CR ensures consistency among judgment factors during the decision-making process, preventing contradictory or inconsistent judgments from negatively impacting the final result. A low CR value (CR < 0.1) indicates good consistency in the judgment matrix, meaning the relative importance of each judgment factor is reasonably reflected. This helps improve the accuracy of decision-making, as decisions are based on a reliable and consistent judgment foundation. Calculating and verifying the consistency ratio enhances the credibility of decisions. In multi-criteria evaluation, the construction and consistency verification of the comprehensive judgment matrix are crucial steps. Calculating the consistency ratio optimizes the multi-criteria evaluation system, ensuring a reasonable allocation and reflection of the relative importance of each evaluation criterion. The calculation and verification of the consistency ratio contribute to the scientific and objective nature of decision-making. It requires decision-makers to follow certain logic and rules when making judgments, avoiding subjective assumptions and arbitrary judgments. This helps ensure the fairness and rationality of the decision outcome. A high consistency ratio (CR ≥ 0.1) indicates poor consistency in the judgment matrix. At this point, decision-makers can adjust and optimize the judgment matrix based on the feedback information provided by the CR value, in order to reduce the CR value and improve consistency. This helps to improve the decision-making process and outcomes.

[0107] In a preferred embodiment of the present invention, step 15, which involves judging the consistency ratio to obtain a judgment result, and step 16, which involves determining the final performance evaluation indicator system based on the judgment result and the weight values ​​of each indicator at each level, may include:

[0108] Determine if the consistency ratio is less than 0.1. If the consistency ratio is less than 0.1, the judgment matrix passes the consistency test. If the consistency ratio is greater than or equal to 0.1, adjust the judgment matrix and repeat the operation of constructing the comprehensive judgment matrix and calculating the consistency ratio until the consistency test is passed and the judgment result is obtained.

[0109] The calculated consistency ratio (CR) is compared with 0.1. If CR < 0.1, the judgment matrix passes the consistency test, indicating good consistency among its elements, allowing for further analysis and decision-making. If CR ≥ 0.1, the judgment matrix fails the consistency test, indicating significant inconsistency among its elements. In this case, the judgment matrix needs adjustment to improve its consistency. Adjustments can include modifying element values, re-collecting expert opinions, or using a more suitable scaling method. The adjusted judgment matrix needs to be reconstructed, and the consistency ratio recalculated. This process of constructing the comprehensive judgment matrix and calculating the consistency ratio is repeated until the judgment matrix passes the consistency test. Once the judgment matrix passes the consistency test, the final judgment results are output, including the weights and rankings of each indicator layer, resulting in the final performance evaluation indicator system.

[0110] Project performance evaluation indicator system

[0111]

[0112] like Figure 2 As shown, embodiments of the present invention also provide a performance evaluation system 20 based on an improved analytic hierarchy process, comprising:

[0113] The acquisition module 21 is used to construct a performance evaluation indicator system containing n levels of indicators; compare the indicators at each level of the performance evaluation indicator system to obtain comparison results; construct multiple judgment matrices based on the comparison results; and integrate the multiple judgment matrices to form a comprehensive judgment matrix.

[0114] Processing module 22 is used to calculate the consistency ratio by comprehensively judging the eigenvectors and the largest eigenvalue of the matrix; judge the consistency ratio to obtain the judgment result; and determine the final performance evaluation index system based on the judgment result and the weight values ​​of each indicator at each level.

[0115] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0116] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0117] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0118] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A performance evaluation method based on an improved analytic hierarchy process, characterized in that, The method includes: Construct a performance evaluation indicator system that includes n levels of indicators; The indicators at each level of the performance evaluation indicator system are compared to obtain the comparison results. Based on the comparison results, multiple judgment matrices are constructed. Multiple judgment matrices are integrated to form a comprehensive judgment matrix; The consistency ratio is calculated by comprehensively judging the eigenvectors and the largest eigenvalue of the matrix. The consistency ratio is evaluated to obtain the evaluation result; Based on the judgment results and the weight values ​​of each indicator at each level, the final performance evaluation indicator system is determined.

2. The performance evaluation method based on the improved analytic hierarchy process according to claim 1, characterized in that, The n-level indicator layer has n≥2, and the indicator names of each level of the indicator layer have been determined.

3. The performance evaluation method based on the improved analytic hierarchy process according to claim 2, characterized in that, The indicators at each level of the performance evaluation indicator system are compared to obtain comparison results. Based on these results, multiple judgment matrices are constructed, including: In each level of the performance evaluation indicator system, two indicators are selected to form a pair for comparison; For each pair of selected indicators, a corresponding scale value is given according to the preset comparison standard to obtain the comparison result of each pair of indicators, including the indicator pair being compared and the scale value information. Based on the comparison results of each pair of indicators, multiple judgment matrices are constructed.

4. The performance evaluation method based on the improved analytic hierarchy process according to claim 3, characterized in that, Multiple judgment matrices are integrated to form a comprehensive judgment matrix, including: Each judgment matrix is ​​standardized to ensure that the comparison scale values ​​are within a uniform range, resulting in a standardized set of judgment matrices. Calculate the position of each comparison scale value in the standardized matrix; Calculate the trust weight based on the standardized location; For each pair of indicators, a weighted average is calculated using the confidence weight and the original comparison scale value to obtain the final comparison scale value between each pair of indicators; The final comparison scale values ​​are imported into the matrix in the order of the index pairs to construct a comprehensive judgment matrix.

5. The performance evaluation method based on the improved analytic hierarchy process according to claim 4, characterized in that, The formula for calculating the final comparison scale value between each pair of indicators is: W f This represents the final comparison scale value between each pair of indicators; m represents the number of comparison scale values; w i R represents the trust weight of the i-th degree; i Represents the trust weight of the i-th degree; α represents the adjustment coefficient; R j This represents the trust weight of the j-th element; i and j represent index variables.

6. The performance evaluation method based on the improved analytic hierarchy process according to claim 5, characterized in that, The formula for calculating the consistency ratio is: Where CR represents the consistency ratio; λ max represents the largest eigenvalue; n represents the order of the judgment matrix; RI represents the average random consistency index.

7. The performance evaluation method based on the improved analytic hierarchy process according to claim 6, characterized in that, The consistency ratio is evaluated to obtain the evaluation result, including: Determine if the consistency ratio is less than 0.

1. If the consistency ratio is less than 0.1, the judgment matrix passes the consistency test. If the consistency ratio is greater than or equal to 0.1, adjust the judgment matrix and repeat the operation of constructing the comprehensive judgment matrix and calculating the consistency ratio until the consistency test is passed and the judgment result is obtained.

8. A performance evaluation system based on an improved analytic hierarchy process, wherein the system implements the method as described in claim 1, characterized in that, include: The acquisition module is used to construct a performance evaluation indicator system containing n levels of indicators. The indicators at each level of the performance evaluation indicator system are compared to obtain the comparison results. Based on the comparison results, multiple judgment matrices are constructed. The multiple judgment matrices are then integrated to form a comprehensive judgment matrix. The processing module is used to calculate the consistency ratio by comprehensively judging the eigenvectors and the largest eigenvalue of the matrix; The consistency ratio is judged to obtain the judgment result; based on the judgment result and the weight values ​​of each indicator at each level, the final performance evaluation indicator system is determined.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images