Power generation project bid invitation scoring factor importance quantification model construction method based on AHP

Through the AHP quantitative model construction method, the problem of insufficient subjective judgment dependence and consistency in the bidding score of power generation projects is solved, and more objective and stable scoring results are achieved, which are suitable for large-scale and complex procurement projects.

CN119026975APending Publication Date: 2024-11-26SHANGHAI HUADIAN ELECTRIC POWER DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410964014.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-18
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The existing bidding scoring methods for power generation projects lack objectivity, consistency and robustness, rely on subjective judgment and are greatly affected by sample capacity, resulting in unstable and inaccurate scoring results.

Method used

Using AHP-based quantitative model construction method, through hierarchical structure model, judgment matrix calculation, weight vector calculation, consistency check and rationality check, objective scoring factor scores are provided, reducing the impact of subjective judgment and improving the consistency of scores. and robustness.

Benefits of technology

It improves the objectivity and consistency of the scoring method, reduces the influence of subjective factors, especially when the sample size is large, which can more accurately reflect the actual situation, and reduces the instability of the scoring results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026975A_ABST
    Figure CN119026975A_ABST
Patent Text Reader

Abstract

The invention discloses a power generation project bid invitation scoring factor importance quantification model construction method based on AHP, and the method comprises the steps: S1, constructing a hierarchical structure model: listing all factors needing to be evaluated according to levels, and listing the factors into a tree-shaped hierarchical structure model; s2, starting from a high level to a low level of the hierarchical structure model, establishing a judgment matrix; s3, calculating a weight vector as the importance of each factor of each layer; s4, consistency verification: detecting the consistency of the judgment matrix to confirm whether the judgment matrix is reasonable or not; s5, determining scores of the scoring factors; and S6, rationality verification: making all score columns into a table, and observing whether the data difference is too great or not and whether the range distribution is reasonable or not. The method provided by the invention is very suitable for factor formulation of purchase projects with large amount and large scale; when a large number of complex factors and a large number of samples are involved, the method can reduce the influence of subjective factors as much as possible.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of coal-fired and gas-fired power generation projects, and is a method for constructing an importance quantification model of scoring factors for bidding of power generation projects based on AHP. Background Art

[0002] Power generation projects, especially coal-fired and gas-fired power generation projects, often involve a large number of goods procurement projects. Taking the F-class combined cycle unit as an example, the equipment purchase cost usually accounts for more than 50% of the total budget, so purchasing high-quality equipment is of great significance to optimizing the project cost.

[0003] In the procurement process, the review of response documents is crucial. There is no room for subjective review of price factors, which account for about 50% of the total score, and technical factors, which account for 35-50%, are the decisive conditions affecting the results. Therefore, it is of considerable practical value to establish a quantitative model to reasonably allocate the scores of scoring factors.

[0004] The existing means of evaluating the importance of scoring factors is mainly through subjective judgment by the procurement initiating profession. Complex procurement projects are jointly judged by multiple professions, and finally points are assigned according to the approximate ranking of importance. To make it simpler, you can refer to and draw on existing templates of similar procurement projects.

[0005] Problems with existing technologies:

[0006] (1) The existing scoring method lacks objectivity: it relies entirely on people’s subjective judgment and work experience, and has no theoretical support or data support;

[0007] (2) The existing scoring methods lack consistency and coordination: when multiple people have different subjective judgments on the importance of factors, there is a lack of scientific methods to coordinate;

[0008] (3) The existing scoring method lacks robustness and is greatly affected by the size of the sample: when the sample size is small, it cannot objectively reflect the actual situation; when the sample size is large, the problem described in (2) will occur. Summary of the invention

[0009] The purpose of the present invention is to provide a method for constructing a quantitative model of the importance of scoring factors in power generation project bidding based on AHP, which mainly solves the following problems:

[0010] (1) It provides a theoretical basis and technical means for the scoring method, which no longer relies entirely on human subjective judgment, but instead adopts a method of a large amount of subjective evaluation + objective calculation;

[0011] (2) The consistency and coordination of the scoring method are enhanced. When the sample differences are large, mathematical methods can be used to scientifically handle them.

[0012] (3) The instability of the scoring method is reduced, and the larger the sample size, the more objectively the problem can be reflected.

[0013] The technical solution of the present invention is:

[0014] The method for constructing a quantitative model of the importance of scoring factors for power generation project bidding based on AHP includes the following steps:

[0015] S1. Construction of hierarchical model: List all factors that need to be evaluated hierarchically and list them into a hierarchical model in a tree diagram;

[0016] S2. Starting from the high level to the low level of the hierarchical model, establish a judgment matrix;

[0017] S3, calculate the weight vector as the importance of each factor in each layer;

[0018] S4, consistency check: for the judgment matrix, check its consistency to confirm whether it is reasonable;

[0019] S5. Confirm the scoring factor scores;

[0020] S6. Reasonableness check: List all the scores in a table to see whether the data differences are too great and whether the range of distribution is reasonable.

[0021] Preferably, all factors to be evaluated are listed hierarchically in S1, which is a hierarchical structure model in a tree diagram.

[0022] Preferably, the method for establishing the judgment matrix in S2 is:

[0023] For the factors at the same level and the same fork, the importance of each factor is evaluated. If the importance of the i-th factor compared to the j-th factor is a ij , then the judgment matrix can be listed as follows:

[0024]

[0025] The values ​​of each element are determined according to the following rules: a ij The numerical value is defined as a subjective judgment of the importance of factor i over factor j, with 1 being equally important, 9 being extremely important, and the rest of the numbers being in the middle; the subjective judgment of the data maintains logical continuity, otherwise it will fail the consistency check.

[0026] Preferably, the method for calculating the weight vector in S3 adopts the sum-product method:

[0027] Normalize the matrix column-wise, doing this for each element:

[0028]

[0029] Then find the eigenvector, calculate the i-th component as follows, and finally combine all components into a standardized eigenvector, which represents the importance of a factor in this layer:

[0030]

[0031] Preferably, the consistency check method in S4 is:

[0032] First, find the largest eigenvalue λ of the matrix max :

[0033]

[0034] Then detect the consistency index CI of subjective logical judgment:

[0035]

[0036] If the CI is less than 0.1, it is considered reasonable, otherwise the judgment matrix is ​​re-discussed;

[0037] When the matrix order is greater than 3, it is corrected according to the consistency ratio CR:

[0038]

[0039] RI is the correction factor.

[0040] Preferably, the method for confirming the scoring factor scores in S5 is:

[0041] The weight vector components of a factor H at the first-level criterion layer and the last-level related factors are recorded as and In each level of the criteria layer there are related weight vector components With a full score of 100, the score of H is calculated by multiplication from the top to the bottom:

[0042]

[0043] Preferably, the method for rationality checking in S6 is:

[0044] All scores are tabulated to see if the data differences are too great and if the range distribution is reasonable. Considering the flexibility of the review, there should be fewer high-scoring factors and more low-scoring factors. A Pareto chart can be made to see if there is a decreasing trend under similar step-size conditions.

[0045] Compared with the existing technology, the beneficial effects of this technical solution are as follows:

[0046] 1. The method of the present invention is very suitable for the formulation of factors for large-scale procurement projects; when a large number of complex factors and a large number of samples are involved, the method can reduce the influence of subjective factors as much as possible.

[0047] 2. The calculation method of the present invention is simple and can be calculated using Excel or other spreadsheets. If a large number of operations are designed, a program can be established using programming software such as Python. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0049] Figure 1 A flowchart of the quantitative model construction method of the present invention;

[0050] Figure 2 It is a hierarchical model diagram;

[0051] Figure 3 It is a typical two-tier hierarchical structure model diagram;

[0052] Figure 4 is a hierarchical structure model diagram in the embodiment;

[0053] Figure 5 1 is a Pareto chart of scores in the embodiment. DETAILED DESCRIPTION

[0054] like Figure 1 As shown in the figure, the construction method of the quantitative model of the importance of the scoring factors of power generation project bidding based on AHP includes the following steps:

[0055] S1. Construction of hierarchical model: List all factors that need to be evaluated hierarchically and list them into a hierarchical model in a tree diagram;

[0056] S2. Starting from the high level to the low level of the hierarchical model, establish a judgment matrix;

[0057] S3, calculate the weight vector as the importance of each factor in each layer;

[0058] S4, consistency check: for the judgment matrix, check its consistency to confirm whether it is reasonable;

[0059] S5. Confirm the scoring factor scores;

[0060] S6. Reasonableness check: List all the scores in a table to see whether the data differences are too great and whether the range of distribution is reasonable.

[0061] S1 lists all factors that need to be evaluated in a hierarchical manner. Figure 2 The hierarchical model is shown in a tree diagram. Figure 3 A typical two-tier hierarchical structure model (i.e. two-level criteria layer) is given, and additional levels can also be added.

[0062] The method of establishing the judgment matrix in S2 is:

[0063] For the factors at the same level and the same fork, the importance of each factor is evaluated. If the importance of the i-th factor compared to the j-th factor is a ij , then the judgment matrix can be listed as follows:

[0064]

[0065] The values ​​of each element are determined according to the following rules: a ij The numerical value is defined as a subjective judgment of the importance of factor i over factor j, with 1 being equally important, 9 being extremely important, and the rest of the numbers being in between. The subjective judgment of the data maintains logical continuity (if A>B and B>C, then A>C), otherwise it will fail the consistency check.

[0066] There are many methods for calculating the weight vector in S3. Considering the numerical accuracy of the final conclusion, the sum-product method is more appropriate.

[0067] Normalize the matrix column-wise, doing this for each element:

[0068]

[0069] Then find the eigenvector, calculate the i-th component as follows, and finally combine all components into a standardized eigenvector, which represents the importance of a factor in this layer:

[0070]

[0071] The consistency check method in S4 is:

[0072] First, find the largest eigenvalue λ of the matrix max :

[0073]

[0074] Then detect the consistency index CI of subjective logical judgment:

[0075]

[0076] If the CI is less than 0.1, it is considered reasonable, otherwise the judgment matrix is ​​re-discussed;

[0077] When the matrix order is greater than 3, it is corrected according to the consistency ratio CR:

[0078]

[0079] RI is the correction factor, please refer to Table 1 for the correction factor.

[0080] Table 1 Correction coefficient values

[0081]

[0082] The method for confirming the scores of the scoring factors in S5 is:

[0083] The weight vector components of a factor H at the first-level criterion layer and the last-level related factors are recorded as and In each level of the criteria layer there are related weight vector components With a full score of 100, the score of H is calculated by multiplication from the top to the bottom:

[0084]

[0085] The method for rationality checking in S6 is:

[0086] All scores are tabulated to see if the data differences are too great and if the range distribution is reasonable. Considering the flexibility of the review, there should be fewer high-scoring factors and more low-scoring factors. A Pareto chart can be made to see if there is a decreasing trend under similar step-size conditions.

[0087] Example 1

[0088] A power plant's gas-steam combined cycle generator project is currently purchasing waste heat boiler equipment. Its technical scoring factors are relatively complex, with a total of 6 major items and 23 minor items, which are quite complex. Therefore, a second-order model was established to calculate the factor scores.

[0089] A questionnaire survey was conducted on 40 employees of the project team. The survey content was a pairwise comparison of the importance of factors in the same group. The reference method was: the questionnaire ranked the importance of factors in the same branch, with the most important one counted as 1 and the least important one counted as 9. The average score of all valid questionnaires was taken, and then the lowest score factor was normalized as a benchmark to obtain the evaluation scores of other factors. Taking the first-level criterion layer as an example, the judgment matrix is:

[0090]

[0091] Calculate the weight vector, with a 11 For example:

[0092]

[0093] Then for the matrix T we have:

[0094]

[0095] Calculate the weight vector, taking ω1 as an example:

[0096]

[0097] Then the weight vector is:

[0098]

[0099] Consistency check:

[0100]

[0101]

[0102] If the check passes, the weight of the top-level factor "Boiler Performance A" is 0.37. The consistency check summary is shown in Table 2.

[0103] Table 2. Example consistency verification data

[0104]

[0105] The consistency checks in Table 2 are all passed, and taking this as an example, all the data can be summarized as shown in Table 3.

[0106] Table 3 Example scoring data

[0107]

[0108]

[0109] Taking "ensuring thermal efficiency A1" as an example, the final weights are:

[0110]

[0111] After calculating all the scores, the rationality check results are shown in Figure 5 The score Pareto chart is shown. The rationality check passed. All the scores in Table 3 are reasonable and the algorithm is closed.

[0112] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with the technology to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any modifications made according to the spirit of the main technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a quantitative model of the importance of scoring factors in power generation project bidding based on AHP, characterized by: Includes steps: S1. Construction of hierarchical model: List all factors that need to be evaluated hierarchically and list them into a hierarchical model in a tree diagram; S2. Starting from the high level to the low level of the hierarchical model, establish a judgment matrix; S3, calculate the weight vector as the importance of each factor in each layer; S4, consistency check: for the judgment matrix, check its consistency to confirm whether it is reasonable; S5. Confirm the scoring factor scores; S6. Reasonableness check: List all the scores in a table to see whether the data differences are too great and whether the range of distribution is reasonable.

2. The method for constructing a quantitative model of importance of scoring factors for power generation project bidding based on AHP according to claim 1 is characterized in that: S1 lists all factors that need to be evaluated hierarchically, which is a hierarchical model in a tree diagram.

3. The method for constructing a quantitative model of importance of scoring factors in power generation project bidding based on AHP according to claim 2 is characterized in that: The method of establishing the judgment matrix in S2 is: For the factors at the same level and the same fork, the importance of each factor is evaluated. If the importance of the i-th factor compared to the j-th factor is a ij , then the judgment matrix can be listed as follows: The values ​​of each element are determined according to the following rules: a ij The numerical value is defined as a subjective judgment of the importance of factor i over factor j, with 1 being equally important, 9 being extremely important, and the rest of the numbers being in the middle; the subjective judgment of the data maintains logical continuity, otherwise it will fail the consistency check.

4. The method for constructing a quantitative model of importance of scoring factors in power generation project bidding based on AHP according to claim 3 is characterized in that: The method for calculating the weight vector in S3 uses the sum-product method: Normalize the matrix column-wise, doing this for each element: Then find the eigenvector, calculate the i-th component as follows, and finally combine all components into a standardized eigenvector, which represents the importance of a factor in this layer:

5. The method for constructing a quantitative model of importance of scoring factors in power generation project bidding based on AHP according to claim 4 is characterized in that: The consistency check method in S4 is: First, find the largest eigenvalue λ of the matrix max : Then detect the consistency index CI of subjective logical judgment: If the CI is less than 0.1, it is considered reasonable, otherwise the judgment matrix is ​​re-discussed; When the matrix order is greater than 3, it is corrected according to the consistency ratio CR: RI is the correction factor.

6. The method for constructing a quantitative model of importance of scoring factors in power generation project bidding based on AHP according to claim 5, characterized in that: The method for confirming the scores of the scoring factors in S5 is: The weight vector components of a factor H at the first-level criterion layer and the last-level related factors are recorded as and In each level of the criteria layer there are related weight vector components With a full score of 100, the score of H is calculated by the cumulative multiplication method from the top to the bottom:

7. The method for constructing a quantitative model of importance of scoring factors in power generation project bidding based on AHP according to claim 6, characterized in that: The method for rationality checking in S6 is: All scores are tabulated to see if the data differences are too great and if the range distribution is reasonable. Considering the flexibility of the review, there should be fewer high-scoring factors and more low-scoring factors. A Pareto chart can be made to see if there is a decreasing trend under similar step-size conditions.