Expert bid evaluation system based on related index data

Abstract

The invention discloses an expert bid evaluation system based on related index data, which relates to the technical field of expert bid evaluation and comprises a historical cooperation data acquisition module, a direct trust index acquisition module, an indirect trust index acquisition module, a trustworthy probability acquisition module, a comprehensive trust index acquisition module, an evaluation expert recommendation module and a clustering bid document calibration module. The technical problems that when experts are selected, historical cooperation data of the experts and enterprises are not fully considered, and the bid evaluation value of indirect cooperation experience is neglected are solved. By comprehensively considering the direct cooperation times, the indirect cooperation times and the work resumes of the experts, the credibility of the experts can be evaluated more objectively, so that subjective deviation is reduced, the professional ability and adaptability of the experts are evaluated more accurately, and the bid evaluation accuracy and quality are improved. The automatic recommendation system can quickly sort the experts according to the comprehensive trust index, so that the efficiency of selecting the experts is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of expert bid evaluation, and in particular to an expert bid evaluation system based on relevant indicator data. Background Art

[0002] In today's complex business environment, enterprises are faced with numerous project bidding and bid evaluation needs. With the intensification of market competition, higher requirements are placed on the fairness, accuracy and efficiency of the bid evaluation process. In various bidding and procurement activities, bid evaluation is a crucial link. For example, in the process of large-scale project management and procurement, it is often necessary to select suppliers or contractors, which usually involves the evaluation of multiple bids. Selecting a suitable team of bid evaluation experts is crucial to ensure the successful execution of the project.

[0003] However, due to the subjectivity and inconsistency of expert selection, traditional methods rely on manual selection or random assignment of experts. Different experts may have different evaluation criteria and preferences, which leads to inconsistency in evaluation results. There is a lack of consideration of experts' historical cooperation data. When selecting experts, their historical cooperation data with enterprises is not fully considered, which cannot ensure the experts' familiarity and adaptability to enterprise projects. At the same time, only considering the number of direct cooperation may ignore the relevant experience and knowledge accumulated by experts through other projects and ignore the value of indirect cooperation experience for evaluation, resulting in uneven evaluation quality and low evaluation efficiency. Based on this, an expert evaluation system based on relevant indicator data is proposed. Summary of the invention

[0004] The purpose of the present invention is to provide an expert bid evaluation system based on relevant indicator data, which solves the technical problem of not fully considering their historical cooperation data with enterprises when selecting experts and ignoring the value of indirect cooperation experience for bid evaluation.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] Expert bid evaluation system based on relevant indicator data, including:

[0007] The historical cooperation data acquisition module obtains the number of direct and indirect cooperation between the enterprise and each candidate assessment expert;

[0008] A direct trust index acquisition module determines the direct trust index between each candidate assessment expert and the enterprise according to the number of direct cooperation between the enterprise and each candidate assessment expert;

[0009] The indirect trust index acquisition module counts the assessment projects jointly participated by each candidate assessment expert and the company's existing cooperative experts, distinguishes the number of parent and friendly business projects and analyzes them, and obtains the indirect trust index between each candidate assessment expert and the enterprise;

[0010] The trust probability acquisition module analyzes the working hours and the number of times each candidate evaluation expert has participated in the evaluation, and obtains the trust probability corresponding to each candidate evaluation expert;

[0011] The comprehensive trust index acquisition module integrates the direct trust index, indirect trust index and trust probability to obtain the comprehensive trust index corresponding to each candidate evaluation expert;

[0012] The evaluation expert recommendation module generates a recommendation list by sorting the comprehensive trust index to recommend the evaluation experts required for the evaluation of the bid;

[0013] The clustered bid calibration module obtains the correlation index coefficients between each bid and other bids based on the direct project index categories of multiple bids to be evaluated, and calibrates each clustered bid corresponding to the multiple bids to be evaluated according to the correlation index coefficients.

[0014] As a further solution of the present invention, the specific method of determining the direct trust index between each candidate evaluation expert and the enterprise is as follows:

[0015] The number of direct cooperation between each candidate assessment expert and the enterprise is obtained, and the number of direct cooperation is defined as the direct trust index Hj corresponding to each candidate assessment expert, where j refers to each candidate assessment expert, j is a positive integer, and j≥1.

[0016] As a further solution of the present invention, the specific method of obtaining the indirect trust index between each candidate evaluation expert and the enterprise is as follows:

[0017] The square of the number of parent projects and the number of friendly projects participated in by each candidate assessment expert in the past two years are marked as Bj and Yj respectively. The sum of the products of the number of parent projects and the number of friendly projects of each candidate assessment expert and the preset parameter factor coefficients θ1 and θ2 is defined as the indirect trust index Jj corresponding to each candidate assessment expert, where 1=θ1+θ2, θ1>θ2.

[0018] As a further solution of the present invention, the specific method of obtaining the trustworthiness probabilities corresponding to each candidate evaluation expert is as follows:

[0019] The sum of the products of the working hours and the number of times each candidate evaluation expert has participated in the evaluation and the preset proportional coefficients β1 and β2 is defined as the trust probability Rj corresponding to each candidate evaluation expert, where β1 and β2 are both preset proportional coefficients, satisfying 1=β1+β2, β2>β1.

[0020] As a further solution of the present invention, the specific method of obtaining the comprehensive trust index corresponding to each candidate evaluation expert is as follows:

[0021] By formula: Zj=(Hj) 2 ×θ3+Jj×θ4+Rj×θ5, calculate and obtain the comprehensive trust index Zj corresponding to each candidate evaluation expert, where θ3, θ4 and θ5 are all preset parameter factor coefficients, satisfying 1=θ3+θ4+θ5, θ3>θ5>θ4.

[0022] As a further solution of the present invention: according to the comprehensive trust index values ​​corresponding to each candidate evaluation expert, the candidate evaluation experts are sorted in descending order, and then a recommendation list of evaluation experts is generated, and the evaluation experts required for the bid to be evaluated are recommended according to the recommendation list of evaluation experts.

[0023] As a further solution of the present invention, the specific method of generating a recommendation list by sorting the comprehensive trust index to recommend the evaluation experts required for the evaluation of the bid is:

[0024] A1: Randomly select one of the bids to be evaluated as the target evaluation bid, and mark the project indicator type corresponding to the target evaluation bid as L1;

[0025] By formula: Calculate the correlation coefficient J between the target evaluation bid and other bids (1，u) ; where L u L1∩L u is the number of the same direct project indicator types corresponding to the target evaluation bid and the remaining bids to be evaluated, where u refers to the remaining bids to be evaluated except the target evaluation bid, u is a positive integer, u≥1;

[0026] A2: Mark the bids to be evaluated whose correlation index coefficient with the target evaluation bid is greater than the preset value YS1 as the same cluster bid W1 together with the target evaluation bid; for the bids to be evaluated whose correlation index coefficient is less than or equal to the preset value YS1, no processing is performed;

[0027] A3: Select any one of the bids to be evaluated whose relevant index coefficient is less than or equal to the preset value YS1 as the target bid, repeat steps A1-A2, and obtain the same cluster bids corresponding to the multiple bids to be evaluated.

[0028] As a further solution of the present invention: the specific method of distinguishing the number of parent and friend projects is as follows:

[0029] Obtain the assessment projects jointly participated by each candidate assessment expert and the company's existing cooperative experts. When the jointly participated assessment project is a project of the company, it is marked as a parent project. When the jointly participated assessment project is not a project of the company, it is marked as a friendly company project. The number of parent projects and friendly company projects participated in by each candidate assessment expert in the past two years is marked as Bj and Yj respectively.

[0030] Beneficial effects of the present invention:

[0031] (1) The present invention can more objectively evaluate the trustworthiness of experts by comprehensively considering the number of direct cooperation, the number of indirect cooperation and the work experience of experts, thereby reducing subjective bias and more accurately evaluating the professional ability and adaptability of experts, thereby improving the accuracy and quality of bid evaluation. The automated recommendation system can quickly sort experts according to the comprehensive trust index, thereby improving the efficiency of selecting experts;

[0032] (2) The present invention, for multiple bids to be evaluated, uses the same clustering bid determination module to classify bids with similar project indicator types into one category for unified bid evaluation, thereby avoiding duplication of work and improving bid evaluation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below in conjunction with the accompanying drawings.

[0034] Figure 1 It is a schematic diagram of the system framework structure of the present invention;

[0035] Figure 2 It is a schematic diagram of the framework structure of the method of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Embodiment 1

[0038] See also Figure 1-Figure 2 As shown, the present invention is an expert bid evaluation system based on relevant indicator data, comprising:

[0039] The historical cooperation data acquisition module acquires the historical cooperation data corresponding to the enterprise and each bid evaluation expert. The historical cooperation data refers to the number of direct cooperation and indirect cooperation between the enterprise and each bid evaluation expert.

[0040] The number of direct cooperation refers to the number of times the enterprise directly cooperates with a bid evaluation expert. For example, if the enterprise and bid evaluation expert A have directly participated in 3 projects together, then the number of direct cooperation between the enterprise and bid evaluation expert A is 3; the number of indirect cooperation refers to the number of times the enterprise has not directly cooperated with a bid evaluation expert, but the bid evaluation expert has participated in the evaluation projects with the company's cooperating experts; for example, bid evaluation expert B has never directly cooperated with the company, but he has participated in 5 other projects with expert C, who has cooperated with the company, then the number of indirect cooperation between bid evaluation expert B and the company is 5;

[0041] The direct trust index acquisition module calculates the direct trust index between each candidate assessment expert and the enterprise according to the number of direct cooperation between the enterprise and each candidate assessment expert. The specific method is as follows:

[0042] Obtain the number of direct cooperations between each candidate assessment expert and the enterprise, and define the number of direct cooperations as the direct trust index Hj corresponding to each candidate assessment expert, where j refers to each candidate assessment expert, j is a positive integer, and j≥1;

[0043] For example, if the enterprise has directly cooperated with the candidate assessment expert E1 once, the number of direct cooperation is 1; if the enterprise has directly cooperated with the candidate assessment expert E2 twice, the number of direct cooperation is 2; if the enterprise has not directly cooperated with the candidate assessment expert E3, the number of direct cooperation is 0;

[0044] The indirect trust index acquisition module calculates the indirect trust index between each candidate assessment expert and the enterprise according to the number of indirect cooperation between the enterprise and each candidate assessment expert. The specific method is as follows:

[0045] Obtain the assessment projects jointly participated by each candidate assessment expert and the experts that have cooperated with our company. If the jointly participated assessment project is a project of our company, it will be marked as a parent project. If the jointly participated assessment project is not a project of our company, it will be marked as a friendly project.

[0046] The number of parent projects and competitor projects that each candidate expert has participated in in the past two years are marked as Bj and Yj respectively;

[0047] For example, there is no direct cooperation between E3 and candidate evaluation expert E5, but candidate evaluation expert E3 and expert E5 have jointly evaluated 12 bids for friendly projects in the past two years. Therefore, candidate evaluation expert E3 has participated in 0 parent projects and 12 friendly projects in the past two years.

[0048] The sum of the square of the number of parent projects and the number of friendly projects of each candidate evaluation expert and the product of the preset parameter factor coefficients θ1 and θ2 is defined as the indirect trust index Jj corresponding to each candidate evaluation expert, that is, through the formula: Jj = (Bj) 2 ×θ1+Yj×θ2, calculate and obtain the indirect trust index Jj corresponding to each candidate evaluation expert, where θ1 and θ2 are both preset parameter factor coefficients, satisfying 1=θ1+θ2, θ1>θ2;

[0049] The trust probability acquisition module obtains the work history of each candidate assessment expert, and analyzes and obtains the trust probability corresponding to each candidate assessment expert based on the working hours and the number of times they participated in the assessment in their work history. The specific method is as follows:

[0050] It should be noted that work experience information needs to be obtained from multiple channels, such as resumes submitted by experts, past review records, and industry databases;

[0051] The working hours and the number of bid evaluations of each candidate evaluation expert are marked as Gj and Cj respectively, and the sum of the products of the working hours and the number of bid evaluations with the preset proportional coefficients β1 and β2 respectively is defined as the trust probability Rj corresponding to each candidate evaluation expert, that is, the trust probability Rj corresponding to each candidate evaluation expert is calculated by the formula: Rj=Gj×β1+Cj×β2, where β1 and β2 are both preset proportional coefficients, satisfying 1=β1+β2, β2>β1;

[0052] The comprehensive trust index acquisition module calculates the comprehensive trust index corresponding to each candidate evaluation expert according to the direct trust index, indirect trust index and trust probability corresponding to each candidate evaluation expert. The specific method is as follows:

[0053] By formula: Zj=(Hj) 2 ×θ3+Jj×θ4+Rj×θ5, calculate and obtain the comprehensive trust index Zj corresponding to each candidate evaluation expert, where θ3, θ4 and θ5 are preset parameter factor coefficients, satisfying 1=θ3+θ4+θ5, θ3>θ5>θ4;

[0054] The assessment expert recommendation module selects the assessment experts according to the comprehensive trust index corresponding to each candidate assessment expert. The specific method is as follows:

[0055] According to the comprehensive trust index values ​​corresponding to each candidate evaluation expert, the candidate evaluation experts are sorted in descending order, and then a recommendation list of evaluation experts is generated. Based on the recommendation list of evaluation experts, the evaluation experts required for the bid to be evaluated are recommended;

[0056] Through the collaborative work of multiple modules, a comprehensive evaluation and recommendation of bid evaluation experts is achieved, which not only takes into account the direct cooperation experience between experts and enterprises, but also takes into account their cooperation experience and professional capabilities in a wider range of fields. Enterprises can select suitable evaluation experts to evaluate bids based on the recommendation list, which provides strong support for enterprises to select high-quality evaluation experts. This method simplifies the expert selection process, improves the efficiency and accuracy of the selection, ensures that high-quality experts are more likely to participate in the bid evaluation, and forms a virtuous circle to improve the accuracy of the bid evaluation results.

[0057] Through the historical cooperation data acquisition module, indirect trust index acquisition module, trust probability acquisition module and comprehensive trust index acquisition module, as well as the evaluation expert recommendation module and the same cluster bid determination module, a comprehensive evaluation of expert capabilities and efficient expert recommendation are achieved, thereby improving the quality and efficiency of the entire bid evaluation process. By comprehensively considering the number of direct cooperation, the number of indirect cooperation and the expert's work experience, the trust of the expert can be evaluated more objectively, thereby reducing subjective bias, and more accurately evaluating the expert's professional ability and adaptability, thereby improving the accuracy and quality of bid evaluation. The automated recommendation system can quickly sort experts according to the comprehensive trust index, improving the efficiency of selecting experts.

[0058] Embodiment 2

[0059] As the second embodiment of the present invention, when the present application is implemented, compared with the first embodiment, the difference between the technical solution of the present embodiment and the first embodiment is that the present embodiment also includes a same clustering bid determination module;

[0060] The clustered bid calibration module, when it is necessary to evaluate multiple bids to be evaluated, obtains the direct project indicator types corresponding to the multiple bids to be evaluated, analyzes the number of direct project indicator types that are the same in each bid and other bids to be evaluated, obtains the correlation indicator coefficients between each bid and other bids according to the analysis results, analyzes the correlation indicator coefficients between each bid and other bids, and obtains the clustered bids corresponding to the multiple bids to be evaluated, the specific method is as follows:

[0061] A1: Randomly select one of the bids to be evaluated as the target evaluation bid, and mark the project indicator type corresponding to the target evaluation bid as L1;

[0062] By formula: Calculate the correlation coefficient J between the target evaluation bid and other bids (1，u) ;

[0063] Where L u L1∩Lu is the number of the same direct project indicator types corresponding to the target evaluation bid and the remaining bids to be evaluated, where u refers to the remaining bids to be evaluated except the target evaluation bid, u is a positive integer, u≥1, and the denominator here takes the maximum number of elements in the two sets to balance the difference in the number of indicators between different bids. In order to avoid the contradiction that the common project indicator type between the target evaluation bid and the remaining bids to be evaluated is 0, the number of common project indicator types between the target evaluation bid and all the remaining bids to be evaluated is initially 1;

[0064] A2: The bids to be evaluated whose correlation index coefficient with the target evaluation bid is greater than the preset value YS1 are marked together with the target evaluation bid as the same cluster bid W1; for the bids to be evaluated whose correlation index coefficient is less than or equal to the preset value YS1, no processing is performed, and the specific value of the preset value YS1 is determined by relevant personnel according to actual needs;

[0065] A3: Select any one of the bids to be evaluated whose relevant index coefficient is less than or equal to the preset value YS1 as the target bid, repeat steps A1-A2, and obtain bids Wr of the same cluster corresponding to the multiple bids to be evaluated, where r refers to different clusters, r is a positive integer, and r≥1;

[0066] When it is necessary to evaluate multiple bids to be evaluated, the same cluster bids will be evaluated uniformly according to the corresponding same cluster bids Wr in the multiple bids to be evaluated. This is conducive to eliminating the interference of other factors and uniformly evaluating the same cluster bids, which is conducive to improving the evaluation efficiency of multiple bids to be evaluated.

[0067] Embodiment 3

[0068] As the third embodiment of the present invention, when the present application is specifically implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments mentioned above for implementation.

[0069] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0070] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. Expert bid evaluation system based on relevant indicator data, characterized by: include: The historical cooperation data acquisition module obtains the number of direct and indirect cooperation between the enterprise and each candidate assessment expert; A direct trust index acquisition module determines the direct trust index between each candidate assessment expert and the enterprise according to the number of direct cooperation between the enterprise and each candidate assessment expert; The indirect trust index acquisition module counts the assessment projects jointly participated by each candidate assessment expert and the company's existing cooperative experts, distinguishes the number of parent and friendly business projects and analyzes them, and obtains the indirect trust index between each candidate assessment expert and the enterprise; The trust probability acquisition module analyzes the working hours and the number of times each candidate evaluation expert has participated in the evaluation, and obtains the trust probability corresponding to each candidate evaluation expert; The comprehensive trust index acquisition module integrates the direct trust index, indirect trust index and trust probability to obtain the comprehensive trust index corresponding to each candidate evaluation expert; The evaluation expert recommendation module generates a recommendation list by sorting the comprehensive trust index to recommend the evaluation experts required for the evaluation of the bid; The clustered bid calibration module obtains the correlation index coefficients between each bid and other bids based on the direct project index categories of multiple bids to be evaluated, and calibrates each clustered bid corresponding to the multiple bids to be evaluated according to the correlation index coefficients.

2. The expert bid evaluation system based on relevant indicator data according to claim 1 is characterized in that: The specific method for determining the direct trust index between each candidate assessment expert and the enterprise is as follows: The number of direct cooperation between each candidate assessment expert and the enterprise is obtained, and the number of direct cooperation is defined as the direct trust index Hj corresponding to each candidate assessment expert, where j refers to each candidate assessment expert, j is a positive integer, and j≥1.

3. The expert bid evaluation system based on relevant indicator data according to claim 2 is characterized in that: The specific method of obtaining the indirect trust index between each candidate evaluation expert and the enterprise is as follows: The square of the number of parent projects and the number of friendly projects participated in by each candidate assessment expert in the past two years are marked as Bj and Yj respectively. The sum of the products of the number of parent projects and the number of friendly projects of each candidate assessment expert and the preset parameter factor coefficients θ1 and θ2 is defined as the indirect trust index Jj corresponding to each candidate assessment expert, where 1=θ1+θ2, θ1>θ2.

4. The expert bid evaluation system based on relevant indicator data according to claim 3 is characterized in that: The specific method of obtaining the trustworthiness probability corresponding to each candidate evaluation expert is as follows: The sum of the products of the working hours and the number of times each candidate evaluation expert has participated in the evaluation and the preset proportional coefficients β1 and β2 is defined as the trust probability Rj corresponding to each candidate evaluation expert, where β1 and β2 are both preset proportional coefficients, satisfying 1=β1+β2, β2>β1.

5. The expert bid evaluation system based on relevant indicator data according to claim 4 is characterized in that: The specific method of obtaining the comprehensive trust index corresponding to each candidate evaluation expert is as follows: By formula: Zj=(Hj) 2 ×θ3+Jj×θ4+Rj×θ5, calculate and obtain the comprehensive trust index Zj corresponding to each candidate evaluation expert, where θ3, θ4 and θ5 are all preset parameter factor coefficients, satisfying 1=θ3+θ4+θ5, θ3>θ5>θ4.

6. The expert bid evaluation system based on relevant indicator data according to claim 5 is characterized in that: According to the comprehensive trust index values ​​corresponding to each candidate evaluation expert, each candidate evaluation expert is sorted in descending order, and then a recommendation list of evaluation experts is generated. Based on the recommendation list of evaluation experts, the evaluation experts required for the evaluation of the bid are recommended.

7. The expert bid evaluation system based on relevant indicator data according to claim 6 is characterized in that: The specific method of generating a recommendation list by sorting the comprehensive trust index to recommend the evaluation experts required for the evaluation of the bid is as follows: A1: Randomly select one of the bids to be evaluated as the target evaluation bid, and mark the project indicator type corresponding to the target evaluation bid as L1; By formula: Calculate the correlation index coefficient J between the target evaluation bid and other bids (1，u) ; where L u L1∩L u is the number of the same direct project indicator types corresponding to the target evaluation bid and the remaining bids to be evaluated, where u refers to the remaining bids to be evaluated except the target evaluation bid, u is a positive integer, u≥1; A2: Mark the bids to be evaluated whose correlation index coefficient with the target evaluation bid is greater than the preset value YS1 as the same cluster bid W1 together with the target evaluation bid; for the bids to be evaluated whose correlation index coefficient is less than or equal to the preset value YS1, no processing is performed; A3: Select any one of the bids to be evaluated whose relevant index coefficient is less than or equal to the preset value YS1 as the target bid, repeat steps A1-A2, and obtain the same cluster bids corresponding to the multiple bids to be evaluated.

8. The expert bid evaluation system based on relevant indicator data according to claim 3 is characterized in that: The specific way to distinguish the number of parent and friendly projects is as follows: Obtain the assessment projects jointly participated by each candidate assessment expert and the company's existing cooperative experts. When the jointly participated assessment project is a project of the company, it is marked as a parent project. When the jointly participated assessment project is not a project of the company, it is marked as a friendly company project. The number of parent projects and friendly company projects participated in by each candidate assessment expert in the past two years is marked as Bj and Yj respectively.

Images