Intelligent bidding and tendering management and control system based on big data

Abstract

The invention discloses an intelligent bidding and tendering management and control system based on big data, relates to the technical field of information management, and solves the problems that an enterprise is not responsible for a large project similar to a bidding project in scale; if the bid invitation project is submitted to the enterprise to be responsible, some risks may exist, so that the bid invitation project cannot be completed according to the quality on schedule; the method comprises the following steps: collecting basic information of a plurality of bidding enterprises, and preprocessing the basic information to obtain enterprise information; scoring the enterprise according to the enterprise information of the enterprise to obtain a credit score; screening the enterprises according to the credit scores to obtain candidate enterprises; bidding document information of the candidate enterprises is obtained, and bidding documents of the candidate enterprises are scored according to the bidding document information; obtaining risk data of the enterprise, and calculating a risk coefficient of the enterprise according to the risk data; calculating a bidding score of the enterprise according to the risk coefficient and the bidding document score; selecting a bid-winning enterprise according to the bidding score; and the accuracy of bidding scoring can be improved.

Description

Technical Field

[0001] The present invention belongs to the field of information management and relates to bidding management and control technology, and specifically is a smart bidding management and control system based on big data. Background Art

[0002] Bidding is a form of competition and transaction in economic activities such as engineering construction, goods trading, property leasing, and intermediary services under market economy conditions; strict supervision of bidding can reduce the occurrence of secret operations; through public information release and standardized processes, all participants can clearly understand the requirements, standards, and review process of the project; control over bidding can improve fairness, stimulate market vitality, and encourage more outstanding companies to participate in bidding; standardized and unified bidding processes help ensure that all participants are treated equally and avoid unfairness caused by information asymmetry; strict review standards and quality control can ensure that the winning project meets or even exceeds the expected quality requirements; through effective management and control, quality problems can be discovered and resolved in a timely manner, reducing the risk of project failure; standardized bidding processes can shorten the time for project preparation and implementation and improve work efficiency; the application of information technology makes the bidding process more convenient and efficient, reducing delays and errors caused by human factors.

[0003] The prior art (invention patent with application number 201911151845X) discloses an analysis method for irregular behaviors in the bidding process based on AI technology, the method comprising: step one, pre-building a learning model based on the behavioral sample data in the bidding process, during the construction period, the results given will be based on the continuously input data, and the results of machine learning will be manually adjusted and optimized, and the behavioral data content and weight values ​​will be optimized to make the prediction results closer to the real situation; step two, importing various relevant behavioral data for result prediction, by analyzing all behaviors in the bidding process and drawing numerical conclusions, it is possible to analyze all behaviors in the bidding process and quickly and efficiently give accurate analysis process data and accurate analysis results; in the bidding process, possible irregular behaviors can be predicted and monitored, and early warnings can be given in time to remind regulatory authorities to intervene as soon as possible.

[0004] Existing technology uses machine learning to analyze irregular behaviors in the bidding process and calculate the credit score of the enterprise. However, in the actual bidding process, there are many companies participating in the bidding. There may be some companies whose bids meet the bidding requirements, but the companies have not been responsible for large projects of similar scale to the bidding projects. If the bidding project is handed over to the company, there may be some risks, resulting in the bidding project not being completed on time and with quality.

[0005] The present invention provides a smart bidding and tendering management and control system based on big data to solve the above technical problems. Summary of the invention

[0006] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a smart bidding and tendering management and control system based on big data, which is used to solve the problem that in the actual bidding process, there are many companies participating in the bidding, and there may be a part of the companies whose bids meet the bidding requirements, but the companies have never been responsible for large projects with a similar scale to the bidding projects; if the bidding projects are handed over to the companies, there may be some risks, resulting in technical problems that the bidding projects cannot be completed on time and with quality.

[0007] To achieve the above-mentioned purpose, the first aspect of the present invention provides a smart bidding and tendering management and control system based on big data, comprising: a bidding and tendering management and control module, and an information collection module and a risk assessment module connected thereto;

[0008] Information collection module: used to collect basic information of several bidding companies through several channels, and pre-process the basic information to obtain company information;

[0009] Bidding control module: used to score enterprises according to their enterprise information to obtain credit scores; screen enterprises according to their credit scores to obtain candidate enterprises; obtain the tender information of candidate enterprises, score the tenders of candidate enterprises according to the tender information, and obtain tender scores;

[0010] Risk assessment module: used to obtain the risk data of the enterprise, calculate the risk coefficient of the enterprise based on the risk data; calculate the bid score of the enterprise based on the risk coefficient and the bid score; and select the winning enterprise based on the bid score.

[0011] Preferably, the preprocessing of basic information to obtain enterprise information includes:

[0012] Retrieve the basic information of the enterprise, including: enterprise name, business scope, historical bidding records, contract performance and financial statement information;

[0013] Use the deduplication function in the data screening tool to delete duplicate information in the basic information; supplement the missing values ​​in the basic information; use statistical methods to identify outliers in the basic information, delete outliers that do not conform to logical thinking, and retain outliers that conform to logical thinking; mark the basic information after processing as corporate information.

[0014] It should be noted that the missing values ​​in the basic information will be supplemented. If there are too many missing values ​​and the error in supplementing is large, the basic information will be re-collected; illogical thinking refers to values ​​that exceed the set range or are not in line with common sense.

[0015] The present invention pre-processes the basic information of an enterprise, can process missing values ​​and abnormal values ​​in the basic information, can ensure the integrity, rationality and effectiveness of the enterprise information, and lays a foundation for subsequent analysis of the enterprise credit.

[0016] Preferably, scoring the enterprise based on the enterprise information to obtain a credit score includes:

[0017] Retrieve the enterprise information of the enterprise and mark the enterprise information as a credit assessment sequence;

[0018] The credit scoring model is called, and the credit assessment sequence is input into the credit scoring model to obtain the credit score of the enterprise; wherein the credit scoring model is constructed based on an artificial intelligence model.

[0019] The present invention utilizes an artificial intelligence model to score the credit of an enterprise, which can comprehensively consider various factors of the enterprise, conduct a comprehensive analysis of the credit of the enterprise, and lay the foundation for the subsequent preliminary screening of bidding enterprises.

[0020] Preferably, the credit scoring model is constructed based on an artificial intelligence model, including:

[0021] Acquire a standard data set; wherein the standard data set includes standard input data consistent with the content attributes of the credit assessment sequence, and standard output data consistent with the content attributes of the credit score;

[0022] Divide the standard data set into a training set, a validation set, and a test set according to a set ratio; use the training set to train the artificial intelligence model; use the validation set to adjust the internal parameters of the trained artificial intelligence model; use the test set to test the adjusted artificial intelligence model to obtain test indicators;

[0023] Determine whether all test indicators are greater than the indicator threshold; if yes, mark the artificial intelligence model as a credit scoring model; if no, retrain the artificial intelligence model; wherein the artificial intelligence model includes a BP neural network model or a convolutional neural network model.

[0024] It should be noted that the division ratio of the standard data set is set based on expert evaluation. When the artificial intelligence model construction fails and needs to be rebuilt, the division ratio of the standard data set will be adjusted; the test indicators include accuracy, recall rate, specificity, F1 score and stability.

[0025] Preferably, the screening of enterprises according to credit scores to obtain candidate enterprises includes:

[0026] Retrieving the credit score of the enterprise; obtaining the score threshold; comparing the credit score of the enterprise with the score threshold;

[0027] When the credit score of an enterprise is greater than the score threshold, the corresponding enterprise is marked as a candidate enterprise; otherwise, the corresponding enterprise is eliminated.

[0028] The present invention compares the credit score of an enterprise with a score threshold. When the credit score of an enterprise is not greater than the score threshold, the bidding enterprise is eliminated, which can exclude enterprises with low credit, and is conducive to ensuring the normal progress of the bidding project.

[0029] Preferably, scoring the bids of candidate enterprises according to the bid information to obtain a bid score includes:

[0030] Retrieve the tender information of the candidate enterprises, conduct a preliminary review of the tenders of the candidate enterprises based on the tender information, and determine whether the standardization of the tenders meets the standards; if yes, mark the corresponding tender as preliminarily qualified; if no, mark the corresponding tender as unqualified;

[0031] Retrieve preliminary qualified bid information; obtain bidding requirements; classify bid information according to bidding requirements to obtain bid evaluation items; the bid evaluation items include: price scoring, technical solution scoring and after-sales service scoring;

[0032] Scoring experts are selected to score the preliminarily qualified bids according to the bid evaluation items to obtain item-by-item scores; the item-by-item scores are integrated according to the set weights to obtain the final scores; the final scores are integrated according to the expert weights to obtain the bid scores of the corresponding bids; among which, the expert weights are set according to the qualifications of the experts.

[0033] It should be noted that the method of selecting scoring experts is determined according to the bidding project. When the bidding project is a general project, the scoring experts are selected by random selection; when the bidding project is highly professional, experts are selected from the list of experts in the corresponding technical profession to score the bids.

[0034] The present invention conducts a preliminary analysis on the standardization of the enterprise's bid documents, screens out bid documents that meet the standards, determines the scoring items of the bid documents, and extracts experts to score the qualified bid documents; the final score of the bid document is integrated according to the expert weight to obtain the bid score of the corresponding bid document, and can score various information in the bid document, ensuring the relative fairness and standardization of the scoring criteria, which is conducive to the accuracy of the bid document scoring.

[0035] Preferably, the expert weight is set according to the expert's qualifications, including:

[0036] Obtain the qualifications of the experts; sort the experts according to their qualifications to obtain an expert ranking table; obtain the number of experts n according to the expert ranking table; determine the weight standard value as 1 / n according to the number of experts;

[0037] Set the expert weight function to f(x) = kx; according to the equation Solving for k, we get Substitute k into the weight function The corresponding expert weights are calculated according to the weight function; where x is the order of the expert in the expert ranking table.

[0038] It should be noted that the order of expert ranking is to rank the expert qualifications from low to high; the higher the expert's qualifications, the larger the corresponding x in the weight function, and the higher the expert weight.

[0039] The present invention ranks experts according to their qualifications, sets a weight function for expert weights, and after solving k in the weight function, determines the expert scores according to the serial numbers in the expert ranking table; the expert weights can be flexibly adjusted according to the number of experts, which is conducive to making the bid scoring more realistic and improving the relative fairness of the bid scoring.

[0040] Preferably, the step of calculating the risk coefficient of the enterprise according to the risk data includes:

[0041] Retrieve the enterprise's risk data; mark the capability difference, capital difference, and enterprise scale difference in the risk data as NC, ZC, and GC respectively;

[0042] The risk coefficient of an enterprise is calculated by the formula FX=α×NC^2+β×ln(ZC+1)+γ×GC; among them, α, β and γ are all proportional coefficients greater than 0.

[0043] It should be noted that the proportional coefficient is set artificially based on actual conditions and experience; when the bidding project has high technical requirements for the enterprise, the focus is on the enterprise's capabilities; when the bidding project has high project cycle requirements for the enterprise, the focus is on the enterprise's scale; when the bidding project has high capital requirements for the enterprise, the focus is on the enterprise's funds.

[0044] The present invention acquires and analyzes the deficiencies of enterprises in terms of capabilities, funds, and scale, calculates the risk coefficient of the enterprise, and can take into account the actual situation of the bidding enterprise itself, laying a foundation for the subsequent calculation of the final bidding score of the enterprise.

[0045] Preferably, the step of calculating the enterprise's bid score based on the risk factor and the bid document score includes:

[0046] Retrieve the enterprise's risk factor and corresponding bid score; mark the enterprise's bid score as BF;

[0047] The enterprise's bid score is calculated using the formula TF=BF-θ×FX; where θ is an adjustment coefficient greater than 0.

[0048] Preferably, the step of selecting a successful bidder based on bid scores includes:

[0049] Retrieve the bidding scores of the enterprises; sort the enterprises from high to low according to the bidding scores to obtain a ranking table of bidding enterprises;

[0050] The first-ranked enterprise will be selected as the winning enterprise according to the ranking list of bidding enterprises; a project contract will be signed with the winning enterprise, and the bidding results will be announced publicly.

[0051] The present invention obtains the bidding score of the enterprise by comprehensive calculation based on the risk coefficient and the bid document score of the enterprise; selects the enterprise with the highest bidding score as the winning enterprise according to the bidding score of the enterprise, and after the winning enterprise signs the project contract, the bidding result is publicized; it is conducive to ensuring that all aspects of the winning enterprise meet the standards, and the bidding results are publicized to ensure the transparency of the bidding process.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The present invention collects basic information of bidding enterprises, pre-processes the basic information of enterprises, can process missing values ​​and abnormal values ​​in the basic information, can ensure the integrity, rationality and effectiveness of enterprise information, and lay the foundation for subsequent analysis of enterprise credit; uses artificial intelligence models to score the credit of enterprises, can comprehensively consider various factors of enterprises, conduct comprehensive analysis of the credit of enterprises, and lay the foundation for subsequent preliminary screening of bidding enterprises; compares the credit score of enterprises with the scoring threshold, and when the credit score of enterprises is not greater than the scoring threshold, the bidding enterprises are eliminated, which can exclude enterprises with low credit, which is conducive to ensuring the normal progress of bidding projects.

[0054] 2. The present invention conducts a preliminary analysis on the standardization of the enterprise's bid documents, screens out bid documents that meet the standards, determines the scoring items of the bid documents, and extracts experts to score qualified bid documents; the final score of the bid document is integrated according to the expert weight to obtain the bid document score of the corresponding bid document, and can score various information in the bid document to ensure the relative fairness and standardization of the scoring criteria, which is conducive to the accuracy of the bid document scoring; the experts are sorted according to their qualifications, a weight function of the expert weight is set, and after solving k in the weight function, the expert score is determined according to the serial number in the expert sorting table; the expert weight can be flexibly adjusted according to the number of experts, which is conducive to Make the bid scoring more realistic and improve the relative fairness of the bid scoring; obtain and analyze the deficiencies of the enterprise in terms of capacity, capital, and scale, calculate the enterprise's risk coefficient, and take into account the actual situation of the bidding enterprise itself, laying the foundation for the subsequent calculation of the enterprise's final bid score; obtain the enterprise's bid score through comprehensive calculation based on the enterprise's risk coefficient and bid score; select the enterprise with the highest bid score as the winning enterprise based on the enterprise's bid score, and after the winning enterprise signs the project contract, publicize the bidding results; it is conducive to ensuring that all aspects of the winning enterprise meet the standards, and publicizing the bidding results can ensure the transparency of the bidding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 A schematic diagram of the overall steps of the system of the present invention;

[0057] Figure 2 A schematic diagram of the enterprise credit evaluation steps of the present invention;

[0058] Figure 3 A schematic diagram of the steps for scoring enterprise bids of the present invention;

[0059] Figure 4 It is a schematic diagram of the enterprise risk assessment and successful bidder screening steps of the present invention. DETAILED DESCRIPTION

[0060] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0061] See also Figure 1 , the first aspect of the present invention provides a smart bidding and tendering management and control system based on big data, including: a bidding and tendering management and control module, and an information collection module and a risk assessment module connected thereto;

[0062] Information collection module: used to collect basic information of several bidding companies through several channels, and pre-process the basic information to obtain company information;

[0063] Bidding control module: used to score enterprises according to their enterprise information to obtain credit scores; screen enterprises according to their credit scores to obtain candidate enterprises; obtain the tender information of candidate enterprises, score the tenders of candidate enterprises according to the tender information, and obtain tender scores;

[0064] Risk assessment module: used to obtain the risk data of the enterprise, calculate the risk coefficient of the enterprise based on the risk data; calculate the bid score of the enterprise based on the risk coefficient and the bid score; and select the winning enterprise based on the bid score.

[0065] See also Figure 2 , collect the basic information of several bidding companies through several channels, among which the basic information includes: company name, business scope, historical bidding records, contract performance and financial statement information; use the deduplication function in the data screening tool to delete the duplicate information in the basic information; supplement the missing values ​​in the basic information; use statistical methods to identify outliers in the basic information, delete the outliers that do not conform to logical thinking, and retain the outliers that conform to logical thinking; mark the basic information after processing as company information.

[0066] It should be noted that several channels include websites, mini-programs, and bidding-related applications.

[0067] For example: suppose that a project of a certain bidding company needs to be tendered externally. After the bidding company publishes the bidding notice, it obtains the basic information of several bidding companies through several channels; uses the deduplication function in the data screening tool to delete duplicate information in the basic information of the company, supplements or re-collects missing values ​​in the historical bidding records, contract performance status and financial statement information in the basic information of the company; deletes or retains outliers; and marks the basic information after processing as company information.

[0068] Retrieve the enterprise information of the enterprise and mark the enterprise information as a credit assessment sequence; call the credit scoring model, input the credit assessment sequence into the credit scoring model, and obtain the credit score of the enterprise; obtain the scoring threshold; compare the credit score of the enterprise with the scoring threshold; when the credit score of the enterprise is greater than the scoring threshold, mark the corresponding enterprise as a candidate enterprise; otherwise, eliminate the corresponding enterprise.

[0069] For example: suppose that the credit of several bidding companies is to be evaluated, and the company information of several bidding companies is retrieved, and the company information is marked as several credit evaluation sequences; the credit scoring model is called, and the credit evaluation sequence is input into the credit scoring model to obtain the credit score of the company, and after comparing the credit score of the company with the scoring threshold, it is concluded that the credit scores of companies D, F and G are less than the scoring threshold, then companies D, F and G are eliminated; the remaining companies are marked as candidate companies.

[0070] It is worth noting that the credit scoring model is built based on artificial intelligence models, including:

[0071] Acquire a standard data set; wherein the standard data set includes standard input data consistent with the content attributes of the credit assessment sequence, and standard output data consistent with the content attributes of the credit score;

[0072] Divide the standard data set into a training set, a validation set, and a test set according to a set ratio; use the training set to train the artificial intelligence model; use the validation set to adjust the internal parameters of the trained artificial intelligence model; use the test set to test the adjusted artificial intelligence model to obtain test indicators;

[0073] Determine whether all test indicators are greater than the indicator threshold; if yes, mark the artificial intelligence model as a credit scoring model; if no, retrain the artificial intelligence model; wherein the artificial intelligence model includes a BP neural network model or a convolutional neural network model.

[0074] It should be noted that the division ratio of the standard data set is set according to expert evaluation. When the artificial intelligence model construction fails and needs to be rebuilt, the division ratio of the standard data set will be adjusted; the test indicators include accuracy, recall rate, specificity, F1 score and stability; the indicator threshold is set according to expert evaluation, and the constructed credit evaluation model needs to be optimized and adjusted regularly, which is conducive to ensuring the accuracy of the credit evaluation model analysis.

[0075] See also Figure 3, obtain the bid information of the candidate enterprises, conduct a preliminary review of the bids of the candidate enterprises based on the bid information, and determine whether the standardization of the bids meets the standards; if yes, mark the corresponding bid as preliminarily qualified; if not, mark the corresponding bid as unqualified; retrieve the preliminarily qualified bid information; obtain the bidding requirements; classify the bid information according to the bidding requirements to obtain the bid evaluation items; among which, the bid evaluation items include: price scoring, technical solution scoring and after-sales service scoring; select scoring experts to score the preliminarily qualified bids according to the bid evaluation items to obtain item-by-item scores; integrate the item-by-item scores according to the set weights to obtain the final score.

[0076] It should be noted that the standards for preliminary review are set by the bidding company based on actual conditions; for example, the standards for preliminary review include whether the supplier meets the requirements, whether the bidding documents are complete, whether the calculations in the bidding documents are correct, etc.

[0077] For example, suppose that a preliminary review of the bids of several candidate companies is conducted, and the integrity, suppliers, and calculations of the bid documents of several companies are judged and analyzed. It is found that company B and company H do not meet the standards. The preliminary review results of company B and company H are marked as unqualified; the preliminary review results of the remaining candidate companies are marked as preliminarily qualified;

[0078] The tender information of the remaining candidate companies is retrieved, and the tender information is divided according to price, technical solution and after-sales service to obtain the evaluation items. Since the project has high technical requirements, experts are randomly selected from the expert list to score the evaluation items of the tender item one by one. According to the bidding project, the weight of each item is set to 0.5 for the technical solution score, 0.3 for the price score, and 0.2 for the after-sales service score. The scores of each item are integrated according to the set proportions to obtain the final expert score for the company.

[0079] Obtain the qualifications of the experts; sort the experts according to their qualifications to obtain an expert ranking table; obtain the number of experts n according to the expert ranking table; determine the weight standard value as 1 / n according to the number of experts;

[0080] Set the expert weight function to f(x) = kx; according to the equation Solving for k, we get Substitute k into the weight function The corresponding expert weights are calculated according to the weight function; wherein x is the order of the experts in the expert ranking table; the final scores are integrated according to the expert weights to obtain the bid scores of the corresponding bids.

[0081] For example: suppose that the bid of enterprise A is to be scored; 4 experts are selected from the expert list to score the bid of enterprise A; the 4 experts are sorted according to their qualifications to obtain an expert ranking table, and the weight of each expert is analyzed according to the weight function; the expert weight of expert 1 is 0.1, the expert weight of expert 2 is 0.2, the expert weight of expert 3 is 0.3, and the expert weight of expert 4 is 0.4;

[0082] The final scores of the bid of Company A from four experts were: Expert 1: 92, Expert 2: 90, Expert 3: 96, Expert 4: 93; the final score of Company A was 93.2.

[0083] See also Figure 4 , obtain the risk data of the enterprise; mark the capacity difference, capital difference and enterprise scale difference in the risk data as NC, ZC and GC respectively; calculate the risk coefficient of the enterprise by the formula FX = α×NC^2+β×ln(ZC+1)+γ×GC; where α, β and γ are all proportional coefficients greater than 0;

[0084] Retrieve the enterprise's risk factor and the corresponding bid score; mark the enterprise's bid score as BF; calculate the enterprise's bid score using the formula TF = BF-θ×FX; where θ is an adjustment coefficient greater than 0;

[0085] Retrieve the bidding scores of the enterprises; sort the enterprises from high to low according to the bidding scores to obtain a ranking list of bidding enterprises; select the first-ranked enterprise as the winning enterprise according to the ranking list of bidding enterprises; sign a project contract with the winning enterprise and publicize the bidding results.

[0086] It should be noted that in this embodiment, the maximum values ​​of the capability difference, capital difference and scale difference are set to 10. After evaluating the capability, capital and scale of the enterprise, the gap between the enterprise capability and the bidding requirements is analyzed, and the capability difference, capital difference and scale difference of the enterprise are obtained based on empirical analysis; the specific difference setting can be set according to the specific bidding project and the information of most bidding companies; the maximum value setting varies depending on the bidding project.

[0087] For example, suppose that the risks of Enterprise A and Enterprise C among the remaining candidate enterprises are evaluated, and the risk data of Enterprise A and Enterprise C are obtained. Based on experience, α, β and γ are set to 1, 10 and 5 respectively; the specific calculation steps are as follows:

[0088] Enterprise A: Capacity difference NC = 4; capital difference ZC = 6; enterprise scale difference GC = 3; then the risk coefficient of enterprise A FX = 41.98;

[0089] Enterprise C: Capacity difference NC = 2; capital difference ZC = 1; enterprise scale difference GC = 2; then the risk coefficient of enterprise A FX = 20.93;

[0090] Calculate the bid scores of Enterprise A and Enterprise C, set θ to 0.5; retrieve the bid scores of Enterprise A and Enterprise C as 93.2 and 90.8 respectively; then calculate that the bid score of Enterprise A is 77.21; the bid score of Enterprise C is 80.335;

[0091] According to the above steps, the risk coefficient is calculated according to the risk data of the enterprise and the bid score of the enterprise is calculated; the enterprises are sorted according to their bid scores to obtain a ranking table of bidding enterprises; the enterprise K ranked first in the ranking table of bidding enterprises is selected as the winning enterprise; the bidding company signs a contract with the winning enterprise K and publicizes the bidding results.

[0092] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0093] The working principle of the present invention is as follows: the present invention collects basic information of several bidding enterprises through several channels, pre-processes the basic information to obtain enterprise information; scores the enterprises according to the enterprise information to obtain credit scores; screens the enterprises according to the credit scores to obtain candidate enterprises; obtains the tender information of the candidate enterprises, scores the tenders of the candidate enterprises according to the tender information to obtain tender scores; obtains the risk data of the enterprises, calculates the risk coefficient of the enterprises according to the risk data; calculates the bid score of the enterprises according to the risk coefficient and the bid score; and selects the winning enterprise according to the bid score.

[0094] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A smart bidding and tendering management and control system based on big data, characterized in that: Bidding control module, and the information collection module and risk assessment module connected to it; Information collection module: used to collect basic information of several bidding companies through several channels, and pre-process the basic information to obtain company information; Bidding control module: used to score enterprises based on their corporate information and obtain credit scores; Screen enterprises according to their credit scores to obtain candidate enterprises; Obtaining the bid information of the candidate enterprises, scoring the bids of the candidate enterprises according to the bid information, and obtaining the bid scores; Risk assessment module: used to obtain the enterprise's risk data and calculate the enterprise's risk coefficient based on the risk data; Calculate the enterprise's bid score based on the risk factor and bid score; The winning bidder will be selected based on the bid scores.

2. According to the big data-based intelligent bidding and tendering management system of claim 1, it is characterized in that: The preprocessing of basic information to obtain enterprise information includes: Retrieve the basic information of the enterprise, including: enterprise name, business scope, historical bidding records, contract performance and financial statement information; Use the deduplication function in the data screening tool to delete duplicate information in the basic information; supplement the missing values ​​in the basic information; use statistical methods to identify outliers in the basic information, delete outliers that do not conform to logical thinking, and retain outliers that conform to logical thinking; mark the basic information after processing as corporate information.

3. According to the big data-based intelligent bidding and tendering management system of claim 1, it is characterized in that: Scoring the enterprise based on the enterprise information to obtain a credit score includes: Retrieve the enterprise information of the enterprise and mark the enterprise information as a credit assessment sequence; The credit scoring model is called, and the credit assessment sequence is input into the credit scoring model to obtain the credit score of the enterprise; wherein the credit scoring model is constructed based on an artificial intelligence model.

4. According to the big data-based intelligent bidding and tendering management system of claim 3, it is characterized in that: The credit scoring model is constructed based on an artificial intelligence model, including: Acquire a standard data set; wherein the standard data set includes standard input data consistent with the content attributes of the credit assessment sequence, and standard output data consistent with the content attributes of the credit score; Divide the standard data set into a training set, a validation set, and a test set according to a set ratio; use the training set to train the artificial intelligence model; use the validation set to adjust the internal parameters of the trained artificial intelligence model; use the test set to test the adjusted artificial intelligence model to obtain test indicators; Determine whether all test indicators are greater than the indicator threshold; if yes, mark the artificial intelligence model as a credit scoring model; if no, retrain the artificial intelligence model; wherein the artificial intelligence model includes a BP neural network model or a convolutional neural network model.

5. According to the big data-based intelligent bidding and tendering management system of claim 3, it is characterized in that: The enterprises are screened according to the credit scores to obtain candidate enterprises, including: Retrieving the credit score of the enterprise; obtaining the score threshold; comparing the credit score of the enterprise with the score threshold; When the credit score of an enterprise is greater than the score threshold, the corresponding enterprise is marked as a candidate enterprise; otherwise, the corresponding enterprise is eliminated.

6. According to the big data-based intelligent bidding and tendering management system of claim 1, it is characterized in that: Scoring the bids of candidate enterprises according to the bid information to obtain bid scores includes: Retrieve the tender information of the candidate enterprises, conduct a preliminary review of the tenders of the candidate enterprises based on the tender information, and determine whether the standardization of the tenders meets the standards; if yes, mark the corresponding tender as preliminarily qualified; if no, mark the corresponding tender as unqualified; Retrieve preliminary qualified bid information; obtain bidding requirements; classify bid information according to bidding requirements to obtain bid evaluation items; the bid evaluation items include: price scoring, technical solution scoring and after-sales service scoring; Scoring experts are selected to score the preliminarily qualified bids according to the bid evaluation items to obtain item-by-item scores; the item-by-item scores are integrated according to the set weights to obtain the final scores; the final scores are integrated according to the expert weights to obtain the bid scores of the corresponding bids; among which, the expert weights are set according to the qualifications of the experts.

7. According to the big data-based intelligent bidding and tendering management system of claim 6, it is characterized in that: The expert weight is set according to the expert's qualifications, including: Obtain the qualifications of the experts; sort the experts according to their qualifications to obtain an expert ranking table; obtain the number of experts n according to the expert ranking table; determine the weight standard value as 1 / n according to the number of experts; Set the expert weight function to f(x) = kx; according to the equation Solving for k, we get Substitute k into the weight function The corresponding expert weights are calculated according to the weight function; where x is the order of the expert in the expert ranking table.

8. The intelligent bidding and tendering management system based on big data according to claim 1 or 6, characterized in that: The step of calculating the risk coefficient of an enterprise according to risk data includes: Retrieve the enterprise's risk data; mark the capability difference, capital difference, and enterprise scale difference in the risk data as NC, ZC, and GC respectively; The risk coefficient of an enterprise is calculated by the formula FX=α×NC^2+β×ln(ZC+1)+γ×GC; among them, α, β and γ are all proportional coefficients greater than 0.

9. The intelligent bidding and tendering management system based on big data according to claim 8 is characterized in that: The calculation of the enterprise's bid score based on the risk factor and the bid document score includes: Retrieve the enterprise's risk factor and corresponding bid score; mark the enterprise's bid score as BF; The enterprise's bid score is calculated using the formula TF=BF-θ×FX; where θ is an adjustment coefficient greater than 0.

10. The intelligent bidding and tendering management system based on big data according to claim 1 is characterized in that: The selection of successful bidders based on bid scoring includes: Retrieve the bidding scores of the enterprises; sort the enterprises from high to low according to the bidding scores to obtain a ranking table of bidding enterprises; The first-ranked enterprise will be selected as the winning enterprise according to the ranking list of bidding enterprises; a project contract will be signed with the winning enterprise, and the bidding results will be announced publicly.

Images