CN115660774A - Material supply chain system credit evaluation method based on block chain - Google Patents

Abstract

A blockchain-based material supply chain system credit evaluation method of the present invention obtains supplier data; respectively obtains obvious risks and potential risks according to the supplier data; quantifies the obvious risks and potential risks respectively to obtain obvious risk grade scores and potential risk grade scores; the deep learning network is used to process the obvious risk grade scores to obtain the first evaluation result, and the subjective weighting method is used to weight the potential risk grade scores to obtain the second evaluation result; according to the first evaluation result and The second evaluation result obtains the credit evaluation result; the supplier data, obvious risks, potential risks, and credit evaluation results are stored on the blockchain through the blockchain, and the evaluation is performed through the obvious risks and potential risks to ensure the accuracy of the evaluation results. Data such as credit evaluation results can be stored on the chain, and the uplink and downlink of data through the blockchain can ensure the privacy, security and openness of the data and avoid tampering.

Description

A credit evaluation method for material supply chain system based on blockchain

technical field

The invention relates to the technical field of credit evaluation, in particular to a credit evaluation method for a material supply chain system based on a block chain.

Background technique

Affected by uncertain factors such as the construction environment and market changes, problems such as operational difficulties, delayed delivery, quality defects, and non-cooperation of suppliers in the process of contract performance have become increasingly prominent, causing risks such as project delays and capital losses, and affecting the economy. Benefits and grid security, but at present, post-event punishment is mainly carried out through bad behavior records, and it is impossible to achieve "foresight" and advanced control.

In the face of a large number of suppliers, due to the lack of scientific classification and objective quantification means, the material department is often unable to analyze each supplier's qualification characteristics, contract performance capabilities, and potential risks one by one. Contract performance risk, and after the risk is discovered, it cannot immediately formulate countermeasures, which will eventually lead to delays in the construction period or loss of assets.

Contents of the invention

In view of this, the present invention proposes a blockchain-based material supply chain system credit evaluation method, which can systematically conduct credit evaluation on suppliers, and store the credit evaluation results on the blockchain for other enterprises to obtain.

Technical scheme of the present invention is realized like this:

A blockchain-based material supply chain system credit evaluation method, comprising the following steps:

Step S1. Obtain supplier data, which includes basic information, historical performance information and change information;

Step S2, obtain the supplier's obvious risk according to the historical performance information, and obtain the supplier's potential risk according to the change information;

Step S3, respectively quantify the obvious risk and the potential risk, and obtain the obvious risk level score and the potential risk level score;

Step S4, use the deep learning network to process the obvious risk grade scores to obtain the first evaluation result, and use the subjective weighting method to perform weighted calculation on the potential risk grade scores to obtain the second evaluation result;

Step S5, obtaining a credit evaluation result according to the first evaluation result and the second evaluation result;

Step S6, store the supplier data, obvious risks, potential risks and credit evaluation results on the blockchain through the blockchain.

Preferably, the suppliers include material suppliers and service suppliers, the basic information includes enterprise type, enterprise nature, business scope, and establishment time, and the historical performance information includes whether the delivery time is on time, delivered products or services Whether the qualification rate and delivery quantity are accurate, the delivery time includes whether the delivery time is on time, whether the delivery time is on time, and whether the logistics is on time, and the change information includes changes in the number of employees, changes in floor space, changes in operating conditions, changes in personnel quality, technical and management level changes, engineering equipment changes.

Preferably, the specific steps of acquiring the supplier's apparent risk based on historical performance information in step S2 include:

Step S21, building an obvious risk database, the obvious risk database contains risk words, and each risk word corresponds to an obvious risk;

Step S22, extract keywords from historical performance information, perform synonymous expansion and approximate expansion on keywords, and obtain keyword combinations;

Step S23, traversing and comparing the keyword combination and the risk words in the obvious risk database, obtaining the risk word most appropriate to the keyword combination, and obtaining the obvious risk according to the risk word.

Preferably, the specific steps of obtaining the apparent risk grade scores and the potential risk grade scores in the step S2 include:

Step S24, constructing an enterprise change database, which includes the change thresholds of various information of the enterprise;

Step S25, calculating the variation according to the variation information;

Step S26 , comparing the variation with the variation threshold, and outputting the variation information corresponding to the variation greater than the variation threshold as a potential risk.

Preferably, the step S2 of obtaining the apparent risk grade score and the potential risk grade score further includes:

Step S27, obtaining credit evaluation results of upstream and downstream enterprises cooperating with suppliers, and obtaining potential risks according to the credit evaluation results of upstream and downstream enterprises.

Preferably, the specific steps of the step S3 are: constructing a quantitative score library, the quantitative score library includes all risks and the grade scores corresponding to the risks, searching for the corresponding risks from the quantitative score library according to the obvious risks and potential risks, and obtaining After the corresponding risk grade score, the obvious risk grade score and the potential risk grade score are output.

Preferably, the step S4 uses a deep learning network to process the apparent risk grade scores, and the specific steps for obtaining the first evaluation result include:

Step S41, building an enterprise risk type library, which includes enterprise risks, the range of risk scores corresponding to each enterprise risk, and enterprise evaluation results;

Step S42, embedding the enterprise risk type library into the deep learning network;

Step S43, the deep learning network identifies the obvious risk grade scores, searches for the enterprise risks corresponding to the risk score range including the obvious risk grade scores, and outputs and adds the enterprise evaluation results corresponding to the enterprise risks to obtain the first evaluation result.

Preferably, the step S4 adopts the subjective weighting method to perform weighted calculation on the potential risk grade scores, and the specific steps for obtaining the second evaluation result include:

Step S44, constructing an expert scoring library, which contains weight factors corresponding to each potential risk;

Step S45 , assigning the weighting factors in the expert scoring database to the potential risk grade scores, and adding up the assigned potential risk grade scores to obtain a second evaluation result.

Preferably, the specific steps of said step S5 include:

Step S51. When the first evaluation result is greater than the second evaluation result, output the first evaluation result as a credit evaluation result;

Step S52, when the second evaluation result is greater than the first evaluation result, multiply the first evaluation result and the second evaluation result by the corresponding reduction factor and then sum, and output the sum value as the credit evaluation result, the first evaluation result The downscaling factor of the result is smaller than the downscaling factor of the second evaluation result.

Preferably, the specific steps of the step S6 are: uploading the supplier data, obvious risks, potential risks and credit evaluation results through the supplier's node server to the chain, and through the blockchain middleware service software on the node server Symmetrically encrypt the data on the chain.

Compared with prior art, the beneficial effect of the present invention is:

The present invention provides a credit evaluation method for the material supply chain system based on blockchain. After collecting the basic information, historical performance information and change information of the supplier, the obvious risk and potential risk of the supplier itself are judged. After obtaining the obvious risks and potential risks, carry out specific quantitative scoring, and then use different methods to process the grade scores according to the risk category, and obtain the evaluation results of each chain, and finally obtain the credit evaluation results of the supplier based on the two evaluation results , and the data contained in the whole process, such as supplier data, obvious risks, potential risks and credit evaluation results, will be stored on the chain through the blockchain, and other enterprises can obtain the credit evaluation of the corresponding suppliers through the blockchain As a result, it is considered whether to cooperate with the company to achieve an objective and complete credit evaluation of suppliers in the supply chain system. At the same time, the way of going up and down the chain through the blockchain can ensure the objectivity, privacy and transparency of the data.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings that need to be used in the description of the embodiments. Obviously, the drawings in the following description are only preferred embodiments of the present invention. For those skilled in the art, other drawings can also be obtained based on these drawings without any creative effort.

Fig. 1 is an overall flow chart of a block chain-based material supply chain system credit evaluation method of the present invention;

Fig. 2 is a flow chart of step S2 of a method for credit evaluation of a blockchain-based material supply chain system of the present invention;

Fig. 3 is a flow chart of step S4 of a method for credit evaluation of a blockchain-based material supply chain system of the present invention;

Fig. 4 is a flow chart of step S5 of a credit evaluation method for a blockchain-based material supply chain system of the present invention.

Detailed ways

In order to better understand the technical content of the present invention, a specific embodiment is provided below, and the present invention is further described in conjunction with the accompanying drawings.

Referring to Figures 1 to 4, the present invention provides a blockchain-based material supply chain system credit evaluation method, including the following steps:

Step S1. Collect and obtain supplier data from various channels such as social credit platform and enterprise information inquiry network. The supplier data includes basic information, historical performance information and change information.

Among them, suppliers include material suppliers and service suppliers. For material suppliers, it is possible to judge whether the supplier has fulfilled the previous agreement with other enterprises based on the quality, quantity and logistics level of the materials. As far as the service category is concerned, credit evaluations can be made for the services it provides, including service time limits, service attitudes, and so on.

In the supplier data, the basic information includes the type of enterprise, the nature of the enterprise, the scope of business, and the time of establishment. The historical performance information includes whether the delivery time is on time, the qualified rate of the delivered product or service, and whether the delivery quantity is accurate. Whether it is on time includes whether the receipt and delivery are on time, whether the logistics is on time, and the change information includes changes in the number of employees, changes in floor space, changes in operating conditions, changes in personnel quality, changes in technology and management levels, and changes in engineering equipment.

The basic information can be used by other companies to understand the basic content of the supplier, and the historical performance information can be used to intuitively understand the cooperation between the supplier and the previous enterprises, so as to judge the credit of the supplier, and the change information can be seen. Changes in strength are used to judge whether subsequent suppliers have the corresponding ability to undertake the order needs of other companies, but the historical performance information and change information contain more content. If the company needs to check the historical performance information and changes of each supplier If the information is verified, more time will be wasted.

Step S2. Obtain the obvious risk of the supplier according to the historical performance information, and obtain the potential risk of the supplier according to the change information. The specific steps are:

Step S21, building an obvious risk database, the obvious risk database contains risk words, and each risk word corresponds to an obvious risk;

Step S22, extract keywords from historical performance information, perform synonymous expansion and approximate expansion on keywords, and obtain keyword combinations;

Step S23, traversing and comparing the keyword combination and the risk words in the obvious risk database, obtaining the risk word most appropriate to the keyword combination, and obtaining the obvious risk according to the risk word.

When obtaining obvious risks, it can be judged directly based on the content contained in historical performance information. For historical performance information, it will include whether the delivery time is on time, the pass rate of delivered products or services, and whether the delivery quantity is accurate. Whether it is on time or not will have punctuality rate, early rate and late rate, which will be reflected in the supplier's complete order information, and different suppliers have different descriptions of orders, so there will be literal differences in historical performance information. For this reason, after extracting keywords from historical performance information, it is necessary to uniformly expand and approximately expand the keywords, and finally obtain the corresponding keyword combinations. In the obvious risk database constructed, there will be risk words, risk Words such as "postponed", "delayed", "unqualified", "defective products", etc., by comparing the combination of keywords and risk words, you can find the closest risk words in the obvious risk database, so that you can Obtain the apparent risk corresponding to the risk word.

When evaluating a supplier's credit, in addition to considering the obvious risks of the supplier, it is also necessary to consider its potential risks. The potential risks are related to the supplier's recent change information, and the change information includes the number of employees. Changes, changes in floor area, changes in operating conditions, changes in personnel quality, changes in technology and management levels, and changes in engineering equipment. By considering the above data to obtain potential risks, step S2 also includes:

Step S24, constructing an enterprise change database, which includes the change thresholds of various information of the enterprise;

Step S25, calculating the variation according to the variation information;

Step S26 , comparing the variation with the variation threshold, and outputting the variation information corresponding to the variation greater than the variation threshold as a potential risk.

Step S27, obtaining credit evaluation results of upstream and downstream enterprises cooperating with suppliers, and obtaining potential risks according to the credit evaluation results of upstream and downstream enterprises.

Firstly, the enterprise change database is constructed according to the corresponding changes. In the enterprise change database, a change threshold is set for each change information in the business process of the enterprise. For example, the turnover of personnel in the past five years should not exceed 20%. After the change information is calculated to obtain the corresponding change amount, compare the change amount with the change threshold. When the change amount is greater than the change threshold, it means that the change information is a potential risk for the supplier. In addition, when considering the potential risk of the supplier , in addition to considering the supplier itself, the credit evaluation results of the upstream and downstream enterprises of the supplier are also considered. If the credit evaluation results of the upstream and downstream enterprises are poor, it may affect the normal order processing of the supplier, which will eventually affect Therefore, obtaining potential risks through multiple channels can ensure the accuracy of the supplier's credit evaluation results.

Step S3, constructing a quantitative score library, the quantitative score library includes all risks and the grade scores corresponding to the risks, according to the obvious risks and potential risks, look up the corresponding risks from the quantitative score library, and after obtaining the grade scores of the corresponding risks, output the obvious Risk Level Score and Potential Risk Level Score.

The final credit evaluation results are not grades or descriptions, but can be quantified by numbers. For this reason, after obtaining the obvious risks and potential risks, the obvious risks and potential risks are converted into corresponding grade scores through the quantitative score library, and the risk The grade score can be used for the calculation of subsequent credit evaluation results.

Step S4, using the deep learning network to process the obvious risk grade scores to obtain the first evaluation result, and using the subjective weighting method to perform weighted calculation on the potential risk grade scores to obtain the second evaluation result.

When calculating the final credit evaluation results, it is not simply to add up the evaluation results corresponding to the obvious risks and potential risks, but to make corresponding selections based on the comparison results. Scores are calculated for evaluation results.

When calculating the first evaluation result corresponding to the apparent risk grade score, a deep learning network is used for processing, and the specific steps include:

Step S41, building an enterprise risk type library, which includes enterprise risks, the range of risk scores corresponding to each enterprise risk, and enterprise evaluation results;

Step S42, embedding the enterprise risk type library into the deep learning network;

Step S43, the deep learning network identifies the obvious risk grade scores, searches for the enterprise risks corresponding to the risk score range including the obvious risk grade scores, and outputs and adds the enterprise evaluation results corresponding to the enterprise risks to obtain the first evaluation result.

The present invention uses a deep learning network to calculate the first evaluation result. Before the calculation, the enterprise risk type library is constructed. The enterprise risk type library contains many types of enterprise risks. The same risk will have many enterprise risks. The range of risk scores contained in an enterprise risk is different from the enterprise evaluation results. When the deep learning network processes the obvious risk grade scores, it first searches the corresponding enterprise risk in the enterprise risk type library according to the obvious risks corresponding to the obvious risk grade scores, and then judges In which risk score range the obvious risk level score falls, the first evaluation result can be correspondingly obtained.

The subjective weighting method is used to carry out weighted calculations on the potential risk grade scores, and the specific steps for obtaining the second evaluation result include:

Step S44, constructing an expert scoring library, which contains weight factors corresponding to each potential risk;

Step S45 , assigning the weighting factors in the expert scoring database to the potential risk grade scores, and adding up the assigned potential risk grade scores to obtain a second evaluation result.

For potential risks, the present invention uses expert scoring to select the weighting factor for each potential risk grade score, and the size of the weighting factor is specifically set according to the type of supply chain, and the weighting factor is assigned to the potential risk grade score After that, they are accumulated to obtain the second evaluation result.

The basis for expert scoring is based on the summary of historical data collected from a large number of supply chain companies and the reference weights obtained after induction. Using historical data as a reference to select the weighting factor can ensure that the selected weighting factor is consistent with Suppliers fit in with each other.

Step S5 comprehensively obtains credit evaluation results according to the first evaluation result and the second evaluation result.

Step S51. When the first evaluation result is greater than the second evaluation result, output the first evaluation result as a credit evaluation result;

Step S52, when the second evaluation result is greater than the first evaluation result, multiply the first evaluation result and the second evaluation result by the corresponding reduction factor and then sum, and output the sum value as the credit evaluation result, the first evaluation result The downscaling factor of the result is smaller than the downscaling factor of the second evaluation result.

For the first evaluation result and the second evaluation result, both are quantifiable data, while for the supplier’s credit evaluation, the potential risk is not the real defect of the supplier, while the obvious risk is the intuitive one. The real defects obtained from previous orders, for this reason, the priority of the first evaluation result is higher than the priority of the second evaluation result, if the first evaluation result is higher than the second evaluation result, the first evaluation result is used as the credit evaluation As a result, when the first evaluation result is smaller than the second evaluation result, the first evaluation result and the second evaluation result are correspondingly weakened and then summed, and the sum value is used as the credit evaluation result. When weakening, the used is the reduction factor, the reduction factor is a constant greater than 0 and less than 1, which can be actually set according to the type of supply chain, and the reduction factor of the first evaluation result is smaller than the reduction factor of the second evaluation result, so that when the two results are weakened, the second The degree of weakening of the second evaluation result is greater than that of the first evaluation result.

Step S6, upload the supplier data, obvious risks, potential risks and credit evaluation results to the chain through the supplier's node server, and perform symmetrical encryption on the chained data through the blockchain middleware service software on the node server.

After obtaining the credit evaluation results, all the relevant data involved in the whole process are stored on the chain, and the relevant enterprises can obtain the credit evaluation results of all suppliers in the supply chain through the blockchain, and can share their basic information, history Visually view performance information, change information, obvious risks and potential risks, etc., so as to understand suppliers in detail.

When the blockchain is used as the data storage and transmission medium, the privacy, openness and transparency of the data can be guaranteed, and there will be no situation where the supplier will tamper with the content on the chain at will.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of the present invention. within the scope of protection.

Claims (10)

1. A block chain-based material supply chain system credit evaluation method, characterized in that, comprising the following steps: Step S1. Obtain supplier data, which includes basic information, historical performance information and change information; Step S2, obtain the supplier's obvious risk according to the historical performance information, and obtain the supplier's potential risk according to the change information; Step S3, respectively quantify the obvious risk and the potential risk, and obtain the obvious risk level score and the potential risk level score; Step S4, use the deep learning network to process the obvious risk grade scores to obtain the first evaluation result, and use the subjective weighting method to perform weighted calculation on the potential risk grade scores to obtain the second evaluation result; Step S5, obtaining a credit evaluation result according to the first evaluation result and the second evaluation result; Step S6, store the supplier data, obvious risks, potential risks and credit evaluation results on the blockchain through the blockchain. 2. A blockchain-based material supply chain system credit evaluation method according to claim 1, wherein said suppliers include material suppliers and service suppliers, and said basic information includes enterprise type , the nature of the enterprise, the scope of business, and the time of establishment. The historical performance information includes whether the delivery time is on time, the qualified rate of delivered products or services, and whether the delivery quantity is accurate. On time, the change information includes changes in the number of employees, changes in floor area, changes in operating conditions, changes in personnel quality, changes in technology and management levels, and changes in engineering equipment. 3. A method for credit evaluation of a blockchain-based material supply chain system according to claim 1, wherein the specific steps of obtaining the apparent risk of the supplier according to historical performance information in the step S2 include: Step S21, building an obvious risk database, the obvious risk database contains risk words, and each risk word corresponds to an obvious risk; Step S22, extract keywords from historical performance information, perform synonymous expansion and approximate expansion on keywords, and obtain keyword combinations; Step S23, traversing and comparing the keyword combination and the risk words in the obvious risk database, obtaining the risk word most appropriate to the keyword combination, and obtaining the obvious risk according to the risk word. 4. A blockchain-based material supply chain system credit evaluation method according to claim 1, characterized in that the step S2 of obtaining obvious risk grade scores and potential risk grade scores comprises: Step S24, constructing an enterprise change database, which includes the change thresholds of various information of the enterprise; Step S25, calculating the variation according to the variation information; Step S26 , comparing the variation with the variation threshold, and outputting the variation information corresponding to the variation greater than the variation threshold as a potential risk. 5. A blockchain-based material supply chain system credit evaluation method according to claim 1, characterized in that the step S2 of obtaining the apparent risk grade score and the potential risk grade score further comprises: Step S27, obtaining credit evaluation results of upstream and downstream enterprises cooperating with suppliers, and obtaining potential risks according to the credit evaluation results of upstream and downstream enterprises. 6. A blockchain-based material supply chain system credit evaluation method according to claim 1, characterized in that, the specific steps of the step S3 are: constructing a quantitative score library, the quantitative score library includes all risk And the grade scores corresponding to the risks, according to the obvious risks and potential risks, find the corresponding risks from the quantitative score database, and after obtaining the grade scores of the corresponding risks, output the obvious risk grade scores and potential risk grade scores. 7. A blockchain-based material supply chain system credit evaluation method according to claim 1, wherein said step S4 uses a deep learning network to process the obvious risk grade scores to obtain the first evaluation result Specific steps include: Step S41, building an enterprise risk type library, which includes enterprise risks, the range of risk scores corresponding to each enterprise risk, and enterprise evaluation results; Step S42, embedding the enterprise risk type library into the deep learning network; Step S43, the deep learning network identifies the obvious risk grade scores, searches for the enterprise risks corresponding to the risk score range including the obvious risk grade scores, and outputs and adds the enterprise evaluation results corresponding to the enterprise risks to obtain the first evaluation result. 8. A blockchain-based material supply chain system credit evaluation method according to claim 1, wherein said step S4 uses a subjective weighting method to perform weighted calculations on potential risk grade scores to obtain the second evaluation Concrete steps for the results include: Step S44, constructing an expert scoring library, which contains weight factors corresponding to each potential risk; Step S45 , assigning the weighting factors in the expert scoring database to the potential risk grade scores, and adding up the assigned potential risk grade scores to obtain a second evaluation result. 9. A blockchain-based material supply chain system credit evaluation method according to claim 1, wherein the specific steps of step S5 include: Step S51. When the first evaluation result is greater than the second evaluation result, output the first evaluation result as a credit evaluation result; Step S52, when the second evaluation result is greater than the first evaluation result, multiply the first evaluation result and the second evaluation result by the corresponding reduction factor and then sum, and output the sum value as the credit evaluation result, the first evaluation result The downscaling factor of the result is smaller than the downscaling factor of the second evaluation result. 10. A blockchain-based material supply chain system credit evaluation method according to claim 1, characterized in that, the specific steps of the step S6 are: supplier data, obvious risks, potential risks and credit evaluation As a result, the data is uploaded to the chain through the supplier's node server, and the data on the chain is encrypted symmetrically through the blockchain middleware service software on the node server.

Images