Data processing system and method of one-product multi-supply aggregation platform

Abstract

The invention provides a data processing system and method for a one-product multi-supply aggregation platform, and the system comprises a data collection module which is responsible for collecting related information of products and suppliers from a plurality of data sources; the data standardization module is used for carrying out standardization processing on the acquired product and supplier data with different formats and different standards; the supplier management module is used for realizing full-life-cycle management of suppliers; the price comparison analysis module is used for performing one-product multi-supply price comparison based on the standardized data; the intelligent decision-making module is used for providing intelligent decision-making suggestions for enterprise purchase by applying a machine learning algorithm and a decision-making model; and the data visualization module is used for visually displaying data processing and analysis results in the form of various charts and reports. According to the invention, a scientific and reasonable purchase plan is made according to the actual situation, a purchase scheme with cost performance is obtained, the purchase cost is effectively reduced, problems of suppliers can be found in time, the purchase risk of enterprises is reduced, and the stability and reliability of a supply chain are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of procurement informatization management, and specifically provides a data processing system and method for a multi-supplier aggregation platform for one product. Background Art

[0002] In today's market environment, enterprise procurement involves numerous categories and a complex supplier system. On the one hand, enterprises need to find products that meet their own needs among the massive products provided by numerous suppliers. This process is time-consuming and laborious and is prone to missing high-quality options. For example, when a manufacturing enterprise purchases raw materials, it needs to screen out products with qualified quality and reasonable prices from numerous raw material suppliers. However, due to the large number of suppliers and diverse product specifications, it is difficult for enterprise procurement personnel to comprehensively evaluate and compare the products of all potential suppliers. On the other hand, enterprises have difficulty ensuring the reasonableness of the prices of the purchased products. They lack effective means to accurately compare the prices of different suppliers and are easily affected by price fluctuations and information asymmetry, resulting in high procurement costs.

[0003] Although some current procurement platforms or tools in the market provide certain product search and price comparison functions, they have many deficiencies. Most platforms can only achieve simple product name or keyword matching and have limited ability to identify and compare prices for non-standard products (such as customized parts, special specification materials, etc.). For example, different suppliers may have differences in the naming and parameter descriptions of the same customized mechanical part, and existing platforms cannot accurately identify these different expressions that are actually the same product, thus unable to conduct effective price comparisons. In addition, existing platforms lack in-depth integration in supplier management and are difficult to obtain key information such as the inventory, delivery date, and after-sales service of suppliers in real time, resulting in enterprises being unable to comprehensively consider the comprehensive strength of suppliers when making procurement decisions and increasing procurement risks.

[0004] A large amount of data is generated during the enterprise procurement process, including product information, supplier information, procurement order information, price history data, etc. However, this data is often scattered in different systems or databases and is difficult to be effectively integrated and analyzed. Without a unified data processing mechanism, enterprises cannot deeply explore the value of data, such as predicting price trends and evaluating supplier performance by analyzing historical procurement data. This makes enterprises lack data support when making procurement decisions, difficult to formulate scientific and reasonable procurement strategies, and unable to achieve refined and intelligent procurement management. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a data processing system and method for a multi-supplier aggregation platform for one product to solve the problems of the prior art described in the background art.

[0006] To solve the above technical problems, the embodiments of the present invention provide the following technical solutions:

[0007] A data processing system for a one-product multi-supplier aggregation platform, comprising:

[0008] A data collection module responsible for collecting product and supplier-related information from multiple data sources, including but not limited to the databases of suppliers, internal procurement records of enterprises, and industry market data platforms;

[0009] A data standardization module that standardizes the collected product and supplier data in different formats and standards;

[0010] A supplier management module that realizes the full life cycle management of suppliers, including supplier access evaluation, supplier performance evaluation, and supplier relationship maintenance;

[0011] A price comparison and analysis module that conducts price comparisons for one product with multiple suppliers based on the standardized data, screens out qualified products from the products of multiple suppliers according to the procurement requirements set by the enterprise, and makes a detailed comparison from multiple dimensions including but not limited to price, cost composition, and service terms;

[0012] An intelligent decision-making module that comprehensively considers the results of price comparison and analysis, supplier performance evaluation data, as well as the enterprise's own procurement strategies and goals, and uses machine learning algorithms and decision-making models to provide intelligent decision-making suggestions for enterprise procurement;

[0013] A data visualization module that visually displays the results of data processing and analysis in the form of various charts and reports.

[0014] Optionally, the data standardization module standardizes the collected product and supplier data in different formats and standards, specifically using natural language processing technology, data mapping algorithms, and rule libraries to convert non-standard product descriptions into a unified standard format.

[0015] Optionally, the access evaluation of suppliers by the supplier management module specifically includes:

[0016] Collect and upload the application materials submitted by suppliers, including but not limited to business licenses, production licenses, and quality management system certifications of enterprises;

[0017] Use optical character recognition (OCR) technology to extract key information from the uploaded business license file, and verify the authenticity and validity of the application materials through the enterprise information query interface or data sharing platform of the administrative department for industry and commerce;

[0018] Extract the certification number and validity period information from the quality management system certification file, check whether the certification type meets the procurement requirements, and whether the validity period of the certification is within the validity period;

[0019] Establish a data interface with a third-party credit assessment agency, request a credit report from the credit assessment agency based on the enterprise information of the supplier, and the credit report includes credit scores, credit ratings, historical default records, and financial status ratings;

[0020] Set the passing criteria and weights for each evaluation factor, and calculate the comprehensive evaluation score based on the evaluation results of each supplier and the set weights;

[0021] Set the initial cooperation level according to the comprehensive evaluation score range, and set the initial credit limit according to the cooperation level.

[0022] Optionally, the supplier management module conducts a performance evaluation of the supplier, specifically including:

[0023] Real-time monitor various performance indicators of the supplier, obtain order delivery status information through docking with the supplier system, and calculate the delivery on-time rate;

[0024] Collect the inspection results of the purchased products by the enterprise quality inspection department, and count the product quality pass rate;

[0025] Track the feedback of the enterprise internal on the after-sales service of the supplier, and evaluate the after-sales service quality;

[0026] Normalize the delivery on-time rate, quality pass rate, and after-sales service quality data and assign corresponding weights, and calculate the comprehensive performance evaluation value.

[0027] Optionally, the supplier management module maintains the supplier relationship, specifically including:

[0028] Establish a supplier information database, including supplier basic information, cooperation history, communication records, and performance evaluation data;

[0029] Obtain the enterprise dynamic information of the supplier through web crawler technology or docking with the information system of the supplier, classify and organize the dynamic information, and associate it with the corresponding supplier records in the supplier information database;

[0030] According to the performance evaluation data and cooperation history of the supplier, use clustering analysis algorithms or custom classification rules to classify the suppliers into different categories;

[0031] Develop a personalized communication plan for each supplier according to the classification and grading of the suppliers;

[0032] Real-time monitor the data in the supplier information database, and automatically send out warning signals when abnormal fluctuations in supplier performance indicators, commercial disputes, or external associated factors are found.

[0033] Optionally, based on the standardized data, the price comparison analysis module conducts price comparison for one product with multiple suppliers, and screens out qualified products from the products of multiple suppliers according to the procurement requirements set by the enterprise. Specifically: Let the set of screened products be C, and the price of product j be P j , the market average price is The price stability index (such as price standard deviation) is σ j ; the cost composition vector is where C jn is a cost factor, and the cost composition weight vector is The service term vector is where S jn is a cost factor, and the service term weight vector is The price weight is ω p , the sum of the cost composition weights is The sum of the service term weights is The comprehensive evaluation score Score of product j j The calculation formula is:

[0034]

[0035] where, represents the advantage of the price relative to the market average price, represents the adjustment of price stability, σ max is the maximum value of the price standard deviation in the screened product set, represents the comprehensive evaluation of the cost composition, represents the comprehensive evaluation of the service terms.

[0036] Optionally, the machine learning algorithm and decision model adopted by the intelligent decision-making module specifically include price trend prediction, supplier performance prediction, and procurement decision-making when executed. Among them:

[0037] (1) The price trend prediction is specifically:

[0038] Collect historical procurement price data, clean and preprocess the data, select the LSTM model, use the processed historical price data and relevant market factor data as input features, and the future price as the target variable. Divide the training set and the test set, use the training set to train the model, and optimize the model performance by adjusting the model parameters to minimize the error between the predicted price and the actual price;

[0039] (2) The supplier performance prediction is specifically:

[0040] Collect the historical performance data of suppliers, perform feature engineering on the data, select the random forest algorithm for supplier performance prediction, divide the processed supplier data into a training set and a test set, use the training set to train the model, optimize the performance of the model on the training set by adjusting the model parameters, use the test set to evaluate the performance of the trained model, adjust the model according to the evaluation results, and input the latest data of new or existing suppliers into the model to predict their future performance;

[0041] (3) Procurement Decision

[0042] Determine the key factors for procurement decisions based on procurement objectives. According to the determined objectives and factors, define utility functions for each factor, convert the actual values of the factors into utility values, where the utility value represents the degree of contribution of the factor to achieving the procurement decision objective. Considering the utility values and weights of all factors comprehensively, calculate the overall utility score for each supplier or procurement plan. The formula for calculating the overall utility score is:

[0043] where U is the overall utility score, ω i is the weight of the i-th factor, and u i is the utility value of the i-th factor;

[0044] Rank the suppliers or procurement plans according to the overall utility score, and the plan with the highest score is the recommended procurement decision.

[0045] Optionally, comprehensively consider the price comparison analysis results, supplier performance evaluation data, and the enterprise's own procurement strategies and objectives, and use machine learning algorithms and decision models to provide intelligent decision-making suggestions for enterprise procurement, specifically:

[0046] Rank the suppliers according to the comprehensive evaluation score. If the overall performance of the supplier with the highest score is significantly better than other suppliers, it is recommended that the enterprise give priority to choosing this supplier, and determine the procurement quantity according to the enterprise's procurement demand and the supplier's production capacity, etc.;

[0047] If the scores of multiple suppliers are similar and all meet the basic requirements, further analyze the advantages and disadvantages of each supplier, and provide different decision-making options in combination with the enterprise's current actual situation;

[0048] If the comprehensive evaluation scores of all suppliers are relatively low or cannot meet the basic requirements of the enterprise, the enterprise re-evaluates the procurement demand, expands the supplier search scope, or negotiates improvement measures with existing suppliers.

[0049] Optionally, the data visualization module displays the parameters of each supplier's products, including but not limited to price, specification parameters, and service terms.

[0050] A method for implementing a data processing system of a multi - supply aggregation platform for one product as described above, comprising the following steps:

[0051] For different data sources, which are not limited to supplier databases, enterprise internal procurement records, and industry market data platforms, data collection is respectively carried out by using API docking, system extraction, and web crawler technology; the collected data is integrated, duplicate data is removed, and it is classified and stored according to dimensions such as product categories and suppliers, and a data index is established;

[0052] The integrated product data is processed. By using natural language processing technology, data mapping algorithms, and rule libraries, non - standardized product names, specification parameters, and attributes are converted into a unified standard format; at the same time, supplier information is standardized, including verifying qualification certificates, unifying contact information formats, and analyzing service terms;

[0053] In the supplier access stage, the application materials submitted by suppliers are collected and reviewed. Combining the reports of third - party credit assessment agencies and industry reputation surveys, the qualifications and reputations of suppliers are comprehensively evaluated to determine access and set initial cooperation levels and credit limits; during the cooperation process, supplier performance indicators are monitored in real time, and performance evaluation and hierarchical adjustment are carried out according to indicators such as on - time delivery rate, product quality pass rate, and after - sales service response speed;

[0054] According to the enterprise's procurement requirements, eligible products and their suppliers are selected from the standardized product data, and price comparison analysis is carried out from multiple dimensions such as price, cost composition, and service terms. At the same time, considering market price fluctuation factors, price trends are predicted; based on the comprehensive price comparison analysis results, supplier performance evaluation data, as well as the enterprise's procurement strategies and goals, machine learning algorithms and decision models are used to provide intelligent decision - making suggestions for enterprise procurement;

[0055] According to user needs and permissions, the data processing and analysis results are presented in an intuitive visual form, including providing product price comparison charts for procurement personnel, generating procurement cost analysis reports and supplier performance evaluation reports for enterprise management, and supporting user - defined reports and charts.

[0056] The beneficial effects of the above - mentioned technical solutions of the present invention are as follows:

[0057] 1. Through data standardization processing, the present invention can accurately identify the same product provided by different suppliers, realizing true price comparison for multiple supplies of one product. Whether it is a standardized product or a non - standardized product, comprehensive and detailed price and service comparisons can be carried out under a unified standard, helping enterprises obtain cost - effective procurement solutions and effectively reducing procurement costs.

[0058] 2. The present invention conducts full - life - cycle management of suppliers, from access assessment to performance monitoring and relationship maintenance, to ensure that the enterprise cooperates with high - quality suppliers. Real - time supplier performance evaluation can promptly detect problems existing in suppliers, prompting suppliers to continuously improve service quality, increase the on - time delivery rate, and improve the qualified rate of product quality. At the same time, the cooperation strategy is dynamically adjusted according to supplier performance, reducing the enterprise's procurement risks and ensuring the stability and reliability of the supply chain.

[0059] 3. Based on big - data collection, integration, and analysis, the system provides data support and intelligent decision - making suggestions for enterprise procurement. By analyzing historical procurement data, market price trends, and supplier performance data, the enterprise can predict changes in procurement costs, evaluate procurement risks, and formulate scientific and reasonable procurement plans according to the actual situation.

[0060] 4. The present invention enables enterprise procurement personnel and management to quickly obtain key information through an intuitive data visualization interface, simplifying the procurement decision - making process. Procurement personnel can complete tasks such as product search, price comparison, and supplier evaluation on the same platform without switching between multiple systems or platforms, greatly improving work efficiency. At the same time, the centralized management and visual display of data enhance the transparency of the procurement process, facilitating internal supervision and management within the enterprise and effectively preventing problems such as procurement corruption. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is the principle block diagram of the data - processing system of the multi - supplier aggregation platform for one product of the present invention;

[0062] Figure 2 is the schematic flow diagram of the decision - making mechanism of the intelligent decision - making module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0063] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0064] As Figure 1 shown, an embodiment of the present invention proposes a data - processing system for a multi - supplier aggregation platform for one product, including:

[0065] A data collection module 101, responsible for collecting product - and - supplier - related information from multiple data sources. The data sources include but are not limited to the databases of suppliers, the enterprise's internal procurement records, and industry market data platforms. The data collection module obtains data through various methods, such as using web crawler technology to scrape public market data, conducting API docking with the supplier system to obtain real - time product information, and extracting historical procurement data from the enterprise's internal system, etc., to ensure the comprehensiveness and timeliness of the data.

[0066] The data standardization module 102 standardizes the product and supplier data collected in different formats and standards. For example, for different names of the same product from different suppliers, it unifies them into a standard name through semantic analysis and predefined mapping rules; it normalizes the product specification parameters to make them comparable. At the same time, it standardizes the supplier information, such as unifying the supplier qualification certification standard, contact information format, etc., to facilitate subsequent data analysis and management.

[0067] The supplier management module 103 realizes the full life cycle management of suppliers. It includes the supplier access assessment to determine whether a supplier meets the enterprise cooperation standard based on multi-dimensional factors such as the supplier's qualification, reputation, production capacity, etc.; the supplier performance assessment to dynamically evaluate and classify manage suppliers by monitoring indicators such as the supplier's on-time delivery rate, product quality pass rate, after-sales service response speed in real time; and the supplier relationship maintenance to maintain close communication with suppliers, timely transmit enterprise requirements and feedback, and promote the continuous optimization of bilateral cooperation.

[0068] The price comparison and analysis module 104 conducts price comparison among multiple suppliers for one product based on the standardized data. According to the enterprise's set procurement requirements (such as product specifications, quantity, delivery period, etc.), it screens out eligible products among the products of multiple suppliers and makes a detailed comparison from multiple dimensions such as price, cost composition (such as raw material cost, transportation cost, processing cost, etc.), service terms (such as warranty period, return and exchange policy, etc.). At the same time, considering the market price fluctuation factor, it predicts the price trend by real-time monitoring of the market conditions and historical price data, provides suggestions on the procurement timing for the enterprise, and helps the enterprise obtain the optimal procurement price.

[0069] The intelligent decision-making module 105 comprehensively considers the price comparison and analysis results, supplier performance assessment data, as well as the enterprise's own procurement strategies and goals (such as cost control goals, quality requirements, inventory management strategies, etc.), and uses machine learning algorithms and decision-making models to provide intelligent decision-making suggestions for enterprise procurement. For example, when the market price of a certain product fluctuates greatly and the inventory level is low, the intelligent decision-making module may suggest that the enterprise appropriately increase the procurement quantity to reduce costs and ensure production continuity; or when the performance assessment of a certain supplier shows a downward trend, it suggests that the enterprise adjust the procurement share or look for alternative suppliers.

[0070] The data visualization module 106 visually displays the results of data processing and analysis in the form of various charts and reports. It provides various charts (such as bar charts, line charts, pie charts, etc.) and reports (such as procurement cost analysis reports, supplier performance assessment reports, price trend reports, etc.), helping enterprise management and procurement personnel quickly understand the information behind the data for timely and accurate decision-making.

[0071] In this embodiment, the data standardization module performs standardization processing on the product and supplier data collected in different formats and standards. Specifically, natural language processing technology, data mapping algorithms, and a rule library are used to convert non-standard product descriptions into a unified standard format. First, the product names are standardized. By establishing a product name dictionary and a semantic analysis model, situations where different names actually refer to the same product are identified. For example, for different expressions such as "computer mainframe", "computer chassis", and "PC mainframe", they are unified into the standard name "computer mainframe". Then, the product specification parameters are standardized. According to the product type and industry standards, the specification parameters with different units and precisions are converted into a unified standard. For example, the length unit is unified to millimeters, the weight unit is unified to kilograms, etc., and the parameter range is normalized. For product attributes (such as color, material, function, etc.), an attribute classification system is established to map various expression methods into the standard attribute classification. For the supplier information, standardization processing is carried out, the validity and authenticity of the supplier qualification certificates are verified, and they are classified according to the unified qualification classification standard. The supplier contact information is sorted out to ensure that the format is unified and accurate. The service terms of the supplier (such as the warranty period, return and exchange policy, etc.) are analyzed and standardized so that fair comparison can be made during the price comparison analysis.

[0072] In this embodiment, the supplier management module conducts an access assessment of the suppliers, specifically including:

[0073] Collect and upload the application materials submitted by the suppliers, including but not limited to the enterprise business license, production license, and quality management system certification. Review and verify these materials, and at the same time, combined with information such as the reports of third-party credit assessment agencies and industry reputation surveys, comprehensively evaluate the qualifications and reputations of the suppliers. According to the evaluation results, determine whether the suppliers are eligible for access, and set their initial cooperation levels and credit limits.

[0074] Use optical character recognition (OCR) technology to extract key information from the uploaded business license file, and verify the authenticity and validity of the application materials through the enterprise information query interface of the administrative department for industry and commerce or the data sharing platform;

[0075] Extract the certification number and validity period information from the quality management system certification file, check whether the certification type meets the procurement requirements, and whether the validity period of the certification is within the validity period;

[0076] Establish a data interface with a third-party credit assessment agency, request a credit report from the credit assessment agency according to the enterprise information of the supplier, and the credit report includes credit scores, credit ratings, historical default records, and financial status ratings;

[0077] Set the passing criteria and weights for each evaluation factor, and calculate the comprehensive evaluation score based on the evaluation results of each supplier and the set weights.

[0078] Set the initial cooperation level according to the comprehensive evaluation score range, and set the initial credit limit according to the cooperation level.

[0079] In this embodiment, the supplier management module evaluates the supplier performance, specifically including:

[0080] Real-time monitor various performance indicators of the supplier, obtain the order delivery status information through the interface with the supplier system, and calculate the on-time delivery rate.

[0081] Collect the inspection results of the purchased products by the enterprise quality inspection department, and count the product quality pass rate.

[0082] Track the feedback of the enterprise internal on the supplier after-sales service, and evaluate the after-sales service quality.

[0083] Normalize the on-time delivery rate, quality pass rate, and after-sales service quality data and assign corresponding weights, and calculate the comprehensive performance evaluation value.

[0084] In this embodiment, the supplier management module maintains the supplier relationship, specifically including:

[0085] Establish a supplier information database, including supplier basic information, cooperation history, communication records, and performance evaluation data.

[0086] Obtain the enterprise dynamic information of the supplier through web crawler technology or interface with the supplier's information system, classify and sort the dynamic information, and associate it with the corresponding supplier record in the supplier information database.

[0087] According to the supplier's performance evaluation data and cooperation history, use clustering analysis algorithm or custom classification rules to classify the suppliers into different categories.

[0088] According to the classification and grading of the suppliers, formulate personalized communication plans for each supplier.

[0089] Real-time monitor the data in the supplier information database, and automatically send out warning signals when abnormal fluctuations in supplier performance indicators, commercial disputes, or external associated factors are found.

[0090] In this embodiment, the price comparison and analysis module conducts price comparison for multiple suppliers of the same product based on standardized data, and screens out qualified products from the products of multiple suppliers according to the procurement requirements set by the enterprise. Specifically, let the set of screened products be C, the price of product j be P j , the market average price be P, and the price stability index (such as price standard deviation) be σ j; The cost composition vector is where C jn is the cost factor, and the cost composition weight vector is The service term vector is where S jn is the cost factor, and the service term weight vector is The price weight is ω p , and the sum of the cost composition weights is The sum of the service term weights is The comprehensive evaluation score Score of product j j The calculation formula is:

[0091]

[0092] where, represents the advantage of the price relative to the market average price, represents the adjustment of price stability, and σ max is the maximum value of the price standard deviation in the screened product set, represents the comprehensive evaluation of the cost composition, represents the comprehensive evaluation of the service terms.

[0093] In this embodiment, the machine learning algorithm and decision model adopted by the intelligent decision-making module specifically include price trend prediction, supplier performance prediction, and procurement decision-making, as Figure 2 shown, which is the schematic diagram of the decision-making mechanism process. Among them:

[0094] (1) The price trend prediction is specifically:

[0095] First, collect historical procurement price data, including price information of different products and different suppliers at different times, as well as relevant market factor data, such as raw material price fluctuations, macroeconomic indicators (such as inflation rate, exchange rate changes, etc.), and industry supply and demand relationship data (such as production volume, sales volume, inventory level, etc.).

[0096] Clean and preprocess the data, handle missing values and outliers, convert relevant data such as dates into appropriate formats (e.g., convert dates into time series-related features such as months, quarters, etc.), and normalize or standardize the data to make different features have similar numerical ranges for easy model training. Select the LSTM model, use the processed historical price data and relevant market factor data as input features, and the future price as the target variable. Divide the training set and the test set, use the training set to train the model, and optimize the model performance by adjusting the model parameters to minimize the error between the predicted price and the actual price. When the model evaluation passes, input the latest market factor data and the current price data into the model to predict the price trend in the future period (e.g., the next month, quarter, etc.), providing a reference for the enterprise's procurement decision-making, such as determining the procurement timing (procure at the price trough) or adjusting the procurement quantity (if the predicted price increases, appropriately increase the procurement volume for reserve).

[0097] (2) Supplier performance prediction specifically includes:

[0098] Collect the historical performance data of suppliers, including the historical records of indicators such as on-time delivery rate, product quality pass rate, and after-sales service response speed, as well as the basic information of suppliers (such as enterprise scale, establishment years, industry category, etc.) and cooperation historical data (such as cooperation years, cumulative transaction amount, number of cooperation projects, etc.). Then, perform feature engineering on the data. For example, perform one-hot encoding on categorical features (such as industry category) and standardize or normalize continuous features. Next, select the random forest algorithm for supplier performance prediction, divide the processed supplier data into a training set and a test set, use the training set to train the model, optimize the model performance on the training set by adjusting the model parameters, use the test set to evaluate the performance of the trained model, adjust the model according to the evaluation results, and input the latest data of new or existing suppliers into the model to predict their future performance, such as predicting the range of indicators such as on-time delivery rate and product quality pass rate in the next quarter or half year.

[0099] (3) Procurement decision-making

[0100] Determine the key factors for procurement decisions based on procurement objectives. According to the determined objectives and factors, define a utility function for each factor. Procurement objectives include cost minimization, quality maximization, supply stability assurance, etc., or a combination of multiple objectives. The key factors affecting procurement decisions include product price, supplier performance (on-time delivery rate, product quality pass rate, after-sales service, etc.), inventory level, market price trend, supplier relationship (cooperation history, degree of trust, etc.), etc. Set corresponding weights for each factor. The determination of weights can be based on the strategic focus and past experience of the enterprise. For example, for a cost-sensitive enterprise, the weight of product price may be relatively high; while for an enterprise with strict quality requirements, the weight of factors related to product quality is relatively large.

[0101] A common method for constructing a decision-making model is to use the multi-attribute utility theory (MAUT). According to the determined objectives and factors, convert the actual values of the factors into utility values. The utility value represents the degree of contribution of the factor to achieving the procurement decision objective. For example, for product price, a lower price corresponds to a higher utility value; for the on-time delivery rate, the higher the on-time rate, the higher the utility value. Considering the utility values and weights of each factor comprehensively, calculate the overall utility score for each supplier or procurement plan. The formula for calculating the overall utility score is:

[0102] where U is the overall utility score, ω i is the weight of the i-th factor, and u i is the utility value of the i-th factor; rank the suppliers or procurement plans according to the overall utility score, and the plan with the highest score is the recommended procurement decision.

[0103] In this embodiment, considering the price comparison analysis results, supplier performance evaluation data, as well as the enterprise's own procurement strategies and objectives comprehensively, use machine learning algorithms and decision-making models to provide intelligent decision-making suggestions for enterprise procurement, specifically:

[0104] Rank the suppliers according to the comprehensive evaluation score. If the overall performance of the supplier with the highest score is significantly better than other suppliers, it is recommended that the enterprise give priority to selecting this supplier, and determine the procurement quantity according to the enterprise's procurement demand and the supplier's production capacity, etc.;

[0105] If the scores of multiple suppliers are similar and all meet the basic requirements, further analyze the advantages and disadvantages of each supplier, and provide different decision-making options in combination with the current actual situation of the enterprise;

[0106] If the comprehensive evaluation scores of all suppliers are relatively low or cannot meet the basic requirements of the enterprise, the enterprise re-evaluates the procurement demand, expands the supplier search scope, or negotiates improvement measures with the existing suppliers.

[0107] In this embodiment, the data visualization module displays the parameters of products from each supplier, including but not limited to price, specification parameters, and service terms, to help quickly select the optimal product. For the enterprise management level, a procurement cost analysis report is generated, presenting the changing trends of procurement costs for different product categories and different time periods, as well as the visualization display of supplier performance evaluation results (such as the ranking of supplier performance scores, radar charts of the completion of each performance indicator, etc.), so that the management can comprehensively understand the procurement business situation and make strategic decisions. At the same time, the visualization module supports users to customize reports and charts. Users can select data dimensions and display methods according to their key concerns to meet personalized data analysis needs.

[0108] The present invention also proposes an implementation method of a data processing system for a multi-supplier aggregation platform for the same product as described above, including the following steps:

[0109] S1. For different data sources including but not limited to supplier databases, enterprise internal procurement records, and industry market data platforms, data collection is respectively carried out by using API docking, system extraction, and web crawler technologies; the collected data is integrated, duplicate data is removed, and it is classified and stored according to dimensions such as product category and supplier, and a data index is established.

[0110] S2. The integrated product data is processed. By using natural language processing technology, data mapping algorithms, and rule libraries, non-standardized product names, specification parameters, and attributes are converted into a unified standard format; at the same time, the supplier information is standardized, including verifying qualification certificates, unifying the contact information format, and parsing service terms.

[0111] S3. At the supplier access stage, the application materials submitted by the supplier are collected and reviewed. Combining the reports of third-party credit assessment agencies and industry reputation surveys, the supplier's qualifications and reputation are comprehensively evaluated to determine access and set the initial cooperation level and credit limit; during the cooperation process, the supplier performance indicators are monitored in real time, and performance evaluation and grading adjustment are carried out according to indicators such as on-time delivery rate, product quality pass rate, and after-sales service response speed.

[0112] S4. According to the enterprise procurement requirements, eligible products and their suppliers are screened out from the standardized product data, and price comparison analysis is carried out from multiple dimensions such as price, cost composition, and service terms. At the same time, considering market price fluctuation factors, the price trend is predicted; based on the comprehensive price comparison analysis results, supplier performance evaluation data, as well as the enterprise procurement strategy and goals, machine learning algorithms and decision-making models are used to provide intelligent decision-making suggestions for enterprise procurement.

[0113] S5. According to user requirements and permissions, present the data processing and analysis results in an intuitive visual form, including providing product price comparison charts for purchasers, generating procurement cost analysis reports and supplier performance evaluation reports for enterprise management, and supporting users to customize reports and charts.

[0114] In summary, through data standardization processing, the present invention can accurately identify the same product provided by different suppliers, realizing price comparison among multiple suppliers for the same product in the true sense. Whether it is a standardized product or a non-standardized product, comprehensive and detailed price and service comparisons can be carried out under a unified standard, helping enterprises obtain cost-effective procurement solutions and effectively reducing procurement costs. For example, in the procurement of electronic components, it can accurately identify chips of different brands but with the same functions and specifications, and compare factors such as their prices, delivery cycles, and quality guarantees, enabling enterprises to make the best choice.

[0115] The system conducts full-life cycle management of suppliers, from access assessment to performance monitoring and relationship maintenance, ensuring that enterprises cooperate with high-quality suppliers. Real-time supplier performance evaluation can promptly detect problems existing in suppliers, prompting suppliers to continuously improve service quality, increase the on-time delivery rate and the qualified rate of product quality. At the same time, adjust cooperation strategies dynamically according to supplier performance, reduce the procurement risks of enterprises, and ensure the stability and reliability of the supply chain. For example, an automobile manufacturing enterprise strictly manages its component suppliers through this system, promptly eliminates unqualified suppliers, and establishes long-term strategic cooperative partnerships with high-quality suppliers, effectively improving product quality and production efficiency.

[0116] Based on big data collection, integration, and analysis, the system provides data support and intelligent decision-making suggestions for enterprise procurement. By analyzing historical procurement data, market price trends, and supplier performance data, enterprises can predict changes in procurement costs, evaluate procurement risks, and formulate scientific and reasonable procurement plans according to actual situations. For example, according to the market price fluctuation trend of raw materials and the enterprise's production plan, the system predicts that the price of a certain raw material will rise, and the enterprise purchases and reserves an appropriate amount of inventory in advance, avoiding cost increases caused by price increases and ensuring the continuity of production.

[0117] The intuitive data visualization interface enables enterprise purchasers and management to quickly obtain key information and simplifies the procurement decision-making process. Purchasers can complete tasks such as product search, price comparison, and supplier evaluation on the same platform without switching between multiple systems or platforms, greatly improving work efficiency. At the same time, the centralized management and visual display of data enhance the transparency of the procurement process, facilitating internal supervision and management within the enterprise and effectively preventing problems such as procurement corruption.

[0118] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A data processing system for a one-product-multiple-supply aggregation platform, characterized in that: include: The data collection module is responsible for collecting product and supplier related information from multiple data sources, including but not limited to the supplier's database, the company's internal procurement records, and the industry market data platform; Data standardization module, which standardizes the collected product and supplier data in different formats and standards; The supplier management module realizes the full life cycle management of suppliers, including supplier admission assessment, supplier performance evaluation, and supplier relationship maintenance; The price comparison analysis module compares prices of multiple suppliers of a product based on standardized data. According to the procurement needs set by the enterprise, it selects qualified products from multiple suppliers and makes detailed comparisons from multiple dimensions including but not limited to price, cost structure, and terms of service. The intelligent decision-making module comprehensively considers the price comparison analysis results, supplier performance evaluation data, and the company's own procurement strategy and goals, and uses machine learning algorithms and decision-making models to provide intelligent decision-making suggestions for corporate procurement; The data visualization module visualizes the results of data processing and analysis in the form of various charts and reports.

2. The data processing system of the one-product-multiple-supply aggregation platform according to claim 1 is characterized in that: The data standardization module standardizes the collected product and supplier data in different formats and standards, and specifically uses natural language processing technology, data mapping algorithms and rule bases to convert non-standardized product descriptions into a unified standard format.

3. The data processing system of the one-product-multiple-supply aggregation platform according to claim 1 is characterized in that: The supplier management module's access assessment of suppliers specifically includes: Collect and upload application materials submitted by suppliers, including but not limited to business licenses, production licenses, and quality management system certifications; Use optical character recognition (OCR) technology to extract key information from the uploaded business license file, and verify the authenticity and validity of the application materials through the enterprise information query interface or data sharing platform of the industrial and commercial administration department; Extract the certification number and validity period information from the quality management system certification documents, check whether the certification type meets the procurement requirements, and whether the certification validity period is within the validity period; Establish a data interface with a third-party credit assessment agency and request a credit report from the credit assessment agency based on the supplier's corporate information. The credit report includes credit score, credit rating, historical default record, and financial status rating; Set the passing criteria and weights for each evaluation factor, and calculate the comprehensive evaluation score based on the supplier's evaluation results and the set weights; The initial cooperation level is set according to the comprehensive evaluation score range, and the initial credit limit is set according to the cooperation level.

4. The data processing system of the one-product-multiple-supply aggregation platform according to claim 1 is characterized in that: The supplier management module evaluates supplier performance, specifically including: Monitor the performance indicators of suppliers in real time, obtain order delivery status information and calculate on-time delivery rate by connecting with supplier systems; Collect the inspection results of the company's quality inspection department on the purchased products and calculate the product quality pass rate; Track the company's internal feedback on suppliers' after-sales services and evaluate the quality of after-sales services; The data on on-time delivery rate, quality pass rate, and after-sales service quality are normalized and weighted accordingly to calculate a comprehensive performance evaluation value.

5. The data processing system of the one-product-multiple-supply aggregation platform according to claim 1 is characterized in that: The supplier management module maintains supplier relationships, specifically including: Establish a supplier information database, including basic supplier information, cooperation history, communication records, and performance evaluation data; Obtain the supplier's enterprise dynamic information through web crawler technology or by connecting with the supplier's information system, classify and organize the dynamic information, and associate it with the corresponding supplier record in the supplier information database; Based on the supplier's performance evaluation data and cooperation history, suppliers are divided into different categories using cluster analysis algorithms or custom classification rules; Develop a personalized communication plan for each supplier based on their classification and grading; Monitor the data in the supplier information database in real time, and automatically issue early warning signals when abnormal fluctuations in supplier performance indicators, business disputes, or external factors are detected.

6. The data processing system of the one-product-multiple-supply aggregation platform according to claim 1 is characterized in that: The price comparison analysis module compares prices of multiple suppliers of a product based on standardized data, and selects qualified products from multiple suppliers according to the purchasing needs set by the enterprise. Specifically, let the filtered product set be C, and the price of product j be P j The average market price is The price stability indicator (such as price standard deviation) is σ j ; The cost component vector is Among them C jn is the cost factor, and the cost composition weight is Terms of Service are for Where S jn is the cost factor, and the service terms weight vector is The price weight is ω p , the sum of the cost component weights is The total weight of the terms of service is Comprehensive evaluation score of product j j The calculation formula is: in, Indicates the price advantage relative to the market average price. represents the adjustment for price stability, σ max To filter the maximum value of the price standard deviation in the product set, Represents a comprehensive evaluation of cost components, Represents a comprehensive evaluation of the terms of service.

7. The data processing system of the one-product-multiple-supply aggregation platform according to claim 1 is characterized in that: The machine learning algorithm and decision model used by the intelligent decision module include price trend prediction, supplier performance prediction and procurement decision when implemented, among which: (1) The specific price trend forecast is as follows: Collect historical purchase price data, clean and preprocess the data, select the LSTM model, use the processed historical price data and related market factor data as input features, and the future price as the target variable. Divide the training set and test set, use the training set to train the model, and optimize the model performance by adjusting the model parameters to minimize the error between the predicted price and the actual price; (2) The supplier performance forecast is as follows: Collect historical performance data of suppliers, perform feature engineering on the data, select the random forest algorithm for supplier performance prediction, divide the processed supplier data into training set and test set, use the training set to train the model, optimize the performance of the model on the training set by adjusting the model parameters, use the test set to evaluate the performance of the trained model, adjust the model according to the evaluation results, input the latest data of new or existing suppliers into the model, and predict their future performance; (3) Procurement Decision Determine the key factors of procurement decision according to the procurement objectives. Define the utility function for each factor according to the determined objectives and factors, convert the actual value of the factor into a utility value. The utility value indicates the contribution of the factor to the realization of the procurement decision objective. Consider the utility value and weight of each factor comprehensively, and calculate the overall utility score of each supplier or procurement plan. The overall utility score calculation formula is: where U is the overall utility score, ω i is the weight of the i-th factor, u i is the utility value of the ith factor; Suppliers or purchasing options are ranked according to their overall utility scores, and the option with the highest score is the recommended purchasing decision.

8. The data processing system of the one-product-multiple-supply aggregation platform according to claim 1 is characterized in that: The above-mentioned comprehensive consideration of the price comparison analysis results, supplier performance evaluation data and the company's own procurement strategy and goals, the use of machine learning algorithms and decision-making models, provides intelligent decision-making suggestions for corporate procurement, specifically: Suppliers are ranked according to their comprehensive evaluation scores. If the supplier with the highest score has significantly better overall performance than other suppliers, the enterprise is advised to give priority to this supplier and determine the purchase volume based on the enterprise's purchase demand and the supplier's production capacity. If multiple suppliers have similar scores and all meet the basic requirements, further analyze the strengths and weaknesses of each supplier and provide different decision-making options based on the company's current actual situation; If the comprehensive evaluation scores of all suppliers are low or fail to meet the basic requirements of the enterprise, the enterprise shall re-evaluate its procurement needs, expand the scope of supplier search or negotiate improvement measures with existing suppliers.

9. The data processing system of the one-product-multiple-supply aggregation platform according to claim 1 is characterized in that: The data visualization module displays the parameters of each supplier's products including but not limited to price, specification parameters, and terms of service.

10. A method for implementing a data processing system of a one-product-multiple-supply aggregation platform as claimed in any one of claims 1 to 9, characterized in that: The following steps are involved: For different data sources including but not limited to supplier databases, internal enterprise procurement records and industry market data platforms, data collection is carried out using API docking, system extraction and web crawler technologies. The collected data is integrated, duplicate data is removed, and classified and stored according to product categories, suppliers and other dimensions to establish data indexes. Process the integrated product data, using natural language processing technology, data mapping algorithms and rule bases to convert non-standardized product names, specifications and attributes into a unified standard format; standardize supplier information, including verifying qualification certificates, unifying contact information formats and parsing service terms; During the supplier admission stage, we collect and review the application materials submitted by suppliers, combine reports from third-party credit assessment agencies and industry reputation surveys, comprehensively assess supplier qualifications and reputation, determine admission and set initial cooperation levels and credit limits; during the cooperation process, we monitor supplier performance indicators in real time, and conduct performance evaluation and grading adjustments based on indicators such as on-time delivery rate, product quality qualification rate, and after-sales service response speed; According to the procurement needs of enterprises, we screen out qualified products and their suppliers from the standardized product data, conduct price comparison analysis from multiple dimensions such as price, cost structure and service terms, and predict price trends by considering market price fluctuations; we combine the price comparison analysis results, supplier performance evaluation data and enterprise procurement strategies and goals, and use machine learning algorithms and decision-making models to provide intelligent decision-making suggestions for enterprise procurement; According to user needs and permissions, the data processing and analysis results are presented in an intuitive visual form, including providing product price comparison charts for purchasing personnel, generating procurement cost analysis reports and supplier performance evaluation reports for corporate management, and supporting user-defined reports and charts.

Images