Collaborative purchasing system and method based on multi-source data fusion and dynamic decision

Abstract

The invention discloses a collaborative purchasing system and method based on multi-source data fusion and dynamic decision in the technical field of supply chain management. The method comprises the following steps: receiving a photovoltaic equipment purchasing order submitted by a purchaser in an order collaborative processing subsystem; a photovoltaic equipment purchase order is provided for a supplier, and a price prediction model is called through a dynamic pricing module to provide a suggested quotation for the supplier according to commodities transmitted to the commodity collaborative management subsystem by the supplier; receiving a photovoltaic equipment purchase order after the supplier modifies the unit price according to the suggested quoted price; and providing the modified photovoltaic equipment purchase order and the commodity related price information to the operator, generating an electronic pricing list based on an audit result of the operator on supplier pricing, and sending the electronic pricing list to the purchaser. The purchase process is transparent and concise, reference suggestions are provided in the aspect of commodity prices, the fund utilization rate is improved through pre-deposit intelligent scheduling, and the multi-party cooperative processing time is shortened on the whole.

Description

Technical Field

[0001] The present invention relates to a collaborative purchasing system and method based on multi-source data fusion and dynamic decision-making, belonging to the technical field of supply chain management. Background Art

[0002] Photovoltaic, or photovoltaic power generation system, is a power generation system that uses the photovoltaic effect of semiconductor materials to convert solar radiation energy into electrical energy. The energy of the photovoltaic power generation system comes from solar energy, which is a clean, safe and renewable energy source. The photovoltaic power generation process does not pollute the environment or damage the ecology. Photovoltaic power generation systems are divided into independent photovoltaic systems and grid-connected photovoltaic systems, which are composed of solar cell arrays, battery packs, charge and discharge controllers, inverters, AC distribution cabinets, solar tracking control systems and other equipment.

[0003] Since the price of photovoltaic products is greatly affected by factors such as market supply and demand and raw material prices, it is difficult for purchasers to lock in costs. The traditional procurement model usually requires multiple steps such as inquiry, price comparison, and negotiation, which is cumbersome and inefficient. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a collaborative procurement system and method based on multi-source data fusion and dynamic decision-making. The procurement process is transparent and concise, relevant reference suggestions are provided for commodity prices, and intelligent scheduling of pre-deposits improves capital utilization, thereby shortening the overall collaborative processing time of multiple parties.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a collaborative procurement method based on multi-source data fusion and dynamic decision-making, comprising: Receive the photovoltaic equipment purchase order submitted by the purchaser in the order collaborative processing subsystem; Providing photovoltaic equipment purchase orders to suppliers, and providing suggested quotations to suppliers by calling the price prediction model through the dynamic pricing module based on the commodities transmitted by the suppliers to the commodity collaborative management subsystem; Accept the photovoltaic equipment purchase order after the supplier has modified the unit price according to the proposed quotation; Provide the operator with the revised photovoltaic equipment purchase order and commodity-related price information, generate an electronic pricing sheet based on the operator's review of the supplier's pricing, and send it to the purchaser; In response to the purchaser confirming the price of the electronic pricing sheet, the pre-deposit management module of the agreement collaboration management subsystem performs pre-deposit deduction on the purchaser's pre-deposit account, otherwise no deduction is made; In response to the supplier confirming receipt of the photovoltaic equipment purchase order, a completion request is sent to the purchaser and the operator, otherwise no request is sent; In response to both the purchaser and the operator confirming the completion request, the photovoltaic equipment purchase order status is marked as completed.

[0006] Furthermore, the proposed quotation includes the total export price and the total market price, and the calculation formula is: In the formula, is the total export price, n is the target quantity, An is the material requirement factor, Pz For export price, is the total market price, Ps is the market price, Py is the entry price, α is the discount factor, β is the export factor, γ is the market factor, Px The source price of the material.

[0007] Furthermore, after parsing the column header text of the external data table through the intelligent distribution module, the commodity collaborative management subsystem calculates the semantic similarity with the system standard field through the word vector model, establishes automatic mapping for the fields whose similarity is not less than the dynamic threshold, and generates a manual correction prompt interface for the fields whose similarity is less than the dynamic threshold, and finally generates an editable Excel template.

[0008] Furthermore, the word vector model formula is: In the formula, For external fields i With standard fields j The comprehensive similarity of is the cosine similarity of word vectors, is the part-of-speech weight coefficient, is the field length difference penalty, , is the dynamic weight coefficient; The dynamic threshold formula is: In the formula, is the dynamic threshold, is the learning rate coefficient, is the confidence of history mapping, is the historical confidence sample size, Logarithmic compensation term for user correction times.

[0009] Furthermore, the order collaborative processing subsystem captures the commodity platform market price data in real time through an automatic price comparator, generates a price fluctuation curve based on the company's historical purchase price, predicts the price trend of raw materials, and assists the purchasing organization to lock in low-priced resources in advance. The calculation formula for the predicted price is: In the formula, for t +1 period predicted price, is the neural network weight coefficient, LSTM is a long short-term memory network, From the time point t − n arrive t The historical price series of is the market sentiment index, ARIMA is the autoregressive integrated moving average model, is the residual correction term.

[0010] Furthermore, the order collaborative processing subsystem integrates the non-price factors of suppliers, generates an optimal comprehensive cost solution based on a comprehensive cost quantification model, dynamically updates the supplier abnormal list database based on a dynamic scoring mechanism for the supplier abnormal list, and intercepts suppliers with abnormal operations and bid-rigging behaviors.

[0011] Furthermore, the comprehensive cost quantification model is: In the formula, TC For the comprehensive cost, is the total number of purchased goods. For the i The purchase price of the goods. For the i The purchase quantity of the following commodities: is the risk cost conversion coefficient, is the total number of risk factors, is the indicator weight determined by the entropy weight method, is a risk factor; The dynamic scoring mechanism for the abnormal supplier list is as follows: In the formula, Score supplier risk; For financial health, is the contract breach density, is the confidence level of bid-rigging behavior, a , b , c Feature weights obtained for training the random forest model.

[0012] In a second aspect, the present invention provides a collaborative procurement system based on multi-source data fusion and dynamic decision-making, comprising: Order receiving module: receiving the photovoltaic equipment purchase order submitted by the purchaser in the order collaborative processing subsystem; Dynamic pricing module: Provides photovoltaic equipment purchase orders to suppliers, and provides suppliers with recommended quotations by calling the price prediction model through the dynamic pricing module based on the commodities transmitted by the suppliers to the commodity collaborative management subsystem; Order modification module: accepts the photovoltaic equipment purchase order after the supplier modifies the unit price according to the suggested quotation; Price review module: Provide the operator with the revised photovoltaic equipment purchase order and commodity-related price information, generate an electronic pricing sheet based on the operator's review of the supplier's pricing, and send it to the purchaser; Pricing confirmation module: in response to the purchaser confirming the pricing of the electronic pricing sheet, the pre-deposit management module of the agreement collaboration management subsystem performs pre-deposit deduction on the purchaser's pre-deposit account, otherwise no deduction is made; Receipt confirmation module: in response to the supplier confirming receipt of the photovoltaic equipment purchase order, a completion request is sent to the purchaser and the operator, otherwise no request is sent; Order completion module: In response to both the purchaser and the operator confirming the completion request, the photovoltaic equipment purchase order status is marked as completed.

[0013] In a third aspect, the present invention provides a collaborative procurement device based on multi-source data fusion and dynamic decision-making, including a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps of any of the methods described above.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of any of the above methods when executed by a processor.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This solution provides a collaborative procurement system and method based on multi-source data fusion and dynamic decision-making. It provides order procurement services for purchasers, suppliers and operators through the agreement collaborative management subsystem, commodity collaborative management subsystem and order collaborative processing subsystem. The procurement process is transparent and concise, and reference suggestions are provided in terms of commodity prices. The intelligent scheduling of pre-deposit improves the utilization rate of funds, and shortens the overall collaborative processing time of multiple parties. 2. The commodity collaborative management subsystem of this solution uses the intelligent distribution module to parse the column header text of the external data table, calculates the semantic similarity with the system standard field through the word vector model, establishes automatic mapping for fields whose similarity is not less than the dynamic threshold, and generates a manual correction prompt interface for fields whose similarity is less than the dynamic threshold. By introducing dynamic weight optimization, transfer learning feedback mechanism, multi-dimensional constraint inheritance, etc., in the data set test, the automatic mapping success rate is greatly improved, and the amount of manual correction after system iteration is significantly reduced, realizing field-level version control and meeting compliance audit requirements; 3. The order collaborative processing subsystem of this solution complements the classical statistical model through deep learning methods, taking into account both short-term market fluctuations and long-term cyclical laws, and significantly improving the robustness of price trend forecasting; through the entropy weight method, implicit cost mapping and game theory model, it transforms non-price risks into quantifiable indicators, breaking through the limitation of the low-price theory in traditional procurement decisions; based on real-time feedback on inventory status, capital liquidity and market volatility, it automatically switches procurement modes to achieve a dynamic balance between cost and risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings: Figure 1 A flowchart of a collaborative procurement method based on multi-source data fusion and dynamic decision-making provided in the first embodiment of the present invention; Figure 2 A photovoltaic purchase and sales flow chart of an e-commerce platform supermarket based on a collaborative purchasing method based on multi-source data fusion and dynamic decision-making provided in the first embodiment of the present invention; Figure 3 An order pricing flow chart of a collaborative procurement method based on multi-source data fusion and dynamic decision-making provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0017] Traditional procurement management systems have problems such as multi-party data silos, delayed pricing decisions, and lengthy financial settlement processes. Especially in the photovoltaic equipment procurement scenario, when it comes to special needs such as supplier dynamic price adjustment, multi-level approval processes, and pre-deposit management, the existing system lacks intelligent coordination mechanisms and real-time data integration capabilities. With the rapid development of the photovoltaic industry, the traditional framework procurement model also needs to be continuously innovated and improved to adapt to market changes and user needs.

[0018] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present invention can be combined with each other without conflict.

[0019] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present invention belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present invention.

[0020] Embodiment 1: See also Figure 1-3 This embodiment discloses a collaborative procurement solution based on multi-source data fusion and dynamic decision-making, which adopts a three-layer architecture design of data layer, business layer and application layer, wherein: the data layer integrates the supplier database, procurement agreement library, commodity information library, pre-deposit account library and market situation database; the business layer includes the agreement management module, commodity distribution module, intelligent pricing module, order collaboration module and dynamic price adjustment engine; the application layer is used to provide the supplier portal, the purchaser workbench, the operator management platform and the financial system interface. Specifically, the core business layer system of this embodiment includes: 1. Protocol Collaboration Management Subsystem 1. Agreement digitization module: Convert offline contract terms into structured agreement templates, including: agreed purchasing parties, sales model (whether to allow bargaining / dynamic pricing), settlement method, logistics time constraint configuration (delivery deadline ± N Daily flexible settings), establish an organizational relationship mapping table.

[0021] 2. Pre-deposit management module: Supports multi-level account system (group-subsidiary account tree): Create the group master account (root node) through the account tree management unit, and the subsidiary generates a secondary account (sub-node) through registration request, and inherits the global permission policy of the root account; the virtual sub-account is generated by the business unit (such as the project team), and a mapping relationship is established with the physical account. Its capital changes are transferred through the bank-enterprise direct connection interface of the physical account. Refined account management and control: The separation of fund ownership and use rights is achieved through the tree-shaped account structure to solve the problem of mixed accounts of multiple legal entities in group enterprises.

[0022] 3. Funds freeze / thaw interface (directly connected to the bank system): When receiving a freeze instruction, the interface module calls the bank system and adds freeze type parameters (such as limiting only spending, limiting all transactions); the thawing operation is implemented through an asynchronous message queue. When the preset thawing conditions are met (the freezing period expires), an thawing request is automatically sent to the bank system and the local account status is updated. The freezing interface directly connected to the bank system reduces the delay in fund control from hours to seconds, avoiding human operational errors.

[0023] 4. Balance warning trigger (configurable threshold): The warning engine periodically polls the account balance data. If it detects a subsidiary AAccount balance is below the threshold X , an early warning event is generated and pushed to the financial personnel's mobile terminal; the early warning rules support logical expression configuration.

[0024] 2. Commodity Collaborative Management Subsystem 1. Intelligent distribution module: Product classification graph builder (supports three-level classification of photovoltaic modules): Generates a three-level classification knowledge graph in the field of photovoltaic modules, where the first-level classification is product type, the second-level classification is technical specifications, and the third-level classification is application scenarios. It also dynamically updates the classification attribute association relationship based on historical transaction data. The product classification graph builder adopts a hybrid ontology construction method.

[0025] Batch import template generator (automatically matching product attribute fields): configured to parse product attribute fields from external data sources, match system preset fields through field similarity algorithm, and automatically generate standardized import templates; the batch import template generator contains a field mapping engine, and its operating logic is: (1) Parse the column header text of the external data table and calculate the semantic similarity with the system standard field through the word vector model. The word vector model formula is: In the formula, For external fields i With standard fields j The comprehensive similarity of (value 0-1); is the cosine similarity of word vectors (basic semantic matching); is the part-of-speech weight coefficient (to strengthen professional terminology matching), when both parties are nouns = 1, noun + verb = 1.5, other combinations = 2; is the field length difference penalty; , is the dynamic weight coefficient ( + =1), initialize =0.7, =0.3. This model introduces a part-of-speech weight mechanism to enhance the matching accuracy for professional nouns; it also increases the penalty for field length differences to avoid " address "and" detailed _ address " and other similar words; RoBERTa - wwm - ext The pre-trained model generates word vectors with domain corpus fine-tuning.

[0026] (2) Automatic mapping is established for fields with similarity ≥ dynamic threshold, and a manual correction prompt interface is generated for fields with similarity < dynamic threshold. The dynamic threshold formula is: In the formula, is the dynamic threshold, is the learning rate coefficient (default 0.05); is the historical mapping confidence (successful mapping = 1, corrected mapping = 0.7, artificial addition = 0.3); is the historical confidence sample size; The logarithmic compensation term for the number of user corrections.

[0027] Execution logic: When ≥ , triggers automatic mapping and generates mapping traceability logs. The similarity is in [0.6, ) interval, the interface displays: recommended mapping path (with confidence star), difference location description (such as: "field length difference of 3 characters, part of speech weight deducted 0.15"). When the similarity is <0.6, manual intervention is forced to display cross-table association suggestions (based on foreign key probability analysis).

[0028] (3) Finally, an editable file with field mapping is generated. Excel Template. By introducing dynamic weight optimization, transfer learning feedback mechanism, multi-dimensional constraint inheritance, etc., the success rate of automatic mapping has been greatly improved in data set testing, the amount of manual corrections after system iteration has been significantly reduced, field-level version control has been achieved, and compliance audit requirements have been met.

[0029] Sales Application Verifier (Price Compliance Pre-review): Pre-review sales applications based on preset price compliance rules, including verifying whether the sales price is lower than the manufacturer's suggested price or the regional minimum price. The Sales Application Verifier has a built-in multi-dimensional verification model, including: (1) Price pre-examination unit: connect to the manufacturer's price release system in real time to obtain the latest guidance price as a benchmark; (2) Regional compliance unit: matching regional price limit policies based on the buyer’s registered place; (3) Special approval channel unit: Generate an exception approval work order for the application that triggers the early warning, and upload a scanned copy of the legal person's authorization letter.

[0030] 2. Dynamic Pricing Module: Establish a dynamic price adjustment mechanism and a multi-factor price model: collect material, market, and cost fluctuation parameters, and generate commodity benchmark prices through a dynamic weight allocation algorithm. The operation logic of the multi-factor coupling price model is to API The interface obtains the standard price of the original material in real time; the sequence analysis method is used to calculate the influence weight of each factor on the price, the discount factor α , export factor β , Market Factors γ (Satisfying 0< α <1;0<{ β , γ}-1≤1); Material requirement factor An , Material source price Px , purchase price Py , export price Pz , market price Ps , total export price and total market price Output price calculation formula:

[0031] 3. Order Collaborative Processing Subsystem Establish a state machine engine: define order status (including waiting to be submitted → waiting for pricing → pricing to be confirmed → priced → waiting to be shipped → waiting for receipt → waiting to be completed → completed → invoiced, etc.) and a state transition rule base (including multiple constraints).

[0032] The intelligent decision-making module combines the procurement and supply chain management needs of enterprises, and realizes full-process automation and risk control through automatic price comparison and multi-dimensional supplier evaluation: Automatic price comparison tool: Dynamic price monitoring and rule triggering, real-time capture of market price data from commodity platforms, etc., combined with the company's historical purchase price to generate a price fluctuation curve, predict raw material price trends, and assist procurement organizations to lock in low-priced resources in advance. The calculation formula for the predicted price is: In the formula, for t +1 period predicted price; is the neural network weight coefficient (adaptively adjusted through Bayesian optimization); LSTM is a long short-term memory network; From the time point t − n arrive t Historical price series of ARIMA It is an autoregressive integrated moving average model; For the market sentiment index (crawl news data BERT Semantic analysis and quantification generation); is the residual correction term (dynamic compensation based on historical error distribution).

[0033] Multi-dimensional supplier evaluation: Integrate non-price factors such as supplier qualifications, performance capabilities, rebate policies, etc., and generate the optimal comprehensive cost solution based on a comprehensive cost quantification model; dynamically update the supplier abnormal list database based on the dynamic scoring mechanism of the supplier abnormal list, and automatically intercept suppliers with abnormal operations and bid-rigging behaviors.

[0034] The comprehensive cost quantification model is: In the formula, TC For the comprehensive cost, is the total number of purchased goods. For the i The purchase price of the goods. For the i The purchase quantity of the following commodities: is the risk cost conversion coefficient, is the total number of risk factors, is the indicator weight determined by the entropy weight method, Risk factors (including qualification level , Delivery delay rate , quality defect rate ).

[0035] The dynamic scoring mechanism for the abnormal supplier list is as follows: In the formula, Score supplier risk; For financial health (based on AltmanZ - score Improved algorithms); is the contract breach density (number of breaches per unit time / total number of transactions); The confidence level of bid-rigging behavior (calculated through analysis of related-party transaction graphs); a , b , c Feature weights obtained for training the random forest model.

[0036] Order completion module: During the execution or after the completion of a PV order, the supplier can initiate an order completion application. After confirmation by the multi-signature verification condition function of the operator and the purchaser, the deposit will be refunded based on the dynamic calculation model for the deposit refund according to the final confirmed amount of the sales order, and the sales order status will be changed to "Completed".

[0037] The dynamic calculation model for pre-deposit refund is: In the formula, Rrefund is the actual refund amount; The initial deposit amount; Cactual The actual amount of the order completed; Cforecast The estimated completion amount of the order; is the time value coefficient (reference LIBOR Dynamic adjustment of interest rates); ΔT The deviation between the actual execution cycle and the planned cycle of the order; is the penalty factor (set in layers according to supplier performance rating); Fpenalty The amount of liquidated damages.

[0038] The multi-signature verification condition function is: In the formula, Verifystatus Verify the status of order completion; For participants k The digital signature verification result (0 / 1); Verify the time validity of the timestamp hash value; sk is the weight index. Core business processes: 1. Order pricing process: (1) The purchaser selects and purchases goods in the order collaborative processing subsystem and submits a photovoltaic equipment purchase order, triggering a pending pricing state; (2) The supplier obtains the order to be priced and performs the following operations: a The supplier obtains the photovoltaic equipment purchase order from the purchaser by logging into the order collaborative processing subsystem, and then transfers the goods in the order to the goods collaborative management subsystem, and calls the price prediction model through the dynamic pricing module to obtain the recommended quotation; b . Refer to the suggested quotation and real-time freight calculation results to modify the unit price of the goods in the order; c .Submit the pricing request to the operator in the photovoltaic pricing management module of the order collaborative processing subsystem.

[0039] (3) Operator review: The operator logs in to the order collaborative processing subsystem, obtains and compares the commodity market guidance price, historical transaction price, and agreement binding price through the automatic price comparator in the order collaborative processing subsystem, displays the price change trend on the manual review interface, determines the sales price of the goods in the order with reference to the above information, and reviews the supplier's pricing. After the review is passed, an electronic pricing sheet is generated.

[0040] (4) The purchaser confirms the pricing in the order management module of the order collaborative processing subsystem. After final confirmation, the deposit management module of the agreement collaborative management subsystem executes the pre-deposit deduction from the purchaser's pre-deposit account.

[0041] Order completion process: (1) The supplier logs in to the order collaborative processing subsystem and can initiate a completion request for orders that have been confirmed to have been received. After confirmation by the operator and the purchaser, the order becomes completed. The following conditions must be met: a .Logistics information verification (matching delivery order number and receipt record); b .The purchaser uploads the quality acceptance certificate in the order collaborative processing subsystem.

[0042] (2) Intelligent refund calculation: After the purchaser completes the confirmation, if the order is completed in part, a refund can be made in the original way according to the proportion.

[0043] Hardware deployment solution: containerized deployment of cloud-native architecture, using Kafka + Flink Realize real-time data stream processing and support 500+ order event processing per second. ‌‌‌‌‌ This technical solution integrates key technologies such as blockchain evidence storage, intelligent algorithm decision-making, and real-time data stream processing to build a collaborative procurement system with self-optimization capabilities, achieving a dynamic balance between procurement efficiency and cost control while ensuring compliance. In summary, this solution uses online processes to ensure that the procurement process is transparent and compliant, improves procurement efficiency, shortens the time for multi-party collaborative processing, and intelligent scheduling of pre-deposits improves capital utilization. Frame-purchased goods have different price adjustment modes, production locations, and transportation costs according to different types of materials. The present invention ensures contract flexibility and traceability.

[0044] Embodiment 2: A collaborative procurement system based on multi-source data fusion and dynamic decision-making can implement a collaborative procurement method based on multi-source data fusion and dynamic decision-making described in Embodiment 1, including: Order receiving module: receiving the photovoltaic equipment purchase order submitted by the purchaser in the order collaborative processing subsystem; Dynamic pricing module: Provides photovoltaic equipment purchase orders to suppliers, and provides suppliers with recommended quotations by calling the price prediction model through the dynamic pricing module based on the commodities transmitted by the suppliers to the commodity collaborative management subsystem; Order modification module: accepts the photovoltaic equipment purchase order after the supplier modifies the unit price according to the suggested quotation; Price review module: Provide the operator with the revised photovoltaic equipment purchase order and commodity-related price information, generate an electronic pricing sheet based on the operator's review of the supplier's pricing, and send it to the purchaser; Pricing confirmation module: in response to the purchaser confirming the pricing of the electronic pricing sheet, the pre-deposit management module of the agreement collaboration management subsystem performs pre-deposit deduction on the purchaser's pre-deposit account, otherwise no deduction is made; Receipt confirmation module: in response to the supplier confirming receipt of the photovoltaic equipment purchase order, a completion request is sent to the purchaser and the operator, otherwise no request is sent; Order completion module: In response to both the purchaser and the operator confirming the completion request, the photovoltaic equipment purchase order status is marked as completed.

[0045] Embodiment three: The embodiment of the present invention also provides a collaborative procurement device based on multi-source data fusion and dynamic decision-making, which can implement the collaborative procurement method based on multi-source data fusion and dynamic decision-making described in the first embodiment, including a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to perform the steps of the following method: Receive the photovoltaic equipment purchase order submitted by the purchaser in the order collaborative processing subsystem; Providing photovoltaic equipment purchase orders to suppliers, and providing suggested quotations to suppliers by calling the price prediction model through the dynamic pricing module based on the commodities transmitted by the suppliers to the commodity collaborative management subsystem; Accept the photovoltaic equipment purchase order after the supplier has modified the unit price according to the proposed quotation; Provide the operator with the revised photovoltaic equipment purchase order and commodity-related price information, generate an electronic pricing sheet based on the operator's review of the supplier's pricing, and send it to the purchaser; In response to the purchaser confirming the price of the electronic pricing sheet, the pre-deposit management module of the agreement collaboration management subsystem performs pre-deposit deduction on the purchaser's pre-deposit account, otherwise no deduction is made; In response to the supplier confirming receipt of the photovoltaic equipment purchase order, a completion request is sent to the purchaser and the operator, otherwise no request is sent; In response to both the purchaser and the operator confirming the completion request, the photovoltaic equipment purchase order status is marked as completed.

[0046] Embodiment 4: The embodiment of the present invention further provides a computer-readable storage medium, which can implement the collaborative procurement method based on multi-source data fusion and dynamic decision-making described in the first embodiment, and stores a computer program thereon, which implements the steps of the following method when executed by a processor: Receive the photovoltaic equipment purchase order submitted by the purchaser in the order collaborative processing subsystem; Providing photovoltaic equipment purchase orders to suppliers, and providing suggested quotations to suppliers by calling the price prediction model through the dynamic pricing module based on the commodities transmitted by the suppliers to the commodity collaborative management subsystem; Accept the photovoltaic equipment purchase order after the supplier has modified the unit price according to the proposed quotation; Provide the operator with the revised photovoltaic equipment purchase order and commodity-related price information, generate an electronic pricing sheet based on the operator's review of the supplier's pricing, and send it to the purchaser; In response to the purchaser confirming the price of the electronic pricing sheet, the pre-deposit management module of the agreement collaboration management subsystem performs pre-deposit deduction on the purchaser's pre-deposit account, otherwise no deduction is made; In response to the supplier confirming receipt of the photovoltaic equipment purchase order, a completion request is sent to the purchaser and the operator, otherwise no request is sent; In response to both the purchaser and the operator confirming the completion request, the photovoltaic equipment purchase order status is marked as completed.

[0047] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.

[0048] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of an all-hardware embodiment, an all-software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of one or more computer-usable storage media (including but not limited to disk storage, CD - ROM , optical storage, etc.).

[0049] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0050] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0051] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1The steps for the functions specified in one or more boxes.

[0052] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A collaborative procurement method based on multi-source data fusion and dynamic decision-making, characterized in that: include: Receive the photovoltaic equipment purchase order submitted by the purchaser in the order collaborative processing subsystem; Providing photovoltaic equipment purchase orders to suppliers, and providing suggested quotations to suppliers by calling the price prediction model through the dynamic pricing module based on the commodities transmitted by the suppliers to the commodity collaborative management subsystem; Accept the photovoltaic equipment purchase order after the supplier has modified the unit price according to the proposed quotation; Provide the operator with the revised photovoltaic equipment purchase order and commodity-related price information, generate an electronic pricing sheet based on the operator's review of the supplier's pricing, and send it to the purchaser; In response to the purchaser confirming the price of the electronic pricing sheet, the pre-deposit management module of the agreement collaboration management subsystem performs pre-deposit deduction on the purchaser's pre-deposit account, otherwise no deduction is made; In response to the supplier confirming receipt of the photovoltaic equipment purchase order, a completion request is sent to the purchaser and the operator, otherwise no request is sent; In response to both the purchaser and the operator confirming the completion request, the photovoltaic equipment purchase order status is marked as completed.

2. The collaborative procurement method based on multi-source data fusion and dynamic decision-making according to claim 1 is characterized in that: The proposed quotation includes the total export price and the total market price, and the calculation formula is: In the formula, is the total export price, n is the target quantity, An is the material requirement factor, Pz For export price, is the total market price, Ps is the market price, Py is the purchase price, α is the discount factor, β is the export factor, γ is the market factor, Px The source price of the material.

3. The collaborative procurement method based on multi-source data fusion and dynamic decision-making according to claim 1 is characterized in that: After the commodity collaborative management subsystem parses the column header text of the external data table through the intelligent distribution module, it calculates the semantic similarity with the system standard field through the word vector model, establishes automatic mapping for the fields whose similarity is not less than the dynamic threshold, and generates a manual correction prompt interface for the fields whose similarity is less than the dynamic threshold, and finally generates an editable Excel template.

4. The collaborative procurement method based on multi-source data fusion and dynamic decision-making according to claim 3 is characterized in that: The word vector model formula is: In the formula, For external fields i With standard fields j The comprehensive similarity of is the cosine similarity of word vectors, is the part-of-speech weight coefficient, is the field length difference penalty, , is the dynamic weight coefficient; The dynamic threshold formula is: In the formula, is the dynamic threshold, is the learning rate coefficient, is the confidence of history mapping, is the historical confidence sample size, The logarithmic compensation term for the number of user corrections.

5. The collaborative procurement method based on multi-source data fusion and dynamic decision-making according to claim 1 is characterized in that: The order collaborative processing subsystem captures the market price data of the commodity platform in real time through the automatic price comparator, generates a price fluctuation curve based on the company's historical purchase price, predicts the price trend of raw materials, and assists the purchasing organization to lock in low-priced resources in advance. The calculation formula for the predicted price is: In the formula, for t +1 period predicted price, is the neural network weight coefficient, LSTM is a long short-term memory network, From the time point t − n arrive t The historical price series of is the market sentiment index, ARIMA is the autoregressive integrated moving average model, is the residual correction term.

6. The collaborative procurement method based on multi-source data fusion and dynamic decision-making according to claim 1 is characterized in that: The order collaborative processing subsystem integrates the non-price factors of suppliers, generates the optimal comprehensive cost solution based on the comprehensive cost quantification model, dynamically updates the supplier abnormal list database based on the dynamic scoring mechanism of the supplier abnormal list, and intercepts suppliers with abnormal operations and bid rigging behaviors.

7. The collaborative procurement method based on multi-source data fusion and dynamic decision-making according to claim 6 is characterized in that: The comprehensive cost quantification model is: In the formula, TC For the comprehensive cost, is the total number of purchased goods. For the i The purchase price of the goods. For the i The purchase quantity of the following commodities: is the risk cost conversion coefficient, is the total number of risk factors, is the indicator weight determined by the entropy weight method, is a risk factor; The dynamic scoring mechanism for the abnormal supplier list is as follows: In the formula, Score supplier risk; For financial health, is the contract breach density, is the confidence level of bid-rigging behavior, a , b , c Feature weights obtained for training the random forest model.

8. A collaborative procurement system based on multi-source data fusion and dynamic decision-making, characterized by: include: Order receiving module: receiving the photovoltaic equipment purchase order submitted by the purchaser in the order collaborative processing subsystem; Dynamic pricing module: Provides photovoltaic equipment purchase orders to suppliers, and provides suppliers with recommended quotations by calling the price prediction model through the dynamic pricing module based on the commodities transmitted by the suppliers to the commodity collaborative management subsystem; Order modification module: accepts the photovoltaic equipment purchase order after the supplier modifies the unit price according to the suggested quotation; Price review module: Provide the operator with the revised photovoltaic equipment purchase order and commodity-related price information, generate an electronic pricing sheet based on the operator's review of the supplier's pricing, and send it to the purchaser; Pricing confirmation module: in response to the purchaser confirming the pricing of the electronic pricing sheet, the pre-deposit management module of the agreement collaboration management subsystem performs pre-deposit deduction on the purchaser's pre-deposit account, otherwise no deduction is made; Receipt confirmation module: in response to the supplier confirming receipt of the photovoltaic equipment purchase order, a completion request is sent to the purchaser and the operator, otherwise no request is sent; Order completion module: In response to both the purchaser and the operator confirming the completion request, the photovoltaic equipment purchase order status is marked as completed.

9. A collaborative procurement device based on multi-source data fusion and dynamic decision-making, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

Images