WO2021003953A1 - 一种用于交易平台实现多元商品智能交易推荐的方法和系统 - Google Patents

一种用于交易平台实现多元商品智能交易推荐的方法和系统 Download PDF

Info

Publication number
WO2021003953A1
WO2021003953A1 PCT/CN2019/120571 CN2019120571W WO2021003953A1 WO 2021003953 A1 WO2021003953 A1 WO 2021003953A1 CN 2019120571 W CN2019120571 W CN 2019120571W WO 2021003953 A1 WO2021003953 A1 WO 2021003953A1
Authority
WO
WIPO (PCT)
Prior art keywords
supply
order
commodity
purchase
transaction
Prior art date
Application number
PCT/CN2019/120571
Other languages
English (en)
French (fr)
Inventor
周永东
Original Assignee
周永东
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 周永东 filed Critical 周永东
Publication of WO2021003953A1 publication Critical patent/WO2021003953A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0603—Catalogue ordering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0605—Supply or demand aggregation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
    • G06Q30/0635—Processing of requisition or of purchase orders

Definitions

  • the present invention generally relates to the field of commodity trading, and more specifically, to a method and system for a trading platform to implement intelligent trading recommendations for multiple commodities.
  • the patent document with application number 201910148206.1 discloses a transaction matching method and system.
  • the transaction matching platform obtains supplier-side goods supply data; the transaction matching platform obtains demand-side goods demand data; the transaction matching platform uploads are sorted and standardized.
  • the data of the big data platform is sent to the big data platform; the big data platform queries and matches the data of the supplier and the demander, and sends matching output suggestions to the transaction matching platform after calculation through the data computing center; the transaction matching platform provides matching output suggestions on the supplier side; the transaction matching platform gives the demand Paired output recommendations for the square end. It can complete the efficient allocation of resources and effectively improve the efficiency of cost control or sales channel expansion.
  • the transaction matching process how to enable buyers to obtain high-quality suppliers to provide satisfactory products, and suppliers obtain the opportunity to provide high-quality buyers with supply, that is, the problem of improving the success rate of transactions has not been solved.
  • the patent document with the application number 201410143963.5 discloses a transaction matching method and system, which processes the transaction reservation submitted by the user and generates a transaction order; the transaction order is transmitted as a message in the first message middleware; Any one server in the matching processing cluster composed of multiple servers reads the transaction order transmitted in the form of a message in the first message middleware, and matches it with the product on sale to generate a result order; The order is processed for business.
  • the distributed architecture can smoothly handle the problems caused by the growth of business volume, ensure the normal operation of the transaction matching system, and reduce the risk of the transaction matching system. This patent also does not solve the problem of increasing the success rate of transactions and the problem of matching specific products.
  • the present invention needs to solve at least one of the following problems, improve the transaction success rate on the trading platform; solve the efficiency problem in the commodity transaction matching process; solve the problem of improving the effectiveness in the commodity transaction matching process; solve the problem of multiple purchasers Suppliers can purchase multiple satisfactory products at the same time.
  • the present invention provides a method for a trading platform to implement multi-commodity transaction recommendation.
  • the method includes the first step S1, obtaining the buyer’s commodity demand information at the purchasing end, forming a purchase order, and obtaining the supplier’s commodity at the supplier end.
  • the second step S2 evaluate the buyer on the purchasing side, determine the priority of the buyer, evaluate the supplier on the supplier side, and determine the priority of the supplier; the third step S3, on the purchasing side Sort the purchase order based on the priority of the purchaser to obtain the purchase order queue, and sort the supply order based on the supplier's priority on the supplier end to obtain the supply order queue; the fourth step S4, based on the purchase order on the purchase end
  • the order of the queue is followed by transaction matching on the purchase order, and the supply order is selected in turn based on the order of the supply order queue at the supplier side, and the transaction matching with the purchase order is performed until the transaction matching of all the purchase orders is completed;
  • the purchase order queue includes : Purchase order C 1 , purchase order C 2 ,..., purchase order C x , where x is a non-empty positive integer;
  • the supply order queue includes: supply order G 1 , supply order G 2 ,..., supply Single G y , where y is a non-empty positive integer.
  • the fourth step S4 includes obtaining the order of the products in the purchase order; matching the products according to the order of the products in the purchase order until all the products in the purchase order are matched.
  • purchase order C i 1, 2, ..., X
  • C is the X-element predicate symbol , C 1 , C 2 ,..., C X are terms,
  • C i is a P-element predicate symbol, M 1 , M 2 , ..., M P are sub-items, then the atomic formula of the generic relationship is C i (M K ) , Where i and k are non-empty positive integers;
  • the matching of the commodity transaction includes that the commodities in the purchase order include: commodity M 1 , commodity M 2 , ..., commodity M p , where p is a non-empty positive integer;
  • the goods in the manifest include: goods N 1 , N 2 , ..., N q , where q is a non-empty positive integer;
  • the first step is to select the goods M k , based on the order of the supply order queue, from each supply order Select a commodity of the same type as the commodity M k to obtain a commodity G j (N r ) with the same supply and demand code;
  • the second sub-step match the commodity G j (N r ) with the commodity M k in the same supply and demand code;
  • Select the next product repeat the first sub-step to the second sub-step, until the matching of all the products M k in the purchase order is completed, where 1 ⁇ k ⁇ p, 1 ⁇ j ⁇ y, 1 ⁇ r ⁇ q.
  • performing transaction matching on the purchase order further includes setting transaction necessary conditions according to the product demand information of the product M k , and using the transaction necessary conditions to filter the supply and demand products G j (N r ) with the same code , Obtain the recommended product sequence; the necessary transaction conditions include at least one of quantity, product quality, and delivery period.
  • using the necessary transaction conditions to filter the supply and demand commodities of the same code includes: a first sub-step: setting a quantity screening threshold according to the commodity demand information of the commodity M k ; said product supply and the same code G j (N r), removing the number is less than the threshold number of screening code with the supply and demand of goods G j (N r), to give a first product screening sequence; and a second sub-step, according to the commodity
  • the commodity demand information of M k sets a quality screening threshold; the quality screening threshold is used to screen the first screened commodity sequence to obtain a second screened commodity sequence; the third sub-step is to set the commodity demand information according to the commodity M k Set a time screening threshold; use the time screening threshold to screen the second screening product sequence to obtain a recommended product sequence; the fourth sub-step, match the recommended product sequence with the product M k .
  • the method further includes updating the purchase order and the supply order every time the matching of a commodity M k is completed, so that the purchase order matches the purchaser's purchase requirements in real time, so that the The supply order matches the supplier's supply capacity in real time.
  • a system for a trading platform to implement multiple commodity trading recommendations including a first device 1 for obtaining commodity demand information from buyers at the purchasing end, forming a purchase order, and providing The cargo side obtains the supplier’s commodity supply information to form a supply order; the second device 2 is used to evaluate the buyer on the purchasing side to determine the priority of the buyer, and to evaluate the supplier on the supplier side to determine the priority of the supplier Level; The third device 3 is used to sort the purchase orders based on the priority of the purchaser on the purchasing end to obtain the purchase order queue, and to sort the supply orders on the supplier end based on the priority of the supplier to obtain the supply order queue The fourth device 4 is used to match the purchase orders in sequence based on the order of the purchase order queue at the purchasing end, and select the supply orders in turn based on the order of the purchase order queue at the supplier end, and match the purchase orders until Complete transaction matching of all purchase orders.
  • the fourth device 4 includes a first module for obtaining the order of the commodities in the purchase order; a second module for matching the commodities in the order of the commodities in the purchase order , Until all commodities in the purchase order are matched.
  • the second module includes a first sub-module, and the first sub-module is used to select the product M k , and based on the order of the supply order queue, the second module includes Commodities of the same type of commodity M k , obtain the commodity G j (N r ) of the same supply and demand code; match the commodity G j (N r ) of the same supply and demand with the commodity M k ; select the next commodity, and repeat the first From the first step to the second step, until the matching of all the goods M k in the purchase order is completed, where 1 ⁇ k ⁇ p, 1 ⁇ j ⁇ y, 1 ⁇ r ⁇ q; wherein, the goods in the purchase order include : Commodity M 1 , Commodity M 2 , ..., Commodity M p , where p is a non-empty positive integer; the commodities in the supply list include: Commodities N 1 , N 2 , ..., N q , where q is a non-empty
  • the second module includes a second sub-module, and the second sub-module is configured to set necessary transaction conditions according to the commodity demand information of the commodity Mk, and use the necessary transaction conditions to filter the Supply and demand the same code commodity Gj(Nr), obtain a recommended commodity sequence; the necessary transaction conditions include at least one of quantity, commodity quality, and delivery period.
  • a method for recommending commodity transactions on a trading platform includes, at the purchasing end, obtaining commodity demand information of buyers to form a purchase order; evaluating buyers to determine their priority; The purchase orders are sorted based on the priority of the purchaser to obtain the purchase order queue; the purchase orders are sequentially matched based on the order of the purchase order queue until the transaction matching of all purchase orders is completed; the purchase order queue includes: purchase order C 1 , Purchase order C 2 ,..., purchase order C x , where x is a non-empty positive integer;
  • a method for recommending commodity transactions on a trading platform includes, at the supplier end, obtaining supplier's commodity supply information to form a supply order; evaluating the supplier and determining the supply Supplier priority level; order the supply orders based on the supplier priority level to obtain the supply order queue; select the supply orders in turn based on the order of the supply order queue, and match the purchase order with the purchase order until the transaction of all purchase orders is completed Match; the supply order queue includes: supply order G 1 , supply order G 2 , ..., supply order G y , where y is a non-empty positive integer.
  • the present invention evaluates suppliers and purchasers before the transaction, determines the purchaser priority level and the supplier priority level, sorts purchase orders and supply orders according to the priority level, and selects high-quality suppliers and purchasers. It is helpful to enable buyers to obtain high-quality goods and purchase goods from high-quality suppliers, thereby facilitating the conclusion of transactions.
  • the matching of supply and demand with the same code makes it possible to quickly screen out similar products from a large number of supply orders, and improve the efficiency of transaction matching recommendation.
  • the invention sorts the screening conditions according to the dependence of external factors, thereby improving the screening efficiency as a whole. Ensure the accuracy of data in real time, avoid multiple purchases and multiple supplies, and ensure the effectiveness of transaction recommendations.
  • Figure 1 is a schematic diagram of the steps of a method for implementing a multi-commodity intelligent transaction recommendation on a trading platform
  • Figure 2 is a schematic diagram of matching the purchase order sequence
  • Figure 3 is a schematic diagram of matching the goods in the purchase order
  • Figure 4 is a schematic diagram of trading matching of commodities
  • Figure 5 is a schematic diagram of the steps for selecting products using preset purchasing conditions.
  • Fig. 6 is a schematic diagram of a system for implementing a multi-commodity intelligent transaction recommendation on a trading platform.
  • Figure 1 shows a schematic diagram of the steps of a method for a trading platform to implement intelligent trading recommendations for multiple commodities.
  • a method for a trading platform to implement a multi-commodity intelligent transaction recommendation including the first step S1, obtain the buyer’s commodity demand information at the purchasing end, form a purchase order, and obtain the supply at the supplier end Supplier’s commodity supply information to form a supply order; the second step S2, evaluate the buyer on the purchasing side, determine the priority of the buyer, evaluate the supplier on the supply side, and determine the priority of the supplier; the third step S3, On the purchasing end, the purchase order is sorted based on the priority of the purchaser to obtain the purchase order queue, and the supply order is sorted based on the supplier's priority on the supplier end to obtain the supply order queue; the fourth step S4, on the purchase end
  • the purchase orders are matched in order based on the order of the purchase order queue, and the supply orders are selected in turn based on the order of the supply order queue at the supplier side, and the transaction is matched with the purchase order until the transaction matching of all the purchase orders is completed;
  • the order queue includes: purchase order C 1 , purchase order C 2 ,
  • the transaction matching of the purchase order refers to the ordering of the purchase order, and the transaction matching of the commodities in each purchase order in turn.
  • the non-empty positive integer means that the purchase order can be one or more; the supply order can also be one or more.
  • the present invention adopts the evaluation of the purchaser, and the selection of high-quality purchasers to give priority to the transaction is beneficial to promote the completion of the transaction. Promote suppliers to maintain good market behavior to play a catalytic role.
  • the purchase order queue includes: evaluating the purchaser, determining the priority purchase level of C i (M K ), using this as the basis for queuing, forming a purchase order queue, the order of purchase of goods, without reducing C i (M K ) Purchasing requirements, buyers give priority to C i (M K ) procurement to obtain the opportunity to give priority to satisfactory suppliers.
  • buyers purchase goods from the system, register in the system, and fill in the registration information in detail. After the purchaser has passed the registration and certification, he will formally become a system commodity purchaser, and he can evaluate the purchaser based on historical transaction records and other information through big data analysis. For buyers with a higher rating, priority can be given to transactions.
  • Figure 2 shows a schematic diagram of matching purchase order sequences.
  • the purchase order queue includes: purchase order C 1 , purchase order C 2 , ..., purchase order C x , where x is a non-empty positive integer.
  • purchase order C 1 The transaction, and then the transaction of purchase order C 2 is carried out in sequence, until all purchase orders in the system end the transaction.
  • the supply order queue includes: evaluating suppliers, determining the priority supplier level of G j (N r ), using this as the basis for queuing, forming a supply order queue, the order of commodity supply, without reducing G j (N r )
  • the comprehensive scoring standard for supply, the supplier gives priority to supply G j (N r ) can obtain the buyer's priority purchase opportunity.
  • Suppliers eligible to participate in system commodity trading activities shall register in the system, fill in and upload the registration information in detail. After the supplier passes the registration certification, it officially becomes the supplier of system products.
  • Suppliers Through historical transaction records and other information, through big data analysis, suppliers can be evaluated, and suppliers with good credit, qualification, and transaction records can be screened. For suppliers with a higher rating, priority transactions can be given.
  • the queuing of C i (M K ) and G j (N r ) is one of the necessary conditions for orderly and intelligently solving the multi-pair quantified commodity transaction mode. According to the needs of the actual application of the system, the commodity transaction queuing can enable high-performing buyers Preferential purchase of high-quality products from high-quality suppliers.
  • the supply order queue includes: supply order G 1 , supply order G 2 , ..., supply order G y , where y is a non-empty positive integer, first of all, the order of goods and purchase order in supply order G 1 Commodities are matched and recommended by transaction, and then the commodities in the supply order G 2 are matched and recommended, and proceed in sequence until a certain commodity in the purchase order is satisfied, or all supply orders are matched.
  • the present invention determines the priority of the purchaser and the priority of the supplier by evaluating the supplier and the purchaser before the transaction, and in the actual transaction, the purchase order and the supply order are sorted according to the priority, and high-quality suppliers are selected. Goods merchants and purchasers help buyers obtain high-quality goods and purchase goods from high-quality suppliers, thereby facilitating the conclusion of transactions.
  • the commodity demand information generally includes commodity price, commodity quality, commodity supplier qualification, commodity quantity, delivery time, etc., which can be set according to the needs of the purchaser.
  • the commodity supply information includes commodity price, commodity quality, commodity supplier qualification, supply quantity, delivery time, etc.
  • Figure 3 shows a schematic diagram of matching commodities in a purchase order.
  • the fourth step S4 includes obtaining the order of the products in the purchase order; matching the products according to the order of the products in the purchase order until all the products in the purchase order are matched.
  • Fig. 4 shows a schematic diagram of transaction matching for commodities.
  • the transaction matching of commodities includes that the commodities in the purchase order include: commodity M 1 , commodity M 2 , ..., commodity M p , where p is a non-empty positive integer;
  • the commodities in include: commodities N 1 , N 2 , ..., N q , where q is a non-empty positive integer;
  • the first sub-step select commodity M k , based on the order of the supply order queue, select from the supply list and Commodities of the same type of the commodity M k , obtain the commodity G j (N r ) of the same code as supply and demand;
  • the second sub-step match the commodity G j (N r ) of the same code as the supply and demand with the commodity M k ; select the next For commodities, repeat the first sub-step to the second sub-step until the matching of all the commodities M k in the purchase order is completed, where 1 ⁇ k ⁇ p, 1 ⁇ j ⁇ y, 1 ⁇ r ⁇ q.
  • the commodities in the purchase order and the commodities in the supply order can be coded according to their types, such as various commodities in the purchase order C i : commodity M 1 , commodity M 2 , ..., commodity M p , Specific to a certain commodity, it can be represented by C i (M k ), which means the kth commodity in the i-th purchase order.
  • C i (M k ) which means the kth commodity in the i-th purchase order.
  • the goods in the supply order G y can be expressed as G y (N r ), which means the rth product in the yth purchase order.
  • the present invention when matching commodities C i (M k ), the same type of matching must be performed first.
  • the present invention is based on the order of the supply order queue, that is, the sequential matching is performed according to the level of the supplier, so that the goods of the high-quality supplier are matched first.
  • the commodity of the same supply and demand code refers to the current commodity C i (M k ) on the current purchase order and the commodity G y (N r ) of the same type on each supply order.
  • the design capacity of C i G j B kr of M k is ⁇ 999, and the P products can accommodate C i G j B kr of the P group, and can accommodate P*999 C i G j B kr of pairings.
  • the channel for matching the supply and demand templates of the comprehensive library between the purchased varieties M k and C i G j B kr corresponding to G j (N r ) is opened.
  • performing transaction matching on the purchase order further includes: setting transaction necessary conditions according to the product demand information of the product M k , and using the transaction necessary conditions to filter the supply and demand products G j (N r ) with the same code , Obtain the recommended product sequence; the necessary transaction conditions include at least one of quantity, product quality, and delivery period.
  • Figure 5 shows a schematic diagram of the steps of selecting products using preset purchasing conditions.
  • Each pair of M k and G j (N r ) corresponding to each C i G j B kr must pass the criteria of the product delivery period, the quantity of purchase and supply, the quality threshold, the qualification threshold and other parameters. After changing the unit of measurement, the orderly correspondence between M k and a family of C i G j B kr G j (N r ) is verified as a valid pairing, and only valid G j (N r ) access matching operation procedures are allowed.
  • the present invention establishes a comprehensive library of data, knowledge, models, graphics, etc., constructs C i (M K ), G j (N r ) transaction templates, and uses transaction templates to express the relationship between C i (M K ) and G j (N r ) Appeals, intelligent matching of commodity transactions; system intelligence scores and points the minimum items of G j (N r ) price GJFr, quality GZFr, and qualification GZZFr, and gives a comprehensive score of G j (N r ) corresponding to each M k GHZ r is summarized, and the scores of supplied goods GZHP r are sorted from high to low.
  • it further includes a first sub-step of setting a quantity screening threshold according to the commodity demand information of the commodity M k ; using the quantity screening threshold to screen the supply and demand commodity G j (N r ) with the same code, and remove The supply and demand commodity G j (N r ) of the same code whose quantity is less than the quantity screening threshold obtains the first screening commodity sequence; the second sub-step, the quality screening threshold is set according to the commodity demand information of the commodity M k ; filtering said quality threshold screening said first screening sequence commodities, commodity obtain a second sequence screening; third sub-step of filtering the threshold requirement of the commodity according to the commodity information M k set time; time filtering using the threshold screening The second screening product sequence is used to obtain a recommended product sequence; the fourth sub-step is to match the recommended product sequence with the product M k .
  • the quantity screening threshold refers to a preset minimum value of supply according to the quantity of a certain commodity that needs to be purchased. For example, the quantity screening threshold is preset to 30% of the purchase quantity, and all supply quantities are lower than the purchase quantity. 30% of the supply and demand of the same size goods will be screened out, that is, for bulk purchase purchase orders, commodity transactions will be concentrated to fewer suppliers as much as possible to reduce the complexity of transactions.
  • the quality screening threshold refers to various restrictions imposed by the purchaser on the attributes of the goods themselves such as the source, weight, size, taste, etc., which can be adjusted according to the needs of the purchaser, or defined according to industry standards or administrative standards.
  • the time screening threshold refers to the limitation of the delivery time, which is set according to the time limitation requirements of the buyer.
  • the quantity requirements are first used for screening, then quality is used for screening, and finally the delivery time is used for screening, and the order cannot be changed.
  • the same code of supply and demand belongs to the screening condition that is the least dependent on the needs of buyers; the quantity requirement is dependent on the needs of buyers compared with the same code of supply and demand; further quality screening should not only meet the needs of buyers, but also comply with national regulations and industry standards And other additional conditions; further, the demand randomness of the delivery time is stronger than the aforementioned screening conditions.
  • the present invention sorts the screening conditions according to the dependence of external factors. Priority factors are less dependent on screening conditions. When screening is performed on screening conditions with less factor dependence, there are more screening targets and simple screening conditions. When the screening conditions are selected, although the screening conditions are complex and changeable, there are fewer screening targets, so that the screening efficiency of each screening process is consistent, and the overall screening efficiency is improved.
  • the method further includes updating the purchase order and the supply order every time the matching of the product M k is completed, so that the purchase order matches the purchaser's purchase requirements in real time, and the supply The manifest matches the supplier’s supply capacity in real time.
  • the updated purchase order refers to removing the products that have been matched from the purchase order, and keeping the unmatched products. Similarly, updating the supply order means removing the supplied goods that have been matched, and using the remaining supplied goods to continue to match the goods in the purchase order.
  • j quantity ykr 0, the purchased quantity of a product C YL , and the supplied quantity of a product G YL .
  • the purchase order quantity i and the purchase order quantity p in the purchase order are also updated in real time.
  • Fig. 6 shows a schematic diagram of a system for implementing a multi-commodity intelligent transaction recommendation on a trading platform.
  • a system for a trading platform to implement smart transaction recommendations for multiple commodities includes a first device 1 for obtaining buyer’s commodity demand information on the purchasing end, forming a purchase order, and obtaining it on the supplier end
  • the supplier’s commodity supply information forms a supply order
  • the second device 2 is used to evaluate the buyer on the purchasing side to determine the priority of the buyer, and to evaluate the supplier on the supply side to determine the priority of the supplier
  • the three devices 3 are used to sort the purchase orders based on the priority of the purchaser on the purchasing end to obtain the purchase order queue, and to sort the supply orders based on the priority of the supplier on the supply end to obtain the supply order queue
  • fourth Device 4 is used to match the purchase orders in sequence based on the order of the purchase order queue at the purchasing end, and select the supply orders in turn based on the order of the purchase order queue at the supplier end, and match the purchase orders with the purchase order until all purchases are completed Single transaction matching.
  • the fourth device 4 includes a first module for obtaining the order of the commodities in the purchase order; a second module for matching the commodities in the order of the commodities in the purchase order , Until all commodities in the purchase order are matched.
  • the second module includes a first sub-module, and the first sub-module is used to select the product M k , and based on the order of the supply order queue, the second module includes Commodities of the same type of commodity M k , obtain the commodity G j (N r ) of the same supply and demand code; match the commodity G j (N r ) of the same supply and demand with the commodity M k ; select the next commodity, and repeat the first From the first step to the second step, until the matching of all the goods M k in the purchase order is completed, where 1 ⁇ k ⁇ p, 1 ⁇ j ⁇ y, 1 ⁇ r ⁇ q; wherein, the goods in the purchase order include : Commodity M 1 , Commodity M 2 , ..., Commodity M p , where p is a non-empty positive integer; the commodities in the supply list include: Commodities N 1 , N 2 , ..., N q , where q is a non-empty
  • the second module includes a second sub-module, and the second sub-module is configured to set necessary transaction conditions according to the commodity demand information of the commodity M k , and use the necessary transaction conditions to filter the The supply and demand commodity G j (N r ) with the same code is used to obtain a recommended commodity sequence; the necessary transaction conditions include at least one of quantity, commodity quality, and delivery period.
  • a method for recommending commodity transactions on a trading platform includes, at the purchasing end, obtaining commodity demand information of buyers to form a purchase order; evaluating buyers to determine their priority; The purchase orders are sorted based on the priority of the purchaser to obtain the purchase order queue; the purchase orders are sequentially matched based on the order of the purchase order queue until the transaction matching of all purchase orders is completed; the purchase order queue includes: purchase order C 1 , Purchase order C 2 ,..., purchase order C x , where x is a non-empty positive integer;
  • a method for recommending commodity transactions on a trading platform includes, at the supplier end, obtaining supplier's commodity supply information to form a supply order; evaluating the supplier and determining the supply Supplier priority level; order the supply orders based on the supplier priority level to obtain the supply order queue; select the supply orders in turn based on the order of the supply order queue, and match the purchase order with the purchase order until the transaction of all purchase orders is completed Match; the supply order queue includes: supply order G 1 , supply order G 2 , ..., supply order G y , where y is a non-empty positive integer.
  • the present invention evaluates suppliers and purchasers before the transaction, determines the purchaser priority level and the supplier priority level, sorts purchase orders and supply orders according to the priority level, and selects high-quality suppliers and purchasers. It is helpful to enable buyers to obtain high-quality goods and purchase goods from high-quality suppliers, thereby facilitating the conclusion of transactions.
  • the matching of supply and demand with the same code makes it possible to quickly screen out similar products from a large number of supply orders, and improve the efficiency of transaction matching recommendation.
  • the invention sorts the screening conditions according to the dependence of external factors, thereby improving the screening efficiency as a whole. Ensure the accuracy of data in real time, avoid multiple purchases and multiple supplies, and ensure the effectiveness of transaction recommendations.
  • Buyers purchase CGDT (C i (M K )) products from the trading platform, register in the trading platform, and fill in the registration information in detail. After the purchaser has passed the registration and certification, he will formally become a system commodity purchaser. After big data information analysis, grade evaluation, and become a VIP purchaser on the trading platform. VIP buyers pay annual membership fees, and membership levels are divided into level 1, level 2, etc.
  • GYDT (G j (N r )) is eligible to participate in system commodity trading activities. Suppliers of GYDT (G j (N r )) register in the trading platform, fill in and upload the registration information in detail. After the supplier has passed the registration and certification, it will formally become a commodity supplier on the platform, and will be graded after big data information analysis.
  • the system sets up a complete, standardized and accurate comprehensive information knowledge expression template for GYDT (G j (N r )) suppliers, including commodities
  • GYDT G j (N r )
  • the three systems of bidding, quality assurance, and supplier qualification are in line with CGDT (C i (M K )) procurement requirements.
  • Table 1 shows the meta-knowledge control strategy parameter configuration and physical meaning description, which is used to guide the system CGDT (C i (M K )) and CGDT (C i (M K )) are fully engaged in multi-target preferred commodity transactions.
  • Trading platform data commodity purchase dynamic information, commodity supply dynamic information, commodity purchase variety, commodity purchase comprehensive score setting, commodity supply variety, commodity transaction data, closed-loop transaction data, transaction results, and purchase contract.
  • the data attributes of the commodity purchase dynamic information serial number, commodity purchase order number, commodity purchase order name, commodity purchase variety.
  • the "product purchasing dynamic information" database generated by the trading platform is the product purchasing information edited by the buyer; the buyer can also push the checked and edited “product purchasing” database information to the "product purchasing dynamic information” database.
  • Commodity purchase variety data attributes serial number, commodity name, commodity number, purchase quantity, purchased quantity, commodity purchase time, commodity delivery time, buyer membership level, commodity purchase comprehensive score setting.
  • Purchased quantity Every time a purchaser makes a purchase and the transaction is successful, the purchase quantity will be reduced by the corresponding purchase quantity.
  • Product purchase comprehensive score setting data attributes product comprehensive evaluation total score, product price weight, product price scoring algorithm, product quality weight, product quality score decomposition, product quality score algorithm, supplier qualification weight, supplier score decomposition, Supplier qualification score algorithm.
  • Product quality score decomposition Buyers can accept the default value set by the trading platform, or they can flexibly set the value of each sub-item of product quality according to the quality score weight according to the product purchase description.
  • Supplier qualification score decomposition Buyers can accept the default value set, or they can flexibly set the value of each sub-item of the supplier score according to the supplier's qualification score weight according to the commodity purchase description.
  • Commodity purchase dynamic information data attributes serial number, commodity supply order number, commodity supply order name, commodity supply variety.
  • the "commodity supply dynamic information" database generated by the trading platform is the product supply information edited by the supplier; the supplier can also push the checked and edited “commodity supply” database information to the "commodity supply dynamic information” database.
  • Product category data attributes serial number, product number, product name, supply quantity, supply standardization quantity, supplied quantity, price, standardization price, product quality, qualification certificate, qualification certificate picture, product delivery time, product Delivery time, supplier membership level.
  • Supplied quantity Every time a supplier participates in a bidding and the transaction is successful, the supply quantity will be reduced by the corresponding purchase quantity.
  • Standardized quantity and price of supply Based on the unit of measurement in the commodity purchase variety database, standardize data processing for the "supply quantity” and “price” parameters in the commodity supply variety database.
  • Commodity transaction data attributes serial number, purchase order queuing, supply order queuing, commodity transaction object, supply unit conversion, supply price score, supply quality score, supply qualification score, summary of supply goods quantitative score, supply Commodity bidding ranking.
  • Purchase order data attributes serial number, purchase order queuing number, commodity purchase order number, purchase order index number, commodity purchase order name.
  • the purchase order of the commodity purchase order is queued in the order of high, medium and low membership level. If the buyer membership level is the same, the buyer belongs to the agreement buyer, and the contract purchase order with a large quantity will be queued first. If it is still the same, it will be queued according to the sequence number.
  • Supply order data attributes serial number, supply order queue number, commodity supply order number, supply order index number, commodity supply order name.
  • the order of supply of the goods supply list is queued in the order of high, medium and low membership levels. If the supplier membership level is the same, the supplier will be ranked first according to the agreement supplier and the number of contract supply orders is large. If it is still the same, it will be queued according to the sequence number.
  • Transaction object data attributes serial number, same code for supply and demand, name of same code for supply and demand, product purchase variety, product supply variety, supply and demand difference percentage (GXCE), transaction result (JYJG).
  • Closed-loop transaction data attributes serial number, purchase order queuing, supply order queuing, purchase order quantity (X), completed purchase order quantity (i), purchase product quantity (p), completed purchase product quantity (k), same code
  • Transaction result data attributes commodity transaction data, purchase quantity verification (C), supply quantity verification (G), supply and demand difference ( ⁇ CG), delivery time verification (TYZ), transaction status (CG).
  • Purchase contract data attributes serial number, purchase contract number, purchase contract legal documents (contract payment terms: direct payment, third-party guarantee, account period, account period treaty, quality commitment, liability for breach of contract, etc.), purchaser (party A), goods Transaction operator (Party B), supplier (Party C), commodity code, supply and demand same code, commodity name, commodity supply quantity, supply order index number, traceability code, unit price, commodity shelf life, contract signing time, delivery Time, delivery company, inspection (product quantity verification, product quality verification, supplier qualification verification).
  • the trading platform In the non-empty multi-body domain of commodity transactions, in order to make all the G j (N r ) participates in C i (M k ) bidding for supply. In the locked C i state, the trading platform generates the P family CiGjBkr for M P according to the following rules:
  • the trading platform executes the C i G j B kr pairing of the Mk family. After completing the C i G j B kr pairing, the trading platform performs measurement conversion, transaction, transaction balance adjustment, etc.
  • the capacity of C i G j B kr for setting the product code M k is ⁇ 999, and P products can accommodate the P group C i G j B kr , and a total of P*999 C i G j B kr can be matched.
  • the unit of measurement for the purchase quantity of a M k and the quantity supplied by the suppliers that can participate in the supply may be different before the transaction, but when the transaction is to be carried out, the unit of measurement must be unified.
  • Cycle 2 Access the closed-loop transaction database, commodity transaction database,
  • Operation 1.0 Access the closed-loop transaction database and commodity transaction database, and enter G j (N r ) measurement unit conversion under the state of locking i, k, and ykr of C i (M k );
  • Operation 1.0.2 If M k measurement unit ⁇ C i G j B kr corresponding G j (N r ) measurement unit, use M k measurement unit as the reference value to convert the supply quantity of C i G j B kr , Unit price, and store the conversion result of G j (N r ) corresponding to this C i G j B kr in the commodity transaction database, and go down;
  • Operation 2.0 Access the closed-loop trading library and identify the current trading status
  • Operation 2.3 When i>x in the closed-loop transaction library C i (M K ), the C i (M K ) transaction matching is completed.
  • the trading platform includes queuing, supply and demand product matching, unit conversion, product supply and demand balance adjustment, product description and supply knowledge template matching, product price score, product quality score, supplier qualification score, Signed purchase contract, cargo transportation and acceptance software modules.
  • the trading platform can simultaneously conduct multi-layer, multi-pair and multi-variety intelligent trading. An effective way to solve the problem is to create a meta-knowledge control strategy.
  • the trading platform has the following aspects:

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种用于交易平台实现多元商品智能交易推荐的方法和系统，在采购端获取采购商的商品需求信息，形成采购单，在供货端获取供货商的商品供应信息，形成供货单（S1）；在采购端评估采购商，确定采购商优先等级，在供货端评估供货商，确定供货商优先等级（S2）；在采购端将采购单基于采购商优先等级进行排序，得到采购单队列，在供货端对供货单基于供货商优先等级进行排序，得到供货单队列（S3）；在采购端基于采购单队列的顺序依次对采购单进行交易匹配，在供货端基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配（S4）。所述方法利于促进交易达成；提高交易匹配推荐和筛选效率；确保推荐有效性。

Description

一种用于交易平台实现多元商品智能交易推荐的方法和系统 技术领域
本发明总体涉及商品交易领域，更具体地，涉及一种用于交易平台实现多元商品智能交易推荐的方法和系统。
背景技术
人机在线交互方式选购单一品种是当前网上商品采购的现状。在商品交易过程中，大宗商品交易现象是普遍存在的，基于商品交易对象多层多元的特性，仅从商品本身的特性进行交易匹配无法满足采购商的多种需求。而且，商品交易的达成受到许多因素的限制，市场行为规范性、商誉等特性良好的采购商和供货商应当得到更好的交易机会，也有利于促进交易的达成。
申请号为201910148206.1的专利文件公开了一种交易匹配的方法和系统，交易匹配平台获取供应商端货物供货数据；交易匹配平台获取需求方端货物需求数据；交易匹配平台上传经过分类整理并标准化的数据至大数据平台；大数据平台查询匹配供需方数据，并通过数据运算中心运算后发送配对输出建议至交易匹配平台；交易匹配平台给出供应商端的匹配输出建议；交易匹配平台给出需求方端的配对输出建议。能够完成资源的高效配置，有效提高成本控制或销售渠道拓展的效率。但是，对于交易匹配过程中如何使采购商获得优质供货商提供满意的商品，供货商获得向优质采购商提供可供货的机会，即提高交易成功率方面的问题没有解决。
申请号为201410143963.5的专利文件，公开了一种交易匹配方法以及系统，对用户提交的交易预约进行业务处理并生成交易订单；将所述交易订单作为消息在第一消息中间件中进行传递；基于多台服务器组成的匹配处理集群当中的任意一台服务器读取所述第一消息中间件中以消息形式传递的交易订单，并将其与在售产品进行匹配，生成结果订单；对所述结果订单进行业务处理。通过分布式架构平滑的处理业务量的增长造成的问题，保证交易匹配系统的正常运行，降低交易匹配系统的风险。此专利也没有解决提高交易 成功率的问题以及具体商品匹配的问题。
发明内容
本发明至少需要解决以下问题之一，在交易平台上提高交易成功率；解决商品交易匹配过程中的效率问题；解决提高商品交易匹配过程中的有效性的问题；解决多家采购商向多家供应商能够同时采购多种满意商品的问题。
本发明提供了一种用于交易平台实现多元商品交易推荐的方法，包括，第一步骤S1，在采购端获取采购商的商品需求信息，形成采购单，在供货端获取供货商的商品供应信息，形成供货单；第二步骤S2，在采购端评估采购商，确定采购商优先等级，在供货端评估供货商，确定供货商优先等级；第三步骤S3，在采购端将采购单基于采购商优先等级进行排序，得到采购单队列，在供货端对供货单基于供货商优先等级进行排序，得到供货单队列；第四步骤S4，在采购端基于采购单队列的顺序依次对采购单进行交易匹配，在供货端基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配；所述采购单队列包括：采购单C 1、采购单C 2、…、采购单C x，其中x为非空正整数；所述供货单队列包括：供货单G 1、供货单G 2、…、供货单G y，其中y为非空正整数。
根据本发明的一个实施方式，所述第四步骤S4包括，获取采购单中的商品排列顺序；按照采购单中商品的排列顺序对商品进行交易匹配，直至采购单中所有商品完成交易匹配。
采购单的表达式为：采购单C i，其中i＝1，2，…，X，C i包含多种采购商品M K，其中K＝1，2，…，P；C为X元谓词符号，C 1，C 2，…，C X为项，C i为P元谓词符号，M 1，M 2，…，M P为子项，则类属关系的原子公式为C i(M K)，其中i、k为非空正整数；
供货单的表达式为：供货单G j,其中j＝1，2，…，Y，G j包含多种供货商品N r，其中r＝1，2，…，Q；G为Y元谓词符号，G 1，G 2，…，G Y为项，G j为Q元谓词符号，N 1，N 2，…，N Q为子项，则类属关系的原子公式为G j(N r)，其中j、r为非空正整数。根据本发明的一个实施方式，所述对商品进行交易匹配包括，所述采购单中的商品包括：商品M 1、商品M 2、…、商 品M p，其中p是非空正整数；所述供货单中的商品包括：商品N 1、N 2、…、N q，其中q是非空正整数；第一分步骤，选取商品M k，基于供货单队列的顺序，从各供货单中选取与所述商品M k同种类的商品，得到供需同码商品G j(N r)；第二分步骤，将所述供需同码商品G j(N r)与所述商品M k匹配；选取下一商品，重复所述第一分步骤至第二分步骤，直至完成采购单中所有商品M k的匹配，其中1≤k≤p，1≤j≤y，1≤r≤q。
根据本发明的一个实施方式，对采购单进行交易匹配还包括，根据商品M k的商品需求信息设定交易必要条件，利用所述交易必要条件筛选所述供需同码商品G j(N r)，获得推荐商品序列；所述交易必要条件包括数量、商品质量、交货期限中的至少一个。
根据本发明的一个实施方式，利用所述交易必要条件筛选所述供需同码商品包括，第一子步骤，根据商品M k的商品需求信息设定数量筛选阈值；利用所述数量筛选阈值筛选所述供需同码商品G j(N r)，去除数量小于所述数量筛选阈值的所述供需同码商品G j(N r)，得到第一筛选商品序列；第二子步骤，根据所述商品M k的商品需求信息设定质量筛选阈值；利用所述质量筛选阈值筛选所述第一筛选商品序列，获得第二筛选商品序列；第三子步骤，根据所述商品M k的商品需求信息设定时间筛选阈值；利用所述时间筛选阈值筛选所述第二筛选商品序列，获得推荐商品序列；第四子步骤，将所述推荐商品序列与所述商品M k匹配。
根据本发明的一个实施方式，所述的方法还包括，每完成一种商品M k的匹配，更新采购单和供货单，使所述采购单实时与采购商的采购需求相符，使所述供货单实时与供货商的供货能力相符。
根据本发明的另一个方面，提供了一种用于交易平台实现多元商品交易推荐的系统，包括，第一装置1，用于在采购端获取采购商的商品需求信息，形成采购单，在供货端获取供货商的商品供应信息，形成供货单；第二装置2，用于在采购端评估采购商，确定采购商优先等级，在供货端评估供货商，确定供货商优先等级；第三装置3，用于在采购端将采购单基于采购商优先等级进行排序，得到采购单队列，在供货端对供货单基于供货商优先等级进行排序，得到供货单队列；第四装置4，用于在采购端基于采购单队列的顺 序依次对采购单进行交易匹配，在供货端基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配。
根据本发明的一个实施方式，所述第四装置4包括，第一模块，用于获取采购单中的商品排列顺序；第二模块，用于按照采购单中商品的排列顺序对商品进行交易匹配，直至采购单中所有商品完成交易匹配。
根据本发明的一个实施方式，所述第二模块包括第一子模块，所述第一子模块，用于选取商品M k，基于供货单队列的顺序，从供货单中选取与所述商品M k同种类的商品，得到供需同码商品G j(N r)；将所述供需同码商品G j(N r)与所述商品M k匹配；选取下一商品，重复所述第一分步骤至第二分步骤，直至完成采购单中所有商品M k的匹配，其中1≤k≤p，1≤j≤y，1≤r≤q；其中，所述采购单中的商品包括：商品M 1、商品M 2、…、商品M p，其中p是非空正整数；所述供货单中的商品包括：商品N 1、N 2、…、N q，其中q是非空正整数。
根据本发明的一个实施方式，所述第二模块包括第二子模块，所述第二子模块，用于根据商品Mk的商品需求信息设定交易必要条件，利用所述交易必要条件筛选所述供需同码商品Gj(Nr)，获得推荐商品序列；所述交易必要条件包括数量、商品质量、交货期限中的至少一个。
根据本发明的一个实施方式，一种用于交易平台的商品交易推荐的方法，包括，在采购端，获取采购商的商品需求信息，形成采购单；评估采购商，确定采购商优先等级；将采购单基于采购商优先等级进行排序，得到采购单队列；基于采购单队列的顺序依次对采购单进行交易匹配，直至完成所有采购单的交易匹配；所述采购单队列包括：采购单C 1、采购单C 2、…、采购单C x，其中x为非空正整数；
根据本发明的一个实施方式，一种用于交易平台的商品交易推荐的方法，包括，在供货端，获取供货商的商品供应信息，形成供货单；评估供货商，确定供货商优先等级；对供货单基于供货商优先等级进行排序，得到供货单队列；基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配；所述供货单队列包括：供货单G 1、供货单G 2、…、供货单G y，其中y为非空正整数。
本发明通过在交易前对供货商和采购商进行评定，确定采购商优先等级和供货商优先等级，按照优先等级将采购单和供货单进行排序，筛选优质供货商和采购商，有助于使采购商获得优质的货物，以及优质供货商的商品被采购，从而促进交易的达成。供需同码的匹配，使得从众多的供货单中能够较为迅速的将同类商品筛选出来，提高交易匹配推荐的效率。本发明按照外界因素依赖情况将筛选条件进行排序，从而整体提高筛选效率。实时确保数据的准确，避免出现多重采购和多重供货的情况，确保交易推荐的有效性。
附图说明
图1是一种用于交易平台实现多元商品智能交易推荐的方法的步骤示意图；
图2是对采购单序列进行匹配的示意图；
图3是对采购单中的商品进行匹配的示意图；
图4是对商品进行交易匹配的示意图；
图5是采用预设采购条件筛选商品的步骤示意图；以及
图6是一种用于交易平台实现多元商品智能交易推荐的系统的示意图。
具体实施方式
下面结合附图对本发明的较佳实施例进行详细阐述，参考标号是指本发明中的组件、技术，以便本发明的优点和特征在适合的环境下实现能更易于被理解。下面的描述是对本发明权利要求的具体化，并且与权利要求相关的其它没有明确说明的具体实现也属于权利要求的范围。
图1示出了一种用于交易平台实现多元商品智能交易推荐的方法的步骤示意图。
如图1所示，一种用于交易平台实现多元商品智能交易推荐的方法，包括，第一步骤S1，在采购端获取采购商的商品需求信息，形成采购单，在供货端获取供货商的商品供应信息，形成供货单；第二步骤S2，在采购端评估采购商，确定采购商优先等级，在供货端评估供货商，确定供货商优先等级；第三步骤S3，在采购端将采购单基于采购商优先等级进行排序，得到采购单 队列，在供货端对供货单基于供货商优先等级进行排序，得到供货单队列；第四步骤S4，在采购端基于采购单队列的顺序依次对采购单进行交易匹配，在供货端基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配；所述采购单队列包括：采购单C 1、采购单C 2、…、采购单C x，其中x为非空正整数；所述供货单队列包括：供货单G 1、供货单G 2、…、供货单G y，其中y为非空正整数。
所述对采购单进行交易匹配是指按照采购单排序，依次对每个采购单内的商品进行交易匹配。
所述非空正整数是指采购单可以是一个或多个；供货单也可以是一个或多个。
在采购活动中，存在多家采购商参与的情况，即存在多个采购单C i，其中i＝1，2，…，X，每个采购单都可以包含多个要采购的商品M K，其中K＝1，2，…，P。
多个采购商参与采购时，由于采购商的资质、诚信度等对交易会产生比较大的影响，本发明采取对采购商进行评估，筛选优质的采购商优先进行交易有利于促进交易的完成以及促进供应商保持良好的市场行为起到促进作用。所述采购单队列包括：对采购商进行评估，确定C i(M K)的优先采购等级，以此作为排队依据，形成采购单队列，商品采购的先后顺序，不降低C i(M K)的采购诉求，采购商优先对C i(M K)采购可以获得满意的供应商优先供货的机会。
例如，采购商从系统中采购商品，在系统中注册，详细填写注册信息。采购商通过注册认证之后，正式成为系统商品采购商，可以通过历史交易记录等信息，通过大数据分析，对采购商进行评定。针对评定等级较高的采购商，可以优先交易。
在对采购单进行排序时，如果某采购商同时有多个采购单，则以采购商递交采购单的时间顺序先进行排序，然后再与其他采购商进行等级排序。
图2示出了对采购单序列进行匹配的示意图。
如图2所示，所述采购单队列包括：采购单C 1、采购单C 2、…、采购单C x，其中x为非空正整数，在进行交易时，首先进行采购单C 1的交易，然后 进行采购单C 2的交易，依次进行，直至系统内所有采购单结束交易。
同样的，存在多个供货商进行供货时，即存在多个供货单G j,其中j＝1，2，…，Y，每个供货单都可以包含多个可供货的商品N r，其中r＝1，2，…，Q。所述供货单队列包括：对供应商进行评估，确定G j(N r)的优先供货商等级，以此作为排队依据，形成供货单队列，商品供货的先后顺序，不降低G j(N r)供货综合评分标准，供货商优先供货G j(N r)可以获得采购商优先采购的机会。
有资格参与系统商品交易活动的供货商在系统注册，详细填写、上传注册信息。供货商通过注册认证之后，正式成为系统商品供货商。可以通过历史交易记录等信息，通过大数据分析，对供货商进行评定，筛选信用、资质、交易记录较好的供货商。针对评定等级较高的供货商，可以优先交易。
C i(M K)、G j(N r)的排队是按顺序智能求解多元对多元量化商品交易模式的必要条件之一，按照系统实际应用的需求，商品交易排队能够让有业绩的采购商优先采购有业绩的供应商的优质商品。
所述供货单队列包括：供货单G 1、供货单G 2、…、供货单G y，其中y为非空正整数，首先对供货单G 1中的商品与采购单的商品进行交易匹配和推荐，然后对供货单G 2中的商品进行交易匹配和推荐，依次进行，直至满足采购单中某项商品完成交易匹配，或者所有供货单完成交易匹配。
本发明通过在交易前对供货商和采购商进行评定，确定采购商优先等级和供货商优先等级，并在实际交易时，按照优先等级将采购单和供货单进行排序，筛选优质供货商和采购商，有助于使采购商获得优质的货物，以及优质供货商的商品被采购，从而促进交易的达成。
所述商品需求信息一般包括：商品价格、商品质量、商品供应商资质、商品数量、交货时间等，可以根据采购商的需求进行设定。
所述商品供应信息包括，商品价格、商品质量、商品供应商资质、供货数量、交货时间等。
图3示出了对采购单中的商品进行匹配的示意图。
如图3所示，所述第四步骤S4包括，获取采购单中的商品排列顺序；按照采购单中商品的排列顺序对商品进行交易匹配，直至采购单中所有商品完 成交易匹配。
对于单个采购单，例如任取采购单，根据采购单中的商品顺序，逐个进行交易匹配，直至采购单中所有商品完成交易匹配后，则此采购单的交易匹配完成，选取下一采购单，按照上述方法再逐个进行交易匹配。
即，在采购单依次进行交易匹配的大循环内，还包括采购单内部的商品依次进行交易匹配循环。
图4示出了对商品进行交易匹配的示意图。
如图4所示，所述对商品进行交易匹配包括，所述采购单中的商品包括：商品M 1、商品M 2、…、商品M p，其中p是非空正整数；所述供货单中的商品包括：商品N 1、N 2、…、N q，其中q是非空正整数；第一分步骤，选取商品M k，基于供货单队列的顺序，从供货单中选取与所述商品M k同种类的商品，得到供需同码商品G j(N r)；第二分步骤，将所述供需同码商品G j(N r)与所述商品M k匹配；选取下一商品，重复所述第一分步骤至第二分步骤，直至完成采购单中所有商品M k的匹配，其中1≤k≤p，1≤j≤y，1≤r≤q。
在本发明中，可以将采购单中的商品和供货单中的商品按照其种类进行编码，如采购单C i中的各种商品：商品M 1、商品M 2、…、商品M p，具体到某个商品，可以用C i(M k)来表示，表示第i个采购单中的第k个商品。同理，供货单G y中的商品可以表示为G y(N r)，表示第y个采购单中的第r个商品。
本发明在对商品C i(M k)进行匹配时，首先要进行同种类的匹配，当C i(M k)与G y(N r)种类相同时，即M＝N时，则将二者进行匹配，即得到供需同码配对。在进行这种匹配时，本发明基于供货单队列的顺序，即按照供货商的等级进行先后的匹配，使得优质供货商的商品优先被匹配。
所述供需同码商品是指对于当前采购单上的当前商品C i(M k)，各供货单上与之相同种类的商品G y(N r)。
供需同码商品C iG jB kr配对算法中，使供需同码配对，在锁定当前采购单C i状态下，设M k＝N，为M k商品匹配一族C iG jB kr，按照供货单队列的顺序，C iG jB kr对应供货单的商品G j(N r)，其中r∈kr，kr为C iG jB kr的个数，由M k配对C iG jB kr生成的G j(N r)不会漏掉在系统中注册的具有供货能力的每 一位供应商。
为M k商品匹配一族C iG jB kr，的过程如下：在锁定的C i状态下，M、N合取原子公式，
Figure PCTCN2019120571-appb-000001
Figure PCTCN2019120571-appb-000002
在M和N合取原子公式中，如果
Figure PCTCN2019120571-appb-000003
M k＝N，则产生一族C iG jB kr，那么，M＝N共计有P族C iG jB kr。
M k的C iG jB kr设计容量≤999，P种商品可以容纳P族C iG jB kr，可容纳配对P*999个C iG jB kr。
利用上述公式的算法，打通了采购品种M k与C iG jB kr相对应的G j(N r)之间展开综合库供、需模板匹配的通道。
本发明提出商品交易元知识控制策略，使得系统能够按照排队顺序独立完成多层多元对多元商品智能匹配；系统获取C i(M k)、G j(N r)的排列顺序，自动计算采购单C i数，其中i＝1、2、…、X是商品采购单的数量，将X作为大循环指针，控制执行X个采购单的排队采购；在锁定C i的状态下，系统自动计算K数，其中K＝1、2、…、P是第i个采购单的采购品种数，将P作为中循环指针，控制执行P种商品采购；在锁定M k状态下，系统自动计算kr数，其中kr∈999表示可容纳G j(N r)参与供货，将kr作为小循环商品交易指针，控制第M k种商品执行第一分步骤配对、第二分步骤验证、第三分步骤匹配交易，保证kr个C iG jB kr所对应验证合格的G j(N r)都有机会为C i(M k)供货。
供需同码的匹配，使得从众多的供货单中能够较为迅速的将同类商品筛 选出来，提高交易匹配推荐的效率。
根据本发明的一个实施方式，对采购单进行交易匹配还包括：根据商品M k的商品需求信息设定交易必要条件，利用所述交易必要条件筛选所述供需同码商品G j(N r)，获得推荐商品序列；所述交易必要条件包括数量、商品质量、交货期限中的至少一个。
图5示出了采用预设采购条件筛选商品的步骤示意图。
如图5所示，在本发明中，基于供需同码匹配的基础上，继续采用其他预设定的采购条件进行筛选供货商品，这种采用单个条件进行多次筛选的方式，使得筛选条件逐步叠加，后续筛选基于前序筛选的基础上进行，减少后续筛选的筛选范围，使后续筛选的计算量明显下降，提高交易匹配推荐的效率。
每一个C iG jB kr所对应的M k与G j(N r)配对都必须经商品交货期限、采购与供货数量、质量门槛、资质门槛等参数的判据，在一致性标化计量单位后，M k与一族C iG jB kr有序对应G j(N r)验证为有效的配对，只允许有效的G j(N r)准入匹配操作流程。
本发明建立数据、知识、模型、图形等综合库，架构C i(M K)、G j(N r)的交易模板，用交易模板表达C i(M K)对G j(N r)的诉求，进行商品交易智能匹配；系统智动为G j(N r)的价格GJFr、质量GZFr、资质GZZFr最小项打分、积分，并对每一个M k所对应的G j(N r)综合得分GHZ r进行汇总、供货商品GZHP r的得分由高至低进行排序。
根据本发明的一个实施方式，还包括第一子步骤，根据商品M k的商品需求信息设定数量筛选阈值；利用所述数量筛选阈值筛选所述供需同码商品G j(N r)，去除数量小于所述数量筛选阈值的所述供需同码商品G j(N r)，得到第一筛选商品序列；第二子步骤，根据所述商品M k的商品需求信息设定质量筛选阈值；利用所述质量筛选阈值筛选所述第一筛选商品序列，获得第二筛选商品序列；第三子步骤，根据所述商品M k的商品需求信息设定时间筛选阈值；利用所述时间筛选阈值筛选所述第二筛选商品序列，获得推荐商品序列；第四子步骤，将所述推荐商品序列与所述商品M k匹配。
所述数量筛选阈值是指根据某种商品所需要采购的数量预先设定一个供 货量的最低值，例如，预先设置所述数量筛选阈值为采购量的30％，所有供货数量低于采购量的30％的供需同码货物将被筛除，即，针对大宗采购的采购单，尽可能的将商品交易集中到较少的供货商范围内，降低交易的复杂程度。
所述质量筛选阈值是指采购商对商品来源、重量、大小、口味等商品本身的属性进行的各种限定，可以根据采购商的需求进行调整也可以根据行业标准或者行政标准等进行限定。
所述时间筛选阈值是指交货时间的限定，是根据采购商的时间限定需求进行的设定。
在本发明中，在供需同码匹配的基础上，先用数量要求进行筛选、再用质量进行筛选、最后用交货时间进行筛选，顺序不可改变。由于供需同码属于对采购商需求依赖最小的筛选条件；数量要求相对于供需同码来说依赖于采购商的需求；进一步的质量筛选除了满足采购商的需求，还要符合国家法规和行业标准等附加条件；进一步的，供货时间的需求随机性较前述各筛选条件更强。
本发明按照外界因素依赖情况将筛选条件进行排序，优先因素依赖少的筛选条件进行筛选，在进行因素依赖少的筛选条件进行筛选时，筛选目标较多，筛选条件简单，在后续进行因素依赖多的筛选条件进行筛选时，虽然筛选条件复杂多变，但筛选目标较少，从而使得各筛选过程的筛选效率前后一致，整体提高筛选效率。
根据本发明的一个实施方式，所述的方法还包括，每完成一次商品M k的匹配，更新采购单和供货单，使所述采购单实时与采购商的采购需求相符，使所述供货单实时与供货商的供货能力相符。
所述更新采购单是指对于已经完成匹配的商品，从采购单中去除，保留未完成匹配的商品。同样的，更新供货单即去除已经完成匹配的供货商品，利用剩余的供货商品继续对采购单中的商品进行匹配。
例如，统计C i(M k)采购单数量i，采购单品种数量p，已完成采购品种k，含同码商品供货单G j的数量kr，已完成含同码商品供货单G j数量ykr＝0，某商品已采购数量C YL，某商品已供货数量G YL。
如果ykr＜kr，则继续供货单中供需同码匹配完毕，直至ykr＝kr，则全部供货单中供需同码匹配完毕，则进行采用预设定的采购条件进行筛选供货商品。
当某商品M k的C YL>G YL时，继续进行供需同码匹配和筛选，直至ykr＝kr，则针对此商品的交易结束。或者当C YL＝G YL时，针对此商品的交易结束。这两种情况对应着商品M k的需求量大于供货量，只完成了部分采购，以及商品M k供货量大于等于采购量，满足完成了这个商品的采购。无论是哪一种情况，都表示针对此商品M k的交易匹配推荐结束。
针对此商品M k的交易匹配推荐结束时，商品M k的采购单中将去除此商品，相应的供货单中将去除已经与此商品匹配完成的商品。
同理，对于采购单中的品种采购单数量i，采购单品种数量p，也进行实时更新。
此时，使采购单的需求和供货单的供应都与真实情况相一致，实时确保数据的准确，避免出现多重采购和多重供货的情况，确保交易推荐的有效性。
图6示出了一种用于交易平台实现多元商品智能交易推荐的系统的示意图。
如图6所示，一种用于交易平台实现多元商品智能交易推荐的系统，包括，第一装置1，用于在采购端获取采购商的商品需求信息，形成采购单，在供货端获取供货商的商品供应信息，形成供货单；第二装置2，用于在采购端评估采购商，确定采购商优先等级，在供货端评估供货商，确定供货商优先等级；第三装置3，用于在采购端将采购单基于采购商优先等级进行排序，得到采购单队列，在供货端对供货单基于供货商优先等级进行排序，得到供货单队列；第四装置4，用于在采购端基于采购单队列的顺序依次对采购单进行交易匹配，在供货端基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配。
根据本发明的一个实施方式，所述第四装置4包括，第一模块，用于获取采购单中的商品排列顺序；第二模块，用于按照采购单中商品的排列顺序对商品进行交易匹配，直至采购单中所有商品完成交易匹配。
根据本发明的一个实施方式，所述第二模块包括第一子模块，所述第一 子模块，用于选取商品M k，基于供货单队列的顺序，从供货单中选取与所述商品M k同种类的商品，得到供需同码商品G j(N r)；将所述供需同码商品G j(N r)与所述商品M k匹配；选取下一商品，重复所述第一分步骤至第二分步骤，直至完成采购单中所有商品M k的匹配，其中1≤k≤p，1≤j≤y，1≤r≤q；其中，所述采购单中的商品包括：商品M 1、商品M 2、…、商品M p，其中p是非空正整数；所述供货单中的商品包括：商品N 1、N 2、…、N q，其中q是非空正整数。
根据本发明的一个实施方式，所述第二模块包括第二子模块，所述第二子模块，用于根据商品M k的商品需求信息设定交易必要条件，利用所述交易必要条件筛选所述供需同码商品G j(N r)，获得推荐商品序列；所述交易必要条件包括数量、商品质量、交货期限中的至少一个。
根据本发明的一个实施方式，一种用于交易平台的商品交易推荐的方法，包括，在采购端，获取采购商的商品需求信息，形成采购单；评估采购商，确定采购商优先等级；将采购单基于采购商优先等级进行排序，得到采购单队列；基于采购单队列的顺序依次对采购单进行交易匹配，直至完成所有采购单的交易匹配；所述采购单队列包括：采购单C 1、采购单C 2、…、采购单C x，其中x为非空正整数；
根据本发明的一个实施方式，一种用于交易平台的商品交易推荐的方法，包括，在供货端，获取供货商的商品供应信息，形成供货单；评估供货商，确定供货商优先等级；对供货单基于供货商优先等级进行排序，得到供货单队列；基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配；所述供货单队列包括：供货单G 1、供货单G 2、…、供货单G y，其中y为非空正整数。
本发明通过在交易前对供货商和采购商进行评定，确定采购商优先等级和供货商优先等级，按照优先等级将采购单和供货单进行排序，筛选优质供货商和采购商，有助于使采购商获得优质的货物，以及优质供货商的商品被采购，从而促进交易的达成。供需同码的匹配，使得从众多的供货单中能够较为迅速的将同类商品筛选出来，提高交易匹配推荐的效率。本发明按照外界因素依赖情况将筛选条件进行排序，从而整体提高筛选效率。实时确保数 据的准确，避免出现多重采购和多重供货的情况，确保交易推荐的有效性。
实施例.
在交易平台的采购端：
交易平台采购端形成采购动态信息CGDT。即，存在多家采购商列出的多个采购单C i，其中i＝1、2、…、X，每个采购单都可以包含多个要采购的商品M K，其中K＝1、2、…、P。采购商从交易平台中采购CGDT(C i(M K))商品，在交易平台中注册，详细填写注册信息。采购商通过注册认证之后，正式成为系统商品采购商，经大数据信息分析，评定等级，并成为交易平台的VIP采购商。VIP采购商按年度缴纳会费，会员级别分1级、2级等。采购商采购CGDT(C i(M K))商品，采用商品价格、商品质量、商品供应商资质的权重分值表述商品综合量化述求，作为对GYDT(G j(N r))量化评分的依据。
在交易平台的供应端：
与CGDT相呼应的供应动态信息GYDT，存在多家供应商列出的多个供货单G j，其中j＝1、2、…、Y，每个供货单都可以包含多个可供货的商品Nr，其中r＝1、2、…、Q。GYDT(G j(N r))有资格参与系统商品交易活动，GYDT(G j(N r))的供应商在交易平台中注册，详细填写、上传注册信息。供应商通过注册认证之后，正式成为平台的商品供应商，经大数据信息分析，评定等级。按照采购商对CGDT(C i(M K))质量保障体系的述求，系统为GYDT(G j(N r))供应商设置一套完整、规范、精准的综合信息知识表达模板，包括商品标价、质量保障、供应商资质三大体系，与CGDT(C i(M K))采购述求相呼应。
为了实现系统采购单的循环C X、采购单内商品匹配循环M P、单个商品匹配循环kr的运行，表1示出了元知识控制策略参数配置及物理意义说明，用于引导系统CGDT(C i(M K))与CGDT(C i(M K))之间全面展开多目标优选商品的交易。
表1.
序号 属性名 缩写符号 数据来源
1 采购动态信息 CGTD 系统需求数据库
2 供应动态信息 GYDT 系统供应数据库
3 商品采购单索引编号 C i 商品交易数据库
4 商品供货单索引编号 G j 商品交易数据库
5 供需同码 C iG jB kr 交易对象数据库
6 采购数量 C L 采购品种数据库
7 已采购数量 C YL 采购品种数据库
8 价格 G D 供货品种数据库
9 采购时间 T C 采购品种数据库
10 采购交货时间 T CJ 采购品种数据库
11 供货数量 G L 供货品种数据库
12 已供货数量 G YL 供货品种数据库
13 供货时间 T G 供货品种数据库
14 供应商交货时间 T GJ 供货品种数据库
15 交货时间验证 T YZ 交易结果库
16 采购数量验证 C 交易结果库
17 供货数量验证 G 交易结果库
18 供需量差 △CG 交易结果库
19 交易状态 CG 交易结果库
20 交易结果 JYJG 商品交易对象数据库
21 同码品种数量 kr 闭环交易库
22 已完成同码品种数量 ykr 闭环交易库
23 采购单数量 X 闭环交易库
24 已采购单数量 i 闭环交易库
25 采购单品种数 P 闭环交易库
26 采购单已采购品种数量 K 闭环交易库
27 暂停交易 ZT 闭环交易库
28 供需差额百分比 GXCE 商品交易对象数据库
交易平台的数据准备：
交易平台数据：商品采购动态信息，商品供应动态信息、商品采购品种、商品采购综合分值设置、商品供货品种、商品交易数据、闭环交易数据、交易结果、采购合同。其中，商品采购动态信息数据属性：序号，商品采购单编号，商品采购单名称，商品采购品种。交易平台生成的“商品采购动态信息”库是由采购商编辑的商品采购信息；采购商也可以将勾选、编辑的“商品采购”库的信息推送到“商品采购动态信息”数据库中。
商品采购品种数据属性：序号，商品名称，商品编号，采购数量，已采 购数量，商品采购时间，商品交货时间，采购商会员级别，商品采购综合分值设置。
已采购数量：采购商每进行一次采购，交易成功，采购数量核减相应的采购数量。
商品采购综合分值设置数据属性：商品综合评价总分，商品价格权重，商品价格得分算法，商品质量权重，商品质量分值分解，商品质量得分算法，供应商资质权重，供应商分值分解，供应商资质得分算法。
商品质量分值分解：采购商可以接受交易平台设值的默认值，也可以根据商品采购述求，按照质量分值权重灵活设置商品质量各子项分值。
供应商资质分值分解：采购商可以接受设置的默认值，也可以根据商品采购述求，按照供应商资质分值权重灵活设置供应商分值各子项分值。
商品采购动态信息数据属性：序号，商品供货单编号，商品供货单名称，商品供货品种。交易平台生成的“商品供应动态信息”库是由供应商编辑的商品供应信息；供应商也可以将勾选、编辑的“商品供应”数据库的信息推送到“商品供应动态信息”数据库中。
商品品种数据属性：序号，商品编号，商品名称，供货数量，供货标化数量，已供货数量，价格，标化价格，商品质量，资质证明，资质证明图片，商品供货时间，商品交货时间，供应商会员级别。
已供货数量：供应商每参与一次竞标，交易成功，供货数量核减相应的采购数量。
供货标化数量及价格：以商品采购品种数据库计量单位为基准，对商品供货品种数据库中的“供货数量”、“价格”参数进行标化数据处理。
商品交易数据属性：序号，采购单排队，供货单排队，商品交易对象，供货计量单位换算，供货价格评分，供货质量评分，供货资质评分，供货商品量化得分汇总，供货商品竞标排名。
采购单数据属性：序号，采购单排队号，商品采购单编号，采购单索引编号，商品采购单名称。
商品采购单的采购顺序按照会员级别高、中、低的顺序进行排队。如果采购商会员级别相同，则按采购商属于协议采购商，且合同采购单数量多的 优先排队，如果仍相同，则按顺序号排队。
供货单数据属性：序号，供货单排队号，商品供货单编号，供货单索引编号，商品供货单名称。
商品供货单的供货顺序按照会员级别高、中、低的顺序进行排队。如果供货商会员级别相同，则按供货商属于协议供货商，且合同供货单数量多的优先排队，如果仍相同，则按顺序号排队。
交易对象数据属性：序号，供需同码，供需同码名称，商品采购品种，商品供货品种，供需差额百分比(GXCE)，交易结果(JYJG)。
闭环交易数据属性：序号，采购单排队，供货单排队，采购单数量(X)，已完成采购单数量(i)，采购品种数量(p)，已完成采购品种数量(k)，同码品种数量(kr)，已完成同码品种数量(ykr)，暂停交易(ZT)。
交易结果数据属性：商品交易数据，采购数量验证(C)，供货数量验证(G)，供需量差(△CG)，交货时间验证(TYZ)，交易状态(CG)。
采购合同数据属性：序号，采购合同号，采购合同法律文书(合同付款条件：直接付款、第三方担保、账期、账期条约，质量承诺、违约责任等)，采购商(甲方)，商品交易运营商(乙方)，供货商(丙方)，商品编码，供需同码，商品名称，商品供货数量，供货单索引号，追溯码，单价，商品保质期，合同签订时间，交货时间，送货公司，验货(商品数量验证，商品质量验证，供货资质验证)。
第一、对采购单和供货单进行排队：
第三装置对C i排队，其中i＝1、2、…、X，C i(M k)的采购顺序按照采购商会员级别高、中、低的顺序进行排队。如果采购商会员级别相同，则按采购商属于协议采购商，且合同采购单数量多的优先排队，如果仍相同，则按顺序号排队。C i(M k)交易先后顺序，不影响C i(M k)的综合采购要求。
第三装置对G j排队，其中j＝1、2、…、Q，G j(N r)的供货顺序按照会员级别高、中、低的顺序进行排队。如果供应商会员级别相同，则按供应商属于协议供货商，且合同供货单数量多的优先排队，如果仍相同，则按顺序号排队。G j(N r)交易先后顺序，不影响G j(N r)的综合得分。
第二、供需同码配对：
在商品交易的非空多体域中，为了让所有满足供需同码条件的
Figure PCTCN2019120571-appb-000004
G j(N r)参与C i(M k)竞标供货，在锁定的C i状态下，交易平台为M P生成P族CiGjBkr的规则如下：
如果
Figure PCTCN2019120571-appb-000005
C i(M k)→(G 1(N r)、G 2(N r)、···、G y(N r))，
则
Figure PCTCN2019120571-appb-000006
C i(M k)＝G j(N r)，
其中K＝1，2，···，P，r＝1，2，···，Q，
i＝1，2，···，X，j＝1，2，···，Y；
那么，P族C iG jB kr配对的谓词类属关系展开式，
C i(M 1，M 2，···，M P)→((G 1(N r)，G 2(N r)，···，G y(N r))
表示C i(M k)中的商品编码蕴含G j(N r)中的商品编码，两者之间可以配成C iG jB kr供需同码，生成C iG jB kr集合码，即以C i(M k)作为源点，生成满足状态矢量空间M＝N条件的P族C iG jB kr。举例说明：当i≤X，k≤P，交易平台执行Mk的一族C iG jB kr配对，每完成一族C iG jB kr配对之后，交易平台执行计量换算、交易、交易平衡调节等流程，完成C i(M k)中的其中一个M k的采购，进入下一个C i(M k+1)的配对；当i≤X，当k＞P时，交易平台完成了一个C i(M k)的采购，进入下一个C i+1(M k)的P族C iG jB kr配对；当i＞X，交易平台完成了CGDT与GYDT的同码配对。
在锁定的Ci状态下，如果
Figure PCTCN2019120571-appb-000007
Mk＝N，则产生一族C iG jB kr，那么M＝N共计有P族C iG jB kr。
设置商品编码M k的C iG jB kr容量≤999，P种商品可以容纳P族C iG jB kr，共计可配对P*999个C iG jB kr。
一个M k的采购数量与可参与供货的供货商所供货物数量的计量单位在交易之前可能不同，但要进行交易时，计量单位必须统一。
第三，商品交易元知识驱动下的闭环交易平衡调节过程：
初始化：访问闭环交易库、商品交易数据库，统计C i(M k)采购单数量(X＝？)，已完成采购单数量(i)＝0，采购单品种数量(p)＝0，已完成采购品种(k)＝0，同码品种C iG jB kr数量(kr)＝0，已完成同码品种C iG jB kr数量(ykr)＝0，已采购数量(C YL)＝0，已供货数量(G YL)＝0；
循环1：访问闭环交易库、商品交易数据库，(1)i+1；(2)如果i＞X，C x(M P)采购结束；(3)如果i≤x，则统计当前C i(M k)采购品种数量P＝？，k＝0，开始C i(M k)采购；
循环2：访问闭环交易库、商品交易数据库，
(1)在锁定C i(M k)的i状态下，K+1；
(2)如果K＞P，转循环1；
(3)如果K≤P，在锁定的C i(M k)的i、k状态下，访问闭环交易库、商品交易数据库，调用《供需同码配对》软件模块，为M k＝G j(N r)产生一族供需同码C iG jB kr，并将该一族C iG jB kr存入商品交易数据库中；
(4)统计C iG jB kr数量kr＝？，kr＝1，2，…,∈999，初始化ykr＝0，JYJG＝00对当前一族C iG jB kr有效性进行验证；
循环3：
(1)在锁定C i(M k)的i、k状态下，ykr+1→ykr；
(2)如果ykr＞kr，则操作1.2；
(3)如果ykr≤kr，对当前ykr状态下的C iG jB kr有效性进行验证；去往规则1.0
规则1.0：访问商品交易数据库，如果C L＞0、C YL≥0，且C L-C YL＞0，则C＝1，C i(M K)对应C iG jB kr品种有采购需求，转规则1.2；
规则1.0.1：访问商品交易数据库，如果C L-C YL≤0，则C＝0，当前C i(M K)对C iG jB kr没有采购需求或C iG jB kr交易结束，从商品交易数据库中清除C iG jB kr，设定CG＝00，转操作2.0；
规则1.1：访问商品交易数据库，如果G L＞0、G YL≥0，且G L-G YL＞0，则G＝1，当前G j(N r)对应C iG jB kr品种能够供货，转规则1.2；
规则1.1.1：如果G L-G YL≤0，则G＝0，G j(N r)对应C iG jB kr不具备交易条件或C iG jB kr交易结束，设定CG＝00，转操作2.0；
规则1.2：访问商品交易数据库，如果规则1.0与规则1.1成立，且T C≥T G，T CJ≥T GJ，则T YZ＝1，转操作1.0；
规则1.2.1：访问商品交易数据库，如果规则1.0与规则1.1成立，且T C＜T G或T CJ＜T GJ，则T YZ＝0，转操作2.0；
操作1.0：访问闭环交易库、商品交易数据库，在锁定C i(M k)的i、k、ykr状态下，进入G j(N r)计量单位换算；
操作1.0.1：如果M k计量单位＝C iG jB kr相对应的G j(N r)计量单位，则不用换算，转操作1.1；
操作1.0.2：如果M k计量单位≠C iG jB kr相对应的G j(N r)计量单位，则以M k计量单位为参考值，换算C iG jB kr的供货数量、单位价格，并将这一个C iG jB kr相对应的G j(N r)换算结果存入商品交易数据库中，往下；
操作1.1：访问闭环交易库、商品交易数据库，判断C iG jB kr的有效性，△CG＝(G L-G YL)-(C L-C YL)，往下；
操作1.1.1如果△CG≥0，T YZ＝1，则设定CG＝11，C iG jB kr所对应的G j(N r)验证为有效的，转操作1.1.3；
操作1.1.2如果△CG＜0，T YZ＝1，满足(C L-C YL)-(G L-G YL)≤(C L-C YL)×GXCE％条件，其中，GXCE在30％～70％范围内可设置，这一个C iG jB kr所对应的G j(N r)验证为有效，则设定CG＝10；不满足(C L-C YL)-(G L-G YL)≤(C L-C YL)×GXCE％条件，设定CG＝00，往下；
操作1.1.3如果ykr≤kr，则转循环3；如果ykr＞kr，M K的一族C iG jB kr验证完毕，对交易结果作判据，往下；
操作1.2：访问闭环交易库、交易结果数据库、采购合同数据库，将数量为kr的C iG jB kr，取满足CG＝11或CG＝10状态的G j(N r)，给M k一族有效C iG jB kr所对应G j(N r)的价格、质量、资质作综合评分，如果交易成功，取综合量化得分最高的G j(N r)的供货商，或由人工干预，从综合量化得分前三名中选择G j(N r)的供货商，设定JYJG＝11；
操作1.2.1当JYJG＝11、CG＝11，M k采购数量仅为一家供应商供货，更新商品交易数据库中的动态数据：已采购数量为C YL+1＝C L-C YL，已供货数量为G YL+1＝(C L-C YL)+G YL，转操作2.0；
操作1.2.2如果在JYJG＝11、CG＝10状态下，M k所需采购数量为多家供应商供货，按得分由高到低排序选择供应商供货，直至M k采购完毕；按供应商供货顺序更新商品交易数据库中的动态数据：首选供应商，已采购数量为C YL+1＝G YL+1+C YL，已供货数量为G YL+1＝G L-G YL；后选供应商，如果CG＝11， 已采购数量为C YL+1＝C L-C YL，已供货数量为G YL+1＝(C L-C YL)+G YL，如果CG＝10，已采购数量为C YL+1＝G YL+1+C YL，已供货数量为G YL+1＝G L-G YL,转操作2.0；
操作1.2.3当JYJG＝11、CG＝11或CG＝10，如果交易失败(例如：供应商对采购商要求的特殊指标不能满足等因素)，JYJG＝00，转操作2.0；
操作2.0：访问闭环交易库，识别当前交易状态；
操作2.1：K≤P、i∈X，进入C i(M K+1)交易，转循环2；
操作2.2：K＞P、i∈X，每当完成一个C i(M K)交易，交易平台便读取闭环交易库ZT标志位，如果ZT＝10，暂停交易，经人工干预ZT＝11后，交易平台在此断点继续运行，进入C i+1(M K)交易，转循环1；
操作2.3：当闭环交易库C i(M K)中的i＞x，则C i(M K)交易匹配完毕。
根据本发明的一种实施例，交易平台由排队、供需商品配对、计量单位换算、商品供需平衡调节、商品述求与供应知识模板匹配、商品价格评分、商品质量评分、供货商资质评分、签订采购合同、货物运输及验收软件模块组成。交易平台能够同时进行多层的多元对多元品种智能化交易，一种有效求解问题的途径就是创建元知识控制策略，交易平台具有如下几个方面的内容：
(1)按照预先制定的优先交易规则，为采购单、供应单交易顺序进行排队，交易平台以排队的顺序进行商品交易；
(2)以采购单中的商品编号为原点，从供货单中逐个为能与采购单存在相同商品编号的品种进行配对，生成“一对多”的供需相同商品编码，即供需同码；
(3)对供需同码的品种是否能够满足交易的必要条件进行验证，经验证符合交易必要条件的供需同码品种才能准入商品交易进程，要求交易平台能够动态调节商品交易供需平衡关系；
(4)对于不具备交易条件的供需同码品种需剔除，为商品采购单中所采购的每一个品种配一族供需同码，验证每一个供需同码品种的有效性，如此循环，依次验证参与交易的全部供需同码品种的有效性；
(5)在商品闭环交易过程中，必须对交易成功和失败的参数进行验证、 更新等。
应该注意的是，上述实施例对本发明进行说明而不是对本发明进行限制，并且本领域技术人员在不脱离所附权利要求的范围的情况下可设计出替换实施例。在权利要求中，不应将位于括号之间的任何参考符号构造成对权利要求的限制。

Claims (10)

  1. 一种用于交易平台实现多元商品智能交易推荐的方法，包括，
    第一步骤(S1)，在采购端获取采购商的商品需求信息，形成采购单，在供货端获取供货商的商品供应信息，形成供货单；
    第二步骤(S2)，在采购端评估采购商，确定采购商优先等级，在供货端评估供货商，确定供货商优先等级；
    第三步骤(S3)，在采购端将采购单基于采购商优先等级进行排序，得到采购单队列，在供货端对供货单基于供货商优先等级进行排序，得到供货单队列；
    第四步骤(S4)，在采购端基于采购单队列的顺序依次对采购单进行交易匹配，在供货端基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配。
  2. 根据权利要求1所述的方法，其中，所述采购单队列包括：采购单C 1、采购单C 2、…、采购单C x，其中x为非空正整数；
    所述供货单队列包括：供货单G 1、供货单G 2、…、供货单G y，其中y为非空正整数；
    采购单的表达式为：采购单C i，其中i＝1，2，…，X，C i包含多种需采购的商品M K，其中K＝1，2，…，P；C为X元谓词符号，C 1，C 2，…，C X为项，C i为P元谓词符号，M 1，M 2，…，M P为子项，则类属关系的原子公式为C i(M K)，其中i、k为非空正整数；
    供货单的表达式为：供货单G j,其中j＝1，2，…，Y，G j包含多种可供货的商品N r，其中r＝1，2，…，Q；G为Y元谓词符号，G 1，G 2，…，G Y为项，G j为Q元谓词符号，N 1，N 2，…，N Q为子项，则类属关系的原子公式为G j(N r)，其中j、r为非空正整数。
  3. 根据权利要求1所述的方法，所述第四步骤S4包括，
    获取采购单中的商品排列顺序；
    按照采购单中商品的排列顺序对商品进行交易匹配，直至采购单中所有 商品完成交易匹配。
  4. 根据权利要求3所述的方法，其中，所述对商品进行交易匹配包括，第一分步骤，选取商品M k，基于供货单队列的顺序，从各供货单中选取与所述商品M k同种类的商品，得到供需同码商品G j(N r)；
    第二分步骤，将所述供需同码商品G j(N r)与所述商品M k匹配；
    选取下一商品，重复所述第一分步骤至第二分步骤，直至完成采购单中所有商品M k的匹配，其中1≤k≤p，1≤j≤y，1≤r≤q。
  5. 根据权利要求4所述的方法，其中，对采购单进行交易匹配还包括，
    根据商品M k的商品需求信息设定交易必要条件，利用所述交易必要条件筛选所述供需同码商品G j(N r)，获得推荐商品序列；
    所述交易必要条件包括数量、商品质量、交货期限中的至少一个。
  6. 根据权利要求5所述的方法，其中，利用所述交易必要条件筛选所述供需同码商品包括，
    第一子步骤，根据商品M k的商品需求信息设定数量筛选阈值；
    利用所述数量筛选阈值筛选所述供需同码商品G j(N r)，去除数量小于所述数量筛选阈值的所述供需同码商品G j(N r)，得到第一筛选商品序列；
    第二子步骤，根据所述商品M k的商品需求信息设定质量筛选阈值；
    利用所述质量筛选阈值筛选所述第一筛选商品序列，获得第二筛选商品序列；
    第三子步骤，根据所述商品M k的商品需求信息设定时间筛选阈值；
    利用所述时间筛选阈值筛选所述第二筛选商品序列，获得推荐商品序列；
    第四子步骤，将所述推荐商品序列与所述商品M k匹配。
  7. 根据权利要求6所述的方法，还包括，
    每完成一种商品M k的匹配，更新采购单和供货单，使所述采购单实时与采购商的采购需求相符，使所述供货单实时与供货商的供货能力相符。
  8. 一种用于交易平台实现多元商品智能交易推荐的系统，包括，
    第一装置(1)，用于在采购端获取采购商的商品需求信息，形成采购单，在供货端获取供货商的商品供应信息，形成供货单；
    第二装置(2)，用于在采购端评估采购商，确定采购商优先等级，在供 货端评估供货商，确定供货商优先等级；
    第三装置(3)，用于在采购端将采购单基于采购商优先等级进行排序，得到采购单队列，在供货端对供货单基于供货商优先等级进行排序，得到供货单队列；
    第四装置(4)，用于在采购端基于采购单队列的顺序依次对采购单进行交易匹配，在供货端基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配。
  9. 一种用于交易平台实现多元商品智能交易推荐的方法，包括，
    在采购端，
    获取采购商的商品需求信息，形成采购单；
    评估采购商，确定采购商优先等级；
    将采购单基于采购商优先等级进行排序，得到采购单队列；
    基于采购单队列的顺序依次对采购单进行交易匹配，直至完成所有采购单的交易匹配；
    所述采购单队列包括：采购单C 1、采购单C 2、…、采购单C x，其中x为非空正整数。
  10. 一种用于交易平台实现多元商品智能交易推荐的方法，包括，
    在供货端，
    获取供货商的商品供应信息，形成供货单；
    评估供货商，确定供货商优先等级；
    对供货单基于供货商优先等级进行排序，得到供货单队列；
    基于供货单队列的顺序依次选取供货单，与采购单进行交易匹配，直至完成所有采购单的交易匹配；
    所述供货单队列包括：供货单G 1、供货单G 2、…、供货单G y，其中y为非空正整数。
PCT/CN2019/120571 2019-07-08 2019-11-25 一种用于交易平台实现多元商品智能交易推荐的方法和系统 WO2021003953A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910611992.4A CN110348963B (zh) 2019-07-08 2019-07-08 一种用于交易平台实现多元商品智能交易推荐的方法和系统
CN201910611992.4 2019-07-08

Publications (1)

Publication Number Publication Date
WO2021003953A1 true WO2021003953A1 (zh) 2021-01-14

Family

ID=68178512

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/120571 WO2021003953A1 (zh) 2019-07-08 2019-11-25 一种用于交易平台实现多元商品智能交易推荐的方法和系统

Country Status (2)

Country Link
CN (1) CN110348963B (zh)
WO (1) WO2021003953A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114820142A (zh) * 2022-06-29 2022-07-29 国能(北京)商务网络有限公司 面向b端采购用户的商品信息推荐方法
CN117422378A (zh) * 2023-10-30 2024-01-19 龙腾数科技术有限公司 一种采购供应的智能化管理方法及系统

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110348963B (zh) * 2019-07-08 2022-06-14 周永东 一种用于交易平台实现多元商品智能交易推荐的方法和系统
CN110908575B (zh) * 2019-12-05 2021-05-04 上海斑马来拉物流科技有限公司 数据处理方法、计算机存储介质及电子设备
CN111401986A (zh) * 2020-02-28 2020-07-10 周永东 一种交易平台的商品交易方法和系统
CN112464086A (zh) * 2020-11-23 2021-03-09 刘蓝田 一种基于大数据根据需求自动匹配并呈现特定供方的方法
CN114904447B (zh) * 2022-05-18 2024-03-12 中冶长天国际工程有限责任公司 一种混匀矿生产方法及系统
CN115293826A (zh) * 2022-10-09 2022-11-04 南京卓博威信息科技有限公司 一种智慧供应链营销系统及评估方法
CN116307881A (zh) * 2023-03-13 2023-06-23 国电南瑞南京控制系统有限公司 一种用于电力物资采购的产品质量监督检测方法
CN117078349A (zh) * 2023-08-30 2023-11-17 薪海科技(上海)有限公司 交易智能匹配平台
CN117670466A (zh) * 2023-11-01 2024-03-08 广州市数商云网络科技有限公司 一种基于多端供销平台的供销关系智能化匹配方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090024518A1 (en) * 2007-07-18 2009-01-22 Alena Pavlisova Automated system of supply and demand pairing
CN104599141A (zh) * 2015-01-09 2015-05-06 西安融成科技有限公司 一种在线知识交易的排队交易系统和方法
CN107274261A (zh) * 2017-06-05 2017-10-20 杭州王道起兮科技有限公司 B2b电子商务供需双边匹配方法及系统
CN107491897A (zh) * 2017-08-31 2017-12-19 四川长虹电器股份有限公司 供应商商品评价的方法
CN110348963A (zh) * 2019-07-08 2019-10-18 周永东 一种用于交易平台实现多元商品智能交易推荐的方法和系统

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8732047B2 (en) * 2008-10-24 2014-05-20 Sciquest, Inc. System and method for contract execution against expressive contracts
CN109816496B (zh) * 2018-11-05 2021-07-06 周永东 一种用于交易匹配的方法和装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090024518A1 (en) * 2007-07-18 2009-01-22 Alena Pavlisova Automated system of supply and demand pairing
CN104599141A (zh) * 2015-01-09 2015-05-06 西安融成科技有限公司 一种在线知识交易的排队交易系统和方法
CN107274261A (zh) * 2017-06-05 2017-10-20 杭州王道起兮科技有限公司 B2b电子商务供需双边匹配方法及系统
CN107491897A (zh) * 2017-08-31 2017-12-19 四川长虹电器股份有限公司 供应商商品评价的方法
CN110348963A (zh) * 2019-07-08 2019-10-18 周永东 一种用于交易平台实现多元商品智能交易推荐的方法和系统

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114820142A (zh) * 2022-06-29 2022-07-29 国能(北京)商务网络有限公司 面向b端采购用户的商品信息推荐方法
CN114820142B (zh) * 2022-06-29 2022-09-16 国能(北京)商务网络有限公司 面向b端采购用户的商品信息推荐方法
CN117422378A (zh) * 2023-10-30 2024-01-19 龙腾数科技术有限公司 一种采购供应的智能化管理方法及系统

Also Published As

Publication number Publication date
CN110348963B (zh) 2022-06-14
CN110348963A (zh) 2019-10-18

Similar Documents

Publication Publication Date Title
WO2021003953A1 (zh) 一种用于交易平台实现多元商品智能交易推荐的方法和系统
CN107507075B (zh) 公共采购大数据价格监测方法
Xu Managing production and procurement through option contracts in supply chains with random yield
US7813961B2 (en) System and method for planning, allocation, and purchasing
Pervin et al. An integrated inventory model with variable holding cost under two levels of trade-credit policy
CN110070244A (zh) 供应商评价方法、系统、电子设备和计算机可读介质
Chu et al. Truthful double auction mechanisms
CN111127198A (zh) 一种电力市场交易服务器、市场主体终端及撮合方法
CN113592207A (zh) 基于机器人的网状供应链决策方法、设备及存储介质
CN111489180A (zh) 参考信息生成方法、系统及装置
CN111833158A (zh) 一种行业定型产品框架式招投标管理平台
Sahoo et al. Selling price dependent demand with allowable shortages model under partially backlogged—deteriorating items
CN110838043A (zh) 商品推荐方法及装置
Winarno et al. Competition and globalization of business to further develop creativity and innovation for the advancement of the company (Case Study at PT. Top Drink World)
US20070219836A1 (en) Shipping planning system
CN109740036A (zh) Ota平台酒店排序方法及装置
Poswal et al. Fuzzy optimization model of two parameter weibull deteriorating rate with quadratic demand and variable holding cost under allowable shortages
TW202025016A (zh) 用以便利商品或服務相關活動進行的系統及方法
Quiroz-Flores et al. Integrated Lean Logistics-Warehousing model to reduce Lead Time in an SME of food sector: A research in Peru
Salhab et al. Dynamic marketing policies with rating-sensitive consumers: A mean-field games approach
CN111275294A (zh) 一种商品信息批量处理分析方法、装置和设备
CN114418318A (zh) 一种采购任务优先级队列的生成方法
CN112734237A (zh) 用于交易业务的配送员匹配方法
Gedam et al. Two stage approach to solve extended transportation problem
US20230080661A1 (en) Dynamic augmenting of relevance rankings using data from external ratings sources

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19936665

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 19936665

Country of ref document: EP

Kind code of ref document: A1