A real-time order ticket analysis system that integrates big data and AI technology
Through a real-time order ticket analysis system integrating big data and AI technology, the problem of single decision types in the existing technology is solved, efficient support for multiple decisions and improved financing flexibility, and reduced corporate financial costs.
Patent Information
- Application Number
- CN202410603247.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-15
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-05-15
AI Technical Summary
The existing order analysis system mainly focuses on material-related decisions, cannot cover other types of decisions, lacks universality, and cannot provide comprehensive and efficient decision-making support to enterprises.
The order ticket real-time analysis system using integrated big data and AI technology, including data collection, data processing, AI analysis and decision support modules. It generates decision suggestions through intelligent learning and real-time analysis modules, supports multiple decision types, and through redemption and receipts The split mechanism improves financing flexibility and efficiency.
It achieves efficient support for a variety of decision-making types, improves financing flexibility and efficiency, reduces the financial costs of the enterprise, and is suitable for all types of decisions, including redemption decisions.
Smart Images

Figure CN118485491B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of electronic digital data processing, and in particular to a real-time analysis system for order tickets that integrates big data and AI technologies. Background Art
[0002] Enterprises need to make various decisions in the process of development. The quality of the decisions will directly affect the development of the enterprise. Order information can reflect the current situation of the enterprise to a certain extent. Therefore, an analysis system is needed that can give decision-making suggestions based on order information, effectively help the enterprise develop smoothly, replace or assist a large number of low-level decision-making implementations, and achieve efficient and stable decision-making.
[0003] The foregoing discussion of the background art is intended only to facilitate an understanding of the present invention. This discussion does not acknowledge or admit that any of the material referred to is part of the common general knowledge.
[0004] Now many order analysis systems have been developed. After a lot of searching and reference, it is found that the existing order analysis systems are such as the system disclosed in the publication number CN114462894B. These system methods generally include: based on the data analysis of historical search records, construct a commodity material feature set, the commodity material feature set includes a number of three-level tree structures, and the three-level tree structures are the commodity material main name, commodity material characteristics and commodity material parameter range from top to bottom; obtain the original order of centralized procurement, match the commonality of the original order of centralized procurement with the various levels of the commodity material main name, determine the replaceable commodity materials from the original order of centralized procurement, and superimpose the required quantity of the replaceable commodity materials with the corresponding required quantity in the original order of centralized procurement. However, the decisions involved in this system are only related to the materials in the order, with a narrow coverage, not universal, and cannot provide help for other types of decisions. Summary of the invention
[0005] The purpose of the present invention is to propose a real-time analysis system for order tickets that integrates big data and AI technology to address the existing deficiencies.
[0006] The present invention adopts the following technical solution:
[0007] A real-time order ticket analysis system integrating big data and AI technology, including a data collection module, a data processing module, an AI analysis module and a decision support module;
[0008] The data collection module is used to collect order information, the data processing module is used to sort and classify the collected data, the AI analysis module is used to perform intelligent analysis on the sorted data, and the decision support module provides decision information based on the analysis results;
[0009] The data acquisition module includes an enterprise docking unit, a privacy processing unit and a data acquisition unit. The enterprise docking unit is used to dock with the enterprise system, the privacy processing unit is used to eliminate privacy information in order data and decision data, and the data acquisition unit is used to acquire and store order data and decision data;
[0010] The data processing module includes a sample interception unit, a decision standardization unit and a sample standardization unit. The sample interception unit is used to intercept part of the information from the order data and decision data of the enterprise as a sample. The decision standardization unit is used to classify and standardize the decision data. The sample standardization unit is used to standardize the sample data.
[0011] The AI analysis module includes an intelligent learning unit and a real-time analysis unit. The intelligent learning unit is used to train samples to obtain an order decision model, and the real-time analysis unit uses the order decision model to analyze real-time orders.
[0012] The decision support module includes a decision screening unit and a decision combining unit, wherein the decision screening unit is used to screen out the decision types to be executed, and the decision combining unit is used to calculate the combination ratio of the decision types;
[0013] Further, the sample interception unit includes a decision base processor, an order matching processor and a sample packaging processor, wherein the decision base processor is used to select a decision set as a base of the sample, the order matching processor selects matching order information based on the sample base, and the sample packaging processor packages the decision set and the order information into a sample;
[0014] Furthermore, the sample standardization unit includes an order feature processor and a benefit evaluation processor, wherein the order feature processor is used to calculate feature information of the preceding order and the following order, and the benefit evaluation processor is used to calculate the economic benefit of the decision set;
[0015] The benefit evaluation processor calculates the benefit value Bf according to the following formula:
[0016] ;
[0017] Where V1 is the total amount of the preceding order, P1 is the unit profit of the preceding order, n1 is the number of units of the preceding order, V2 is the total amount of the following order, P2 is the unit profit of the following order, and n2 is the number of units of the following order;
[0018] Furthermore, the real-time analysis unit includes a real-time information collection processor, a model execution processor and a feature sorting processor. The real-time information collection processor is used to collect real-time orders that need to be analyzed and calculate feature information of the orders. The model execution processor is used to input decision data and feature information of different types and quantitative values into the order decision model for operation. The feature sorting processor sorts the output results of the order decision model and sends them to the decision support module.
[0019] Further, the decision screening unit includes a type feature processor and a screening calculation processor, the type feature processor is used to perform feature calculation on the output results of the same decision type, and the screening calculation processor performs screening calculation on the decision type based on the type feature;
[0020] The type feature processor calculates the feature value C of the decision type according to the following formula:
[0021] ;
[0022] Among them, a i Refers to the ith quantized value, Bf(a i ) refers to the benefit value of output feature information at the i-th quantization value, is the quantization interval, Bf0 is the benefit threshold, and La is the quantization length at which the benefit value exceeds the benefit threshold;
[0023] Furthermore, the decision types in the decision support module include redemption decisions. By introducing a flexible redemption mechanism, downstream enterprises are allowed to flexibly choose the redemption time and amount according to their own capital needs and supply chain conditions. At the same time, by introducing a bill splitting mechanism, downstream enterprises are allowed to split supply chain bills into multiple small bills according to order requirements and capital conditions, and redeem them gradually. The platform provides intelligent splitting algorithms and redemption strategies to ensure that the split bill amount matches the redemption plan, thereby improving the flexibility and efficiency of financing and reducing the financial costs of enterprises.
[0024] The beneficial effects achieved by the present invention are:
[0025] This system collects a large amount of sample data through big data, and trains the sample data through AI technology to obtain an order decision model. The order decision model can predict subsequent order information based on real-time order information and arbitrary decision information output, and calculate the benefit information corresponding to the decision. Finally, multiple decision types are combined to give current decision recommendations. The whole process is efficient and applicable to all decision types. The decision types of this system include redemption decisions, which can split large bills into multiple small bills for analysis and redemption, which can improve the flexibility and efficiency of financing and reduce the financial costs of enterprises.
[0026] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and description and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 It is a schematic diagram of the overall structural framework of the present invention;
[0028] Figure 2 This is a schematic diagram of the data acquisition module of the present invention;
[0029] Figure 3 This is a schematic diagram of the data processing module of the present invention;
[0030] Figure 4 This is a schematic diagram of the real-time analysis unit of the present invention;
[0031] Figure 5 This is a schematic diagram of the decision-making screening unit of the present invention;
[0032] Figure 6 It is a schematic diagram of the ticket supply platform framework of the present invention. DETAILED DESCRIPTION
[0033] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual sizes. It is stated in advance. The following embodiments will further explain the relevant technical contents of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention.
[0034] Embodiment 1: This embodiment provides a real-time analysis system for order tickets that integrates big data and AI technology. Figure 1 , including data collection module, data processing module, AI analysis module and decision support module;
[0035] The data collection module is used to collect order information, the data processing module is used to sort and classify the collected data, the AI analysis module is used to perform intelligent analysis on the sorted data, and the decision support module provides decision information based on the analysis results;
[0036] The data acquisition module includes an enterprise docking unit, a privacy processing unit and a data acquisition unit. The enterprise docking unit is used to dock with the enterprise system, the privacy processing unit is used to eliminate privacy information in order data and decision data, and the data acquisition unit is used to acquire and store order data and decision data;
[0037] The data processing module includes a sample interception unit, a decision standardization unit and a sample standardization unit. The sample interception unit is used to intercept part of the information from the order data and decision data of the enterprise as a sample. The decision standardization unit is used to classify and standardize the decision data. The sample standardization unit is used to standardize the sample data.
[0038] The AI analysis module includes an intelligent learning unit and a real-time analysis unit. The intelligent learning unit is used to train samples to obtain an order decision model, and the real-time analysis unit uses the order decision model to analyze real-time orders.
[0039] The decision support module includes a decision screening unit and a decision combining unit, wherein the decision screening unit is used to screen out the decision types to be executed, and the decision combining unit is used to calculate the combination ratio of the decision types;
[0040] The sample interception unit includes a decision base processor, an order matching processor and a sample packaging processor, wherein the decision base processor is used to select a decision set as a base of a sample, the order matching processor selects matching order information based on the sample base, and the sample packaging processor packages the decision set and the order information into a sample;
[0041] The sample standardization unit includes an order feature processor and a benefit evaluation processor, wherein the order feature processor is used to calculate the feature information of the preceding order and the following order, and the benefit evaluation processor is used to calculate the economic benefit of the decision set;
[0042] The benefit evaluation processor calculates the benefit value Bf according to the following formula:
[0043] ;
[0044] Where V1 is the total amount of the preceding order, P1 is the unit profit of the preceding order, n1 is the number of units of the preceding order, V2 is the total amount of the following order, P2 is the unit profit of the following order, and n2 is the number of units of the following order;
[0045] The real-time analysis unit includes a real-time information collection processor, a model execution processor and a feature sorting processor. The real-time information collection processor is used to collect real-time orders that need to be analyzed and calculate the feature information of the orders. The model execution processor is used to input decision data and feature information of different types and quantitative values into the order decision model for operation. The feature sorting processor sorts the output results of the order decision model and sends them to the decision support module.
[0046] The decision screening unit includes a type feature processor and a screening calculation processor, wherein the type feature processor is used to perform feature calculation on the output results of the same decision type, and the screening calculation processor performs screening calculation on the decision type based on the type feature;
[0047] The type feature processor calculates the feature value C of the decision type according to the following formula:
[0048] ;
[0049] Among them, a i Refers to the ith quantized value, Bf(a i ) refers to the benefit value of output feature information at the i-th quantization value, is the quantization interval, Bf0 is the benefit threshold, and La is the quantization length at which the benefit value exceeds the benefit threshold.
[0050] Embodiment 2: This embodiment includes all the contents of Embodiment 1, and provides an order ticket real-time analysis system integrating big data and AI technology, including a data collection module, a data processing module, an AI analysis module and a decision support module;
[0051] The data collection module is used to collect order information, the data processing module is used to sort and classify the collected data, the AI analysis module is used to perform intelligent analysis on the sorted data, and the decision support module provides decision information based on the analysis results;
[0052] Combination Figure 2 The data acquisition module includes an enterprise docking unit, a privacy processing unit and a data acquisition unit. The enterprise docking unit is used to dock with the enterprise system, the privacy processing unit is used to eliminate the privacy information in the order data and the decision data, and the data acquisition unit is used to acquire and store the order data and the decision data;
[0053] Combination Figure 3The data processing module includes a sample interception unit, a decision standardization unit and a sample standardization unit. The sample interception unit is used to intercept part of the information from the order data and decision data of the enterprise as a sample. The decision standardization unit is used to classify and standardize the decision data. The sample standardization unit is used to standardize the sample data;
[0054] The AI analysis module includes an intelligent learning unit and a real-time analysis unit. The intelligent learning unit is used to train samples to obtain an order decision model, and the real-time analysis unit uses the order decision model to analyze real-time orders.
[0055] The decision support module includes a decision screening unit and a decision combining unit, wherein the decision screening unit is used to screen out the decision types to be executed, and the decision combining unit is used to calculate the combination ratio of the decision types;
[0056] The enterprise docking unit includes an enterprise search processor and a communication connection processor, wherein the enterprise search processor is used to retrieve the network address of the enterprise, and the communication connection processor is used to connect with the retrieved network address and transmit data;
[0057] The privacy processing unit includes a privacy identification processor and a code replacement processor, wherein the privacy identification processor is used to identify the proper name, and the code replacement processor is used to create a code for the proper name and replace the proper name in the data with the code;
[0058] The data acquisition unit includes an order information acquisition processor and a decision information acquisition processor, wherein the order information acquisition processor is used to acquire the order information of the enterprise, and the decision information acquisition processor is used to acquire the decision information of the enterprise;
[0059] The sample interception unit includes a decision base processor, an order matching processor and a sample packaging processor, wherein the decision base processor is used to select a decision set as a base of a sample, the order matching processor selects matching order information based on the sample base, and the sample packaging processor packages the decision set and the order information into a sample;
[0060] The decision base processor adds decisions with time closeness into the same set. The time closeness means that the time interval between two decisions does not exceed the close threshold. The time interval between the first decision and the last decision in a decision set is called the base width, denoted by T. The decision time of the first decision in the decision set is called the base time.
[0061] The order matching processor calculates the lead time according to the following formula and post-time :
[0062] ;
[0063] ;
[0064] Among them, T0 is the basic time, is the pre-factor, is the time efficiency coefficient;
[0065] The order matching processor regards the order information within the lead time before the base time as the lead order, and regards the order information within the post time after the base time as the post order;
[0066] The decision standardization unit includes a decision classification processor and a decision quantification processor, wherein the decision classification processor is used to classify each decision, and the decision quantification processor is used to quantify the degree of each decision;
[0067] The sample standardization unit includes an order feature processor and a benefit evaluation processor, wherein the order feature processor is used to calculate the feature information of the preceding order and the following order, and the benefit evaluation processor is used to calculate the economic benefit of the decision set;
[0068] The benefit evaluation processor calculates the benefit value Bf according to the following formula:
[0069] ;
[0070] Where V1 is the total amount of the preceding order, P1 is the unit profit of the preceding order, n1 is the number of units of the preceding order, V2 is the total amount of the following order, P2 is the unit profit of the following order, and n2 is the number of units of the following order;
[0071] The unit total amount V, unit profit P and unit order number n are all obtained by dividing the total amount, profit and order number by time. The unit total amount, unit profit and unit order number are the characteristic information of the order;
[0072] The sample standardization unit converts V1, P1, n1, and decision set data and Bf as the data information of a sample;
[0073] Tp represents the decision category, a is the decision quantification value, i Represents the i-th decision data in the decision set;
[0074] The intelligent learning unit includes a sample input control processor, a learning optimization processor and a model output processor, wherein the input control processor is used to input samples into the learning optimization processor for training and control and manage the input process, the learning optimization processor is used to perform nonlinear calculation processing on the data in the samples and continuously optimize the parameters, and the model output processor outputs an order decision model based on the optimized parameters;
[0075] The learning optimization processor adopts the existing machine learning technology framework for training and learning;
[0076] The order decision model takes single decision data and order feature information as input and outputs order feature information, which is expressed as follows: ;
[0077] Combination Figure 4 The real-time analysis unit includes a real-time information collection processor, a model execution processor and a feature sorting processor. The real-time information collection processor is used to collect real-time orders that need to be analyzed and calculate the feature information of the orders. The model execution processor is used to input decision data and feature information of different types and quantitative values into the order decision model for operation. The feature sorting processor sorts the output results of the order decision model and sends them to the decision support module.
[0078] The model execution processor sets the decision data at the same quantization interval;
[0079] Combination Figure 5 , the decision screening unit includes a type feature processor and a screening calculation processor, the type feature processor is used to perform feature calculation on the output results of the same decision type, and the screening calculation processor performs screening calculation on the decision type based on the type feature;
[0080] The type feature processor calculates the feature value C of the decision type according to the following formula:
[0081] ;
[0082] Among them, a i Refers to the ith quantized value, Bf(a i ) refers to the benefit value of output feature information at the i-th quantization value, is the quantization interval, Bf0 is the benefit threshold, and La is the quantization length at which the benefit value exceeds the benefit threshold;
[0083] The type feature processor sorts the decision types from large to small according to the feature values;
[0084] The screening calculation processor calculates the cumulative characteristic value Ac(m) according to the following formula:
[0085] ;
[0086] Among them, C i’ represents the characteristic value of the i'th decision type, and m is the cumulative number;
[0087] When the cumulative characteristic value is just greater than the cumulative threshold, the first m decision types are used as the screened decision types;
[0088] The decision combination unit includes a weight calculation processor and a quantization allocation processor, wherein the weight calculation processor is used to calculate the weight value of the decision type, and the quantization allocation processor is used to calculate the quantization allocation value of each decision type:
[0089] The weight calculation processor calculates the basic value Q of the i'th decision type according to the following formula: i’ :
[0090] ;
[0091] Among them, Bf i’ Represents the maximum benefit value of the i'th decision type
[0092] The weight calculation processor calculates the weight value k of the i'th decision type according to the following formula: i’ :
[0093] ;
[0094] The quantization allocation processor calculates the quantization value a(i') of the i'th decision type according to the following formula:
[0095] ;
[0096] Among them, a max Indicates the quantitative value when the benefit value is maximum;
[0097] The recommended decision information is composed of decision type and quantitative value;
[0098] The decision types in the decision support module include redemption decisions. By introducing a flexible redemption mechanism, downstream enterprises are allowed to flexibly choose the redemption time and amount according to their own capital needs and supply chain conditions. At the same time, by introducing a bill splitting mechanism, downstream enterprises are allowed to split supply chain bills into multiple small bills according to order needs and capital conditions, and gradually redeem the goods. The platform provides intelligent splitting algorithms and redemption strategies to ensure that the bill amount after splitting matches the redemption plan, thereby improving the flexibility and efficiency of financing and reducing the financial costs of enterprises;
[0099] Example 3: This system involves a ticket supply platform, combined with Figure 6, including the registration configuration center, the middle platform microservice group, the access encryption machine, the gateway, the service cluster, the group IAM, the storage and operation management backend. Among them, the service cluster is the core part of the ticket supply platform, including the ticket service group and the transaction service group. The ticket service group is used to process the registration and verification of the ticket, and the transaction service group is used to process the redemption of the ticket;
[0100] The workflow of the ticket service group includes the following steps:
[0101] S1. Initiate financing application;
[0102] S2. The ticket is successfully issued after the information is verified;
[0103] S3. Verify the ticket registration;
[0104] S4. Sign for the bill;
[0105] S5. Recharge the virtual account;
[0106] The workflow of the transaction service group includes the following steps:
[0107] S21. Check balance and place order;
[0108] S22, initiate a redemption application;
[0109] S23. Verify the application information and complete endorsement and receipt;
[0110] S24, initiate payment application;
[0111] S25, confirm the deduction and settle the bill;
[0112] The i and i' appearing in the above text are ordinal numbers used to indicate sequence numbers.
[0113] The contents disclosed above are only preferred feasible embodiments of the present invention, and do not limit the protection scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the protection scope of the present invention. In addition, the elements therein can be updated as technology develops.
Claims
1. A real-time order ticket analysis system integrating big data and AI technology, characterized in that: It includes data collection module, data processing module, AI analysis module and decision support module; The data collection module is used to collect order information, the data processing module is used to sort and classify the collected data, the AI analysis module is used to perform intelligent analysis on the sorted data, and the decision support module provides decision information based on the analysis results; The data collection module includes an enterprise docking unit, a privacy processing unit and a data acquisition unit. The enterprise docking unit is used to dock with the enterprise system, the privacy processing unit is used to eliminate the privacy information in the order data and decision data, and the data acquisition unit is used to acquire and store the order data and decision data; The data processing module includes a sample interception unit, a decision standardization unit and a sample standardization unit. The sample interception unit is used to intercept part of the information from the order data and decision data of the enterprise as a sample. The decision standardization unit is used to classify and standardize the decision data. The sample standardization unit is used to standardize the sample data. The AI analysis module includes an intelligent learning unit and a real-time analysis unit. The intelligent learning unit is used to train samples to obtain an order decision model, and the real-time analysis unit uses the order decision model to analyze real-time orders. The decision support module includes a decision screening unit and a decision combining unit, wherein the decision screening unit is used to screen out the decision types to be executed, and the decision combining unit is used to calculate the combination ratio of the decision types; The sample interception unit includes a decision base processor, an order matching processor and a sample packaging processor, wherein the decision base processor is used to select a decision set as a base of a sample, the order matching processor selects matching order information based on the sample base, and the sample packaging processor packages the decision set and the order information into a sample; The decision screening unit includes a type feature processor and a screening calculation processor, wherein the type feature processor is used to perform feature calculation on the output results of the same decision type, and the screening calculation processor performs screening calculation on the decision type based on the type feature; The type feature processor calculates the feature value C of the decision type according to the following formula: ; Among them, a i Refers to the ith quantized value, Bf(a i ) refers to the benefit value of output feature information at the i-th quantization value, is the quantization interval, Bf0 is the benefit threshold, and La is the quantization length at which the benefit value exceeds the benefit threshold; The type feature processor sorts the decision types from large to small according to the feature values; The screening calculation processor calculates the cumulative characteristic value Ac(m) according to the following formula: ; in, Indicates The characteristic value of the decision type, m is the cumulative number; When the cumulative characteristic value is just greater than the cumulative threshold, the first m decision types are used as the screened decision types; The decision combination unit includes a weight calculation processor and a quantization allocation processor, wherein the weight calculation processor is used to calculate the weight value of the decision type, and the quantization allocation processor is used to calculate the quantization allocation value of each decision type: The weight calculation processor calculates the first Base value for each decision type : ; in, Indicates The maximum benefit value of a decision type The weight calculation processor calculates the first The weight of the decision type : ; The quantization allocation processor calculates the first Quantitative value of decision type : ; Among them, a max Indicates the quantitative value when the benefit value is maximum; The recommended decision information is composed of decision type and quantitative value; The sample standardization unit includes an order feature processor and a benefit evaluation processor, wherein the order feature processor is used to calculate the feature information of the preceding order and the following order, and the benefit evaluation processor is used to calculate the economic benefit of the decision set; The benefit evaluation processor calculates the benefit value Bf according to the following formula: ; Among them, V1 is the unit total amount of the preceding order, P1 is the unit profit of the preceding order, n1 is the unit order number of the preceding order, V2 is the unit total amount of the subsequent order, P2 is the unit profit of the subsequent order, and n2 is the unit order number of the subsequent order.
2. The real-time analysis system for order tickets integrating big data and AI technology as claimed in claim 1, characterized in that: The real-time analysis unit includes a real-time information acquisition processor, a model execution processor and a feature sorting processor. The real-time information acquisition processor is used to collect real-time orders that need to be analyzed and calculate the feature information of the orders. The model execution processor is used to input decision data and feature information of different types and quantitative values into the order decision model for execution. The feature sorting processor sorts the output results of the order decision model and sends them to the decision support module.
Citation Information
Cited By
Large media advertisement charging profit data intelligent monitoring method and system based on AI
CN120873919A
AI-based big media advertisement charging profit data intelligent monitoring method and system
CN120873919B
