Supply chain bill pricing abnormity early warning system based on artificial intelligence large model

Abstract

The invention provides a supply chain bill pricing abnormity early warning system based on an artificial intelligence large model, and relates to the field of electric digital data processing, the supply chain bill pricing abnormity early warning system comprises a bill information management module, an enterprise information management module, a bill pricing analysis module and an early warning prompt processing module, the enterprise information management module is used for managing credit information of an enterprise, the bill pricing analysis module is used for intelligently analyzing a discount interest rate, and the early warning prompt processing module executes a corresponding early warning operation based on a pricing analysis result; according to the system, bill pricing is intelligently analyzed through big data, abnormal bill pricing is detected, and an enterprise is helped to better manage bill operation.

Description

Technical Field

[0001] The present invention relates to the field of electronic digital data processing, and in particular to a supply chain bill pricing anomaly early warning system based on an artificial intelligence large model. Background Art

[0002] In modern supply chain finance, bills, as important payment and financing tools, are widely used in corporate transactions. However, due to the complexity of the bill market, their pricing is influenced by a variety of factors, including market supply and demand fluctuations, credit risk, and macroeconomic conditions. This leads to significant volatility in bill pricing. Furthermore, because supply chain transactions involve multiple parties, some companies may exploit bill arbitrage and fraudulent transactions to gain undue advantage, potentially leading to financial risks. Therefore, a system is needed to automatically analyze bill pricing to ensure that corporate interests are not harmed.

[0003] Numerous ticket systems have been developed. After extensive research and reference, I discovered existing ticket systems, such as the one disclosed in Publication No. CN118863893B. These systems generally include a front-end application module, a back-end management module, a service processing module, and a block recording module. The front-end application module allows users to log in to the platform and access application information; the back-end management module receives and manages application information; the service processing module executes specific ticket services; and the block recording module records ticket cycle information. However, these systems only manage information on ticket circulation and are unable to analyze the rationality of ticket prices. Summary of the Invention

[0004] The purpose of this invention is to address the existing deficiencies and propose a supply chain bill pricing anomaly early warning system based on an artificial intelligence large model.

[0005] The present invention adopts the following technical solutions:

[0006] A supply chain bill pricing anomaly warning system based on an artificial intelligence large model, including a bill information management module, an enterprise information management module, a bill pricing analysis module, and an early warning prompt processing module;

[0007] The bill information management module is used to manage bill circulation information, the enterprise information management module is used to manage enterprise credit information, the bill pricing analysis module is used to perform intelligent analysis of discount rates, and the early warning processing module performs corresponding early warning operations based on the pricing analysis results;

[0008] The bill information management module includes a bill status management unit, a bill circulation recording unit, and a bill archiving and statistics unit. The bill status management unit is used to manage the life cycle status of the bill, the bill circulation recording unit is used to record and store each bill circulation information, and the bill archiving and statistics unit is used to archive the bills of the completed cycle and count the common parameters of all archived bills.

[0009] The enterprise information management module includes an enterprise information registration unit, an enterprise information collection unit, and a dynamic credit evaluation unit. The enterprise information registration unit is used to register the enterprise related to the bill in the system, the enterprise information collection unit is used to collect the enterprise's operating information, and the dynamic credit evaluation unit is used to evaluate and record the enterprise's credit ability.

[0010] The bill pricing analysis module includes a bill discount monitoring unit, a related information collection unit, and a multi-dimensional feature analysis unit. The bill discount monitoring unit is used to monitor the bill discount status in real time, the related information collection unit is used to collect data information related to the discounted bills, and the multi-dimensional feature analysis unit is used to perform intelligent analysis of bill prices.

[0011] The early warning prompt processing module includes an early warning grading unit, an early warning locking unit and an early warning execution unit. The early warning grading unit is used to store detailed information of the early warning level. The early warning locking unit locks the corresponding early warning level based on the price analysis result. The early warning execution unit executes corresponding safety measures based on the early warning level.

[0012] Furthermore, the multidimensional feature analysis unit includes a discount timing processor, a big data comparison processor and an enterprise adjustment processor. The discount timing processor is used to perform timing analysis on the circulation process of the bill, the big data comparison processor is used to compare the big data information with the analysis results, and the enterprise adjustment processor adjusts the discount data based on the enterprise credit.

[0013] Furthermore, the discount timing processor calculates the discount rate d of the i-th discount according to the following formula: i :

[0014]

[0015] Among them, P i is the i-th discount price, F is the face value of the bill, t i is the number of days remaining for the i-th discount;

[0016] The discount time series processor calculates the average discount rate and discount standard deviation of the data including the current discount data and the data excluding the current discount data, and calculates the reasonable value S1 of the time series according to the following formula:

[0017]

[0018] Where d is the current discount rate, is the average discount rate including the current discount data, is the average discount rate excluding the current discount data, σ1 is the standard deviation of discount including the current discount data, and σ2 is the standard deviation of discount excluding the current discount data.

[0019] Furthermore, the big data comparison processor obtains all discount data consistent with the current discount type and calculates the average discount rate and the discount standard deviation σ3, and calculate the historical reasonable value S2 according to the following formula:

[0020]

[0021] The enterprise adjustment processor calculates the adjustment coefficient k according to the following formula:

[0022]

[0023] Among them, V is the credit score of the enterprise, and M is the total credit score.

[0024] Furthermore, the warning locking unit includes an analysis result receiving processor, a warning mapping processor, and a bill locking processor. The analysis result receiving processor is used to receive bill price analysis data, the warning mapping processor is used to map to a corresponding warning level, and the bill locking processor is used to add a warning tag to the corresponding bill to lock it.

[0025] The early warning mapping processor calculates the risk value Q of the current discount price according to the following formula:

[0026]

[0027] Among them, P0 is the current discount price and t0 is the remaining days of the current discount.

[0028] The beneficial effects achieved by the present invention are:

[0029] This system obtains big data information and conducts intelligent analysis of bill prices from three perspectives: pricing changes of the bill itself, historical comparisons of similar bills, and the company's own credit capabilities. Based on the analysis results, it obtains risk information on bill pricing and takes corresponding protective measures, which can effectively protect the interests of the company in a timely manner.

[0030] 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 illustration and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the overall structural framework of the present invention;

[0032] Figure 2 This is a schematic diagram of the bill information management module of the present invention;

[0033] Figure 3 This is a schematic diagram of the enterprise information management module of the present invention;

[0034] Figure 4 This is a schematic diagram of the bill pricing analysis module of the present invention;

[0035] Figure 5 This is a schematic diagram of the early warning processing module of the present invention;

[0036] Figure 6 This is a comparison data table of anomaly detection accuracy of the present invention. DETAILED DESCRIPTION

[0037] 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 in actual size. It is stated in advance. The following embodiments will further explain the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0038] Example 1.

[0039] This embodiment provides a supply chain bill pricing anomaly warning system based on an artificial intelligence large model, combined with Figure 1 , including bill information management module, enterprise information management module, bill pricing analysis module and early warning prompt processing module;

[0040] The bill information management module is used to manage bill circulation information, the enterprise information management module is used to manage enterprise credit information, the bill pricing analysis module is used to perform intelligent analysis of discount rates, and the early warning processing module performs corresponding early warning operations based on the pricing analysis results;

[0041] The bill information management module includes a bill status management unit, a bill circulation recording unit, and a bill archiving and statistics unit. The bill status management unit is used to manage the life cycle status of the bill, the bill circulation recording unit is used to record and store each bill circulation information, and the bill archiving and statistics unit is used to archive the bills of the completed cycle and count the common parameters of all archived bills.

[0042] The enterprise information management module includes an enterprise information registration unit, an enterprise information collection unit, and a dynamic credit evaluation unit. The enterprise information registration unit is used to register the enterprise related to the bill in the system, the enterprise information collection unit is used to collect the enterprise's operating information, and the dynamic credit evaluation unit is used to evaluate and record the enterprise's credit ability.

[0043] The bill pricing analysis module includes a bill discount monitoring unit, a related information collection unit, and a multi-dimensional feature analysis unit. The bill discount monitoring unit is used to monitor the bill discount status in real time, the related information collection unit is used to collect data information related to the discounted bills, and the multi-dimensional feature analysis unit is used to perform intelligent analysis of bill prices.

[0044] The early warning prompt processing module includes an early warning grading unit, an early warning locking unit and an early warning execution unit. The early warning grading unit is used to store detailed information of the early warning level. The early warning locking unit locks the corresponding early warning level based on the price analysis result. The early warning execution unit executes corresponding safety measures based on the early warning level.

[0045] The multidimensional feature analysis unit includes a discount timing processor, a big data comparison processor and an enterprise adjustment processor. The discount timing processor is used to perform timing analysis on the circulation process of the bill, the big data comparison processor is used to compare the big data information with the analysis results, and the enterprise adjustment processor adjusts the discount data based on the enterprise credit.

[0046] The discount timing processor calculates the discount rate d of the i-th discount according to the following formula: i :

[0047]

[0048] Among them, P i is the i-th discount price, F is the face value of the bill, t i is the number of days remaining for the i-th discount;

[0049] The discount time series processor calculates the average discount rate and discount standard deviation of the data including the current discount data and the data excluding the current discount data, and calculates the reasonable value S1 of the time series according to the following formula:

[0050]

[0051] Where d is the current discount rate, is the average discount rate including the current discount data, is the average discount rate excluding the current discount data, σ1 is the standard deviation of discount including the current discount data, and σ2 is the standard deviation of discount excluding the current discount data.

[0052] The big data comparison processor obtains all discount data consistent with the current discount type and calculates the average discount rate and the discount standard deviation σ3, and calculate the historical reasonable value S2 according to the following formula:

[0053]

[0054] The enterprise adjustment processor calculates the adjustment coefficient k according to the following formula:

[0055]

[0056] Among them, V is the enterprise's credit score, and M is the total credit score.

[0057] The warning locking unit includes an analysis result receiving processor, a warning mapping processor, and a bill locking processor. The analysis result receiving processor is used to receive bill price analysis data, the warning mapping processor is used to map to a corresponding warning level, and the bill locking processor is used to add a warning tag to the corresponding bill to lock it.

[0058] The early warning mapping processor calculates the risk value Q of the current discount price according to the following formula:

[0059]

[0060] Among them, P0 is the current discount price and t0 is the remaining days of the current discount.

[0061] Example 2.

[0062] This embodiment includes all the contents of the first embodiment, and provides a supply chain bill pricing anomaly warning system based on an artificial intelligence large model, including a bill information management module, an enterprise information management module, a bill pricing analysis module, and a warning prompt processing module;

[0063] The bill information management module is used to manage bill circulation information, the enterprise information management module is used to manage enterprise credit information, the bill pricing analysis module is used to perform intelligent analysis of discount rates, and the early warning processing module performs corresponding early warning operations based on the pricing analysis results;

[0064] Combine Figure 2 The bill information management module includes a bill status management unit, a bill circulation recording unit, and a bill archiving and statistics unit. The bill status management unit is used to manage the life cycle status of the bill, the bill circulation recording unit is used to record and store each bill circulation information, and the bill archiving and statistics unit is used to archive the bills of the completed cycle and count the common parameters of all archived bills;

[0065] Combine Figure 3 The enterprise information management module includes an enterprise information registration unit, an enterprise information collection unit, and a dynamic credit evaluation unit. The enterprise information registration unit is used to register the enterprise related to the bill in the system, the enterprise information collection unit is used to collect the enterprise's operating information, and the dynamic credit evaluation unit is used to evaluate and record the enterprise's credit ability;

[0066] Combine Figure 4 The bill pricing analysis module includes a bill discount monitoring unit, a related information collection unit, and a multi-dimensional feature analysis unit. The bill discount monitoring unit is used to monitor the bill discount status in real time, the related information collection unit is used to collect data information related to the discounted bills, and the multi-dimensional feature analysis unit is used to perform intelligent analysis of the bill price.

[0067] Combine Figure 5 The warning prompt processing module includes a warning grading unit, a warning locking unit and a warning execution unit. The warning grading unit is used to store detailed information of the warning level. The warning locking unit locks the corresponding warning level based on the price analysis result. The warning execution unit executes the corresponding safety measures based on the warning level.

[0068] The bill status management unit includes a bill creation processor, a status synchronization processor, and an expiration timer processor. The bill creation processor is used to create a management space for each new bill, the status synchronization processor is used to synchronize status information when the bill is circulated, and the expiration timer processor is used to count down each bill.

[0069] The bill circulation recording unit includes a circulation event collector, a time and space stamp generator, and a circulation path recorder. The circulation event collector is used to collect bill circulation events, the time and space stamp generator is used to generate the time and space information of the circulation events, and the circulation path recorder is used to construct a bill flow map.

[0070] The bill archiving and statistics unit includes an archiving rule configurator, an archiving tag manager, and a statistical report generator. The archiving rule configurator is used to set archiving rule information, the archiving tag manager is used to archive bills and generate corresponding tags for management, and the statistical report generator is used to generate statistical information.

[0071] The enterprise information registration unit includes an enterprise information register, a bill enterprise detector, and a basic information collector. The enterprise information register is used to store the basic information of the enterprise. The bill enterprise detector is used to detect whether the associated enterprise of the new bill has been registered. The basic information collector is used to collect the basic information of unregistered enterprises.

[0072] The enterprise information collection unit includes a crawling address register, an operation information collector, and an operation information register. The crawling address register is used to store the data crawling address of each registered enterprise. The operation information collector is used to collect the operation information of the enterprise from the crawling address. The operation information register is used to store the operation information of the enterprise.

[0073] The dynamic credit evaluation unit includes a credit index library, a credit scoring processor, and a scoring control processor. The credit index library is used to store index information for evaluating credit, the credit scoring processor is used to perform credit scoring on the enterprise, and the scoring control processor is used to control the scoring timing.

[0074] The bill discount monitoring unit includes a discount event sensor, a task generation processor, and a task encoding processor. The discount event sensor is used to trigger the task generation processor when a bill discount occurs. The task generation processor generates a corresponding analysis task set based on the bill discount information. The task encoding processor is used to encode the analysis task set.

[0075] The bill discount monitoring unit further includes an anomaly detection processor, which detects discount rate mutations and discounting behaviors with abnormal discount transaction times by embedding a lightweight large model;

[0076] The associated information collection unit includes a task receiving processor, a data information collector, and a data aggregation processor. The task receiving processor is used to receive an analysis task set. The data information collector obtains corresponding bill information and enterprise information based on the analysis task. The data aggregation processor is used to pair and aggregate the collected data information based on the task code.

[0077] The associated information collection unit also includes a graph analysis processor, which collects and preprocesses structured and unstructured data from different sources to make them suitable for large-scale model processing, and uses the named entity recognition and relationship extraction capabilities of the large-scale model to extract key entities and their relationships from unstructured data and map them to structured data. It uses a resource description framework or a graph database to store various entities such as enterprises, bills, markets, and their relationships, generates graph structured data, forms a cross-modal bill knowledge graph, and visualizes transaction patterns and risk propagation paths in the supply chain;

[0078] The multi-dimensional feature analysis unit includes a discount time series processor, a big data comparison processor, and an enterprise adjustment processor. The discount time series processor is used to perform time series analysis on the bill circulation process. The big data comparison processor is used to compare big data information with the analysis results. The enterprise adjustment processor adjusts the discount data based on the enterprise credit.

[0079] The discount timing processor calculates the discount rate d for the i-th discount according to the following formula: i :

[0080]

[0081] Among them, P i is the i-th discount price, F is the face value of the bill, t i is the number of days remaining for the i-th discount;

[0082] The discount time series processor calculates the average discount rate and discount standard deviation of the data including the current discount data and the data excluding the current discount data, and calculates the reasonable value S1 of the time series according to the following formula:

[0083]

[0084] Where d is the current discount rate, is the average discount rate including the current discount data, is the average discount rate excluding the current discount data, σ1 is the standard deviation of discount including the current discount data, and σ2 is the standard deviation of discount excluding the current discount data;

[0085] The big data comparison processor obtains all discount data consistent with the current discount type and calculates the average discount rate and the discount standard deviation σ3, and calculate the historical reasonable value S2 according to the following formula:

[0086]

[0087] The big data comparison processor also retrieves supply chain network information through big data, including unstructured data such as upstream and downstream corporate news content and public opinion dynamics, and assigns weights to text features through the large model attention mechanism. Combined with the knowledge graph relationship generated by the graph analysis processor, high-precision abnormal target monitoring is achieved. At the same time, a joint reasoning framework of a multimodal large model is adopted to establish a trustworthy evaluation matrix from the dimensions of language consistency, data timeliness, and information authority to evaluate the credibility of information such as news and public opinion. When the credibility is greater than the threshold, the position relationship between the abnormal target and the discount enterprise in the supply chain is analyzed to obtain a logical relationship, and then the logical relationship is sent to the early warning processing module;

[0088] The enterprise adjustment processor calculates the adjustment coefficient k according to the following formula:

[0089]

[0090] Among them, V is the credit score of the enterprise, and M is the total credit score;

[0091] The credit score of an enterprise can adopt any scoring method in the existing technology. This system has no specific requirements for the scoring method. For example, the following simple scoring method:

[0092] The credit score range is 1 to 100, so the total credit score M is 100, and the credit score value is the product of the percentage of the company's debts paid off within the specified period and 100;

[0093] The warning classification unit includes a rule editing processor, a warning rule register and a rule calling processor. The rule editing processor is used to edit the warning classification rules. The warning rule register is used to store the warning classification rules. The rule calling processor is used to call the warning classification rules.

[0094] The warning locking unit includes an analysis result receiving processor, a warning mapping processor, and a bill locking processor. The analysis result receiving processor is used to receive bill price analysis data, the warning mapping processor is used to map to a corresponding warning level, and the bill locking processor is used to add a warning tag to the corresponding bill to lock it.

[0095] The early warning mapping processor calculates the risk value Q of the current discount price according to the following formula:

[0096]

[0097] Where P0 is the current discount price and t0 is the number of days remaining for the current discount;

[0098] The early warning execution unit includes a response strategy library, a handling instruction distributor, and a handling feedback collector. The response strategy library is used to store response strategies corresponding to different early warning levels. The handling instruction distributor sends execution instructions based on the response strategies. The handling feedback collector is used to collect feedback information after handling.

[0099] The i appearing in the above text is an ordinal number used to indicate a sequence number and has no actual meaning.

[0100] Some code information of this system is as follows:

[0101]

[0102] self.bill_archive = [] #Archived bill

[0103] def manage_bill_status(self,bill_id,status):

[0104] """Manage the lifecycle status of a bill"""

[0105] self.bill_status[bill_id]=status

[0106] def record_bill_transaction(self,bill_id,transaction_info):

[0107] """Record bill circulation information"""

[0108] self.bill_records.append({"bill_id":bill_id,

[0109] "transaction_info":transaction_info})

[0110] def archive_bill(self,bill_id,details):

[0111] """Archive processing and statistics of archived bills"""

[0112] self.bill_archive.append({"bill_id":bill_id,"details":details})

[0113] class EnterpriseInfoManagement:

[0114] """Enterprise Information Management Module"""

[0115] def__init__(self):

[0116] self.enterprise_data = {} #Enterprise information storage

[0117] self.credit_scores = {} # Dynamic credit score

[0118] def register_enterprise(self,enterprise_id,enterprise_info):

[0119] """Enterprise registration"""

[0120] self.enterprise_data[enterprise_id]=enterprise_info

[0121] def collect_enterprise_info(self,enterprise_id,operational_data):

[0122]

[0123]

[0124]

[0125] The contents disclosed above are only preferred feasible embodiments of the present invention and do not limit the scope of protection of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention description and drawings are included in the scope of protection of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. A supply chain bill pricing anomaly warning system based on an artificial intelligence large model, characterized by: It includes bill information management module, enterprise information management module, bill pricing analysis module and early warning processing module; The bill information management module is used to manage bill circulation information, the enterprise information management module is used to manage enterprise credit information, the bill pricing analysis module is used to perform intelligent analysis of discount rates, and the early warning processing module performs corresponding early warning operations based on the pricing analysis results; The bill information management module includes a bill status management unit, a bill circulation recording unit, and a bill archiving and statistics unit. The bill status management unit is used to manage the life cycle status of the bill, the bill circulation recording unit is used to record and store each bill circulation information, and the bill archiving and statistics unit is used to archive the bills of the completed cycle and count the common parameters of all archived bills. The enterprise information management module includes an enterprise information registration unit, an enterprise information collection unit, and a dynamic credit evaluation unit. The enterprise information registration unit is used to register the enterprise related to the bill in the system, the enterprise information collection unit is used to collect the enterprise's operating information, and the dynamic credit evaluation unit is used to evaluate and record the enterprise's credit ability. The bill pricing analysis module includes a bill discount monitoring unit, a related information collection unit, and a multi-dimensional feature analysis unit. The bill discount monitoring unit is used to monitor the bill discount status in real time, the related information collection unit is used to collect data information related to the discounted bills, and the multi-dimensional feature analysis unit is used to perform intelligent analysis of bill prices. The early warning prompt processing module includes an early warning grading unit, an early warning locking unit and an early warning execution unit. The early warning grading unit is used to store detailed information of the early warning level. The early warning locking unit locks the corresponding early warning level based on the price analysis result. The early warning execution unit executes corresponding safety measures based on the early warning level.

2. The supply chain bill pricing anomaly warning system based on an artificial intelligence large model as claimed in claim 1 is characterized in that: The multidimensional feature analysis unit includes a discount timing processor, a big data comparison processor and an enterprise adjustment processor. The discount timing processor is used to perform timing analysis on the circulation process of the bill, the big data comparison processor is used to compare the big data information with the analysis results, and the enterprise adjustment processor adjusts the discount data based on the enterprise credit.

3. The supply chain bill pricing anomaly warning system based on an artificial intelligence large model as claimed in claim 2 is characterized in that: The discount timing processor calculates the discount cost rate d of the i-th discount according to the following formula: i : Among them, P i is the i-th discount price, F is the face value of the bill, t i is the number of days remaining for the i-th discount; The discount time series processor calculates the average discount rate and discount standard deviation of the data including the current discount data and the data excluding the current discount data, and calculates the reasonable value S1 of the time series according to the following formula: Where d is the current discount rate, is the average discount rate including the current discount data, is the average discount rate excluding the current discount data, σ1 is the standard deviation of discount including the current discount data, and σ2 is the standard deviation of discount excluding the current discount data.

4. The supply chain bill pricing anomaly warning system based on an artificial intelligence large model as claimed in claim 3 is characterized by: The big data comparison processor obtains all discount data consistent with the current discount type and calculates the average discount rate and the discount standard deviation σ3, and calculate the historical reasonable value S2 according to the following formula:

5. The supply chain bill pricing anomaly warning system based on an artificial intelligence large model as claimed in claim 4 is characterized in that: The enterprise adjustment processor calculates the adjustment coefficient k according to the following formula: Among them, V is the credit score of the enterprise, and M is the total credit score.

6. The supply chain bill pricing anomaly warning system based on an artificial intelligence large model as claimed in claim 5 is characterized in that: The warning locking unit includes an analysis result receiving processor, a warning mapping processor, and a bill locking processor. The analysis result receiving processor is used to receive bill price analysis data, the warning mapping processor is used to map to a corresponding warning level, and the bill locking processor is used to add a warning tag to the corresponding bill for locking. The early warning mapping processor calculates the risk value Q of the current discount price according to the following formula: Among them, P0 is the current discount price and t0 is the remaining days of the current discount.

Images