Store sales statistical method based on pos machine

Abstract

The invention discloses a pos machine-based store sales statistical method, and relates to the technical field of financial auditing. A POS machine terminal, a cloud data processing module, a user side module and an agent side module are included, dynamic profit sharing is realized through an intelligent algorithm, the transaction amount is considered, and multi-dimensional factors such as transaction time and market fluctuation are combined, so that the profit sharing efficiency is improved. And the profit sharing strategy is more fair and reasonable. The dynamic adjustment mechanism can stimulate the agent to input more resources in a high-traffic period or a large-amount transaction, thereby promoting the improvement of the overall sales, providing a reliable data tracing and auditing approach for the merchant and the agent, optimizing the user experience and the operation convenience, and bringing significant commercial value for the merchant and the agent.

Description

Technical Field

[0001] The present invention relates to the technical field of financial auditing, and in particular to a store sales statistics method based on a POS machine. Background Art

[0002] Current store sales statistics rely primarily on manually exporting transaction data from traditional POS machines and manually aggregating it into Excel spreadsheets. Agent commission calculations are typically based on a fixed ratio, unable to dynamically adjust based on transaction size, time period, and other factors, resulting in insufficient agent incentives. Furthermore, regarding data security, transaction information is often transmitted in plain text, posing a risk of interception. Existing systems lack real-time performance, with commission results often delayed by 1-3 days. Furthermore, users are unable to view detailed information, leading to frequent disputes. Therefore, we propose a POS-based store sales statistics method. Summary of the Invention

[0003] The purpose of the present invention is to solve the problems of rigid profit-sharing model, inability to adapt to high-frequency and multi-dimensional transaction scenarios, lengthy data links, prone to human errors, and lack of intelligent risk control mechanisms. The present invention provides a store sales statistics method based on POS machines.

[0004] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0005] A method for calculating store sales based on a POS machine, comprising:

[0006] The POS terminal has a built-in transaction data collection module, data encryption module, and communication module. It is used to collect transaction data in real time, including transaction amount, time, merchant ID, and product type, and encrypts it using the AES-256 algorithm before uploading it to the cloud server.

[0007] The cloud data processing module receives encrypted data, decrypts it, stores it in a distributed database, and calculates the total transaction amount of the user end and the agent's profit sharing income through intelligent algorithms;

[0008] The user-side module provides a mobile APP visualization interface that displays merchant transaction amounts, transaction trend charts, daily / weekly / monthly statistics, and supports data export in CSV or PDF formats;

[0009] The agent-side module displays the transaction data of the merchants under its jurisdiction, including the commission ratio and commission amount, and supports real-time withdrawal of commission income and dynamic configuration of commission rules.

[0010] Furthermore, the operation of the data encryption module includes:

[0011] Input value: original transaction data in JSON format, including merchant_id, amount, and timestamp fields;

[0012] Encryption method: AES-256-CBC mode is used, and the key is dynamically generated using HMAC-SHA256 based on the POS machine hardware ID and the current date.

[0013] Output value: Encrypted ciphertext and initialization vector (IV), transmitted to the cloud via HTTPS protocol; ensure that man-in-the-middle attacks are prevented during data transmission, and the key is updated daily to avoid long-term leakage risks.

[0014] Furthermore, the intelligent algorithm is a dynamic profit-sharing model based on linear regression, specifically including:

[0015] Input values: transaction amount T (unit: RMB), preset profit sharing ratio R0, transaction time t (accurate to the hour for analyzing time period sensitivity), and historical transaction volatility coefficient σ calculated based on the standard deviation of transactions over the past 30 days.

[0016] Calculation formula: Dynamic profit sharing ratio R = R0 + β1*T + β2*t + β3*σ, where β1, β2, and β3 are regression coefficients trained using historical data;

[0017] Output value: dynamic profit sharing amount P = T * R;

[0018] Automatically adjust commissions based on transaction size and time, incentivizing agents to earn higher returns during high-traffic periods or large transactions.

[0019] Furthermore, the training method of the regression coefficients β1, β2, and β3 of the dynamic profit sharing model is:

[0020] Training data: historical transaction records, including T, t, σ, and R_actual;

[0021] Loss function: mean square error (MSE), i.e. Loss = Σ(R_predicted - R_actual)^2;

[0022] Optimizer: Adam algorithm is used, the learning rate is set to 0.001, and it is iterated 1000 times until convergence;

[0023] Optimize the profit-sharing model through continuous learning and adapt to market changes.

[0024] Furthermore, the intelligent algorithm also includes an abnormal transaction filtering step.

[0025] Input values: single transaction amount T, merchant's historical average daily transaction amount μ, merchant's industry average transaction amount I_avg;

[0026] Calculation method: If T > 3*(μ + I_avg), it is marked as an abnormal transaction and the manual review process is triggered.

[0027] Output value: A list of abnormal transactions and a confidence score (0 - 100 points). The scoring formula is Score = (T - μ) / I_avg * 100.

[0028] Prevent money laundering or false transactions and ensure the authenticity of profit sharing data.

[0029] Furthermore, each time the data is updated in the agent - side module, the cloud generates a global version number V. If the agent's local version V_local < V, the latest data is forcibly pulled to resolve version number conflicts. Only the changed amount, Delta data, is transmitted, reducing bandwidth consumption. This ensures that when the agent reconnects to the network after being offline, the un - settled profit sharing amount is automatically synchronized.

[0030] Furthermore, the interaction functions of the user - side module include displaying transaction distributions by hour / day dimensions, generating a scalable heat map layer using the D3.js library, viewing the details of profit sharing calculations by clicking on a single transaction, and pushing APP notifications and SMS reminders when the daily transaction amount is below the threshold.

[0031] Furthermore, the POS terminal reads membership card information through the NFC module, scans product QR codes with the camera, obtains the transaction geographical location using GPS, adopts the STM32F4 series processor, enters the sleep mode when there is no transaction, supports both 4G and Wi - Fi communications, and automatically switches to the optimal network to ensure the success rate of data transmission.

[0032] Furthermore, the method for dynamically configuring the profit sharing rules in the agent - side module is to customize the profit sharing strategy through a rule engine, set the trigger conditions for the profit sharing ratio, support inputting mathematical expressions and verifying the syntax legality, to meet the different differential sharing requirements of different agents.

[0033] Furthermore, the data export module supports API interfaces and blockchain evidence - storing functions. The API interface provides RESTful APIs for third - party systems to call. The blockchain evidence - storing function writes the daily transaction hash value into the Hyperledger Fabric chain to ensure the immutability of data, meet the merchant's financial audit requirements, and support agents to connect the profit sharing data to their own ERP systems.

[0034] The beneficial effects of the present invention are as follows:

[0035] 1. This invention uses the AES-256 algorithm to encrypt transaction data, decrypting and storing it in the cloud. This effectively prevents data leakage and tampering during transmission, providing solid data security for merchants and agents. This significantly improves the security and reliability of transaction data. Dynamic profit sharing is achieved through an intelligent algorithm that not only considers the transaction amount but also multiple factors such as transaction time and market fluctuations, making the profit sharing strategy more fair and reasonable. This dynamic adjustment mechanism incentivizes agents to invest more resources during high-traffic periods or for large transactions, thereby driving overall sales growth. The design of the user-side and agent-side modules fully considers user experience and operational convenience. The user-side provides a rich visual interface and data export functions to help merchants better understand their business status. The agent-side supports real-time viewing and profit withdrawal, as well as dynamic configuration of profit sharing rules, providing great convenience for agents. The introduction of blockchain evidence storage ensures the integrity and immutability of transaction data. Daily transaction hash values are written into the Hyperledger Fabric blockchain, providing merchants and agents with a reliable data traceability and auditing method, optimizing user experience and operational convenience, and bringing significant commercial value to merchants and agents.

[0036] 2. By automatically adjusting the profit sharing ratio based on the size of the transaction and the time period in which the transaction occurs, the present invention can effectively incentivize agents to obtain higher returns during high-traffic periods or when handling large transactions. This mechanism is designed to encourage agents to participate more actively in the transaction process during those critical periods, thereby increasing overall transaction activity. At the same time, this dynamic adjustment of the profit sharing ratio also helps agents to more reasonably balance the relationship between risk and return when facing transactions with different risk levels, ensuring that their returns match the risks they bear, thereby promoting the stable development of agents and the continued growth of their business as a whole.

[0037] 3. Through continuous learning and self-optimization, the present invention continuously improves and adjusts the profit-sharing model. This continuous learning process enables the model to adapt to the ever-changing and evolving market, ensuring it stays current with the latest trends and dynamics. This ensures that the profit-sharing model maintains its predictive accuracy regardless of changing market conditions, ensuring efficient profit-sharing operations in a variety of market environments. This continuous optimization and adjustment of the model is key to maintaining competitiveness and profitability in a highly competitive market. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below.

[0039] The present invention provides a store sales statistics method based on a POS machine, comprising:

[0040] The POS terminal has a built-in transaction data collection module, data encryption module, and communication module. It is used to collect transaction data in real time, including transaction amount, time, merchant ID, and product type, and encrypts it using the AES-256 algorithm before uploading it to the cloud server.

[0041] The cloud data processing module receives encrypted data, decrypts it, stores it in a distributed database such as MongoDB, and calculates the total transaction amount of the user end and the agent's profit sharing income through intelligent algorithms;

[0042] The user-side module provides a mobile APP visualization interface that displays merchant transaction amounts, transaction trend charts, daily / weekly / monthly statistics, and supports data export in CSV or PDF formats;

[0043] The agent-side module displays the transaction data of the merchants under its jurisdiction, including the commission ratio and commission amount, and supports real-time withdrawal of commission income and dynamic configuration of commission rules.

[0044] In this embodiment, preferably, the operations of the data encryption module include:

[0045] The input value is the original transaction data in JSON format, which includes the following fields: merchant_id (merchant ID), amount (transaction amount), and timestamp (timestamp);

[0046] The encryption method uses AES-256-CBC mode, an advanced symmetric encryption algorithm that provides strong security. To further enhance security, the key is not statically generated, but dynamically generated using the HMAC-SHA256 algorithm based on the POS device hardware ID and the current date, ensuring that the key used for each transaction is unique.

[0047] The output value is the encrypted ciphertext and the initialization vector (IV), both of which are important components of the encryption process. The encrypted data is transmitted to the cloud via the HTTPS protocol. The HTTPS protocol can provide an encrypted channel to ensure the security of data during transmission and prevent man-in-the-middle attacks.

[0048] To ensure the security of data transmission, the key is updated daily. This can avoid the potential risk of leakage caused by the key remaining unchanged for a long time, and further enhance the security of the data transmission process.

[0049] In this embodiment, preferably, the intelligent algorithm is a dynamic profit-sharing model based on linear regression.

[0050] Inputs include: transaction amount T (in RMB); the pre-set commission ratio R0; and transaction time t, where t must be accurate to the hourly level. These inputs are used to analyze timeframe sensitivity. Additionally, based on the past 30 days of trading data, we calculate a historical trading volatility coefficient σ, which reflects trading volatility.

[0051] The dynamic profit sharing ratio is calculated as follows: R = R0 + β1*T + β2*t + β3*σ. In this formula, β1, β2, and β3 are regression coefficients derived from historical data training. They correspond to the weights of the transaction amount, transaction time, and historical transaction volatility, respectively.

[0052] The output value is the dynamic profit sharing amount P, which is equal to the transaction amount T multiplied by the dynamic profit sharing ratio R. In this way, the profit sharing amount will automatically adjust according to the changes in transaction amount and time.

[0053] By automatically adjusting the commission ratio based on transaction size and time of day, agents are effectively incentivized to earn higher returns during high-traffic periods or when handling large transactions. This mechanism is designed to encourage agents to actively participate in transactions during these critical periods, thereby increasing overall trading activity. Furthermore, this dynamic adjustment of the commission ratio helps agents more effectively balance risk and reward when dealing with transactions of varying risk levels, ensuring their returns align with the risks they assume, thereby promoting the stable development of agents and the continued growth of their business.

[0054] In this embodiment, preferably, the training method of the regression coefficients β1, β2, and β3 of the dynamic profit sharing model is:

[0055] Training data: Use historical transaction records as training data. These records contain detailed information about time T, transaction volume t, price fluctuation σ, and actual return R_actual.

[0056] Loss function: The mean square error (MSE) is selected as the loss function, and the prediction accuracy of the model is measured by calculating the sum of the squares of the difference between the predicted return R_predicted and the actual return R_actual, that is,

[0057] Loss=Σ(R_predicted-R_actual)^2;

[0058] Optimizer: During model training, we used the Adam algorithm, a widely used adaptive learning rate optimization algorithm, as the optimizer. We set the learning rate to 0.001 and iterated the model 1,000 times until the model converged.

[0059] Through continuous learning and self-optimization, the profit-sharing model can be continuously improved and adjusted. This ongoing learning process enables the model to adapt to the constant changes and evolution of the market, ensuring it stays current with the latest trends and dynamics. This ensures that the profit-sharing model maintains its predictive accuracy regardless of changing market conditions, ensuring efficient profit-sharing operations in a variety of market environments. This continuous optimization and adjustment of the model is key to maintaining competitiveness and profitability in a highly competitive market.

[0060] In this embodiment, preferably, the intelligent algorithm further includes an abnormal transaction filtering step.

[0061] Input values include: single transaction amount T, merchant's historical average daily transaction amount μ, and the average transaction amount of the merchant's industry I_avg;

[0062] The specific calculation method is: if the single transaction amount T is greater than the sum of three times the merchant's historical average daily transaction amount μ and the average transaction amount of the merchant's industry I_avg, that is, T>3*(μ+I_avg), then the transaction will be marked as an abnormal transaction and the manual review process will be triggered;

[0063] The output will include a list of abnormal transactions and a confidence score for each abnormal transaction, ranging from 0 to 100. The score is calculated as Score = (T-μ) / I_avg*100, where T represents the single transaction amount, μ represents the merchant's historical average daily transaction amount, and I_avg represents the average transaction amount of the merchant's industry.

[0064] The purpose of this mechanism is to prevent money laundering or false transactions, ensure the authenticity of profit-sharing data, and thus maintain the health and stability of the entire financial transaction system.

[0065] In this embodiment, preferably, in this system, whenever the agent-side module performs a data update operation, the cloud server will generate a new global version number V. Once it is detected that the version number V_local stored locally by the agent is smaller than the version number V on the cloud, the system will automatically perform a mandatory operation, namely, to pull the latest data from the cloud. This mechanism effectively solves the problem of version number conflicts. In order to further optimize performance and reduce network bandwidth consumption, the system only transmits data that has changed since the last synchronization, which is the so-called Delta data. In this way, when the agent completes a period of work offline, once the network is reconnected, the system will be able to automatically synchronize all unsettled commission amounts to ensure data integrity and consistency.

[0066] In this embodiment, preferably, a rich set of interactive functions are provided in the user-side module to help users better understand their transaction data. These functions include the ability to display the distribution of transactions by hour or day, and use the powerful D3.js library to generate a scalable heat map, so that users can intuitively see the hot spots of the data. In addition, users can view the profit calculation details of any transaction by clicking on the heat map. These details include key data such as the basic ratio R0 and the dynamic adjustment item β1*T, so that users can clearly understand the profitability of each transaction. In order to ensure that users do not miss any important transaction information, an intelligent reminder system is also designed. When the daily transaction volume is lower than the set threshold, such as lower than 50% of the average transaction volume in the past 7 days, the system will automatically push APP notifications and SMS reminders to users to ensure that users can take appropriate measures in time.

[0067] In the present embodiment, preferably, in the present invention, the POS terminal device can read the information in the membership card through the integrated NFC module, thereby realizing fast and accurate member identification. In addition, the terminal is also equipped with a camera that can scan the QR code on the product, thereby quickly obtaining product information. In order to record the specific location of the transaction, the POS terminal is also built-in with a GPS module, which can accurately obtain the geographical location information of the transaction. In terms of hardware, the POS terminal adopts a powerful STM32F4 series processor to ensure the efficiency and stability of the equipment operation. In order to reduce energy consumption, when the POS terminal is in a non-transaction state, it will automatically enter sleep mode, and the power consumption at this time is less than 0.1W, which greatly saves energy. In addition, the POS terminal also supports two communication modes, 4G and Wi-Fi, and can automatically switch to the optimal network connection according to the current network conditions, ensuring a high success rate and stability of data transmission.

[0068] In this embodiment, preferably, in the present invention, the proposed method for dynamic configuration of the agent-side module profit-sharing rules involves the use of a rule engine to achieve customization of the profit-sharing strategy. In this way, the profit-sharing ratio and trigger conditions can be flexibly set. For example, a condition can be set as "When the merchant's monthly transaction volume exceeds 100,000 yuan, the profit-sharing ratio will increase by 2%." In addition, the method supports users to enter mathematical expressions and verify the grammatical validity of these expressions. For example, the expression R=R0+log(T) / 10 can be used to calculate the profit-sharing value. Such a design can meet the needs of different agents to implement differentiated profit-sharing strategies based on their business characteristics and needs.

[0069] In this embodiment, preferably, the data export module involved in the present invention has an API interface and a blockchain evidence storage function, which are provided to the third-party system in the form of a RESTful API for calling. For example, data can be obtained by accessing a URL such as / export? format=csv&date=2023-01-01 through a GET request. In addition, the module also includes a blockchain evidence storage function, which can record and write daily transaction hash values into the Hyperledger Fabric blockchain, thereby ensuring the non-tamperability of the data. Such a design meets the needs of merchants in financial auditing and ensures the authenticity and integrity of the data. At the same time, the module also supports agents to connect profit-sharing data to their own ERP system, making data management more flexible and efficient.

[0070] The working principle and use process of the present invention: When the system is used,

[0071] The POS terminal captures transaction details in real time, including the transaction amount, time, merchant ID, and item type. This data is then encrypted using the AES-256 Advanced Encryption Standard to ensure security during transmission and prevent unauthorized access. The encrypted data, along with an initialization vector (IV), is then sent to a cloud server via secure HTTPS.

[0072] In the cloud, the data processing module first decrypts this encrypted data and then stores it in an efficient distributed database. Using a carefully designed intelligent algorithm, this module rapidly calculates the total transaction amount on the client side and the commission income that the agent deserves. This intelligent algorithm not only considers the transaction amount but also the timing of the transaction and historical market fluctuations, dynamically adjusting the commission ratio to ensure a fair and efficient profit-sharing strategy.

[0073] On the user side, a user-friendly mobile app provides a rich visual interface. Merchants can easily view their transaction amounts and transaction trend charts. These charts can be displayed on a daily, weekly, or monthly basis to help merchants better understand their business performance. In addition, users can also export data to CSV or PDF formats for further analysis and archiving.

[0074] For agents, the agent-side module provides a comprehensive overview of transaction data, including key information such as commission ratios and commission amounts. Agents can view and withdraw their commission earnings in real time, and can also utilize dynamic configuration features to flexibly adjust commission rules based on market changes. This high level of flexibility ensures that agent interests are aligned with market demand.

[0075] To ensure data integrity and immutability, the data export module not only provides a RESTful API for third-party systems but also incorporates blockchain-based evidence storage. Daily transaction hashes are written to the Hyperledger Fabric blockchain, meeting merchants' financial audit requirements while providing agents with a convenient way to seamlessly integrate profit sharing data into their own ERP systems.

[0076] By encrypting transaction data using the AES-256 algorithm and decrypting and storing it in the cloud, this effectively prevents data leakage and tampering during transmission, providing solid data security for merchants and agents. This significantly improves the security and reliability of transaction data. Dynamic profit sharing is achieved through an intelligent algorithm that not only considers transaction amount but also multiple factors such as transaction time and market volatility, resulting in a more fair and reasonable profit sharing strategy. This dynamic adjustment mechanism incentivizes agents to invest more resources during high-traffic periods or for large-value transactions, thereby boosting overall sales. Both the client and agent modules are designed with user experience and operational convenience in mind. The client offers a rich visual interface and data export functionality, helping merchants better understand their business performance. The agent side supports real-time viewing and withdrawal of profit sharing, as well as dynamic configuration of profit sharing rules, providing significant convenience for agents. The integration of blockchain evidence storage ensures the integrity and immutability of transaction data. The daily transaction hash values are written into the Hyperledger Fabric blockchain, providing merchants and agents with a reliable data traceability and auditing method, optimizing user experience and operational convenience, and bringing significant commercial value to merchants and agents.

[0077] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A store sales statistics method based on POS machine, characterized in that: include: The POS terminal has a built-in transaction data collection module, data encryption module, and communication module. It is used to collect transaction data in real time, including transaction amount, time, merchant ID, and product type, and encrypts it using the AES-256 algorithm before uploading it to the cloud server. The cloud data processing module receives encrypted data, decrypts it, stores it in a distributed database, and calculates the total transaction amount of the user end and the agent's profit sharing income through intelligent algorithms; The user-side module provides a mobile APP visualization interface that displays merchant transaction amounts, transaction trend charts, daily / weekly / monthly statistics, and supports data export in CSV or PDF formats; The agent-side module displays the transaction data of the merchants under its jurisdiction, including the commission ratio and commission amount, and supports real-time withdrawal of commission income and dynamic configuration of commission rules.

2. A method for calculating store sales based on a POS machine according to claim 1, characterized in that: The operations of the data encryption module include: Input value: original transaction data in JSON format, including merchant_id, amount, and timestamp fields; Encryption method: AES-256-CBC mode is used, and the key is dynamically generated using HMAC-SHA256 based on the POS machine hardware ID and the current date. Output value: encrypted ciphertext and initialization vector (IV), transmitted to the cloud via HTTPS protocol; Ensure that man-in-the-middle attacks are prevented during data transmission, and keys are updated daily to avoid long-term leakage risks.

3. A method for calculating store sales based on POS machines according to claim 1, characterized in that: The intelligent algorithm is a dynamic profit-sharing model based on linear regression, which specifically includes: Input values: transaction amount T (unit: RMB), preset profit sharing ratio R0, transaction time t (accurate to the hour for analyzing time period sensitivity), and historical transaction volatility coefficient σ calculated based on the standard deviation of transactions over the past 30 days. Calculation formula: Dynamic profit sharing ratio R = R0 + β1*T + β2*t + β3*σ, where β1, β2, and β3 are regression coefficients trained using historical data; Output value: dynamic profit sharing amount P = T * R; Automatically adjust commissions based on transaction size and time, incentivizing agents to earn higher returns during high-traffic periods or large transactions.

4. The method for calculating store sales based on a POS machine according to claim 3, characterized in that: The training method of the regression coefficients β1, β2, and β3 of the dynamic profit sharing model is as follows: Training data: historical transaction records, including T, t, σ, and R_actual; Loss function: mean square error (MSE), i.e. Loss = Σ(R_predicted - R_actual)^2; Optimizer: Adam algorithm is used, the learning rate is set to 0.001, and it is iterated 1000 times until convergence; Optimize the profit-sharing model through continuous learning and adapt to market changes.

5. The method for calculating store sales based on a POS machine according to claim 3, characterized in that: The intelligent algorithm also includes an abnormal transaction filtering step. Input values: single transaction amount T, merchant's historical average daily transaction amount μ, merchant's industry average transaction amount I_avg; Calculation method: If T>3*(μ+I_avg), it is marked as an abnormal transaction and triggers the manual review process; Output value: abnormal transaction list and confidence score (0-100 points), the scoring formula is Score = (T- μ) / I_avg*100; Prevent money laundering or false transactions and ensure the authenticity of profit-sharing data.

6. The method for calculating store sales based on a POS machine according to claim 1, characterized in that: Each time the data is updated on the agent side module, the cloud generates a global version number V. If the local version V_local of the agent is less than V, the latest data is forcibly pulled to resolve version number conflicts. Only the changed amount, Delta data, is transmitted, reducing bandwidth consumption. Ensure that when the agent reconnects after going offline, the un-settled profit sharing amount is automatically synchronized.

7. The method for calculating store sales based on a POS machine according to claim 1, characterized in that: The interaction functions of the user side module include displaying transaction distributions by hour / day dimension, generating a scalable heat map layer using the D3.js library, viewing the details of profit sharing calculations by clicking on a single transaction, and pushing APP notifications and SMS reminders when the daily transaction amount is below the threshold.

8. The method for calculating store sales based on a POS machine according to claim 1, characterized in that: The POS terminal reads membership card information through the NFC module, scans product QR codes with the camera, obtains the transaction geographical location using GPS, adopts an STM32F4 series processor, enters the sleep mode when there is no transaction, and supports both 4G and Wi-Fi communications, automatically switching to the optimal network to ensure the success rate of data transmission.

9. The method for calculating store sales based on a POS machine according to claim 1, characterized in that: The method for dynamically configuring the profit sharing rules of the agent side module is to customize the profit sharing strategy through a rule engine, set the trigger conditions for the profit sharing ratio, support inputting mathematical expressions and verifying the syntax legality, to meet the differentiated sharing requirements of different agents.

10. The method for calculating store sales based on a POS machine according to claim 1, characterized in that: The data export module supports API interfaces and blockchain deposit and verification functions. The API interface provides RESTful APIs for third-party systems to call. The blockchain deposit and verification function writes the daily transaction hash value into the Hyperledger Fabric chain to ensure the immutability of data, meet the merchant's financial audit requirements, and support agents to connect the profit sharing data to their own ERP systems.