Product distribution application software system

Abstract

The invention discloses a product distribution application software system, which relates to the field of product distribution management systems and comprises a data acquisition and processing module for acquiring and preprocessing data related to distribution, a feature extraction module for performing feature extraction and integration on the preprocessed data to form comprehensive sales features, a market demand prediction module, a product distribution management module and a product distribution management module. The system comprises a pre-processing module, a market demand prediction model establishing module based on the pre-processed data, a market demand prediction result generating module, a strategy adjustment module, automatic distribution data adjustment according to the market demand prediction result, and a block chain recording module for recording the adjusted distribution data on a block chain. According to the method, the market demand model is established, the stable trend of historical data and the dynamic change of the market demand are captured at the same time, and finally a more comprehensive and accurate market demand prediction result is generated.

Description

Technical Field

[0001] The present invention relates to the field of product distribution management systems, in particular to a product distribution application software system. Background Art

[0002] With the popularity of e-commerce platforms, product distribution systems are becoming increasingly complex. In order to improve distribution efficiency, optimize inventory management, and predict market demand, companies are increasingly demanding distribution management systems. Traditional distribution management systems usually rely on manual operations or semi-automatic data processing methods, and their functions are relatively limited. In recent years, distribution management systems based on big data, artificial intelligence, and the Internet of Things have gradually emerged, improving the intelligence level of the distribution process through dynamic data analysis, sales forecasting algorithms, and other means.

[0003] The shortcomings of traditional technology are mainly reflected in the lack of in-depth analysis of historical data and full consideration of external factors when predicting market demand. Static forecasting models cannot accurately reflect the dynamic changes in market demand. In addition, the security and transparency of distribution data are still pain points in the industry. Traditional systems have the risk of tampering in data recording and transaction processing, making it difficult to ensure the credibility and integrity of data. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a product distribution application software system to solve the problem that when predicting market demand, there is a lack of in-depth analysis of historical data and sufficient consideration of external factors, and the static prediction model cannot accurately reflect the dynamic changes of market demand.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] The present invention provides a product distribution application software system, which comprises:

[0008] Data collection and processing module, collects distribution-related data and performs pre-processing;

[0009] The feature extraction module extracts and integrates the features of the preprocessed data to form comprehensive sales features;

[0010] The market demand forecasting module establishes a market demand forecasting model based on the preprocessed data and generates market demand forecasting results;

[0011] Strategy adjustment module, automatically adjusts distribution data according to market demand forecast results;

[0012] The blockchain recording module records the adjusted distribution data on the blockchain;

[0013] Smart contract generation module, which automatically generates smart contracts based on distribution data on the blockchain;

[0014] Execution module: When the triggering conditions of the smart contract are met, the contract is automatically executed to complete the profit sharing settlement;

[0015] The distribution monitoring module monitors the key indicators of each distribution channel in real time. When an abnormality is detected in a distribution channel, the intelligent early warning mechanism is triggered to generate abnormality detection results.

[0016] The intelligent early warning module automatically adjusts the distribution data according to the abnormal detection results in the intelligent early warning and records it in the blockchain.

[0017] As a preferred solution of the product distribution application software system of the present invention, the data related to distribution is collected and pre-processed, which specifically includes the following steps:

[0018] Obtain historical sales data, market dynamics data, consumer behavior data, and external factors data through APIs;

[0019] Clean, process outliers and standardize the collected data;

[0020] The processed data is organized into a structured format and stored.

[0021] As a preferred solution of the product distribution application software system of the present invention, the following steps are specifically included: extracting and integrating the features of the pre-processed data to form comprehensive sales features:

[0022] Automatically extract temporal features in time series from preprocessed data;

[0023] Based on historical sales data, historical lagged features are generated through lag operations;

[0024] Organize the external factor data into external factor features;

[0025] Interact consumer behavior data with historical sales data to generate interactive features;

[0026] Integrate time characteristics, historical lag characteristics, external factor characteristics and interaction characteristics to form comprehensive sales characteristics.

[0027] As a preferred solution of the product distribution application software system of the present invention, wherein: based on the pre-processed data, a market demand forecasting model is established to generate a market demand forecasting result, which specifically includes the following steps:

[0028] A time step is set based on the comprehensive sales characteristics, and an exponential smoothing model is used to perform exponential smoothing on the comprehensive sales characteristics with the set time step to obtain smoothed lag data;

[0029] Use the weighted summation method to aggregate the smoothed lagged data to obtain the static market demand forecast value;

[0030] Introduce dynamic change rate analysis to calculate the change rate of comprehensive sales characteristics, and apply integral technology to accumulate the change rate of comprehensive sales characteristics to obtain dynamic market demand forecast values;

[0031] Combining the static market demand forecast value with the dynamic market demand forecast value to form a complete market demand forecast value, the expression is:

[0032]

[0033] Among them, Y t (r,c) represents the demand forecast value of region r and channel c at time t, M represents the number of lagged time steps, i represents the index, and X t-i represents the comprehensive sales characteristics of the past i-th time step, λ i represents the attenuation coefficient, t represents the current time, γ i represents the weighting coefficient, d t represents the integration over time t, represents partial derivative.

[0034] As a preferred solution of the product distribution application software system of the present invention, the distribution data is automatically adjusted according to the market demand forecast result, which specifically includes the following steps:

[0035] The market demand forecast values ​​for each region and each channel are counted to determine the inventory demand, and the expression is:

[0036] S 需求 (r,c,t)=Y t (r,c)+α×σ t (r,c);

[0037] Among them, S 需求 (r,c,t) represents the inventory demand of region r and channel c at the current time t, t represents the current time, σ t (r,c) represents the standard deviation of demand forecast, and α represents the confidence interval coefficient;

[0038] Adjust inventory allocation for each region and channel based on inventory needs;

[0039] According to the market demand forecast value, the market demand growth rate of each region and each channel is calculated, and the expression is:

[0040]

[0041] Among them, G t(r,c) represents the market demand growth rate of region r and channel c at the current time t, t-1 represents the previous time point, Y t-1 (r,c) represents the market demand growth rate of region r and channel c at the previous time point;

[0042] The price is adjusted according to the market demand growth rate, and the expression is:

[0043] P 调整 (r,c,t)=P 当前 (r,c)×(1+β×G t (r,c));

[0044] Among them, P 调整 (r,c,t) represents the adjusted price of region r and channel c at the current time t, P 当前 (r,c) represents the current price in region r and channel c at the current time t, and β represents the price adjustment coefficient.

[0045] As a preferred solution of the product distribution application software system of the present invention, the adjusted distribution data is recorded on the blockchain, which specifically includes the following steps:

[0046] Structuring and organizing the distribution data and packaging it into transaction data packages;

[0047] Use asymmetric encryption technology to encrypt transaction data packets and broadcast them to the blockchain network as transactions;

[0048] The nodes in the blockchain network receive the transaction and conduct verification and consensus;

[0049] Once consensus is reached, the transaction data will be written into the next block of the blockchain.

[0050] As a preferred solution of the product distribution application software system of the present invention, wherein: automatically generating a smart contract based on the distribution data on the blockchain specifically includes the following steps:

[0051] Access and extract distribution data stored on the blockchain;

[0052] Based on the distribution scenario, select a smart contract template, which includes an inventory adjustment contract, a price adjustment contract, a promotion strategy contract, and a profit sharing settlement contract;

[0053] The distribution data extracted from the blockchain is automatically filled into the smart contract template to generate the corresponding contract terms;

[0054] Digitally sign the generated smart contract;

[0055] The signed smart contract is broadcast as a transaction to the blockchain network. The nodes in the blockchain network will verify the contract and package it into the next block of the blockchain through a consensus mechanism.

[0056] As a preferred solution of the product distribution application software system of the present invention, when the triggering conditions of the smart contract are met, the contract is automatically executed to complete the profit sharing settlement, which specifically includes the following steps:

[0057] Real-time monitoring of distribution channels. When the conditions that trigger the contract terms are met, the smart contract automatically enters the execution state and calculates the specific profit sharing amount for each participant.

[0058] Based on the specific profit amount, the allocation of funds is executed directly on the blockchain.

[0059] As a preferred solution of the product distribution application software system of the present invention, the key indicators of each distribution channel are monitored in real time. When an abnormality in the distribution channel is detected, the intelligent early warning mechanism is triggered to generate an abnormality detection result. Specifically, the following steps are included:

[0060] Set safety thresholds based on inventory levels and sales;

[0061] Real-time monitoring of inventory levels and sales in each distribution channel. When the inventory levels and sales drop below the set safety stock threshold, an abnormal warning will be issued.

[0062] Generate anomaly detection reports based on anomaly warnings.

[0063] As a preferred solution of the product distribution application software system of the present invention, the distribution data is automatically adjusted according to the abnormal detection results in the intelligent early warning, and recorded in the blockchain at the same time, which specifically includes the following steps:

[0064] Based on the anomaly detection results generated in the intelligent early warning, the distribution data is automatically adjusted according to the cause of the anomaly;

[0065] After making adjustments, continuously monitor the adjusted data;

[0066] All adjustment operations and adjustment results are recorded on the blockchain.

[0067] The beneficial effects of the present invention are as follows: by establishing a market demand model that simultaneously captures the steady trend of historical data and the dynamic changes in market demand, the static prediction results enhance the sensitivity to historical data, making the prediction results more responsive to past market behaviors; and the dynamic prediction results can make more agile predictions of future market changes, ultimately generating more comprehensive and accurate market demand prediction results; in addition, recording distribution data on the blockchain enhances data security and transparency, greatly improving the credibility of data management. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0069] Figure 1 This is a schematic diagram of the product distribution application software system in Example 1.

[0070] Figure 2 This is a schematic diagram of the market demand forecasting model in Example 1. DETAILED DESCRIPTION

[0071] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0072] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0073] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0074] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a product distribution application software system, comprising the following steps:

[0075] Data collection and processing module, collects distribution-related data and performs pre-processing;

[0076] Obtain historical sales data, market dynamics data, consumer behavior data, and external factors data through APIs;

[0077] Historical sales data refers to historical sales records from ERP, covering sales volume, sales amount, payment collection and other information for each time period, each channel and each region. Historical sales data can reflect the sales trend of products in different time and regions.

[0078] Market dynamics data refers to information such as market competition, industry dynamics, market share, and changes in consumer preferences.

[0079] Consumer behavior data reflects consumers’ purchasing preferences, browsing habits, purchase frequency, return rates, etc.

[0080] External factor data refers to macroeconomic data (such as GDP, unemployment rate, inflation rate, etc.), weather data, holiday information, policy changes, etc.

[0081] Eliminate invalid or incomplete data, process outliers in the data (such as sudden increase or decrease in sales, abnormal prices, etc.), and standardize the data so that it can be compared and analyzed on a unified scale during subsequent modeling.

[0082] Organize data from different sources and dimensions in a unified format and store them in an efficient database for subsequent quick access and calculation.

[0083] The feature extraction module extracts and integrates the features of the preprocessed data to form comprehensive sales features;

[0084] Extract temporal features from timestamps.

[0085] Time characteristics include seasonal characteristics, weekly characteristics, hourly characteristics and holiday characteristics.

[0086] It can be further explained that by extracting time features of different granularities, the cyclical changes in market demand can be captured.

[0087] Through the lag operation, historical lag features are generated, which reflect the impact of past sales on current market demand.

[0088] For example, sales volume in the past 1 day, 7 days, 30 days, or average sales volume in the past 7 days;

[0089] It can be further explained that by introducing historical lag features, the impact of historical sales on current market demand can be captured, especially sales fluctuations in the short term.

[0090] External factors have a significant impact on market demand, and are collected and organized as external factor features. For example, weather, promotional activities, and consumer price index.

[0091] It is further explained that the introduction of external factor characteristics can take into account the impact of the external environment on market demand and enhance the comprehensiveness and accuracy of the forecasting model.

[0092] Collect the interaction between consumer behavior data and historical sales data, such as browsing, clicking, and adding to shopping cart, and combine consumer behavior data with historical sales data to generate interaction features.

[0093] It is further explained that by generating interactive features, it is possible to deeply analyze the relationship between consumer behavior and sales data, capture the potential patterns in consumer purchasing decisions, and help companies identify which products have potential market demand and which marketing strategies (such as promotions, price adjustments, etc.) may increase conversion rates.

[0094] After extracting time features, historical lag features, external factor features, and interaction features, these features are integrated to form comprehensive sales features. Comprehensive sales features are a multi-dimensional, multi-level feature set that includes various factors related to market demand.

[0095] It is further explained that by integrating the features, a more comprehensive market demand forecast input is formed. This fusion of multi-dimensional data improves the model's ability to perceive changes in market demand, making the forecast results more accurate and reliable.

[0096] The market demand forecasting module establishes a market demand forecasting model based on the preprocessed data and generates market demand forecasting results;

[0097] The time step setting is a key step in market demand forecasting, which determines the time window for analyzing past sales data. For example, the system can set different time steps such as 1 day, 7 days, and 30 days to reflect short-term, medium-term, and long-term market demand fluctuations respectively.

[0098] After completing the design of the time step, the exponential smoothing model is used to smooth the comprehensive sales characteristics.

[0099] Exponential smoothing model is a commonly used time series processing method, which better captures the trend of current market demand by giving greater weight to recent data and smaller weight to more distant data.

[0100] Exponential smoothing introduces a decay coefficient so that the weight of the data gradually decays as the time step increases. Therefore, the most recent sales data has a greater impact on the current demand forecast, while older data has a smaller impact.

[0101] The smoothed lagged data reflects the impact of the comprehensive sales characteristics of different time steps on the current market demand. In order to obtain the static market demand forecast value, the smoothed lagged data is weighted and summed. The weighted summation method calculates the contribution of each time step to the current market demand by assigning different weights to the data of different time steps, and obtains the static market demand forecast value.

[0102] By introducing dynamic change rate analysis to calculate the change rate of comprehensive sales characteristics, the fluctuation of market demand can be dynamically monitored. At the same time, the integration operation is used to accumulate the change trend of comprehensive sales characteristics, and finally the dynamic market demand forecast value is obtained;

[0103] A complete market demand forecast needs to consider both static and dynamic factors. Therefore, the static market demand forecast value and the dynamic market demand forecast value are combined to generate a more accurate and comprehensive market demand forecast result. The expression is:

[0104]

[0105] Among them, Y t (r,c) represents the market demand forecast value of region r and channel c at time t, M represents the number of lagged time steps, i represents the index, and X t-i represents the comprehensive sales characteristics of the past i-th time step, λ i represents the attenuation coefficient, t represents the current time, γ i represents the weighting coefficient, dt represents the integration over time t, represents partial derivative.

[0106] Strategy adjustment module, automatically adjusts distribution data according to market demand forecast results;

[0107] The inventory demand for each region and channel is calculated using the market demand forecast value and the standard deviation of the demand forecast. The expression is:

[0108] S 需求 (r,c,t)=Y t (r,c)+α×σ t (r,c);

[0109] Among them, S 需求 (r,c,t) represents the inventory demand of region r and channel c at the current time t, t represents the current time, σ t (r,c) represents the standard deviation of the demand forecast, which is used to reflect the uncertainty of the forecast. α represents the confidence interval coefficient, which represents the enterprise's requirement for inventory security. A larger α value means that the enterprise hopes to have a higher inventory to cope with demand fluctuations.

[0110] For example, for an electronic product in the online channel in Beijing, the market demand forecast value = 1000, the standard deviation = 50, and the confidence interval coefficient α set by the enterprise = 1.96 (indicating a 95% confidence level). At this time, the inventory demand of the online channel in Beijing is 1000 + 1.96 × 50 = 1098

[0111] In this case, it is recommended that online channels in the Beijing area maintain at least 1,098 units of inventory to cope with future demand fluctuations.

[0112] Once the inventory requirements for each region and channel are determined, inventory allocation will be automatically adjusted based on the demand in each region. By analyzing the market demand forecast and inventory requirements for each region, more inventory will be allocated to regions and channels with high demand, while inventory allocation will be reduced for regions and channels with relatively low demand.

[0113] By calculating the market demand growth rate, companies can understand the market demand trends in each region and channel. The formula for calculating the market demand growth rate based on the market demand forecast is:

[0114]

[0115] Among them, G t (r,c) represents the market demand growth rate of region r and channel c at the current time t, t-1 represents the previous time point, Y t-1 (r,c) represents the market demand growth rate of region r and channel c at the previous point in time.

[0116] For example, in the offline channels in Shanghai, the current market demand forecast value of a certain electronic product = 1200, and the market demand forecast value at the previous time point = 1000, then the market demand growth rate of the offline channels in Shanghai at this time is 0.2 (ie 20%).

[0117] This shows that in Shanghai's offline channels, market demand has increased by 20%. Companies can quickly adjust their inventory and pricing strategies based on this trend.

[0118] By combining the market demand growth rate, the product price of each region and channel is automatically adjusted to maximize profits or promote sales. The formula for price adjustment is:

[0119] P 调整 (r,c,t)=P 当前 (r,c)×(1+β×G t (r,c));

[0120] Among them, P 调整 (r,c,t) represents the adjusted price of region r and channel c at the current time t, P 当前(r,c) represents the current price in region r and channel c at the current time t, and β represents the price adjustment coefficient, which determines the intensity of the impact of the market demand growth rate on the price.

[0121] For example, in the online channels in Guangzhou, the current price of a smartphone is RMB 2,000. Through market demand forecasting, it is calculated that the market demand growth rate of the online channels in Guangzhou is 15%, and the price adjustment coefficient β = 0.5, so the adjusted price is 2,000 × (1 + 0.5 × 0.15) = 2,150.

[0122] It is further explained that by combining the market demand forecast value and the standard deviation of demand fluctuations, it is possible to ensure that the inventory level in each region and channel is reasonable, and to optimize inventory allocation and improve supply chain efficiency based on the demand forecast of different regions and channels. In addition, by calculating the demand growth rate, the changing trend of market demand is dynamically monitored to implement a dynamic pricing strategy.

[0123] The blockchain recording module records the adjusted distribution data on the blockchain;

[0124] After the adjusted distribution data is generated, the distribution data fields are defined to determine the distribution data that needs to be recorded in the blockchain.

[0125] For example, region indicates the geographical area to which the distribution data belongs. Channel indicates the sales channel to which the distribution data belongs (online, offline, third-party platform, etc.). Inventory quantity indicates the adjusted inventory data, and so on.

[0126] The defined fields are organized into structured data objects and packaged into a transaction data packet, ready to be transmitted in the blockchain network.

[0127] Each node or user (such as distribution) participating in the blockchain network generates a pair of public and private keys. The public key is used for encryption and the private key is used for decryption.

[0128] Before broadcasting the transaction, the distributor will encrypt the transaction data packet using the recipient's public key. This ensures that only the recipient holding the corresponding private key (such as a node in a blockchain network) can decrypt the transaction data packet. And to ensure the authenticity of the data, the sender will also sign the transaction data packet with his or her own private key. Anyone can verify the signature with the sender's public key to ensure that the data packet was indeed sent by the sender and has not been tampered with.

[0129] The encrypted data packet will be broadcast as a transaction to all nodes in the blockchain network through the peer-to-peer network.

[0130] The node uses the sender's public key to verify the digital signature of the transaction data packet to ensure that the transaction does come from the legitimate sender and that the transaction content has not been tampered with.

[0131] Once the verification is passed, the blockchain network will reach consensus on the transaction through a consensus algorithm.

[0132] The role of the consensus mechanism is to ensure that all nodes in the network reach a consensus on the authenticity of the transaction and prevent malicious nodes from doing evil in the network.

[0133] For example, node A receives a transaction and verifies the signature using the sender’s public key to confirm that the transaction is legitimate.

[0134] Node A passes the transaction to other nodes, and multiple nodes verify it.

[0135] When enough (such as a majority) nodes confirm that the transaction is legitimate, consensus is reached and the transaction is considered valid.

[0136] Once consensus is reached, the transaction will be recorded in the next block of the blockchain.

[0137] It is further explained that recording the adjusted distribution data on the blockchain can significantly improve the transparency, security and trust of supply chain management.

[0138] Smart contract generation module, which automatically generates smart contracts based on distribution data on the blockchain;

[0139] Extract distribution data by connecting to nodes in the blockchain network.

[0140] According to different distribution scenarios, select the corresponding smart contract template. Smart contracts are automated protocols that can automatically execute operations according to preset conditions, including: inventory adjustment contracts, price adjustment contracts, promotion strategy contracts and profit-sharing settlement contracts.

[0141] Inventory adjustment contracts are used to adjust inventory in different regions and channels, and automatically perform inventory replenishment or transfer based on demand forecasts and inventory levels.

[0142] Price adjustment contracts are used to dynamically adjust prices based on changes in market demand, ensuring that pricing strategies are flexible in response to market fluctuations.

[0143] The promotion strategy contract is used to manage the launch, end and automated promotion execution of promotion activities, ensuring that promotion information is correctly delivered to various channels.

[0144] The profit-sharing settlement contract is used to automate the distribution of profit shares in a multi-party distribution network, ensuring that the profit distribution of all parties is transparent and fair, and automatically executing settlement.

[0145] For example, if the current scenario is inventory adjustment, select the inventory adjustment contract template to adjust the inventory. The template includes the adjusted region or channel, adjusted inventory quantity, adjustment time, and adjustment reason.

[0146] After selecting the appropriate smart contract template, the distribution data extracted from the blockchain will be automatically filled into the template to generate the corresponding contract terms. This process is fully automated to ensure the accuracy and consistency of the data.

[0147] For example, based on the current scenario, an inventory adjustment contract is selected, and the distribution data is automatically filled into the smart contract template. The data filled in are region: Shanghai, channel: online, adjusted inventory quantity: 1,500 pieces, and adjustment reason: promotion activity launched.

[0148] The generated inventory adjustment smart contract is that the current inventory of products in the online channel in Shanghai is 1,000 pieces. Due to the launch of the promotion, the inventory is adjusted to 1,500 pieces. The reason for this adjustment operation is the increase in market demand and the increase in inventory demand caused by the promotion.

[0149] Digitally sign the generated smart contract and broadcast the signed smart contract as a transaction to the blockchain network.

[0150] The node receives the smart contract transaction for verification. After the verification is passed, a consensus is reached through the consensus mechanism and it is packaged into the next block of the blockchain, becoming an unalterable on-chain record.

[0151] It is further explained that through smart contracts, operations such as inventory adjustment, price adjustment, promotion management and profit-sharing settlement can be executed automatically, reducing human intervention and improving operational efficiency.

[0152] Execution module: When the triggering conditions of the smart contract are met, the contract is automatically executed to complete the profit sharing settlement;

[0153] Real-time monitoring of sales data, inventory changes, order processing status and other information of each distribution channel. It can ensure that after the goods are sold, the inventory and order status are updated in time, and continuously detect whether the triggering conditions of the smart contract are met.

[0154] Trigger conditions include, but are not limited to, sales reaching a certain threshold, a batch of goods being sold out, the end of a promotion, and the arrival of the booking time point.

[0155] When an event that meets the triggering conditions is monitored, the smart contract will automatically enter the execution state.

[0156] For example, a certain brand of electronic products is sold through multiple distribution channels, and the agreed profit-sharing mechanism is that the brand receives 50% of the sales, distributor A receives 30%, and retailer B receives 20%.

[0157] When the total sales of distributor A reaches 1 million yuan (trigger condition) within a certain period of time, the smart contract will be automatically called to execute profit-sharing settlement.

[0158] The execution logic is that the profit-sharing rules set in the smart contract will automatically calculate the profit amounts of each party based on the distribution data.

[0159] The brand's profit share: 1 million yuan × 50% = 500,000 yuan, distributor A's profit share: 1 million yuan × 30% = 300,000 yuan, retailer B's profit share: 1 million yuan × 20% = 200,000 yuan

[0160] The profit sharing amount will be automatically transferred to the digital wallet or bank account of each participant based on the calculated profit sharing amount.

[0161] Further explanation: the detailed records of each fund allocation (such as amount, time, recipient wallet address) will be permanently recorded on the blockchain and can be traced back and audited at any time to ensure that all transactions comply with the predetermined profit-sharing rules and contract terms. At the same time, due to the decentralized nature of the blockchain, the allocation of funds can be completed in a short period of time without relying on third-party institutions for fund settlement, which greatly shortens the fund settlement cycle.

[0162] The distribution monitoring module monitors the key indicators of each distribution channel in real time. When an abnormality is detected in a distribution channel, the intelligent early warning mechanism is triggered to generate abnormality detection results.

[0163] Set safety thresholds based on inventory levels and sales.

[0164] For example, if the average daily sales volume of a product is 100 pieces and the lead time is 5 days, the safety stock threshold can be set to 500 pieces to ensure that the inventory is sufficient to support sales demand before the supply arrives.

[0165] If the daily sales of a channel is 50,000 yuan, the safety threshold of sales can be set to no less than 30,000 yuan per day. Once it falls below this threshold, a warning will be issued.

[0166] By querying the inventory database of the distribution channel, the current inventory level is obtained. The inventory data is accurate to the quantity of specific products and updated in real time. At the same time, the sales of each channel are tracked in real time, including the specific time, amount, products sold and other information of each sale.

[0167] All inventory and sales data are transmitted to the central unit in real time through the data interface. The central unit continuously detects these data and compares them with the set safety threshold. Once the inventory or sales volume falls below the safety threshold, an abnormal warning is triggered.

[0168] When it is detected that the inventory level or sales of a distribution channel falls below the preset safety threshold, an abnormal warning will be automatically issued. This warning can be passed on to relevant personnel and management in a variety of ways.

[0169] Generate detailed anomaly detection reports based on anomalies.

[0170] For example, distribution channel: Shanghai region, distribution channel A, exception type: Product X inventory level is 400 pieces, which is lower than the safety stock threshold of 500 pieces. The sales volume for the day is 20,000 yuan, which is lower than the daily sales threshold of 30,000 yuan. Cause analysis: The recent increase in demand and the delay in inventory replenishment have led to insufficient inventory; the decline in sales may be related to the promotion activities of competitors. Recommended measures: Immediately transfer 100 pieces of product X from other warehouses to channel A. Start a promotional activity and offer a 10% discount to stimulate sales.

[0171] Further explanation: insufficient inventory is a common problem in distribution management, which may lead to stock-outs, thus affecting sales and customer satisfaction. By setting safety stock thresholds and monitoring them in real time, companies can issue early warnings when inventory is about to run out, ensuring timely replenishment to avoid stock-outs and loss of sales opportunities. An abnormal decline in sales may indicate changes in market demand or other operational problems. By monitoring sales in real time and setting safety thresholds, companies can promptly detect abnormal fluctuations in sales and take quick measures to adjust sales strategies or conduct promotions to prevent further sales declines.

[0172] Intelligent early warning module, which automatically adjusts distribution data based on the abnormal detection results in intelligent early warning and records them in the blockchain;

[0173] Analyze the anomaly detection results and adjust the corresponding distribution data, such as inventory, product prices, etc., based on the anomaly detection results.

[0174] Continuously monitor the adjusted inventory level to ensure that replenishment or transfer operations can be executed in time and the inventory level returns to above the safety stock threshold. For example, check whether the inventory in Shanghai area has recovered to more than 500 pieces after replenishment.

[0175] In order to ensure the transparency and traceability of the distribution data adjustment process, each adjustment operation and its results are recorded on the blockchain. The decentralization and immutability of the blockchain ensure the security and transparency of all adjustment records.

[0176] Further explanation: Traditional distribution management usually relies on manual inspection and adjustment, which may delay the resolution of problems. Through automated intelligent early warning and distribution data adjustment, abnormal situations of inventory and sales can be responded to immediately, avoiding a lot of manual operations and decision-making time.

[0177] In summary, the present invention achieves this by: establishing a market demand model that simultaneously captures the steady trend of historical data and the dynamic changes in market demand; static forecast results enhance the sensitivity to historical data, making the forecast results more responsive to past market behaviors; and dynamic forecast results can make more agile predictions of future market changes, ultimately generating more comprehensive and accurate market demand forecast results; in addition, recording distribution data on the blockchain enhances data security and transparency, greatly improving the credibility of data management.

[0178] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a product distribution application software system is provided.

[0179] This embodiment builds a product distribution application software, combining multiple modules such as data collection, feature extraction, market demand forecasting, strategy adjustment, blockchain records, and smart contract generation, to demonstrate its innovative advantages in market demand forecasting, inventory management, price adjustment, etc. Two comparison technologies, comparison technology A and comparison technology B, are specifically compared, representing two typical market demand forecasting and inventory management solutions.

[0180] Comparative Technology A (Traditional Market Demand Forecasting Method Based on Linear Regression): Comparative Technology A uses a linear regression model and relies on historical sales data to forecast market demand. This technology is based only on time series data of sales, and fails to effectively integrate multi-dimensional factors such as external factors, lag effects, and consumer behavior, resulting in low forecast accuracy when the market environment changes, especially when faced with rapidly changing demands, and insufficient response.

[0181] Comparative Technique B (Static Inventory Management Method): Comparative Technique B uses fixed inventory management rules and usually makes inventory adjustments based on average sales over a period of time. This method lacks the ability to respond immediately to changes in market demand and is prone to inventory backlogs or shortages, especially when market demand fluctuates greatly.

[0182] Experimental subjects: This experiment selected four distribution channels (online shopping malls, physical stores, wholesalers and agents) in three regions (A, B, C) to compare the performance of the solution of the present invention, comparative technology A and comparative technology B in the same sales cycle.

[0183] Experimental steps:

[0184] Data collection and processing: We collected historical sales data, market dynamics data, consumer behavior data, and external factors data for the past three months from the company's ERP system through the API interface. The data covers both peak and normal sales periods. After data cleaning, outlier processing, and standardization, we ensured its accuracy and consistency.

[0185] Feature extraction: extract time series features (such as month, quarter), historical lag features, external factor features (such as weather, policy changes) and consumer behavior features (such as purchase frequency, return rate) from the preprocessed data. These features are integrated into comprehensive sales features as the basis for subsequent market demand forecasting.

[0186] Market demand forecast: The present invention uses an exponential smoothing model combined with dynamic change rate analysis to generate market demand forecasts for each region and channel. Unlike the linear regression model of comparative technology A, the present invention not only considers the time series of sales, but also integrates the characteristics of multiple dimensions, making the forecast results more accurate. In particular, for market demands with large fluctuations, the present invention captures market changes through dynamic change rate analysis. Thus, the market demand forecast results are obtained.

[0187] Strategy adjustment: According to the market demand forecast results, the present invention automatically adjusts inventory and prices. Inventory adjustment is calculated by the following formula:

[0188] S 需求 (r,c,t)=Y t (r,c)+α×σ t (r,c)

[0189] According to the market demand forecast value, the present invention calculates the market demand growth rate, and its expression is:

[0190]

[0191] The price is adjusted according to the market demand growth rate, and the expression is:

[0192] P 调整 (r,c,t)=P 当前 (r,c)×(1+β×G t (r,c));

[0193] In contrast, technology B relies on static inventory rules and cannot be dynamically adjusted according to changes in market demand, resulting in delayed inventory and price adjustments.

[0194] Blockchain records and smart contracts: The distribution data after strategic adjustment is recorded on the blockchain through asymmetric encryption technology to ensure data security and immutability. Smart contracts are automatically executed according to trigger conditions (such as inventory reaching a threshold) to complete profit settlement and inventory adjustment. Comparative Technology A and Comparative Technology B do not use blockchain and smart contracts, so data transparency and transaction security are lower.

[0195] The details are shown in Table 1 below:

[0196] Table 1 Distribution comparison table

[0197]

[0198]

[0199] By analyzing the data in the table, the advantages of the present invention in market demand forecasting, inventory adjustment, price adjustment, etc. are revealed.

[0200] Market demand forecast: The present invention can more accurately forecast market demand through the integration of multi-dimensional features. For example, the historical average monthly sales of the online mall in region A is 1,243 pieces, and the present invention predicts sales of 1,351 pieces, while the comparative technology A only predicts 1,305 pieces. It fails to fully incorporate external factors (such as seasonality, promotional activities, etc.), resulting in forecast deviations. In the physical store in region B, the present invention predicts sales of 1,918 pieces, while the comparative technology A predicts sales of 1,883 pieces, showing its lack of adaptability to changes in market demand.

[0201] Inventory management: The present invention significantly optimizes inventory management. Taking the wholesaler in region C as an example, the inventory of Comparative Technology B is 2,379 pieces, which is significantly higher than the market forecast of 2,091 pieces, resulting in inventory backlog. The present invention reduces the inventory to 2,132 pieces through inventory adjustment, avoiding the problem of overstocking. In the online mall in region A, the inventory of Comparative Technology B is 1,288 pieces, which is lower than the market forecast of 1,351 pieces, which may lead to out-of-stock. The present invention increases the inventory to 1,403 pieces through adjustment to ensure sufficient inventory.

[0202] Price adjustment: Through the analysis of the market demand growth rate, the present invention can flexibly adjust the price. For example, the online mall price in region A is adjusted from 96 yuan to 104 yuan, and the wholesaler price in region C is adjusted from 149 yuan to 157 yuan. This adjustment not only responds to the growth of market demand, but also effectively increases corporate profits. In contrast, the price adjustment of technology B is delayed, and it fails to respond to market fluctuations in a timely manner, which may lead to profit losses.

[0203] Market demand growth rate: By comparing the historical market demand growth rate and the adjusted market demand growth rate, it can be seen that the present invention has better prediction ability for market demand than comparative techniques A and B. For example, the historical demand growth rate of agents in region C is 5%, and the adjusted growth rate of the present invention reaches 6%. In contrast, comparative technique A only predicts an increase of 5.3%, indicating that it lacks integration of external factors and consumer behavior data.

[0204] Advantages of blockchain records and smart contracts: The present invention records distribution data and generates smart contracts through blockchain technology. The immutability and transparency of blockchain ensure the security of data, while smart contracts reduce the risk of manual operations through automatic triggering mechanisms. Comparative technologies A and B do not use blockchain technology, so there is a large gap in data security and transaction transparency.

[0205] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A product distribution application software system, characterized by: include, Data collection and processing module, collects distribution-related data and performs pre-processing; The feature extraction module extracts and integrates the features of the preprocessed data to form comprehensive sales features; The market demand forecasting module establishes a market demand forecasting model based on the preprocessed data and generates market demand forecasting results; Strategy adjustment module, automatically adjusts distribution data according to market demand forecast results; The blockchain recording module records the adjusted distribution data on the blockchain; Smart contract generation module, which automatically generates smart contracts based on distribution data on the blockchain; Execution module: When the triggering conditions of the smart contract are met, the contract is automatically executed to complete the profit sharing settlement; The distribution monitoring module monitors the key indicators of each distribution channel in real time. When an abnormality is detected in a distribution channel, the intelligent early warning mechanism is triggered to generate abnormality detection results. The intelligent early warning module automatically adjusts the distribution data according to the abnormal detection results in the intelligent early warning and records it in the blockchain.

2. The product distribution application software system according to claim 1, characterized in that: Collect distribution-related data and pre-process it, which includes the following steps: Obtain historical sales data, market dynamics data, consumer behavior data, and external factors data through APIs; Clean, process outliers and standardize the collected data; The processed data is organized into a structured format and stored.

3. The product distribution application software system as claimed in claim 2, characterized in that: Extract and integrate the features of the preprocessed data to form comprehensive sales features, which specifically includes the following steps: Automatically extract temporal features in time series from preprocessed data; Based on historical sales data, historical lagged features are generated through lag operations; Organize the external factor data into external factor features; Interact consumer behavior data with historical sales data to generate interactive features; Integrate time characteristics, historical lag characteristics, external factor characteristics and interaction characteristics to form comprehensive sales characteristics.

4. The product distribution application software system as claimed in claim 3, characterized in that: Based on the preprocessed data, a market demand forecasting model is established to generate market demand forecasting results, which specifically includes the following steps: A time step is set based on the comprehensive sales characteristics, and an exponential smoothing model is used to perform exponential smoothing on the comprehensive sales characteristics with the set time step to obtain smoothed lag data; Use the weighted summation method to aggregate the smoothed lagged data to obtain the static market demand forecast value; Introduce dynamic change rate analysis to calculate the change rate of comprehensive sales characteristics, and apply integral technology to accumulate the change rate of comprehensive sales characteristics to obtain dynamic market demand forecast values; Combining the static market demand forecast value with the dynamic market demand forecast value to form a complete market demand forecast value, the expression is: Among them, Y t (r,c) represents the demand forecast value of region r and channel c at time t, M represents the number of lagged time steps, i represents the index, and X t-i represents the comprehensive sales characteristics of the past i-th time step, λ i represents the attenuation coefficient, t represents the current time, γ i represents the weighting coefficient, dt represents the integration over time t, represents partial derivative.

5. The product distribution application software system according to claim 4, characterized in that: According to the market demand forecast results, the distribution data is automatically adjusted, which includes the following steps: The market demand forecast values ​​for each region and each channel are counted to determine the inventory demand, and the expression is: S 需求 (r,c,t)=Y t (r,c)+α×σ t (r,c); Among them, S 需求 (r,c,t) represents the inventory demand of region r and channel c at the current time t, t represents the current time, σ t (r,c) represents the standard deviation of demand forecast, and α represents the confidence interval coefficient; Adjust inventory allocation for each region and channel based on inventory needs; According to the market demand forecast value, the market demand growth rate of each region and each channel is calculated, and the expression is: Among them, G t (r,c) represents the market demand growth rate of region r and channel c at the current time t, t-1 represents the previous time point, Y t-1 (r,c) represents the market demand growth rate of region r and channel c at the previous time point; The price is adjusted according to the market demand growth rate, and the expression is: P 调整 (r,c,t)=P 当前 (r,c)×(1+β×G t (r,c)); Among them, P 调整 (r,c,t) represents the adjusted price of region r and channel c at the current time t, P 当前 (r,c) represents the current price in region r and channel c at the current time t, and β represents the price adjustment coefficient.

6. The product distribution application software system according to claim 5, characterized in that: The adjusted distribution data is recorded on the blockchain, which specifically includes the following steps: Structuring and organizing the distribution data and packaging it into transaction data packages; Use asymmetric encryption technology to encrypt transaction data packets and broadcast them to the blockchain network as transactions; The nodes in the blockchain network receive the transaction and conduct verification and consensus; Once consensus is reached, the transaction data will be written into the next block of the blockchain.

7. The product distribution application software system according to claim 6, characterized in that: Automatically generate smart contracts based on distribution data on the blockchain, which includes the following steps: Access and extract distribution data stored on the blockchain; Based on the distribution scenario, select a smart contract template, which includes an inventory adjustment contract, a price adjustment contract, a promotion strategy contract, and a profit sharing settlement contract; The distribution data extracted from the blockchain is automatically filled into the smart contract template to generate the corresponding contract terms; Digitally sign the generated smart contract; The signed smart contract is broadcast as a transaction to the blockchain network. The nodes in the blockchain network will verify the contract and package it into the next block of the blockchain through a consensus mechanism.

8. The product distribution application software system according to claim 7, characterized in that: When the triggering conditions of the smart contract are met, the contract is automatically executed to complete the profit sharing settlement, which includes the following steps: Real-time monitoring of distribution channels. When the conditions that trigger the contract terms are met, the smart contract automatically enters the execution state and calculates the profit share of each participant. Based on the profit amount, the distribution of funds is executed directly on the blockchain.

9. The product distribution application software system according to claim 8, characterized in that: Real-time monitoring of key indicators of each distribution channel. When anomalies in the distribution channel are detected, the intelligent early warning mechanism is triggered to generate anomaly detection results. The specific steps include the following: Set safety thresholds based on inventory levels and sales; Real-time monitoring of inventory levels and sales in each distribution channel. When the inventory levels and sales drop below the set safety stock threshold, an abnormal warning will be issued. Generate anomaly detection reports based on anomaly warnings.

10. The product distribution application software system according to claim 9, characterized in that: According to the abnormal detection results in the intelligent early warning, the distribution data is automatically adjusted and recorded in the blockchain. The specific steps include the following: Based on the anomaly detection results generated in the intelligent early warning, the distribution data is automatically adjusted according to the cause of the anomaly; After making adjustments, continuously monitor the adjusted data; All adjustment operations and adjustment results are recorded on the blockchain.

Images