Retail enterprise inventory collaborative management system based on smart order fulfillment and cross-channel demand prediction

Abstract

The invention relates to the technical field of enterprise operation management, in particular to a retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand prediction, which comprises an order management module, an intelligent order fulfillment module, a demand prediction module, an inventory collaborative management module, a data analysis module and a user management module. Compared with a simple scheme in the prior art that the historical logistics cycle is directly adopted as the predicted arrival time, and the scheme cannot accurately reflect the current logistics progress and the actual situation, the scheme of the invention adopts a more refined prediction formula; the logistics average period T, the current logistics progress P and the coefficients beta and gamma corresponding to the logistics average period T and the current logistics progress P are comprehensively considered, so that the arrival time can be predicted more accurately, more reliable logistics information is provided for customers, and the customer experience and credibility are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise operation management, and in particular to a retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting. Background Art

[0002] In today's retail enterprises, inventory management and order fulfillment are key links to ensure smooth business operations and customer satisfaction. However, traditional inventory management and order fulfillment methods often rely on historical data and experience judgment, which is difficult to accurately reflect the real-time changes in the market and the actual needs of customers.

[0003] Especially in terms of order fulfillment, existing technologies usually directly use historical logistics cycles as the basis for estimated delivery time. However, this method has obvious limitations. On the one hand, it cannot accurately reflect the actual situation of the current logistics progress. Unexpected factors such as logistics congestion and weather impact may cause the logistics time to be extended or shortened. On the other hand, it cannot make personalized predictions based on the specific needs and characteristics of different orders, such as the type of goods, quantity, shipping address and delivery address of the order.

[0004] In addition, in terms of cross-channel demand forecasting, existing technologies are often based on simple weighted average or single-factor forecasting models, which cannot fully consider the impact of channel characteristics and market dynamics on demand. This results in inaccurate forecasting results that cannot meet the needs of retail companies for refined operations.

[0005] Therefore, in order to overcome the shortcomings of the existing technology and improve the accuracy and efficiency of retail enterprise inventory management and order fulfillment, the present invention proposes a retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting. By comprehensively considering multiple factors, the system achieves accurate prediction of order delivery time and scientific prediction of cross-channel demand, providing strong support for the operation optimization and competitiveness improvement of retail enterprises. Summary of the invention

[0006] In order to overcome the problems raised in the above-mentioned background technology, the present invention proposes a retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting.

[0007] The technical solution of the present invention is: a retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting, including:

[0008] The order management module is used to track and manage the entire process of customer orders, including order receipt, processing, tracking, analysis and optimization;

[0009] Smart order fulfillment module, which is used to manage the entire business process from receiving a sales order to delivering the goods to the customer;

[0010] Demand forecasting module, used to forecast demand across all channels based on historical sales data, market trends, and promotional activities;

[0011] Inventory collaborative management module, used to achieve real-time sharing of inventory information and intelligent early warning, and optimize inventory management;

[0012] Data analysis module, which is used to collect and analyze order data, sales data, and inventory data to provide decision support for retail enterprises;

[0013] The user management module is used to manage the user information of the system, including user registration, login, permission management and personal information management.

[0014] Preferably, the order management module includes the following steps when tracking and managing the entire process of customer orders:

[0015] S11: Order reception: the system receives order information from online and offline channels, where the order information includes product details, quantity, delivery address, and payment method;

[0016] S12: Order review, automatically reviewing the completeness and accuracy of order information, including checking the validity of product inventory and delivery address;

[0017] S13: Order processing: splitting, merging and reordering approved orders according to the set delivery requirements and logistics strategies;

[0018] S14: Order tracking, real-time update of order status, including product picking, packaging, delivery, transportation and receipt;

[0019] S15: Order analysis, collect order data and conduct statistical analysis.

[0020] As a preferred option, the smart order fulfillment module manages the entire business process from receiving a sales order to delivering the goods to the customer, including the following steps:

[0021] S21: Order reception and confirmation, receiving order information from the order management module, and performing confirmation and pre-processing;

[0022] S22: Inventory pre-occupancy: According to the product information in the order, the corresponding inventory is pre-occupied to ensure that the order can be shipped on time;

[0023] S23: Order splitting and dispatching: splitting and dispatching orders according to product inventory, logistics strategy and customer needs;

[0024] S24: Logistics tracking and updating: real-time tracking of order logistics status, updating order information, and providing customers with accurate logistics information and estimated delivery time;

[0025] S25: Exception handling, identifying and handling exceptions during order fulfillment.

[0026] As a preferred option, the smart order fulfillment module provides customers with accurate logistics information and estimated delivery time, and predicts the estimated delivery time using the following formula:

[0027]

[0028] Among them, S is the predicted delivery period, that is, the time required to complete the delivery, T is the average logistics cycle, that is, the time required for delivery to similar logistics addresses in historical data, P is the current logistics progress, P is expressed in percentage, β is the coefficient of the average logistics cycle T, and γ is the coefficient of the current logistics progress P.

[0029] Preferably, the smart order fulfillment module includes the following steps when identifying and handling abnormal situations in the order fulfillment process:

[0030] S31: Establish a logistics model, divide the entire logistics process into multiple processes, and establish a logistics prediction model;

[0031] S32: Parameter input, determining the parameters of each process in the logistics prediction model according to the shipping address and the receiving address, and inputting them into the logistics prediction model;

[0032] S33: Data comparison, comparing the actual logistics information with the logistics prediction model. When the error between the actual logistics information and the logistics prediction model exceeds a threshold, an early warning signal is issued;

[0033] S34: Error analysis, analyzing the actual logistics information according to the logistics forecast model to determine the errors that occurred;

[0034] S35: Take corrective measures. For specific problems that arise, adopt targeted corrective measures to solve the problems.

[0035] Preferably, when the entire logistics process is divided into multiple processes and a logistics prediction model is established, the following steps are included:

[0036] S41: Data acquisition, acquiring historical freight data, delivery information data, and receipt information data;

[0037] S42: Process prediction: predicting the entire logistics process based on the shipping information data and the receiving information data, wherein the predicted data includes the logistics transfer stations and key nodes of logistics;

[0038] S43: Model building: Based on the acquired data and predicted data, a logistics prediction model is established using a neural network algorithm.

[0039] Preferably, the demand forecasting module includes the following steps when forecasting the demand for all channels based on historical sales data, market trends and promotional activities:

[0040] S51: Data collection, collecting sales data and related data from all channels, including historical sales data, market trend data and promotion activity data;

[0041] S52: Data cleaning and preprocessing: identifying and processing duplicate data, abnormal data, and missing data in the data, and converting the data into a unified format;

[0042] S53: Model construction and training: select a suitable forecasting model and use the collected data for training to build a demand forecasting model. The selectable forecasting models include exponential smoothing model, ARIMA model, support vector machine-based model, random forest model and neural network model.

[0043] S54: Demand forecasting, using the trained model to predict cross-channel demand;

[0044] S55: Result analysis: analyzing the prediction results and analyzing the problems existing in the prediction results, wherein the problems include duplicate channel data and abnormal data.

[0045] Preferably, when the demand forecasting module forecasts the demand for all channels based on historical sales data, market trends and promotional factors, the principle formula for forecasting the demand for all channels is:

[0046]

[0047] Among them, Y is the demand of all channels, ω i is the weight of the ith channel, that is, the importance of the ith channel, X i is the demand for goods in the ith channel, where in, is the daily sales volume of the channel, α is the activity intensity coefficient, that is, the impact of promotional activities on sales, and Δ is the market trend coefficient, which indicates the impact of market trends on sales.

[0048] Preferably, the inventory collaborative management module includes the following steps when realizing real-time sharing of inventory information and intelligent early warning and optimizing inventory management:

[0049] S61: Inventory data collection, collecting inventory data from various channels, including commodity inventory quantity, inventory status, and inventory location;

[0050] S62: Sales volume and promotion activity analysis, analyzing the historical sales data of each channel and the current trend, and evaluating the impact of promotion activities on sales volume;

[0051] S63: Channel weight adjustment and priority sorting: adjust the inventory call volume of each channel based on sales volume, promotion effect and sales volume change trend factors;

[0052] S64: Inventory warning and adjustment: According to the set inventory warning threshold, the inventory is warned and the inventory is adjusted according to the current demand and predicted change.

[0053] As a preferred method, when adjusting the inventory call volume of each channel according to sales volume, promotion effect and sales volume change trend factors, the rules used are:

[0054] A11: When adjusting the inventory call volume of each channel, first set the priority. The inventory call volume of the channel within the set priority is its predicted sales volume.

[0055] A12: Determine the weight of each channel based on channel type and historical sales data, and allocate the inventory call volume for each channel from the inventory volume outside the priority level;

[0056] A13: The inventory call volume for each channel is updated at a set time period.

[0057] Beneficial effects of the present invention:

[0058] 1. Compared with the existing technology that directly uses the historical logistics cycle as the simple solution for the estimated delivery time, which has the disadvantage of not being able to accurately reflect the current logistics progress and actual situation, this solution adopts a more refined prediction formula, which comprehensively considers the average logistics cycle T, the current logistics progress P and their corresponding coefficients β and γ, so as to more accurately predict the delivery time, provide customers with more reliable logistics information, and improve customer experience and trust;

[0059] 2. Compared with the existing technology that uses logistics forecasting solutions based on empirical rules or simple statistics, which have the disadvantages of low forecasting accuracy, poor adaptability and inability to effectively cope with complex and changeable logistics environments, this solution uses data acquisition, process forecasting and neural network algorithm modeling to establish an intelligent logistics forecasting model. It can make full use of historical freight, delivery and receipt information data, accurately predict the transfer stations and key nodes of the entire logistics process, significantly improve the accuracy and flexibility of logistics forecasting, and provide strong technical support for optimizing logistics management and improving customer experience;

[0060] 3. Compared with the existing technology that adopts a unified standard or a simple proportion to allocate inventory call volume, the solution has the disadvantage of being unable to flexibly respond to different channel characteristics and sales dynamic changes. This solution adopts an inventory call volume adjustment rule based on priority setting, channel weight ratio and time period update, which can more accurately match the actual demand and sales trend of each channel, and realize efficient and flexible inventory call. This solution not only significantly improves inventory turnover efficiency, but also reduces inventory costs, enhances the retail enterprise's ability to respond to market changes, and provides a strong guarantee for the enterprise's operational optimization and competitiveness improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 Shown is a schematic diagram of the structure of the retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting of the present invention;

[0062] Figure 2 What is shown is a workflow diagram of the smart order fulfillment module in the retail enterprise inventory collaborative management system based on smart order fulfillment and cross-channel demand forecasting of the present invention. DETAILED DESCRIPTION

[0063] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0064] See also Figure 1-2 The present invention provides an embodiment: a retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting, comprising:

[0065] The order management module is used to track and manage the entire process of customer orders, including order receipt, processing, tracking, analysis and optimization;

[0066] Smart order fulfillment module, which is used to manage the entire business process from receiving a sales order to delivering the goods to the customer;

[0067] Demand forecasting module, used to forecast demand across all channels based on historical sales data, market trends, and promotional activities;

[0068] Inventory collaborative management module, used to achieve real-time sharing of inventory information and intelligent early warning, and optimize inventory management;

[0069] Data analysis module, which is used to collect and analyze order data, sales data, and inventory data to provide decision support for retail enterprises;

[0070] The user management module is used to manage the user information of the system, including user registration, login, permission management and personal information management.

[0071] As described above, the present invention realizes the full process tracking and optimization of retail enterprise orders, real-time sharing and intelligent early warning of inventory, and accurate demand forecasting based on multi-dimensional data by integrating order management, intelligent order fulfillment, demand forecasting, inventory collaborative management, data analysis and user management modules, which effectively improves operational efficiency, reduces inventory costs, and provides strong support for the strategic decision-making of enterprises, while ensuring the security of the system and the effective management of user information.

[0072] Preferably, the order management module includes the following steps when tracking and managing the entire process of customer orders:

[0073] S11: Order reception: the system receives order information from online and offline channels, where the order information includes product details, quantity, delivery address, and payment method;

[0074] S12: Order review, automatically reviewing the completeness and accuracy of order information, including checking the validity of product inventory and delivery address;

[0075] S13: Order processing: splitting, merging and reordering approved orders according to the set delivery requirements and logistics strategies;

[0076] S14: Order tracking, real-time update of order status, including product picking, packaging, delivery, transportation and receipt;

[0077] S15: Order analysis, collect order data and conduct statistical analysis.

[0078] As described above, the present invention realizes all-round management from order receipt, review, processing, tracking to analysis through automated processes, ensuring the integrity and accuracy of order information, effectively improving the efficiency and flexibility of order processing, while updating the order status in real time, enhancing customer experience, and providing companies with valuable market insights and operational optimization basis through statistical analysis of order data.

[0079] As a preferred option, the smart order fulfillment module manages the entire business process from receiving a sales order to delivering the goods to the customer, including the following steps:

[0080] S21: Order reception and confirmation, receiving order information from the order management module, and performing confirmation and pre-processing;

[0081] S22: Inventory pre-occupancy: According to the product information in the order, the corresponding inventory is pre-occupied to ensure that the order can be shipped on time;

[0082] S23: Order splitting and dispatching: splitting and dispatching orders according to product inventory, logistics strategy and customer needs;

[0083] S24: Logistics tracking and updating: real-time tracking of order logistics status, updating order information, and providing customers with accurate logistics information and estimated delivery time;

[0084] S25: Exception handling, identifying and handling exceptions during order fulfillment.

[0085] As described above, the present invention ensures the efficient operation of the entire chain from order receipt to delivery to customers by automating the processes of order reception, inventory pre-occupancy, order splitting and dispatching, logistics tracking and exception handling, thereby effectively improving order fulfillment efficiency and reducing inventory occupancy. At the same time, it provides real-time logistics information and exception handling capabilities, significantly enhancing customer experience and satisfaction.

[0086] As a preferred option, the smart order fulfillment module provides customers with accurate logistics information and estimated delivery time, and predicts the estimated delivery time using the following formula:

[0087]

[0088] Among them, S is the predicted delivery period, that is, the time required to complete the delivery, T is the average logistics cycle, that is, the time required for delivery to similar logistics addresses in historical data, P is the current logistics progress, P is expressed in percentage, β is the coefficient of the average logistics cycle T, and γ is the coefficient of the current logistics progress P.

[0089] As mentioned above, compared with the existing technology that directly uses the historical logistics cycle as the simple solution for the estimated delivery time, this solution has the disadvantage of not being able to accurately reflect the current logistics progress and actual situation. This solution uses a more refined prediction formula that comprehensively considers the average logistics cycle T, the current logistics progress P and their corresponding coefficients β and γ, so that the delivery time can be predicted more accurately, providing customers with more reliable logistics information, and improving customer experience and trust.

[0090] Preferably, the smart order fulfillment module includes the following steps when identifying and handling abnormal situations in the order fulfillment process:

[0091] S31: Establish a logistics model, divide the entire logistics process into multiple processes, and establish a logistics prediction model;

[0092] S32: Parameter input, determining the parameters of each process in the logistics prediction model according to the shipping address and the receiving address, and inputting them into the logistics prediction model;

[0093] S33: Data comparison, comparing the actual logistics information with the logistics prediction model. When the error between the actual logistics information and the logistics prediction model exceeds a threshold, an early warning signal is issued;

[0094] S34: Error analysis, analyzing the actual logistics information according to the logistics forecast model to determine the errors that occurred;

[0095] S35: Take corrective measures. For specific problems that arise, adopt targeted corrective measures to solve the problems.

[0096] As described above, compared with the exception handling solutions in the prior art that directly rely on manual monitoring and processing, there are shortcomings such as slow response speed, low processing efficiency and difficulty in preventing potential problems. This solution adopts an intelligent exception handling solution that establishes logistics models, parameter input, data comparison, error analysis and takes corrective measures. It can monitor logistics information in real time, promptly discover and accurately locate anomalies, and quickly take targeted corrective measures, which effectively improves the reliability and efficiency of order fulfillment, reduces the cost of exception handling, and significantly improves customer experience.

[0097] Preferably, when the entire logistics process is divided into multiple processes and a logistics prediction model is established, the following steps are included:

[0098] S41: Data acquisition, acquiring historical freight data, delivery information data, and receipt information data;

[0099] S42: Process prediction: predicting the entire logistics process based on the shipping information data and the receiving information data, wherein the predicted data includes the logistics transfer stations and key nodes of logistics;

[0100] S43: Model building: Based on the acquired data and predicted data, a logistics prediction model is established using a neural network algorithm.

[0101] As mentioned above, compared with the existing technology that uses logistics forecasting solutions based on empirical rules or simple statistics, which have the disadvantages of low prediction accuracy, poor adaptability and inability to effectively cope with complex and changeable logistics environments, this solution uses data acquisition, process prediction and neural network algorithm modeling to establish an intelligent logistics forecasting model. It can make full use of historical freight, delivery and receipt information data, accurately predict the transfer stations and key nodes of the entire logistics process, significantly improve the accuracy and flexibility of logistics forecasting, and provide strong technical support for optimizing logistics management and improving customer experience.

[0102] Preferably, the demand forecasting module includes the following steps when forecasting the demand for all channels based on historical sales data, market trends and promotional activities:

[0103] S51: Data collection, collecting sales data and related data from all channels, including historical sales data, market trend data and promotion activity data;

[0104] S52: Data cleaning and preprocessing: identifying and processing duplicate data, abnormal data, and missing data in the data, and converting the data into a unified format;

[0105] S53: Model construction and training: select a suitable forecasting model and use the collected data for training to build a demand forecasting model. The selectable forecasting models include exponential smoothing model, ARIMA model, support vector machine-based model, random forest model and neural network model.

[0106] S54: Demand forecasting, using the trained model to predict cross-channel demand;

[0107] S55: Result analysis: analyzing the prediction results and analyzing the problems existing in the prediction results, wherein the existing problems include duplicate channel data and abnormal data.

[0108] As described above, the present invention comprehensively collects omni-channel sales and related data, performs strict data cleaning and preprocessing to ensure data quality, and then uses a variety of advanced forecasting models for construction and training to achieve accurate forecasting of cross-channel demand. This solution not only improves the accuracy of the forecast, but also timely discovers and handles problems in the forecast through result analysis, such as duplicate channel data and abnormal data, providing strong data support for retail enterprises' inventory management and market strategies.

[0109] Preferably, when the demand forecasting module forecasts the demand for all channels based on historical sales data, market trends and promotional factors, the principle formula for forecasting the demand for all channels is:

[0110]

[0111] Among them, Y is the demand of all channels, ω i is the weight of the ith channel, that is, the importance of the ith channel, X i is the demand for goods in the ith channel, where in, is the daily sales volume of the channel, α is the activity intensity coefficient, that is, the impact of promotional activities on sales, and Δ is the market trend coefficient, which indicates the impact of market trends on sales.

[0112] As mentioned above, compared with the existing technology that uses simple weighted average or single factor to predict omni-channel demand, this solution has the disadvantages of insufficient prediction accuracy and failure to fully consider the characteristics of each channel and the impact of market dynamics. This solution uses an omni-channel demand prediction formula that comprehensively considers channel weights, daily sales volume, activity intensity coefficients, and market trend coefficients, which can more accurately reflect the actual demand and market changes of each channel, thereby providing more accurate demand prediction results. This solution not only improves the accuracy and reliability of the prediction, but also provides a more scientific basis for retail companies' inventory management and market strategy formulation.

[0113] Preferably, the inventory collaborative management module includes the following steps when realizing real-time sharing of inventory information and intelligent early warning and optimizing inventory management:

[0114] S61: Inventory data collection, collecting inventory data from various channels, including commodity inventory quantity, inventory status, and inventory location;

[0115] S62: Sales volume and promotion activity analysis, analyzing the historical sales data of each channel and the current trend, and evaluating the impact of promotion activities on sales volume;

[0116] S63: Channel weight adjustment and priority sorting: adjust the inventory call volume of each channel based on sales volume, promotion effect and sales volume change trend factors;

[0117] S64: Inventory warning and adjustment: According to the set inventory warning threshold, the inventory is warned and the inventory is adjusted according to the current demand and predicted change.

[0118] As described above, the present invention realizes real-time sharing and intelligent early warning of inventory information by comprehensively collecting inventory data from various channels, deeply analyzing sales trends and the impact of promotional activities, and flexibly adjusting channel proportions and inventory call priorities. This solution not only significantly improves the level of refinement of inventory management, but also dynamically adjusts inventory according to actual demand and predicted changes, effectively avoiding inventory backlogs and out-of-stock risks, ensuring efficient and balanced inventory turnover, and providing strong support for operational optimization and cost control of retail enterprises.

[0119] As a preferred method, when adjusting the inventory call volume of each channel according to sales volume, promotion effect and sales volume change trend factors, the rules used are:

[0120] A11: When adjusting the inventory call volume of each channel, first set the priority. The inventory call volume of the channel within the set priority is its predicted sales volume.

[0121] A12: Determine the weight of each channel based on channel type and historical sales data, and allocate the inventory call volume for each channel from the inventory volume outside the priority level;

[0122] A13: The inventory call volume for each channel is updated at a set time period.

[0123] As mentioned above, compared with the existing technology that adopts a unified standard or a simple proportion to allocate inventory call volume, this solution has the disadvantage of not being able to flexibly respond to different channel characteristics and sales dynamic changes. This solution adopts an inventory call volume adjustment rule based on priority setting, channel weight ratio and time period update, which can more accurately match the actual demand and sales trend of each channel, and achieve efficient and flexible inventory call. This solution not only significantly improves inventory turnover efficiency, but also reduces inventory costs, enhances the retail enterprise's ability to respond to market changes, and provides a strong guarantee for the enterprise's operational optimization and competitiveness improvement.

[0124] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of those skilled in the art without departing from the spirit of the present invention.

Claims

1. A retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting; characterized by: Included are: The order management module is used to track and manage the entire process of customer orders, including order receipt, processing, tracking, analysis and optimization; Smart order fulfillment module, which is used to manage the entire business process from receiving a sales order to delivering the goods to the customer; Demand forecasting module, used to forecast demand across all channels based on historical sales data, market trends, and promotional activities; Inventory collaborative management module, used to achieve real-time sharing of inventory information and intelligent early warning, and optimize inventory management; Data analysis module, which is used to collect and analyze order data, sales data, and inventory data to provide decision support for retail enterprises; The user management module is used to manage the user information of the system, including user registration, login, permission management and personal information management.

2. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 1, characterized in that: The order management module includes the following steps when tracking and managing the entire process of customer orders: S11: Order reception: the system receives order information from online and offline channels, where the order information includes product details, quantity, delivery address, and payment method; S12: Order review, automatically reviewing the completeness and accuracy of order information, including checking the validity of product inventory and delivery address; S13: Order processing: splitting, merging and reordering approved orders according to the set delivery requirements and logistics strategies; S14: Order tracking, real-time update of order status, including product picking, packaging, delivery, transportation and receipt; S15: Order analysis, collect order data and conduct statistical analysis.

3. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 2, characterized in that: The smart order fulfillment module manages the entire business process from receiving a sales order to delivering the goods to the customer, including the following steps: S21: Order reception and confirmation, receiving order information from the order management module, and performing confirmation and pre-processing; S22: Inventory pre-occupancy: According to the product information in the order, the corresponding inventory is pre-occupied to ensure that the order can be shipped on time; S23: Order splitting and dispatching: splitting and dispatching orders according to product inventory, logistics strategy and customer needs; S24: Logistics tracking and updating: real-time tracking of order logistics status, updating order information, and providing customers with accurate logistics information and estimated delivery time; S25: Exception handling, identifying and handling exceptions during order fulfillment.

4. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 3 is characterized by: When providing customers with accurate logistics information and estimated delivery time, the smart order fulfillment module predicts the estimated delivery time using the following formula: Among them, S is the predicted delivery period, that is, the time required to complete the delivery, T is the average logistics cycle, that is, the time required for delivery to similar logistics addresses in historical data, P is the current logistics progress, P is expressed in percentage, β is the coefficient of the average logistics cycle T, and γ is the coefficient of the current logistics progress P.

5. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 4, characterized in that: The smart order fulfillment module includes the following steps when identifying and handling abnormal situations in the order fulfillment process: S31: Establish a logistics model, divide the entire logistics process into multiple processes, and establish a logistics prediction model; S32: Parameter input, determining the parameters of each process in the logistics prediction model according to the shipping address and the receiving address, and inputting them into the logistics prediction model; S33: Data comparison, comparing the actual logistics information with the logistics prediction model. When the error between the actual logistics information and the logistics prediction model exceeds a threshold, an early warning signal is issued; S34: Error analysis, analyzing the actual logistics information according to the logistics forecast model to determine the errors that occurred; S35: Take corrective measures. For specific problems that arise, adopt targeted corrective measures to solve the problems.

6. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 5, characterized in that: When dividing the entire logistics process into multiple processes and establishing a logistics forecasting model, the following steps are included: S41: Data acquisition, acquiring historical freight data, delivery information data, and receipt information data; S42: Process prediction: predicting the entire logistics process based on the shipping information data and the receiving information data, wherein the predicted data includes the logistics transfer stations and key nodes of logistics; S43: Model building: Based on the acquired data and predicted data, a logistics prediction model is established using a neural network algorithm.

7. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 6, characterized in that: The demand forecasting module predicts omni-channel demand based on historical sales data, market trends, and promotional factors, including the following steps: S51: Data collection, collecting sales data and related data from all channels, including historical sales data, market trend data and promotion activity data; S52: Data cleaning and preprocessing: identifying and processing duplicate data, abnormal data, and missing data in the data, and converting the data into a unified format; S53: Model construction and training: select a suitable forecasting model and use the collected data for training to build a demand forecasting model. The selectable forecasting models include exponential smoothing model, ARIMA model, support vector machine-based model, random forest model and neural network model. S54: Demand forecasting, using the trained model to predict cross-channel demand; S55: Result analysis: analyzing the prediction results and analyzing the problems existing in the prediction results, wherein the existing problems include duplicate channel data and abnormal data.

8. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 7, characterized in that: When the demand forecasting module forecasts the demand for all channels based on historical sales data, market trends and promotional factors, the principle formula for forecasting the demand for all channels is: Among them, Y is the demand of all channels, ω i is the weight of the ith channel, that is, the importance of the ith channel, X i is the demand for goods in the ith channel, where in, is the daily sales volume of the channel, α is the activity intensity coefficient, that is, the impact of promotional activities on sales, and Δ is the market trend coefficient, which indicates the impact of market trends on sales.

9. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 8, characterized in that: The inventory collaborative management module includes the following steps to achieve real-time sharing of inventory information and intelligent early warning and optimize inventory management: S61: Inventory data collection, collecting inventory data from various channels, including commodity inventory quantity, inventory status, and inventory location; S62: Sales volume and promotion activity analysis, analyzing the historical sales data of each channel and the current trend, and evaluating the impact of promotion activities on sales volume; S63: Channel weight adjustment and priority sorting: adjust the inventory call volume of each channel based on sales volume, promotion effect and sales volume change trend factors; S64: Inventory warning and adjustment: According to the set inventory warning threshold, the inventory is warned and the inventory is adjusted according to the current demand and predicted change.

10. The retail enterprise inventory collaborative management system based on intelligent order fulfillment and cross-channel demand forecasting according to claim 9, characterized in that: When adjusting the inventory call volume for each channel based on sales volume, promotional activity effects, and sales volume change trends, the rules used are: A11: When adjusting the inventory call volume of each channel, first set the priority. The inventory call volume of the channel within the set priority is its predicted sales volume. A12: Determine the weight of each channel based on channel type and historical sales data, and allocate the inventory call volume for each channel from the inventory volume outside the priority level; A13: The inventory call volume for each channel is updated at a set time period.

Images