Cross-border storage consignment and sale collaborative scheduling management method and system

Abstract

The invention relates to the technical field of warehouse management, in particular to a cross-border warehouse consignment and sale collaborative scheduling management method and system. Obtaining historical goods transfer logistics data of the target goods in the target sales area, training a goods transfer speed predictor according to the historical goods transfer logistics data, and predicting and obtaining predicted goods transfer time required by current goods transfer; obtaining historical sales data of the target goods in the target sales area, training a sales data predictor according to the historical sales data, and predicting to obtain daily predicted sales volume within preset sales time; based on the predicted sales volume and the predicted cargo transfer time consumption, the minimum stock volume is obtained, a stock coefficient is configured, and based on the minimum stock volume and the stock coefficient, a dynamic stock alert threshold value is configured; and obtaining the current stock quantity of the target sales area, judging whether the current stock quantity is lower than a stock warning threshold value, and configuring a transfer plan according to a judgment result. And the operation efficiency and economic benefits of cross-border storage consignment sale are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of warehouse management, in particular to a cross-border warehouse consignment cooperative scheduling management method and system. BACKGROUND

[0002] In the traditional cross-border warehouse consignment management, inventory management and restocking arrangement are usually relied on manual experience or simple statistical methods. Generally, the inventory quantity is roughly determined according to the past sales situation, and when the inventory is reduced to a certain extent, the restocking request is manually initiated, and the quantity and time of restocking are mainly dependent on the subjective judgment of the staff. There are technical problems of rigid inventory management and unreasonable restocking plan. SUMMARY

[0003] The present application provides a cross-border warehouse consignment cooperative scheduling management method and system to solve the technical problems of rigid inventory management and unreasonable restocking plan in the prior art.

[0004] The technical solution of the present application to solve the above technical problems is as follows: In a first aspect, the present application provides a cross-border warehouse consignment cooperative scheduling management method, comprising: obtaining historical restocking logistics data of a target product in a target sales area, training a restocking speed predictor according to the historical restocking logistics data, and predicting a predicted restocking time required for current restocking; obtaining historical sales data of the target product in the target sales area, training a sales data predictor according to the historical sales data, and predicting a predicted sales quantity per day within a preset sales time, wherein the time length of the preset sales time is greater than the predicted restocking time; based on the predicted sales quantity and the predicted restocking time, obtaining a minimum inventory quantity, configuring a reserve inventory coefficient, and based on the minimum inventory quantity and the reserve inventory coefficient, configuring a dynamic inventory alert threshold; obtaining a current inventory quantity of the target sales area, judging whether the current inventory quantity is lower than the inventory alert threshold, and configuring a restocking plan according to the judgment result.

[0005] Optionally, obtaining historical restocking logistics data of a target sales area, training a restocking speed predictor according to the historical restocking logistics data, and predicting a predicted restocking time required for current restocking, comprises: obtaining historical restocking logistics data within a historical time, including restocking information and restocking time; using machine learning to build the restocking speed predictor, and training the restocking speed predictor with the historical restocking logistics data until convergence; inputting the current restocking information into the restocking speed predictor to obtain the predicted restocking time.

[0006] Optionally, the historical sales data of the target goods in the target sales area is obtained, and a sales data predictor is trained according to the historical sales data to predict a daily predicted sales volume within a preset sales time, including: obtaining the historical sales data of the target goods in the target sales area, including promotion information and daily sales volume; using machine learning to build the sales data predictor, and training the sales data predictor with the historical sales data until convergence; inputting the current promotion information and the current date into the sales data predictor to obtain the daily predicted sales volume within the preset sales time, wherein the time length of the preset sales time is greater than the predicted restocking time.

[0007] Optionally, based on the predicted sales volume and the predicted restocking time, a minimum inventory is obtained, and a reserve inventory coefficient is configured, and a dynamic inventory alert threshold is configured based on the minimum inventory and the reserve inventory coefficient, including: calculating the total sales volume within the predicted restocking time range as the minimum inventory according to the predicted daily predicted sales volume within the preset sales time; configuring the reserve inventory coefficient according to the type of the target goods, and taking the product of the reserve inventory coefficient and the minimum inventory as the inventory alert threshold.

[0008] Optionally, the current inventory of the target sales area is obtained, and it is judged whether the current inventory is lower than the inventory alert threshold, and a restocking plan is configured according to the judgment result, including: obtaining the current inventory of the target sales area; if the current inventory is not lower than the inventory alert threshold, no restocking is performed; if the current inventory is lower than the inventory alert threshold, a restocking plan is configured according to the current inventory, the predicted sales volume and the predicted restocking time, and restocking is performed.

[0009] Optionally, the current inventory of the target sales area is obtained, including: counting the inventory of the target goods in the current target sales area as the in-warehouse inventory, which is included in the current inventory; counting the quantity of the target goods in multiple batches being transported to the target sales area and the in-transit restocking information as the multiple in-transit inventories; calling the restocking speed predictor to predict the in-transit predicted restocking time of the multiple in-transit inventories, and calculating the predicted in-warehouse time of the multiple in-transit inventories to the target sales area according to the in-transit predicted restocking time and the in-transit restocking information; when the predicted in-warehouse time of a certain in-transit inventory is earlier than the predicted restocking time, the quantity of the target goods in the in-transit inventory is counted into the current inventory, otherwise, the quantity of the target goods in the in-transit inventory is not counted into the current inventory.

[0010] Optionally, if the current inventory is lower than the inventory alert threshold, a restocking plan is configured according to the current inventory and the minimum inventory, and restocking is performed, including: calculating the difference between the current inventory and the minimum inventory as the minimum restocking quantity, and performing restocking according to the restocking quantity which is not less than the minimum restocking quantity.

[0011] In a second aspect, the present application provides a cross-border warehousing consignment collaborative scheduling management system, comprising: The time consumption prediction module is configured to obtain historical order-fulfillment logistics data of the target product in the target sales region, train an order-fulfillment speed predictor based on the historical order-fulfillment logistics data, and predict a predicted order-fulfillment time consumption required for current order fulfillment. The daily sales prediction module is configured to obtain historical sales data of the target product in the target sales region, train a sales data predictor based on the historical sales data, and predict a predicted daily sales volume within a preset sales time, wherein the preset sales time is longer than the predicted order-fulfillment time consumption. The inventory threshold configuration module is configured to obtain a minimum inventory quantity based on the predicted daily sales volume and the predicted order-fulfillment time consumption, configure a reserve inventory coefficient, and configure a dynamic inventory alert threshold based on the minimum inventory quantity and the reserve inventory coefficient. The order-fulfillment plan configuration module is configured to obtain a current inventory quantity, determine whether the current inventory quantity is lower than the inventory alert threshold, and configure an order-fulfillment plan according to the determination result.

[0012] By implementing the present application, the historical order-fulfillment logistics data of the target product in the target sales region can be obtained, the order-fulfillment speed predictor can be trained based on the historical order-fulfillment logistics data, and the predicted order-fulfillment time consumption required for current order fulfillment can be predicted, thereby providing a key time reference for subsequent inventory management and order-fulfillment plan making, and avoiding inventory management confusion caused by inaccurate order-fulfillment time estimation, such as overstocking caused by early order fulfillment or stockout caused by late order fulfillment. By implementing the present application, the historical sales data of the target product in the target sales region can be obtained, the sales data predictor can be trained based on the historical sales data, and the predicted daily sales volume within a preset sales time can be predicted, wherein the preset sales time is longer than the predicted order-fulfillment time consumption, thereby providing data support for calculating the minimum inventory quantity, helping to prepare for inventory in advance, responding to sales demand, reducing inventory risks caused by inaccurate sales prediction, ensuring sufficient inventory to meet sales during order fulfillment by the minimum inventory quantity, and making inventory management more flexible and accurate by the dynamic inventory alert threshold, thereby avoiding stockout and preventing excessive inventory from occupying funds and warehouse space. By implementing the present application, the minimum inventory quantity can be obtained based on the predicted daily sales volume and the predicted order-fulfillment time consumption, the reserve inventory coefficient can be configured, and the dynamic inventory alert threshold can be configured based on the minimum inventory quantity and the reserve inventory coefficient. By implementing the present application, the current inventory of the target sales area can be obtained, it is determined whether the current inventory is lower than the inventory alert threshold, and the inventory adjustment plan is configured according to the determination result, so that it can be determined in time whether the inventory is sufficient, the inventory is adjusted in time when the inventory is insufficient, and the inventory is not blindly adjusted when the inventory is sufficient, so that the inventory is reasonably controlled, the continuity of sales is ensured, and unnecessary inventory adjustment cost is reduced.

[0013] In summary, by implementing the present application, the inventory backlog and the risk of out-of-stock can be effectively reduced while ensuring the continuity of goods supply in the target sales area, the allocation of inventory resources is optimized, and the operating efficiency and economic benefits of cross-border warehousing consignment are improved. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 A flowchart of a cross-border warehousing consignment collaborative scheduling management method provided by the present application is shown. Figure 2 A structure diagram of a cross-border warehousing consignment collaborative scheduling management system provided by the present application is shown.

[0015] In the drawings, the components represented by the numbers are as follows: The inventory adjustment time consumption prediction module 11, the daily sales prediction module 12, the inventory threshold configuration module 13, and the inventory adjustment plan configuration module 14. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0017] In the description of the present application, the terms "first" and "second" are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first" and "second" can explicitly or implicitly include one or more features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0018] In the description of the present application, the term "for example" is used to mean "serving as an example, instance, or illustration." Any embodiment described as "for example" in the present application is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present application. It will be apparent, however, to one skilled in the art that the present application can be practiced without using these specific details. In other instances, well-known structures and processes are not elaborated upon in order to avoid unnecessary detail, which can obscure the description of the present application. Thus, the present application is not intended to be limited by the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0019] As shown in Embodiment I, Figure 1 The embodiment of the present application provides a cross-border warehousing consignment collaborative scheduling management method, which comprises the following steps: S100: obtaining historical inventory flow data of a target product in a target sales area, training an inventory speed predictor according to the historical inventory flow data, and predicting a predicted inventory time required for current inventory; S200: obtaining historical sales data of the target product in the target sales area, training a sales data predictor according to the historical sales data, and predicting a predicted sales volume per day within a preset sales time, wherein the time length of the preset sales time is greater than the predicted inventory time; S300: obtaining a minimum inventory quantity based on the predicted sales volume and the predicted inventory time, configuring a reserve inventory coefficient, and configuring a dynamic inventory alert threshold based on the minimum inventory quantity and the reserve inventory coefficient; S400: obtaining a current inventory quantity of the target sales area, judging whether the current inventory quantity is lower than the inventory alert threshold, and configuring an inventory plan according to the judgment result.

[0020] In step S100 of the embodiment of the present application, the historical inventory flow data of the target sales area is obtained, the inventory speed predictor is trained according to the historical inventory flow data, and the predicted inventory time required for current inventory is predicted, which comprises the following steps: Obtaining historical inventory flow data within a historical time, which comprises inventory information and inventory time; Using machine learning to build the inventory speed predictor, and training the inventory speed predictor with the historical inventory flow data until convergence; Inputting the current inventory information into the inventory speed predictor to obtain the predicted inventory time.

[0021] In the embodiments of the present application, the inventory adjustment speed predictor is obtained to predict the inventory adjustment time, so as to accurately predict the time required for current inventory adjustment and provide a key time dimension basis for subsequent inventory management and inventory adjustment planning. By training the prediction model based on historical inventory adjustment logistics data, the inventory adjustment time can be estimated in advance based on current inventory adjustment information, so as to ensure that the time difference from inventory adjustment to warehousing is fully considered when formulating the minimum inventory, inventory alert threshold and inventory adjustment plan, thereby avoiding out-of-stock due to untimely inventory adjustment or inventory accumulation due to excessive storage.

[0022] To achieve the above steps, first, historical inventory adjustment logistics data in a historical time period is needed, including inventory adjustment information and inventory adjustment time. The historical time period is a period in which a plurality of inventory adjustments have been made and is not less than one year to cover various seasons and holidays. The inventory adjustment information is various information related to inventory adjustment, including transportation mode (railway, highway, air transportation), product quantity, inventory adjustment initiation date, etc. The inventory adjustment time is the time consumed from the start of inventory adjustment to the entry of the product into the target sales area warehouse. Not less than 1000 pieces of inventory adjustment information and corresponding inventory adjustment time are prepared as historical inventory adjustment logistics data, and 20% of them are extracted as a validation set, and the rest are used as a training set to train the inventory adjustment speed predictor.

[0023] Further, machine learning is needed to build the inventory adjustment speed predictor. For the task type predicted by the inventory adjustment speed predictor, gradient boosting trees can be selected to build the inventory adjustment speed predictor. Specifically, the input features can include transportation mode, product quantity, inventory adjustment initiation date, and the output feature is the inventory adjustment time.

[0024] In the building of the inventory adjustment speed predictor, the number of trees is 100-200 to control the model complexity and avoid overfitting. The maximum tree depth is 5-8 layers to limit the complexity of a single tree and balance the fitting ability and generalization ability. The learning rate is set to 0.01-0.1 to control the contribution weight of each tree and gradually optimize the model. The leaf node minimum sample weight is set to 10-20 to avoid overfitting caused by too small sample size of the leaf node. The subsampling ratio is set to 0.7-0.9, that is, a part of samples are randomly selected for training in each iteration to enhance the robustness of the model. The number of training rounds is 50-100, and is dynamically adjusted according to the convergence of the model to avoid overtraining.

[0025] In the continuous 10 rounds of iterations, the prediction error of the model on the validation set, i.e. the average value of the absolute value of the difference between the actual order picking time and the predicted order picking time, decreases by less than 0.5%, the model is judged to be converged, and the order picking speed predictor is obtained. The current order picking information is input into the order picking speed predictor, and the predicted order picking time can be obtained. For example, if the order picking goods are sports shoes, the current order picking information is air transportation, 500 pairs, and the order picking initiation date is June 25, 2025, inputting the above data into the order picking speed predictor can output the order picking time of the batch of goods, such as 3 days.

[0026] It should be noted that the above order picking speed predictor is only used to predict the order picking time of a specific target product on a specific order picking route, such as order picking sports shoes on the Shanghai-Los Angeles order picking route. If it is necessary to expand the application scope of the order picking speed predictor, multiple order picking speed predictors can be trained for different order picking routes and product categories. Or increase the number of features of the sample data used to train the order picking speed predictor, such as adding route features and adding product category features, etc. The method of expanding the application scope of the order picking speed predictor is well known in the art, and will not be described here.

[0027] In step S200 of the embodiment of the present application, the historical sales data of the target product in the target sales area is obtained, and the sales data predictor is trained according to the historical sales data to obtain the predicted daily sales volume within the preset sales time, including: The historical sales data of the target product in the target sales area is obtained, including promotion information and daily sales volume; The sales data predictor is built by using machine learning, and the sales data predictor is trained with historical sales data until convergence; The current promotion information and the current date are input into the sales data predictor to obtain the predicted daily sales volume within the preset sales time, wherein the time length of the preset sales time is greater than the predicted order picking time.

[0028] In the embodiment of the present application, the sales data predictor is obtained, which aims to accurately predict the daily sales volume within the preset sales time by analyzing and modeling the historical sales data of the target product in the target sales area. Accurate sales volume prediction can avoid cost increase due to inventory accumulation, prevent out-of-stock situations from occurring, and ensure the smooth progress of sales business.

[0029] To achieve the above steps, first, the historical sales data of the target product in the target sales area needs to be obtained. The historical sales data is the daily sales volume and promotion information of the target product in the target sales area at the historical time, wherein the historical time is not less than one year to ensure coverage of different sales scenarios, such as different seasons and different sales under different promotion activities, so that the model learns more comprehensive sales rules. The promotion information includes the duration of the promotion activity, the discount ratio, and the corresponding promotion information is labeled for the daily sales volume. Finally, not less than 1000 historical sales data are collected, and 20% of them are extracted as a validation set, and the rest are used as a training set to train the sales data predictor.

[0030] In the construction of the sales data predictor, according to the characteristics of sales data prediction, a random forest algorithm can be selected as the core architecture. Random forest can effectively handle multiple types of data, has good fitting ability for nonlinear relationships, and can adapt to the influence of complex factors such as promotion information on sales volume.

[0031] Specifically, in the random forest, the number of trees can be set to 100-200. A larger number of trees can improve the stability and accuracy of the model, but too many will increase the calculation time. This range can balance the calculation cost while ensuring performance. The maximum tree depth can be set to 5-8 layers. Limiting the depth of the tree can prevent overfitting of the model and avoid the model being too complex to learn the details of the training data, affecting the generalization ability. The minimum sample split size can be set to 5-10. This parameter indicates the minimum number of samples each child node must contain when splitting. If it is less than this number, the node will not be split, which helps to prevent overfitting of the model. The feature selection ratio can be set to 0.6-0.8, that is, when building each tree, a portion of the features are randomly selected to find the best split point. This ratio determines how many features are selected each time, and a suitable ratio can allow the model to learn under different feature combinations to enhance the model's generalization ability.

[0032] Finally, the historical sales data is used to train the completed sales data predictor for 50-100 rounds. In the last 10 rounds of training, the average absolute error (MAE) of the model on the validation set decreases by less than 0.01, indicating that the performance of the model has improved very little, and the sales data predictor can be judged to be converged. The sales data predictor is obtained, and the current promotion information and the current date are input into the sales data predictor to obtain the predicted daily sales volume within the preset sales time.

[0033] For example, assuming the target product is a certain brand of sports shoes, the current date for prediction is June 25, 2025, and there is a promotional activity during the preset sales time, which lasts from June 28 to July 7, with a discount ratio of 70%. By inputting the current date and promotional information, the predicted daily sales volume during the preset sales time can be output. Assuming the preset sales time is 5 days, the output is the predicted daily sales volume from June 25, 2025 to June 29, 2025, such as 50, 60, 55, 60, and 64.

[0034] The time length of the preset sales time is greater than the predicted restocking time. The essence is to let the predicted sales cycle completely contain the time required for restocking, avoiding sales interruption due to lack of goods during restocking. Assuming that the predicted restocking time is 3 days, i.e., 3 days are required from the start of restocking to the arrival of goods in the warehouse; the preset sales time should be greater than 3 days, for example, set to 5 days, i.e., predict the daily sales volume for the next 5 days. In short, this setting is to make inventory planning and restocking planning forward-looking in time, balance the time difference between "restocking time" and "sales demand", and is the key prerequisite for dynamic inventory management.

[0035] In step S300 of the embodiments of the present application, based on the predicted sales volume and the predicted restocking time, the minimum inventory is obtained, and the reserve inventory coefficient is configured. Based on the minimum inventory and the reserve inventory coefficient, a dynamic inventory alert threshold is configured, including: According to the prediction, the predicted daily sales volume during the preset sales time is obtained, and the total sales volume within the predicted restocking time is calculated as the minimum inventory; According to the type of target product, the reserve inventory coefficient is configured, and the product of the reserve inventory coefficient and the minimum inventory is taken as the inventory alert threshold.

[0036] The core purpose of this step is to determine the minimum inventory standard to ensure sales continuity and set a dynamic inventory alert threshold to provide a clear quantitative basis for restocking decisions. By combining the predicted sales volume and the predicted restocking time to calculate the minimum inventory, it is ensured that there is enough inventory to meet sales demand during restocking. Configuring the reserve inventory coefficient and generating the inventory alert threshold can cope with sales fluctuations, restocking delays, and other uncertainties, avoiding the risk of lack of goods, while achieving dynamic optimization of inventory.

[0037] Specifically, first, the predicted daily sales volume in the preset sales time is obtained according to the prediction, and the total sales volume in the time range of the predicted restocking time is calculated as the minimum inventory, for example, if the predicted restocking time is 5 days, and the predicted sales volume of the 5 days in the preset sales time is 50, 60, 55, 60 and 64 respectively, then the minimum inventory = 50+60+55+60+64 = 289, which ensures that the inventory in the 5 days of restocking can cover the daily sales volume.

[0038] Further, the reserve inventory coefficient is configured according to the target product category, and the product of the reserve inventory coefficient and the minimum inventory is taken as the inventory alert threshold. Specifically, the reserve inventory coefficient is set according to the characteristics of the target product, such as damageability, sales volatility, restocking difficulty, etc., which is usually a value greater than 1. For example, for fast-moving consumer goods with large sales volatility, the coefficient can be set to 1.5; for daily necessities with stable demand, the coefficient can be set to 1.2. Then the reserve inventory coefficient is multiplied by the minimum inventory, and the result is the inventory alert threshold.

[0039] For example, if the minimum inventory is 289 and the reserve inventory coefficient is 1.3, then the inventory alert threshold = 289 x 1.3 = 357.7, and the integer is 358. When the current inventory is less than 358, the restocking mechanism is triggered.

[0040] In the above manner, the inventory alert threshold varies dynamically with the "predicted sales volume", "predicted restocking time" and "product category", achieving flexibility and accuracy of inventory management.

[0041] In step S400 of the embodiments of the present application, the current inventory of the target sales area is obtained, and it is judged whether the current inventory is lower than the inventory alert threshold, and a restocking plan is configured according to the judgment result, including: The current inventory of the target sales area is obtained; If the current inventory is not lower than the inventory alert threshold, no restocking is performed; If the current inventory is lower than the inventory alert threshold, a restocking plan is configured according to the current inventory, the predicted sales volume and the predicted restocking time, and restocking is performed.

[0042] In the embodiments of the present application, the core purpose of step S400 is to make a decision on whether to restock by comparing the current inventory with the inventory alert threshold, and to develop a corresponding restocking plan, so as to achieve dynamic matching of inventory and sales demand. Specifically, by judging whether the current inventory is lower than the alert threshold, it is determined whether to start restocking: when the inventory is sufficient, invalid restocking is avoided to reduce costs, and when the inventory is insufficient, restocking is performed in time to prevent out-of-stock, so as to ultimately ensure the rationality of the inventory and the continuity of the sales in the cross-border warehousing consignment process.

[0043] The current inventory of the target sales area is obtained, including: The inventory of the target goods in the target sales area is counted as the in-warehouse inventory, which is added to the current inventory. The quantity of the target goods in the batches being transported to the target sales area and the in-transit transportation information are counted as the in-transit inventory. The transportation speed predictor is called to predict the in-transit predicted transportation time of the in-transit inventory, and the predicted arrival time of the in-transit inventory in the target sales area is calculated based on the in-transit predicted transportation time and the in-transit transportation information. If the predicted arrival time of a batch of in-transit inventory is earlier than the predicted transportation time, the quantity of the target goods in the batch of in-transit inventory is added to the current inventory, otherwise, it is not added to the current inventory.

[0044] The inventory of the target goods in the target sales area is counted as the in-warehouse inventory, which is added to the current inventory.

[0045] Further, the quantity of the target goods in the batches being transported to the target sales area and the in-transit transportation information are counted as the in-transit inventory. First, all in-transit transportation information of the target goods being transported to the target sales area is collected, and the in-transit transportation information corresponding to the in-transit inventory is obtained, i.e., the transportation mode (railway, highway, air transportation), the quantity of goods, and the transportation initiation date information. The trained transportation speed predictor is called to input the in-transit transportation information of each batch of in-transit inventory, obtain the in-transit predicted transportation time of each batch, and calculate the predicted arrival time.

[0046] For example, the transportation initiation date of batch A of in-transit inventory is June 20, and the predicted transportation time is 3 days, so the predicted arrival time is June 23; the transportation initiation date of batch B of in-transit inventory is June 15, and the predicted transportation time is 15 days, so the predicted arrival time is June 30.

[0047] Further, the predicted arrival time of each batch of in-transit inventory is compared with the "predicted transportation time" of the current transportation plan. If the predicted arrival time is earlier than the current predicted transportation time, it means that the batch of in-transit inventory can be in-warehouse before the current transportation arrives, and can be used to meet the sales demand, so the quantity is added to the current inventory; otherwise, it means that the batch of in-transit inventory cannot timely supplement the inventory, and is not added to the current inventory.

[0048] For example, assuming that the current predicted order delivery time is 5 days (i.e., the goods are ordered now and delivered after 5 days), if the current date is June 25, in the above example, if the predicted arrival time of batch A is June 23 (earlier than June 30), the in-transit inventory quantity of batch A, such as 500 pairs, is included in the current inventory; if the predicted arrival time of batch B is June 30 (later than June 30), the in-transit inventory quantity of batch B, such as 800 pairs, is not included in the current inventory.

[0049] If the quantity of the already-arrived inventory is 300 pairs and the in-transit inventory quantity that meets the condition is 500 pairs, then the current inventory quantity = 300 + 500 = 800 pairs.

[0050] If the current inventory quantity is lower than the inventory alert threshold, the order plan is configured according to the current inventory quantity and the minimum inventory quantity, and the goods are ordered, including: The difference between the current inventory quantity and the minimum inventory quantity is calculated as the minimum order quantity, The goods are ordered in an order quantity that is not less than the minimum order quantity.

[0051] In the embodiments of the present application, the method for calculating the minimum order quantity is to calculate the difference between the current inventory quantity and the minimum inventory quantity, and the minimum order quantity represents the quantity of goods that needs to be ordered at least to meet the continuous sales demand.

[0052] For example, if the minimum inventory quantity is 165 pieces and the current inventory quantity is 100 pieces, the minimum order quantity = 165 - 100 = 65 pieces.

[0053] The actual order quantity is determined according to the standard that is not less than the minimum order quantity. The order quantity can be equal to the minimum order quantity, or can be greater than the minimum order quantity. Factors such as preventing out-of-stock and reducing order frequency can be considered for comprehensive determination. For example, when the minimum order quantity is 65 pieces, 80 pieces can be actually ordered.

[0054] So far, the complete process of the cross-border warehousing consignment collaborative scheduling management method has been completed.

[0055] Embodiment two, as Figure 2 shown, based on the same inventive concept as the system method provided in embodiment one, the embodiments of the present application further provide a cross-border warehousing consignment collaborative scheduling management method system, which comprises: An order delivery time prediction module 11 is configured to obtain historical order logistics data of target goods in a target sales area, train an order speed predictor according to the historical order logistics data, and predict a predicted order delivery time required for current ordering; The daily sales prediction module 12 is configured to obtain historical sales data of a target product in a target sales area, train a sales data predictor based on the historical sales data, and predict a predicted daily sales volume in a preset sales time, wherein the preset sales time is longer than the predicted restocking time; The inventory threshold configuration module 13 is configured to obtain a minimum inventory based on the predicted daily sales volume and the predicted restocking time, and configure a dynamic inventory alert threshold based on the minimum inventory and a reserve inventory coefficient. The restocking plan configuration module 14 is configured to obtain a current inventory, determine whether the current inventory is lower than the inventory alert threshold, and configure a restocking plan according to the determination result.

[0056] Further, the restocking time prediction module 11 includes the following execution steps: Obtain historical restocking logistics data including restocking information and restocking time in a historical time period; Use machine learning to build the restocking speed predictor and train the restocking speed predictor with the historical restocking logistics data until convergence; Input the current restocking information into the restocking speed predictor to obtain the predicted restocking time.

[0057] Further, the daily sales prediction module 12 includes the following execution steps: Obtain historical sales data of a target product in a target sales area, including promotion information and daily sales volume; Use machine learning to build the sales data predictor and train the sales data predictor with the historical sales data until convergence; Input the current promotion information and the current date into the sales data predictor to obtain the predicted daily sales volume in a preset sales time, wherein the preset sales time is longer than the predicted restocking time.

[0058] Further, the inventory threshold configuration module 13 includes the following execution steps: According to the predicted daily sales volume in the preset sales time, calculate the total sales volume in the predicted restocking time as the minimum inventory; According to the type of the target product, configure a reserve inventory coefficient, and multiply the reserve inventory coefficient by the minimum inventory to obtain the inventory alert threshold.

[0059] Further, the restocking plan configuration module 14 includes the following execution steps: Obtain the current inventory of the target sales area; If the current inventory is not lower than the inventory alert threshold, no restocking is performed; If the current inventory is lower than the inventory alert threshold, a restocking plan is configured according to the current inventory, the predicted sales and the predicted restocking time, and restocking is performed.

[0060] The current inventory of the target sales area is obtained, including: The inventory of the target goods in the target sales area is counted as the in-warehouse inventory, and is added to the current inventory. The quantity of the target goods in the multiple batches being transported to the target sales area and the in-transit restocking information are counted as the multiple in-transit inventories. The restocking speed predictor is called to predict the in-transit predicted restocking time of the multiple in-transit inventories, and the predicted in-warehouse time of the multiple in-transit inventories arriving at the target sales area is calculated according to the in-transit predicted restocking time and the in-transit restocking information. If the predicted in-warehouse time of a batch of in-transit inventory is earlier than the predicted restocking time, the quantity of the target goods in the batch of in-transit inventory is added to the current inventory, otherwise, the quantity of the target goods in the batch of in-transit inventory is not added to the current inventory.

[0061] If the current inventory is lower than the inventory alert threshold, a restocking plan is configured according to the current inventory and the minimum inventory, and restocking is performed, including: The difference between the current inventory and the minimum inventory is calculated as the minimum restocking quantity. Restocking is performed according to a restocking quantity that is not less than the minimum restocking quantity.

[0062] It should be noted that the description of each of the above embodiments focuses on different aspects, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0063] Those skilled in the art should understand that the embodiments of the present application can provide a method, a system or a computer program product. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0064] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts.

[0065] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts.

[0066] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts.

[0067] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments described and shown, without departing from the spirit and scope of the application.

[0068] Obviously, many modifications and variations of the present application are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims, the application can be practiced otherwise than as specifically described.

Claims

1. A cross-border warehousing consignment collaborative scheduling management method, characterized in that: The method comprises: Obtain historical cargo transfer logistics data for the target product in the target sales area, and train a cargo transfer speed predictor based on the historical cargo transfer logistics data to predict the cargo transfer time required to obtain the current cargo transfer; Obtain historical sales data of the target product in the target sales area, train a sales data predictor based on the historical sales data, and predict the daily sales volume within a preset sales period, wherein the length of the preset sales period is greater than the predicted stock transfer time; Based on the predicted sales volume and the predicted time required to transfer goods, a minimum inventory level is obtained, and a reserve inventory coefficient is configured; based on the minimum inventory level and the reserve inventory coefficient, a dynamic inventory alert threshold is configured; Obtain the current inventory level of the target sales area, determine whether the current inventory level is lower than the inventory warning threshold, and configure a stock transfer plan based on the determination result.

2. The cross-border warehousing consignment collaborative scheduling management method according to claim 1 is characterized in that: Obtain historical cargo transfer logistics data for the target sales area, train a cargo transfer speed predictor based on the historical cargo transfer logistics data, and predict the cargo transfer time required for the current cargo transfer, including: Obtain historical cargo transfer logistics data within a historical period, including cargo transfer information and transfer time; Using machine learning, the cargo transfer speed predictor is built, and the cargo transfer speed predictor is trained using historical cargo transfer logistics data until convergence; The current transfer information is input into the transfer speed predictor to obtain the predicted transfer time.

3. The cross-border warehousing consignment collaborative scheduling management method according to claim 1 is characterized in that: Obtain historical sales data for the target product in the target sales area, train a sales data predictor based on the historical sales data, and predict the daily sales volume within a preset sales period, including: Obtain historical sales data for target products in target sales areas, including promotional information and daily sales volume; Using machine learning to build the sales data predictor, and training the sales data predictor using historical sales data until convergence; The current promotion information and the current date are input into the sales data forecaster to obtain the predicted sales volume for each day within the preset sales time, wherein the length of the preset sales time is greater than the predicted stock transfer time.

4. The cross-border warehousing consignment collaborative scheduling management method according to claim 1 is characterized in that: Based on the predicted sales volume and the predicted time required to transfer goods, a minimum inventory level is obtained, and a reserve inventory coefficient is configured. Based on the minimum inventory level and the reserve inventory coefficient, a dynamic inventory alert threshold is configured, including: Obtain the daily sales forecast within the preset sales period based on the forecast, and calculate the total sales volume within the predicted time frame for stock transfers as the minimum inventory level; According to the target commodity type, a reserve inventory coefficient is configured, and the product of the reserve inventory coefficient and the minimum inventory quantity is used as the inventory warning threshold.

5. The cross-border warehousing consignment collaborative scheduling management method according to claim 1 is characterized in that: Obtain the current inventory level of the target sales area, determine whether the current inventory level is lower than the inventory warning threshold, and configure a stock transfer plan based on the determination result, including: Get the current inventory quantity in the target sales area; If the current inventory level is not lower than the inventory warning threshold, no stock transfer will be performed; If the current inventory level is lower than the inventory warning threshold, a stock transfer plan is configured based on the current inventory level, predicted sales volume, and predicted stock transfer time, and the stock transfer is carried out.

6. The cross-border warehousing consignment collaborative scheduling management method according to claim 5 is characterized in that: Get the current inventory quantity in the target sales area, including: Count the inventory of the target products in the current target sales area and include it in the current inventory as the inventory received; Count the quantity of target goods being transferred to target sales areas and in-transit transfer information for multiple batches as in-transit inventory; Retrieving the goods transfer speed predictor, predicting the in-transit predicted goods transfer time of the multiple batches of in-transit inventory, and calculating the predicted warehousing time for the multiple batches of in-transit inventory to arrive at the target sales area based on the in-transit predicted goods transfer time and the in-transit goods transfer information; When the predicted warehousing time of a batch of in-transit inventory is earlier than the predicted transfer time, the quantity of the target goods of this batch of in-transit inventory is included in the current inventory; otherwise, it is not included in the current inventory.

7. The cross-border warehousing consignment collaborative scheduling management method according to claim 5 is characterized in that: If the current inventory level is lower than the inventory warning threshold, a stock transfer plan is configured based on the current inventory level and the minimum inventory level, and stock transfers are carried out, including: Calculate the difference between the current inventory and the minimum inventory as the minimum transfer quantity. Goods shall be transferred in accordance with a transfer quantity that is not less than the stated minimum transfer quantity.

8. A cross-border warehousing consignment collaborative scheduling management method system, characterized in that: The system comprises: A goods transfer time prediction module is used to obtain historical goods transfer logistics data of the target goods in the target sales area, train a goods transfer speed predictor based on the historical goods transfer logistics data, and predict the goods transfer time required for the current goods transfer; A daily sales forecasting module is used to obtain historical sales data of the target product in the target sales area, train a sales data forecaster based on the historical sales data, and predict the daily sales volume within a preset sales period, wherein the length of the preset sales period is greater than the predicted stock transfer time; An inventory threshold configuration module, configured to obtain a minimum inventory level based on the predicted sales volume and the predicted inventory transfer time, configure a reserve inventory coefficient, and configure a dynamic inventory warning threshold based on the minimum inventory level and the reserve inventory coefficient; The stock transfer plan configuration module is used to obtain the current inventory level, determine whether the current inventory level is lower than the inventory warning threshold, and configure the stock transfer plan based on the judgment result.

Images