Dynamic warehouse-dividing logistics optimization method and system based on multi-source data and intelligent algorithm

Abstract

The invention relates to the technical field of logistics, in particular to a dynamic warehouse-dividing logistics optimization method and system based on multi-source data and an intelligent algorithm. Comprising the following steps: S1, obtaining the related data of the sub-warehouse inventory, and carrying out the state classification of each sub-warehouse according to the related data of the sub-warehouse inventory; s2, obtaining the first conversion capability of each sub-warehouse, and obtaining the supply score of each sub-warehouse for the conventional commodities by using a conventional commodity scoring formula according to the first conversion capability and the inventory related data of each sub-warehouse; and S3, obtaining the second conversion capability of each sub-warehouse, and obtaining the supply score of each sub-warehouse for the blind box commodities by using a blind box commodity scoring formula according to the second conversion capability and the inventory related data of each sub-warehouse. According to the method, the inventory data of the conventional commodities and the blind box commodities in each sub-warehouse are obtained, and the warehouse state is divided into four inventory states according to the preset standard number, so that an accurate basis is provided for subsequent decision making.

Description

Technical Field

[0001] The present invention relates to the field of logistics technology, and in particular to a dynamic warehouse distribution logistics optimization method and system based on multi-source data and intelligent algorithms. Background Art

[0002] With the rapid development of e-commerce and new retail models, consumers' personalized and diversified demands for goods are continuously increasing. The traditional single-warehouse model has been difficult to meet the requirements of real-time, efficient, and accurate distribution. Currently, most logistics systems mainly rely on static warehouse distribution strategies, simply mapping user requests to the nearest warehouse for shipment, lacking comprehensive consideration of multi-source information such as the real-time inventory status of each warehouse and user preferences. Moreover, with the continuous innovation of the e-commerce model, blind box products have become an important means to attract users. However, blind boxes and regular products have different characteristics in inventory management and logistics warehouse distribution. When users choose blind boxes, there is randomness and unpredictability, making it difficult to accurately schedule through traditional deterministic replenishment strategies; the inventory levels of regular products and blind box products in the warehouse fluctuate violently with the sales rhythm. If not balanced in time, it is easy to lead to out-of-stock or overstock. Summary of the Invention

[0003] In order to overcome the drawback of lacking consideration for the warehouse distribution and shipment of regular products and blind box products, the present invention provides a dynamic warehouse distribution logistics optimization method and system based on multi-source data and intelligent algorithms.

[0004] The technical implementation solution of the present invention is: a dynamic warehouse distribution logistics optimization method based on multi-source data and intelligent algorithms, including the following steps: S1: Obtain data related to the warehouse inventory, and classify the status of each warehouse according to the data related to the warehouse inventory; S2: Obtain the first conversion ability of each warehouse, and use the regular product scoring formula according to the first conversion ability and the data related to the warehouse inventory of each warehouse to obtain the supply score of each warehouse for regular products; S3: Obtain the second conversion ability of each warehouse, and use the blind box product scoring formula according to the second conversion ability and the data related to the warehouse inventory of each warehouse to obtain the supply score of each warehouse for blind box products; S4: According to the warehouse status of each warehouse, the supply score of each warehouse for regular products, and the supply score of each warehouse for blind box products, use the warehouse goods decision formula to obtain the warehouse goods score; S5: Obtain the warehouse recommendation value according to the data related to the warehouse inventory and the warehouse goods score of each warehouse, and make a warehouse shipment recommendation according to the warehouse recommendation value.

[0005] Preferably, obtaining the data related to the inventory of each sub-warehouse and classifying the status of each sub-warehouse according to the data related to the inventory of each sub-warehouse includes: obtaining the data related to the inventory of each sub-warehouse, where the data related to the inventory of each sub-warehouse includes the quantity of regular goods in the sub-warehouse, the standard quantity of regular goods in the sub-warehouse, the quantity of blind box goods in the sub-warehouse, the standard quantity of blind box goods in the sub-warehouse, and the location data of the sub-warehouse, and classifying the sub-warehouse status of each sub-warehouse according to the data related to the inventory of each sub-warehouse to obtain the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status.

[0006] Preferably, classifying the sub-warehouse status of each sub-warehouse according to the data related to the inventory of each sub-warehouse to obtain the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status includes: where the first inventory status is that the quantity of regular goods in the sub-warehouse is greater than or equal to the standard quantity of regular goods in the sub-warehouse, and the quantity of blind box goods in the sub-warehouse is greater than or equal to the standard quantity of blind box goods in the sub-warehouse; the second inventory status is that the quantity of regular goods in the sub-warehouse is less than the standard quantity of regular goods in the sub-warehouse, and the quantity of blind box goods in the sub-warehouse is greater than or equal to the standard quantity of blind box goods in the sub-warehouse; the third inventory status is that the quantity of regular goods in the sub-warehouse is greater than or equal to the standard quantity of regular goods in the sub-warehouse, and the quantity of blind box goods in the sub-warehouse is less than the standard quantity of blind box goods in the sub-warehouse; the fourth inventory status is that the quantity of regular goods in the sub-warehouse is less than the standard quantity of regular goods in the sub-warehouse, and the quantity of blind box goods in the sub-warehouse is less than the standard quantity of blind box goods in the sub-warehouse.

[0007] Preferably, obtaining the first conversion ability of each sub-warehouse and using the regular goods scoring formula according to the first conversion ability and the data related to the inventory of each sub-warehouse to obtain the supply score of each sub-warehouse for regular goods includes: the first conversion ability is the maximum conversion ability of the sub-warehouse to convert blind box goods into regular goods, and using the regular goods scoring formula to obtain the regular goods supply score of the i-th sub-warehouse, where the regular goods scoring formula is: ; In the formula, is the supply score of the i-th sub-warehouse for regular goods; is the quantity of regular goods in the i-th sub-warehouse; is the standard quantity of regular goods in the i-th sub-warehouse; p is the user replacement preference weight, and the value range is [0, 1]; is the maximum conversion ability of the i-th sub-warehouse to convert blind box goods into regular goods; is the first inventory status of the i-th sub-warehouse; is the second inventory status of the i-th sub-warehouse; is the third inventory status of the i-th sub-warehouse; is the fourth inventory status of the i-th sub-warehouse.

[0008] Preferably, the p is the user replacement preference weight, including: obtaining the user's historical replacement product data and real-time replacement product data, using the historical replacement product data as training data to train a Transformer model to obtain a preference weight model, and inputting the real-time replacement product data into the preference weight model to obtain the user's user replacement preference weight, where the larger p is, the more the user prefers to replace regular products with blind box shipments, and the larger 1-p is, the more the user prefers to replace blind boxes with regular product shipments.

[0009] Preferably, obtaining the second conversion ability of each distribution center, and using the blind box product scoring formula according to the second conversion ability and the relevant inventory data of each distribution center to obtain the supply score of each distribution center for blind box products, including: the second conversion ability is the maximum conversion ability of the distribution center to convert regular products into blind box products, and using the blind box product scoring formula to obtain the blind box product supply score of the i-th distribution center, where the blind box product scoring formula is: ; In the formula, is the supply score of the i-th distribution center for blind box products; is the number of blind box products in the i-th distribution center; is the standard number of blind box products in the i-th distribution center; p is the user replacement preference weight, and the value range is [0,1]; is the maximum conversion ability of the i-th distribution center to convert regular products into blind box products; is the first inventory status of the i-th distribution center; is the second inventory status of the i-th distribution center; is the third inventory status of the i-th distribution center; is the fourth inventory status of the i-th distribution center.

[0010] Preferably, according to the distribution center status of each distribution center, the supply score of each distribution center for regular products, and the supply score of each distribution center for blind box products, using the distribution center goods decision formula to obtain the distribution center goods score, including: the distribution center status is the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status, where the distribution center goods decision formula is: ; In the formula, is the distribution center goods score of the i-th distribution center; is the weight coefficient of regular products; is the weight coefficient of blind box products; is the supply score of the i-th distribution center for regular products; is the supply score of the i-th distribution center for blind box products.

[0011] Preferably, obtaining a recommended value for each warehouse based on the relevant data of the warehouse inventory and the scores of the goods in each warehouse, and making a recommendation for warehouse shipment according to the recommended value for each warehouse, includes: obtaining the recommended value for each warehouse by using a warehouse recommendation formula based on the relevant data of the warehouse inventory and the scores of the goods in each warehouse, and performing a shipment process for the target user for the warehouse whose recommended value for each warehouse is greater than a preset threshold according to the recommended value for each warehouse.

[0012] Preferably, obtaining the recommended value for each warehouse by using a warehouse recommendation formula based on the relevant data of the warehouse inventory and the scores of the goods in each warehouse, includes: The warehouse recommendation formula is: ; In the formula, is the recommended value for each warehouse; is the distance between the target user and the i-th warehouse; is the score of the goods in the i-th warehouse; cost is the cost of shipping from the i-th warehouse to the target user; , are the weight adjustment coefficients of the warehouse recommendation formula; is the adjustment parameter of the warehouse recommendation formula.

[0013] A dynamic warehouse logistics optimization system based on multi-source data and intelligent algorithms further includes: A data processing module, configured to obtain relevant data of the warehouse inventory and classify the status of each warehouse according to the relevant data of the warehouse inventory; A status classification module, configured to classify the warehouse status of each warehouse according to the relevant data of the warehouse inventory, and obtain a first inventory status, a second inventory status, a third inventory status, and a fourth inventory status; A first score calculation module, configured to obtain the first conversion ability of each warehouse and obtain the supply score of each warehouse for regular goods by using a regular commodity scoring formula based on the first conversion ability and the relevant data of the warehouse inventory of each warehouse; A preference weight calculation module, configured to input real-time replacement data into a preference weight model to obtain the user's replacement preference weight; A second score calculation module, configured to obtain the second conversion ability of each warehouse and obtain the supply score of each warehouse for blind box goods by using a blind box commodity scoring formula based on the second conversion ability and the relevant data of the warehouse inventory of each warehouse; A final score calculation module, configured to obtain the score of the goods in the warehouse by using a warehouse goods decision formula based on the warehouse status of each warehouse, the supply score of each warehouse for regular goods, and the supply score of each warehouse for blind box goods; A shipment recommendation module, configured to obtain a recommended value for each warehouse based on the relevant data of the warehouse inventory and the scores of the goods in each warehouse, and make a recommendation for warehouse shipment according to the recommended value for each warehouse.

[0014] Beneficial effects

[0015] 1. The present invention obtains the inventory data of regular products and blind box products in each sub-warehouse, and divides the warehouse status into four inventory statuses according to a preset standard quantity, so as to immediately grasp the inventory health of each sub-warehouse and provide an accurate basis for subsequent decision-making. 2. The first and second conversion ability indicators are introduced to measure the ability of the sub-warehouse to convert blind box products into regular products and regular products into blind box products respectively. By using the regular product scoring formula or the blind box product scoring formula, the supply scores of regular products and blind box products are obtained, and the sub-warehouse shipping strategy is dynamically adjusted to improve inventory utilization and flexibility. 3. Based on the training of historical and real-time replacement product data by the Transformer model, the user replacement preference weights are output in real time, and the supply strategy can be intelligently adjusted according to the user's preference for regular products and blind box products, enhancing user satisfaction and increasing the repurchase rate. 4. By using the sub-warehouse goods decision formula and the sub-warehouse recommendation formula, a balance is achieved among the supply capacity score, logistics cost, and distribution distance, significantly reducing the logistics cost, shortening the distribution time limit, and improving the overall distribution efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is the method flow chart of the dynamic sub-warehouse logistics optimization of the present invention; Figure 2 is the system structure schematic diagram of the dynamic sub-warehouse logistics optimization of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following further illustrates the above solution with specific embodiments. It should be understood that these embodiments are used to illustrate the present application and are not limited to restricting the scope of the present application. The implementation conditions adopted in the embodiments can be further adjusted according to the conditions of specific manufacturers, and the implementation conditions not specified are usually the conditions in conventional experiments.

[0018] Embodiment 1: A dynamic sub-warehouse logistics optimization method based on multi-source data and intelligent algorithms, as Figure 1 shown, includes the following steps: S1: Obtain the sub-warehouse inventory related data, and classify the status of each sub-warehouse according to the sub-warehouse inventory related data; Obtain the sub-warehouse inventory related data, where the sub-warehouse inventory related data includes the quantity of regular products in the sub-warehouse, the standard quantity of regular products in the sub-warehouse, the quantity of blind box products in the sub-warehouse, the standard quantity of blind box products in the sub-warehouse, and the sub-warehouse location data. Classify the sub-warehouse status of each sub-warehouse to obtain the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status.

[0019] It should be noted that by calling the inventory query interface from the warehouse management system interface of each sub-warehouse, the following data fields are obtained: the quantity of regular goods in the sub-warehouse, the standard quantity of regular goods in the sub-warehouse, the quantity of blind box goods in the sub-warehouse, the standard quantity of blind box goods in the sub-warehouse, and the sub-warehouse location data; among them, the standard quantity of regular goods in the sub-warehouse and the standard quantity of blind box goods in the sub-warehouse are obtained through historical experience and historical test data.

[0020] Among them, the first inventory status is that the quantity of regular goods in the sub-warehouse is greater than or equal to the standard quantity of regular goods in the sub-warehouse, and the quantity of blind box goods in the sub-warehouse is greater than or equal to the standard quantity of blind box goods in the sub-warehouse; The second inventory status is that the quantity of regular goods in the sub-warehouse is less than the standard quantity of regular goods in the sub-warehouse, and the quantity of blind box goods in the sub-warehouse is greater than or equal to the standard quantity of blind box goods in the sub-warehouse; The third inventory status is that the quantity of regular goods in the sub-warehouse is greater than or equal to the standard quantity of regular goods in the sub-warehouse, and the quantity of blind box goods in the sub-warehouse is less than the standard quantity of blind box goods in the sub-warehouse; The fourth inventory status is that the quantity of regular goods in the sub-warehouse is less than the standard quantity of regular goods in the sub-warehouse, and the quantity of blind box goods in the sub-warehouse is less than the standard quantity of blind box goods in the sub-warehouse.

[0021] It should be noted that the first inventory status : greater than or equal to ; greater than or equal to , example: if the sub-warehouse = 1200, = 1000; = 700, = 500; then the condition is met and it is determined as ; at this time, both the regular goods and blind box goods in this sub-warehouse are in sufficient supply and are given priority to undertake the delivery of various orders; The second inventory status : less than ; greater than or equal to , example: if the sub-warehouse = 800, = 1000; = 700, = 500; then the condition is met and it is determined as ; at this time, the inventory of regular goods in this sub-warehouse is low and the inventory of blind boxes is sufficient. Consider giving priority to allocating blind box goods or considering the conversion strategy from blind box goods to regular goods; The third inventory status : greater than or equal to ; less than , example: if the sub-warehouse = 1200, = 1000; = 400, = 500; Then the conditions are met and it is determined as ; At this time, the inventory of regular products in this sub - warehouse is sufficient, while the inventory of blind boxes is insufficient. Consider preferentially allocating regular products or triggering the conversion strategy from regular products to blind box products; The fourth inventory status : Less than ; Less than , Example: If the sub - warehouse = 800, = 1000; = 400, = 500; Then the conditions are met and it is determined as , indicating that the inventory of both types of products in this sub - warehouse is insufficient. The delivery priority should be appropriately reduced in subsequent decisions, or the inventory should be replenished.

[0022] S2: Obtain the first conversion ability of each sub - warehouse, and use the regular product scoring formula based on the first conversion ability and the inventory - related data of each sub - warehouse to obtain the supply scores of each sub - warehouse for regular products; The first conversion ability is the maximum conversion ability of the sub - warehouse to convert blind box products into regular products. Use the regular product scoring formula to obtain the regular product supply score of the i - th sub - warehouse, where the regular product scoring formula is: ; In the formula, is the supply score of the i - th sub - warehouse for regular products; is the quantity of regular products in the i - th sub - warehouse; is the standard quantity of regular products in the i - th sub - warehouse; p is the user replacement preference weight, and its value range is [0, 1]; is the maximum conversion ability of the i - th sub - warehouse to convert blind box products into regular products; is the first inventory status of the i - th sub - warehouse; is the second inventory status of the i - th sub - warehouse; is the third inventory status of the i - th sub - warehouse; is the fourth inventory status of the i - th sub - warehouse.

[0023] It should be noted that the maximum conversion ability of each sub - warehouse to convert blind box products into regular products is read from the sub - warehouse intelligent operation module, and the supply scores of each sub - warehouse for regular products are obtained using the regular product scoring formula.

[0024] Obtain the historical replacement data and real-time replacement data of the user. Use the historical replacement data as training data to train a Transformer model to obtain a preference weight model. Input the real-time replacement data into the preference weight model to obtain the user's replacement preference weight of the user, where the larger p is, the more the user prefers to replace regular goods with blind box shipments, and the larger 1 - p is, the more the user prefers to replace blind boxes with regular goods shipments.

[0025] It should be noted that the historical replacement data: collect the "regular ⇄ blind box" replacement requests submitted by users among each distribution center in the past 3 months, including user ID, order time, replacement direction, and success rate; real-time replacement data: the real-time incoming daily replacement operation log, including the user's current intention and the product categories before and after replacement; construct a Transformer model with the historical replacement data as the training set. After training, obtain a preference weight prediction model. Input the feature vector of the real-time replacement data into the above preference weight model, and the model outputs the user's replacement preference weight. When p is larger, it means that the user is more inclined to replace "regular goods with blind box goods" (that is, prefers blind box shipments); when 1 - p is larger, it means that the user is more willing to replace "blind box goods with regular goods" (that is, prefers regular shipments).

[0026] S3: Obtain the second conversion ability of each distribution center, and use the blind box product scoring formula according to the second conversion ability and the relevant inventory data of each distribution center to obtain the supply score of each distribution center for blind box products; The second conversion ability is the maximum conversion ability of the distribution center to convert regular goods into blind box goods. Use the blind box product scoring formula to obtain the blind box product supply score of the i-th distribution center, where the blind box product scoring formula is: ; In the formula, is the supply score of the i-th distribution center for blind box products; is the number of blind box products in the i-th distribution center; is the standard number of blind box products in the i-th distribution center; p is the user replacement preference weight, and the value range is [0, 1]; is the maximum conversion ability of the i-th distribution center to convert regular goods into blind box goods; is the first inventory status of the i-th distribution center; is the second inventory status of the i-th distribution center; is the third inventory status of the i-th distribution center; is the fourth inventory status of the i-th distribution center.

[0027] It should be noted that the maximum conversion ability of each sub-warehouse to convert regular goods into blind box goods is read from the sub-warehouse intelligent operation module, and the blind box goods score of each sub-warehouse is obtained using the blind box goods scoring formula; the actual conversion records of "blind box goods to regular goods" and "regular goods to blind box goods" in the past 6-12 months are exported from the warehouse management system and the order management system, including conversion batches, conversion quantities, conversion success rates, time consumption, and costs; the packaging line capacity, personnel proficiency, and equipment status of each sub-warehouse are regularly evaluated to obtain the maximum conversion volume supported per unit time; thus, the first conversion ability and the second conversion ability are obtained.

[0028] S4: According to the sub-warehouse status of each sub-warehouse, the supply score of each sub-warehouse for regular goods, and the supply score of each sub-warehouse for blind box goods, the sub-warehouse goods score is obtained using the sub-warehouse goods decision formula; The sub-warehouse status is the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status, and the sub-warehouse goods decision formula is: ; In the formula, is the sub-warehouse goods score of the i-th sub-warehouse; is the weight coefficient of regular goods; is the weight coefficient of blind box goods; is the supply score of the i-th sub-warehouse for regular goods; is the supply score of the i-th sub-warehouse for blind box goods.

[0029] It should be noted that by combining the calculated scores of regular goods and blind box goods, the process of "obtaining the sub-warehouse goods score using the sub-warehouse goods decision formula according to the sub-warehouse status of each sub-warehouse, the supply score of each sub-warehouse for regular goods, and the supply score of each sub-warehouse for blind box goods" is explained. The supply scores of the two types of goods, regular and blind box, are linearly combined according to the preset weight coefficients to generate the sub-warehouse goods score; as the key basis for subsequent sub-warehouse recommendation and shipping decisions, it realizes the dynamic optimization scheduling of multi-source data and intelligent algorithms.

[0030] S5: Obtain the sub-warehouse recommendation value based on the sub-warehouse inventory-related data and the sub-warehouse goods score, and make a sub-warehouse shipping recommendation according to the sub-warehouse recommendation value.

[0031] According to the sub-warehouse inventory-related data and the sub-warehouse goods score, the sub-warehouse recommendation value is obtained using the sub-warehouse recommendation formula, and according to the sub-warehouse recommendation value, the sub-warehouses with sub-warehouse recommendation values greater than the preset threshold are used to ship goods to the target users.

[0032] It should be noted that the distance between the sub-warehouse and the target user is calculated by the geographic information system; the average value of the actual consumption on highways and routes in each region is taken as the cost from the sub-warehouse to the target user according to the fixed value of the cost per unit distance; the sub-warehouse recommendation value is obtained using the sub-warehouse recommendation formula, and the recommendation value of each sub-warehouse is compared with the preset threshold. If the recommendation value of the sub-warehouse is greater than the preset threshold, the sub-warehouse is included in the shipping candidate list for the target user; if the recommendation value of the sub-warehouse is less than or equal to the preset threshold, the sub-warehouse for shipping is not considered for the time being.

[0033] The sub-warehouse recommendation formula is: ; In the formula, is the sub-warehouse recommendation value; is the distance between the target user and the i-th sub-warehouse; is the score of the goods in the i-th sub-warehouse; cost is the cost from the i-th sub-warehouse to the target user; 、 are the weight adjustment coefficients of the sub-warehouse recommendation formula; is the adjustment parameter of the sub-warehouse recommendation formula.

[0034] It should be noted that based on the scores of regular and blind box supplies, combined with logistics costs and distance factors, the intuitive sub-warehouse recommendation value is calculated using the sub-warehouse recommendation formula, providing a quantitative basis for the final shipping decision.

[0035] Example 2: On the basis of Example 1, a dynamic sub-warehouse logistics optimization method based on multi-source data and intelligent algorithms, as Figure 2 shown, further includes: A data processing module, configured to obtain data related to the sub-warehouse inventory and classify the status of each sub-warehouse according to the data related to the sub-warehouse inventory; A status classification module, configured to classify the sub-warehouse status of each sub-warehouse according to the data related to the sub-warehouse inventory, and obtain the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status; A first score calculation module, configured to obtain the first conversion ability of each sub-warehouse and use the regular commodity scoring formula according to the first conversion ability and the data related to the sub-warehouse inventory of each sub-warehouse to obtain the supply scores of each sub-warehouse for regular commodities; A preference weight calculation module, configured to input the real-time replacement data into the preference weight model to obtain the user replacement preference weight of the user; A second score calculation module, configured to obtain the second conversion ability of each sub-warehouse and use the blind box commodity scoring formula according to the second conversion ability and the data related to the sub-warehouse inventory of each sub-warehouse to obtain the supply scores of each sub-warehouse for blind box commodities; The final score calculation module is used to obtain the score of the goods in the bin using the bin goods decision formula based on the bin status of each bin, the supply scores of each bin for regular goods, and the supply scores of each bin for blind box goods; The delivery recommendation module is used to obtain the bin recommendation value based on the bin inventory related data and the scores of the goods in each bin, and make bin delivery recommendations according to the bin recommendation value.

[0036] As mentioned above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms, characterized in that, Including the following steps: S1: Obtain the data related to the inventory of each sub-warehouse, and classify the status of each sub-warehouse according to the data related to the inventory of each sub-warehouse; S2: Obtain the first conversion ability of each sub-warehouse, and use the conventional commodity scoring formula according to the first conversion ability and the data related to the inventory of each sub-warehouse to obtain the supply scores of each sub-warehouse for conventional commodities; S3: Obtain the second conversion ability of each sub-warehouse, and use the blind box commodity scoring formula according to the second conversion ability and the data related to the inventory of each sub-warehouse to obtain the supply scores of each sub-warehouse for blind box commodities; S4: According to the sub-warehouse status, the supply scores of each sub-warehouse for conventional commodities, and the supply scores of each sub-warehouse for blind box commodities, use the sub-warehouse goods decision formula to obtain the sub-warehouse goods scores; S5: Obtain the sub-warehouse recommendation values according to the data related to the inventory of each sub-warehouse and the sub-warehouse goods scores, and make sub-warehouse shipping recommendations according to the sub-warehouse recommendation values.

2. The dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms according to claim 1, wherein, The obtaining of the data related to the inventory of each sub-warehouse and the classification of the status of each sub-warehouse according to the data related to the inventory of each sub-warehouse include: obtaining the data related to the inventory of each sub-warehouse, where the data related to the inventory of each sub-warehouse includes the quantity of conventional commodities in the sub-warehouse, the standard quantity of conventional commodities in the sub-warehouse, the quantity of blind box commodities in the sub-warehouse, the standard quantity of blind box commodities in the sub-warehouse, and the sub-warehouse location data, and classifying the sub-warehouse status of each sub-warehouse according to the data related to the inventory of each sub-warehouse to obtain the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status.

3. The dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms according to claim 2, characterized in that, The classification of the sub-warehouse status of each sub-warehouse according to the data related to the inventory of each sub-warehouse to obtain the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status includes: Where the first inventory status is that the quantity of conventional commodities in the sub-warehouse is greater than or equal to the standard quantity of conventional commodities in the sub-warehouse, and the quantity of blind box commodities in the sub-warehouse is greater than or equal to the standard quantity of blind box commodities in the sub-warehouse; The second inventory status is that the quantity of conventional commodities in the sub-warehouse is less than the standard quantity of conventional commodities in the sub-warehouse, and the quantity of blind box commodities in the sub-warehouse is greater than or equal to the standard quantity of blind box commodities in the sub-warehouse; The third inventory status is that the quantity of conventional commodities in the sub-warehouse is greater than or equal to the standard quantity of conventional commodities in the sub-warehouse, and the quantity of blind box commodities in the sub-warehouse is less than the standard quantity of blind box commodities in the sub-warehouse; The fourth inventory status is that the quantity of conventional commodities in the sub-warehouse is less than the standard quantity of conventional commodities in the sub-warehouse, and the quantity of blind box commodities in the sub-warehouse is less than the standard quantity of blind box commodities in the sub-warehouse.

4. A dynamic warehousing logistics optimization method based on multi-source data and intelligent algorithms according to claim 1, characterized in that Obtaining the first conversion ability of each sub-warehouse, and using a conventional commodity scoring formula based on the first conversion ability and relevant data of the inventory in each sub-warehouse to obtain the supply scores of each sub-warehouse for conventional commodities, including: the first conversion ability is the maximum conversion ability of the sub-warehouse to convert blind box commodities into conventional commodities, and the conventional commodity cargo score of the i-th sub-warehouse is obtained using the conventional commodity scoring formula, where the conventional commodity scoring formula is: ; Wherein, is the supply score of the i-th bin for regular products; is the quantity of regular products in the i-th bin; is the standard quantity of regular products in the i-th bin; p is the user replacement preference weight, and its value range is [0, 1]; is the maximum conversion ability of the i-th bin to convert blind box products into regular products; is the first inventory status of the i-th bin; is the second inventory status of the i-th bin; is the third inventory status of the i-th bin; is the fourth inventory status of the i-th bin.

5. The dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms according to claim 4, characterized in that The p is the user replacement preference weight, including: obtaining the historical replacement data and real-time replacement data of the user, using the historical replacement data as training data to train the Transformer model to obtain the preference weight model, inputting the real-time replacement data into the preference weight model to obtain the user replacement preference weight of the user, where the larger p is, the more the user prefers to replace conventional commodities with blind box shipments, and the larger 1 - p is, the more the user prefers to replace blind boxes with conventional commodity shipments.

6. The dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms according to claim 1, wherein, Obtaining the second conversion ability of each sub-warehouse, and using the blind box product scoring formula based on the second conversion ability and the relevant data of the inventory of each sub-warehouse to obtain the supply score of each sub-warehouse for the blind box product, including: the second conversion ability is the maximum conversion ability of the sub-warehouse to convert regular products into blind box products, and the blind box product cargo score of the i-th sub-warehouse is obtained using the blind box product scoring formula, where the blind box product scoring formula is: ; Wherein, is the supply score of the blind box products for the i-th bin; is the number of blind box products in the i-th bin; is the standard number of blind box products in the i-th bin; p is the user replacement preference weight, and the value range is [0, 1]; is the maximum conversion ability of converting regular products into blind box products in the i-th bin; is the first inventory status of the i-th bin; is the second inventory status of the i-th bin; is the third inventory status of the i-th bin; is the fourth inventory status of the i-th bin.

7. A dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms according to claim 1, characterized in that Obtaining the score of goods in the warehouse using the formula for decision-making on goods in the warehouse according to the status of each warehouse compartment, the supply score of each warehouse compartment for regular goods, and the supply score of each warehouse compartment for blind box goods, including: the status of the warehouse compartment is the first inventory status, the second inventory status, the third inventory status, and the fourth inventory status, where the formula for decision-making on goods in the warehouse is: ; Wherein, is the score of the goods in the i-th bin; is the weight coefficient of regular commodities; is the weight coefficient of blind box commodities; is the supply score of the i-th bin for regular commodities; is the supply score of the i-th bin for blind box commodities.

8. A dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms according to claim 1, characterized in that Obtaining a warehouse recommendation value based on the relevant data of the warehouse inventory and the scores of goods in each warehouse, and performing warehouse shipping recommendations according to the warehouse recommendation value, including: using a warehouse recommendation formula to obtain a warehouse recommendation value based on the relevant data of the warehouse inventory and the scores of goods in each warehouse, and according to the warehouse recommendation value, performing shipping processing on the target user for the warehouses whose warehouse recommendation values are greater than a preset threshold.

9. The dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms according to claim 8, characterized in that The obtaining of the bin recommendation value using the bin recommendation formula based on the relevant data of the bin inventory and the scores of the goods in each bin includes: The bin recommendation formula is: ; Wherein, is the recommended value of the sub-warehouse; is the distance between the target user and the i-th sub-warehouse; is the score of the goods in the i-th sub-warehouse; cost is the cost from the i-th sub-warehouse to the target user; 、 are the weight adjustment coefficients of the sub-warehouse recommendation formula; is the adjustment parameter of the sub-warehouse recommendation formula.

10. A dynamic warehouse allocation logistics optimization system based on multi-source data and intelligent algorithms, according to any one of claims 1-9, a dynamic warehouse allocation logistics optimization method based on multi-source data and intelligent algorithms, characterized in that, Further including: A data processing module, configured to obtain relevant data of the warehouse inventory and classify the status of each warehouse according to the relevant data of the warehouse inventory; A status classification module, configured to classify the warehouse status of each warehouse according to the relevant data of the warehouse inventory to obtain a first inventory status, a second inventory status, a third inventory status, and a fourth inventory status; A first score calculation module, configured to obtain the first conversion ability of each warehouse and use a regular commodity scoring formula to obtain the supply scores of each warehouse for regular commodities according to the first conversion ability and the relevant data of the warehouse inventory of each warehouse; A preference weight calculation module, configured to input real-time replacement product data into a preference weight model to obtain the user replacement preference weight of the user; A second score calculation module, configured to obtain the second conversion ability of each warehouse and use a blind box commodity scoring formula to obtain the supply scores of each warehouse for blind box commodities according to the second conversion ability and the relevant data of the warehouse inventory of each warehouse; A final score calculation module, configured to obtain the warehouse goods score by using a warehouse goods decision formula according to the warehouse status of each warehouse, the supply scores of each warehouse for regular commodities, and the supply scores of each warehouse for blind box commodities; A shipping recommendation module, configured to obtain a warehouse recommendation value according to the relevant data of the warehouse inventory and the scores of goods in each warehouse, and perform warehouse shipping recommendations according to the warehouse recommendation value.

Images