A warehouse logistics data processing method and system

Abstract

The present invention discloses a warehousing logistics data processing method and system, and relates to the field of logistics management technology. A warehousing logistics data processing system includes a warehousing goods incoming quality inspection module, a warehousing goods incoming distribution module, a warehousing goods logistics data storage module, a warehousing goods outbound module and a warehousing goods data analysis module. The present invention can effectively determine the storage location of goods and improve the utilization efficiency of warehouse space by extracting and analyzing the feature of the incoming warehousing goods point cloud, combining warehousing partitions and digital twin technology; integrating warehousing partitions, shelf information and logistics data, and realizing dynamic management and optimization of inventory information through updating and combination; through the analysis of the goods fluctuation quantity table and inventory information, future purchase demand can be predicted, providing guidance for the adjustment of warehousing shelves at the next warehousing, helping to reduce inventory backlogs and out-of-stock situations, and improving the flexibility of warehousing management.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and in particular to a warehousing logistics data processing method and system. Background Art

[0002] Traditional warehousing logistics often lacks planning. There is no planning for the incoming storage partitions and storage shelves for the incoming goods, which leads to the situation that the goods cannot be found when they are out of the warehouse; and when the goods are put into the warehouse, the prepared storage shelves cannot meet the current warehousing needs, causing the goods to appear in multiple places in the warehouse at one time, wasting manpower; at the same time, all warehousing logistics data is not partitioned and stored, resulting in low efficiency in querying data and finding goods, wasting manpower.

[0003] Therefore, we urgently need to provide a warehousing logistics data processing method and system for warehousing planning using digital twin technology. Summary of the invention

[0004] The present invention aims to provide a warehousing logistics data processing method and system for planning warehousing using digital twin technology, so as to improve the efficiency of warehousing goods in and out and the efficiency of searching goods.

[0005] A method for processing warehousing logistics data, comprising:

[0006] S1. Obtain the logistics data of the warehoused goods to be entered, and obtain the logistics data of the warehoused goods to be entered, which includes the goods information, consignor information, transportation information, transportation status and cost information; perform quantity verification and quality inspection on the warehoused goods to be entered, and obtain the loss situation of the warehoused goods to be entered; update the logistics data of the warehoused goods to be entered based on the loss situation of the warehoused goods to be entered, and obtain new logistics data of the warehoused goods to be entered; establish point cloud data for the warehoused goods to be entered, and obtain the point cloud of the warehoused goods to be entered;

[0007] S2. Extract features from the point cloud of the warehouse goods to be stored, obtain volume features of the warehouse goods to be stored, and determine the storage partition based on the cargo information in the logistics data of the warehouse goods to be stored; determine the storage shelves in the storage partition to which the warehouse goods to be stored belong according to the cargo information in the logistics data of the warehouse goods to be stored, the recommended storage shelves and the volume features of the warehouse goods to be stored; the recommended storage shelves are matched according to the last predicted warehouse goods purchase situation;

[0008] S3. Integrate the storage partitions and the storage shelves in the storage partitions to which the stored goods to be stored belong, the point cloud of the stored goods to be stored, and the logistics data of the stored goods to be stored to obtain the inventory information of the stored goods that have been stored; combine the inventory information of the stored goods that have been stored with the inventory information of the stored goods that have been stored for the same storage goods to obtain new inventory information of the stored goods that have been stored;

[0009] S4. Obtain outbound storage goods information, which includes the quantity and name of goods; match the corresponding storage goods with the inbound storage goods inventory information based on the name of the goods; locate the position using the storage code of the goods in the inbound storage goods inventory information, and select outbound storage goods from the first inbound storage goods in the inbound storage goods inventory information according to the quantity of goods; update the inbound storage goods inventory information based on the outbound storage goods information to obtain new inbound storage goods inventory information;

[0010] S5. Extract all outbound storage goods information from the previous period to this period and the changes in the goods quantity in the current warehouse storage goods inventory information at preset time intervals q to obtain a goods fluctuation quantity table; perform data analysis on the goods fluctuation quantity table and the warehouse storage goods inventory information of the previous x times to obtain the predicted warehouse storage goods purchase situation; adjust the idle storage shelves at the next time of storage according to the predicted warehouse storage goods purchase situation to obtain recommended storage shelves.

[0011] As a preferred technical solution of the present invention, the specific steps of step S1 include:

[0012] S11. Obtain the logistics data of the goods to be stored in the warehouse, and obtain the logistics data of the goods to be stored in the warehouse, which includes the goods information, consignor information, transportation information, transportation status, and cost information; the goods information includes the name of the goods to be stored in the warehouse, the quantity of the goods, and the unit weight of the goods; the consignor information includes the consignor and the place of shipment; the transportation information includes the transportation distance; the transportation status includes the transportation time; the cost information includes the total cost and transportation cost of the goods to be stored in the warehouse;

[0013] S12. Use RFID technology to obtain the number of warehouse goods to be stored, and obtain the real number of warehouse goods to be stored; weigh the warehouse goods to be stored, and obtain the total weight of the warehouse goods to be stored; based on the total weight of the warehouse goods to be stored, the real number of the warehouse goods to be stored, and the unit weight of the goods, it can be determined whether they are consistent; if they are not consistent, record the real number of the warehouse goods to be stored, and feedback the abnormal situation of the goods to the consignor; otherwise, no operation;

[0014] S13. Take p images of the goods to be stored in the warehouse to obtain a set of images of the goods to be stored in the warehouse; based on the image set of the goods to be stored in the warehouse and the goods quality inspection model, obtain a list of unqualified stored goods; based on the actual number of goods to be stored in the warehouse and the list of unqualified stored goods, obtain the loss of goods to be stored in the warehouse;

[0015] S14. Update the logistics data of the warehoused goods to be stored based on the loss of the warehoused goods to be stored, and obtain new logistics data of the warehoused goods to be stored;

[0016] S15. Establish point cloud data for any of the warehoused goods to be stored, and obtain a point cloud of the warehoused goods to be stored;

[0017] The cargo quality inspection model described in step S13 includes an image preprocessing layer, a feature recognition layer and a result output layer;

[0018] The image preprocessing layer is used to preprocess the image set of warehouse goods to be stored in the warehouse, and obtain the preprocessed image set of warehouse goods to be inspected;

[0019] The feature recognition layer is used to perform feature recognition on the pre-processed warehouse goods image set to be inspected and mark abnormal values ​​to obtain abnormal warehouse goods images; the abnormal warehouse goods images are marked with the locations of unqualified warehouse goods;

[0020] The result output layer is used to output abnormal warehouse goods images. Based on the abnormal warehouse goods images, the abnormal warehouse goods are removed from the warehouse goods to be stored, and a list of unqualified warehouse goods is obtained;

[0021] The YOLOv3 model is used as the basic model of the cargo inspection model. A number of labeled images of stored cargo to be inspected are collected to obtain a training set of images of stored cargo to be inspected. The training set of images of stored cargo to be inspected is input into the initial model of cargo inspection, and the model is trained with the annotation information as the target to obtain the annotation accuracy. If the annotation accuracy is within the error standard range, the trained initial model of cargo inspection is output to obtain the cargo inspection model. Otherwise, the training set of images of stored cargo to be inspected is used to continue model training.

[0022] As a preferred technical solution of the present invention, the specific steps of step S2 are:

[0023] S21. Extracting features from the point cloud of the goods to be stored in the warehouse to obtain volume features of the goods to be stored in the warehouse;

[0024] S22. Match the goods name in the goods information in the warehouse goods table to obtain the storage partition corresponding to the goods name;

[0025] S23. Based on the recommended storage shelves, the quantity of goods in the goods information and the volume characteristics of the goods to be stored in the warehouse are fitted with the digital twin technology, and the storage shelves in the storage partition to which the goods to be stored in the warehouse belong are output;

[0026] S24. Perform warehousing operations on the goods to be stored based on the storage partitions and the storage shelves in the storage partitions to which the goods to be stored belong.

[0027] As a preferred technical solution of the present invention, the specific steps of step S23 include:

[0028] S231. Obtain a point cloud image of a recommended storage shelf and obtain point cloud data of the recommended storage shelf;

[0029] S232. Perform point cloud segmentation on the recommended storage point cloud data according to the volume characteristics of the storage cargo to be stored in the warehouse, and obtain a number of point cloud storage cargo blocks based on the volume characteristics of the storage cargo to be stored in the warehouse;

[0030] S233. According to the quantity of goods in the goods information, a point cloud storage cargo block is selected from a number of point cloud storage cargo blocks based on the volume characteristics of the goods to be stored as the storage location of the goods to be stored, and according to the storage shelf to which the selected point cloud storage cargo block belongs, the storage shelf corresponding to the goods to be stored is obtained.

[0031] As a preferred technical solution of the present invention, the specific steps of step S3 include:

[0032] S31. Digitally encode the storage partitions and the storage shelves in the storage partitions to which the goods to be stored belong to obtain the goods storage code;

[0033] S32. Compress the point cloud of the goods to be stored in the warehouse to obtain a compressed point cloud of the goods;

[0034] S33. The logistics data of the warehoused goods to be stored is split to obtain the main warehouse goods inventory information and the secondary warehouse goods inventory information; the main warehouse goods inventory information includes the goods information, the goods storage code and the compressed goods point cloud; the secondary warehouse goods inventory information includes the consignor information, the transportation information, the transportation status and the cost information;

[0035] S34. The main warehouse goods inventory information and the secondary warehouse goods inventory information are partitioned and stored to obtain the warehoused warehouse goods inventory information; the warehoused warehouse goods inventory information is combined with the warehoused warehouse goods inventory information of the same warehouse goods to obtain new warehoused warehouse goods inventory information.

[0036] As a preferred technical solution of the present invention, the specific steps of step S5 include:

[0037] S51. Extract all the outbound storage goods information and the inbound storage goods inventory information from the previous period to this period at every preset time interval q to obtain the goods quantity table;

[0038] S52. Analyze the quantity of goods in the goods fluctuation quantity table and the inventory information of the goods already in storage to obtain the goods outbound rate;

[0039] S53. Input the cargo outbound rate and the inventory information of the warehoused cargo that has been in stock for the previous x times into the purchase forecast analysis model for data analysis to obtain the forecast of the purchase of warehoused cargo;

[0040] S54. Obtain the recommended storage shelves for the next warehousing according to the predicted warehouse goods purchase situation.

[0041] As a preferred technical solution of the present invention, the purchase forecast analysis model includes a data processing layer, a data fitting layer, an analysis and forecast layer, and a result output layer;

[0042] The data processing layer is used to pre-process the inventory information of the warehoused goods that have been put into storage for the previous x times, and extract the inventory information of the secondary warehoused goods and the corresponding quantity of goods in the inventory information of the warehoused goods that have been put into storage for the previous x times;

[0043] The data fitting layer is used to perform data fitting based on the inventory information of the previous x secondary storage goods in the storage goods inventory information and the corresponding quantity of goods to obtain the goods cost relationship;

[0044] The analysis and prediction layer is used to perform multi-objective optimization based on the relationship between goods costs and the goods outbound rate, and output the predicted warehouse goods incoming situation;

[0045] The result output layer is used to obtain recommended storage shelves based on the predicted storage goods purchase situation;

[0046] In the data fitting layer, data fitting is performed based on the inventory information of the previous x secondary storage goods and the quantity of goods in the warehoused storage goods inventory information, and F(r)=f(k, t, m) is obtained; f(k, t, m) is the relationship between the cost of goods, k is the distance of a single transportation, t is the time of a single transportation, m is the quantity of goods in a single transportation, and F(r) is the cost of a single transportation;

[0047] In the analysis and prediction layer, the maximum profit of goods is defined as the goal, and the quantity of goods transported in a single trip is adjusted by constraints; the quantity of goods transported in a single trip fluctuates within β-γ to β+γ; the maximum number of iterations is set; simulation calculations are performed based on the relationship between goods costs and the rate of goods leaving the warehouse to obtain the fitness; when the maximum number of iterations is reached, the quantity of goods transported in a single trip whose fitness meets the goal is output, and the predicted warehouse goods purchase situation is obtained; β represents the baseline value of the quantity of goods transported in a single trip, and γ represents the fluctuation value of the constraint adjustment;

[0048] In the result output layer, the volume is estimated based on the predicted warehouse cargo purchase situation to obtain the predicted warehouse cargo volume estimate; the volume is fitted using digital twin technology based on the predicted warehouse cargo volume estimate and the standard warehouse shelves to obtain the recommended warehouse shelves for the next warehousing.

[0049] A warehousing logistics data processing system, comprising:

[0050] The warehouse goods incoming quality inspection module is used to deploy the goods quality inspection model, obtain the logistics data of the warehouse goods to be entered, and obtain the logistics data of the warehouse goods to be entered, which includes the goods information, consignor information, transportation information, transportation status, and cost information; perform quantity verification and quality inspection on the warehouse goods to be entered, and obtain the loss situation of the warehouse goods to be entered; update the logistics data of the warehouse goods to be entered based on the loss situation of the warehouse goods to be entered, and obtain new logistics data of the warehouse goods to be entered; establish point cloud data for the warehouse goods to be entered, and obtain the point cloud of the warehouse goods to be entered;

[0051] The storage goods inbound allocation module is used to extract features from the point cloud of the storage goods to be inbound, obtain the volume features of the storage goods to be inbound, and determine the storage partition based on the goods information in the logistics data of the storage goods to be inbound; determine the storage shelves in the storage partition to which the storage goods to be inbound belong according to the goods information in the logistics data of the storage goods to be inbound, the recommended storage shelves and the volume features of the storage goods to be inbound; the recommended storage shelves are matched according to the last predicted storage goods purchase situation;

[0052] The storage goods logistics data storage module is used to integrate the storage partitions and the storage shelves in the storage partitions to which the storage goods to be stored belong, the point cloud of the storage goods to be stored, and the logistics data of the storage goods to be stored, so as to obtain the inventory information of the stored goods that have been stored; the inventory information of the stored goods that have been stored is combined with the inventory information of the stored goods that have been stored of the same storage goods, so as to obtain new inventory information of the stored goods that have been stored;

[0053] The warehouse goods outbound module is used to obtain outbound warehouse goods information, which includes the quantity and name of goods; match the corresponding warehouse goods with the warehoused warehouse goods inventory information based on the name of the goods; locate the position by using the warehouse code of the goods in the inventory information of the warehoused warehouse goods, and select the outbound warehouse goods from the first warehoused warehouse goods in the warehoused warehouse goods inventory information according to the quantity of goods; update the warehoused warehouse goods inventory information based on the outbound warehouse goods information to obtain new warehoused warehouse goods inventory information;

[0054] The storage cargo data analysis module is used to deploy the purchase prediction analysis model. At preset time intervals q, all the outbound storage cargo information from the previous period to this period and the changes in the cargo quantity in the current storage cargo inventory information are extracted to obtain a cargo fluctuation quantity table; data analysis is performed on the cargo fluctuation quantity table and the previous x storage cargo inventory information in the storage cargo inventory information to obtain the predicted storage cargo purchase situation; the idle storage shelves at the next time of storage are adjusted according to the predicted storage cargo purchase situation to obtain recommended storage shelves.

[0055] The present invention has the following advantages:

[0056] 1. The present invention can effectively determine the storage location of goods and improve the utilization efficiency of warehouse space by extracting and analyzing the feature of the point cloud of warehouse goods to be entered, combining warehouse partitioning and digital twin technology; integrating warehouse partitioning, shelf information and logistics data to obtain inventory information of warehouse goods that have been entered, and through updating and combining, dynamic management and optimization of inventory information are realized; through the analysis of the goods fluctuation quantity table and inventory information, future purchase demand can be predicted, and guidance is provided for the adjustment of warehouse shelves at the next time of entry, which helps to reduce inventory backlogs and out-of-stock situations and improve the flexibility and efficiency of warehouse management.

[0057] 2. The present invention can ensure the integrity and accuracy of logistics data by obtaining the logistics data of the warehoused goods to be stored and conducting quantity verification and quality inspection, which is helpful to improve the quality of warehoused goods; by checking the loss of the warehoused goods to be stored, the damage or loss of the goods can be discovered in time, and the logistics data can be updated, thereby improving the accuracy and reliability of management; based on the matching of goods names and the positioning of inventory information, the rapid retrieval and selection of outbound goods is realized, the outbound process is optimized, and the outbound efficiency is improved.

[0058] 3. The present invention uses digital twin technology to perform volume fitting, can match the predicted cargo volume with standard storage shelves, optimize space utilization, and ensure that each cargo can find a suitable storage location, thereby maximizing the use of storage space. By optimizing space utilization and reducing errors, the cost of warehouse management can be reduced, and additional costs caused by insufficient or excessive storage can be avoided, thereby improving management efficiency and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A schematic diagram of the structure of a warehousing logistics data processing system adopted in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Example

[0061] The storage and delivery of warehouse goods are based on warehouse logistics data. Therefore, taking a type of warehouse goods as an example, a warehouse logistics data processing method is adopted, including:

[0062] S1. Obtain the logistics data of the warehoused goods to be entered, and obtain the logistics data of the warehoused goods to be entered, which includes the goods information, consignor information, transportation information, transportation status and cost information; perform quantity verification and quality inspection on the warehoused goods to be entered, and obtain the loss situation of the warehoused goods to be entered; update the logistics data of the warehoused goods to be entered based on the loss situation of the warehoused goods to be entered, and obtain new logistics data of the warehoused goods to be entered; establish point cloud data for the warehoused goods to be entered, and obtain the point cloud of the warehoused goods to be entered; the logistics data of the warehoused goods to be entered is provided by the sender of the goods; a number of millimeter-wave radars are installed in the warehouse, and the millimeter-wave radars are used to scan the warehoused goods to be entered, and obtain the point cloud of the warehoused goods to be entered;

[0063] The specific steps of step S1 include:

[0064] S11. Obtain the logistics data of the goods to be stored in the warehouse, and obtain the logistics data of the goods to be stored in the warehouse, which includes the goods information, consignor information, transportation information, transportation status, and cost information; the goods information includes the name of the goods to be stored in the warehouse, the quantity of the goods, and the unit weight of the goods; the consignor information includes the consignor and the place of shipment; the transportation information includes the transportation distance; the transportation status includes the transportation time; the cost information includes the total cost and transportation cost of the goods to be stored in the warehouse;

[0065] S12. Use RFID technology to obtain the number of goods to be stored in the warehouse, and obtain the real number of goods to be stored in the warehouse; weigh the goods to be stored in the warehouse to obtain the total weight of the goods to be stored in the warehouse; based on the total weight of the goods to be stored in the warehouse, the real number of the goods to be stored in the warehouse and the unit weight of the goods, it can be judged whether they are consistent; if they are inconsistent, record the real number of goods to be stored in the warehouse, and feedback the abnormal situation of the goods to the consignor; otherwise, no operation; RFID technology can accurately obtain the number of goods to be stored, and weighing provides the actual weight of the goods. Combining the two can ensure the accuracy of the number of goods; by comparing the total weight and the actual number of goods to be stored, if there is a discrepancy, the abnormal situation can be recorded and fed back to the consignor in time, avoiding the situation where abnormal goods are mistakenly stored, which helps to ensure the accuracy of the number of goods;

[0066] S13. Take p images of the goods to be stored in the warehouse to obtain a set of images of the goods to be stored in the warehouse; based on the image set of the goods to be stored in the warehouse and the goods quality inspection model, obtain a list of unqualified stored goods; based on the actual number of goods to be stored in the warehouse and the list of unqualified stored goods, obtain the loss of goods to be stored in the warehouse;

[0067] By identifying substandard goods through the cargo quality inspection model, possible quality problems can be discovered and eliminated in a timely manner, thus improving the overall quality level of incoming goods. By identifying substandard goods and calculating the loss, timely measures can be taken, such as returns or repairs, to reduce losses caused by substandard goods, thus helping to reduce costs.

[0068] S14. Update the logistics data of the warehoused goods to be stored based on the loss of the warehoused goods to be stored, and obtain new logistics data of the warehoused goods to be stored;

[0069] S15. Establish point cloud data for any of the warehoused goods to be stored, and obtain a point cloud of the warehoused goods to be stored;

[0070] The cargo quality inspection model described in step S13 includes an image preprocessing layer, a feature recognition layer and a result output layer;

[0071] The image preprocessing layer is used to preprocess the image set of warehouse goods to be stored in the warehouse, and obtain the preprocessed image set of warehouse goods to be inspected;

[0072] The feature recognition layer is used to perform feature recognition on the pre-processed warehouse goods image set to be inspected and mark abnormal values ​​to obtain abnormal warehouse goods images; the abnormal warehouse goods images are marked with the locations of unqualified warehouse goods;

[0073] The result output layer is used to output abnormal warehouse goods images. Based on the abnormal warehouse goods images, the abnormal warehouse goods are removed from the warehouse goods to be stored, and a list of unqualified warehouse goods is obtained;

[0074] The YOLOv3 model is used as the basic model of the cargo inspection model. A number of labeled warehouse cargo images to be inspected are collected to obtain a training set of warehouse cargo images to be inspected. The training set of warehouse cargo images to be inspected is input into the initial model of cargo inspection to train the model with the labeled information as the target to obtain the labeling accuracy. If the labeling accuracy is within the error standard range, the trained initial model of cargo inspection is output to obtain the cargo inspection model. Otherwise, the training set of warehouse cargo images to be inspected is used to continue model training.

[0075] S2. Extract features from the point cloud of the warehouse goods to be stored, obtain volume features of the warehouse goods to be stored, and determine the storage partition based on the cargo information in the logistics data of the warehouse goods to be stored; determine the storage shelves in the storage partition to which the warehouse goods to be stored belong according to the cargo information in the logistics data of the warehouse goods to be stored, the recommended storage shelves and the volume features of the warehouse goods to be stored; the recommended storage shelves are matched according to the last predicted warehouse goods purchase situation;

[0076] The specific steps of step S2 include:

[0077] S21. Extract features from the point cloud of the goods to be stored in the warehouse to obtain volume features of the goods to be stored in the warehouse; based on the point cloud volume, the subsequent use of digital twin technology can be guaranteed;

[0078] S22. Based on the name of the goods in the goods information, the goods in the warehouse goods master table are matched to obtain the storage partition corresponding to the name of the goods; the warehouse goods master table contains all the incoming goods in the warehouse and their corresponding storage partitions. The goods are stored in different partitions according to different goods types, which helps to improve the comfort of the goods storage space and also provides convenience for subsequent search of goods;

[0079] S23. Based on the recommended storage shelves, the quantity of goods in the goods information and the volume characteristics of the goods to be stored in the warehouse are fitted with the digital twin technology, and the storage shelves in the storage partition to which the goods to be stored in the warehouse belong are output;

[0080] S24. Performing warehousing operations on the warehousing goods to be stored based on the warehousing partitions and the warehousing shelves in the warehousing partitions to which the warehousing goods to be stored belong;

[0081] The specific steps of step S23 include:

[0082] S231. Obtain a point cloud image of a recommended storage shelf and obtain point cloud data of the recommended storage shelf;

[0083] The recommended storage shelves are matched with the last predicted storage goods purchase situation; the establishment of point cloud data for the recommended storage shelves can facilitate the subsequent simulation volume segmentation;

[0084] S232. Perform point cloud segmentation on the recommended storage point cloud data according to the volume characteristics of the storage cargo to be stored in the warehouse, and obtain a number of point cloud storage cargo blocks based on the volume characteristics of the storage cargo to be stored in the warehouse;

[0085] S233. According to the quantity of goods in the goods information, a point cloud storage goods block is selected from a number of point cloud storage goods blocks based on the volume characteristics of the goods to be stored as the storage location of the goods to be stored, and the storage shelf corresponding to the goods to be stored is obtained according to the storage shelf to which the selected point cloud storage goods block belongs;

[0086] The number of point cloud storage cargo blocks is determined by the number of goods to be stored in the warehouse, and the simulation of the storage situation can save the time of subsequent transportation of goods into the warehouse; the recommended storage point cloud data is segmented according to the volume characteristics of the storage goods to be stored in the warehouse, and the volume of goods can be matched with the shelf, optimizing the space utilization rate, ensuring that each product can find a suitable storage location, and reducing the waste of storage space; according to the quantity in the cargo information, the appropriate storage location is selected from the segmented point cloud storage cargo blocks, which can reduce the spatial fragmentation of the goods during the storage process, improve storage efficiency, and facilitate subsequent pickup and management; through automated point cloud data processing and the selection of cargo storage locations, the manual operation cost can be reduced and the efficiency and accuracy of storage operations can be improved;

[0087] S3. Integrate the storage partitions and the storage shelves in the storage partitions to which the stored goods to be stored belong, the point cloud of the stored goods to be stored, and the logistics data of the stored goods to be stored to obtain the inventory information of the stored goods that have been stored; combine the inventory information of the stored goods that have been stored with the inventory information of the stored goods that have been stored for the same storage goods to obtain new inventory information of the stored goods that have been stored;

[0088] The specific steps of step S3 include:

[0089] S31. Digitally encode the storage partitions and the storage shelves in the storage partitions to which the goods to be stored belong to obtain the goods storage code; the digital encoding method refers to the storage partition - storage shelf encoding to obtain the goods storage code;

[0090] S32. Compressing the point cloud of the goods to be stored in the warehouse to obtain a compressed point cloud of the goods; compressing the point cloud data is to save storage space;

[0091] S33. The logistics data of the warehoused goods to be stored is split to obtain the main warehouse goods inventory information and the secondary warehouse goods inventory information; the main warehouse goods inventory information includes the goods information, the goods storage code and the compressed goods point cloud; the secondary warehouse goods inventory information includes the consignor information, the transportation information, the transportation status and the cost information;

[0092] S34. The inventory information of the primary stored goods and the inventory information of the secondary stored goods are partitioned and stored to obtain the inventory information of the stored goods that have been put into storage; the inventory information of the stored goods that have been put into storage is combined with the inventory information of the stored goods that have been put into storage of the same stored goods to obtain new inventory information of the stored goods that have been put into storage; the inventory information of the primary and secondary stored goods is partitioned and stored, and the stored information of the same stored goods is combined, which helps to integrate data and improve the efficiency and accuracy of data management; the partitioned storage of the inventory information of the stored goods can make the query more efficient, because targeted searches can be performed according to the primary and secondary classifications, which improves the query speed and accuracy;

[0093] S4. Obtain outbound storage goods information, which includes the quantity and name of goods; match the corresponding storage goods with the inbound storage goods inventory information based on the name of the goods; locate the position using the storage code of the goods in the inbound storage goods inventory information, and select outbound storage goods from the first inbound storage goods in the inbound storage goods inventory information according to the quantity of goods; update the inbound storage goods inventory information based on the outbound storage goods information to obtain new inbound storage goods inventory information;

[0094] S5. Extract all the outbound storage goods information from the previous period to this period and the changes in the quantity of goods in the current storage goods inventory information at preset time intervals q to obtain a goods fluctuation quantity table; perform data analysis on the goods fluctuation quantity table and the storage goods inventory information of the previous x times in the storage goods inventory information to obtain a predicted storage goods purchase situation; adjust the idle storage shelves at the next storage according to the predicted storage goods purchase situation to obtain recommended storage shelves; the values ​​of q and x are set by professionals according to actual conditions;

[0095] The specific steps of step S5 include:

[0096] S51. Extract all the outbound storage goods information and the goods quantity information in the warehoused storage goods inventory information from the previous period to this period at every preset time interval q to obtain a goods fluctuation quantity table; the goods fluctuation quantity table stores the changes in the quantity of all goods from the previous period to this period; for example, +10, -20, etc., positive numbers represent warehoused goods, negative numbers represent outbound goods, and include the quantity of goods in the warehoused storage goods inventory information, which is the current total inventory;

[0097] S52. Analyze the goods quantity in the goods fluctuation quantity table and the goods inventory information of the warehoused goods to obtain the goods outbound rate; the goods outbound rate is obtained by comparing the total outbound quantity and the total inventory quantity plus the inbound quantity, which represents the goods outbound rate;

[0098] S53. Input the cargo outbound rate and the inventory information of the warehoused cargo that has been in stock for the previous x times into the purchase forecast analysis model for data analysis to obtain the forecast of the purchase of warehoused cargo;

[0099] S54. Obtain the recommended storage shelves for the next warehousing according to the predicted storage goods purchase situation;

[0100] The purchase forecast analysis model includes data processing layer, data fitting layer, analysis and forecast layer, and result output layer;

[0101] The data processing layer is used to pre-process the inventory information of the warehoused goods that have been put into storage for the previous x times, and extract the inventory information of the secondary warehoused goods and the corresponding quantity of goods in the inventory information of the warehoused goods that have been put into storage for the previous x times;

[0102] The data fitting layer is used to perform data fitting based on the inventory information of the previous x secondary storage goods in the storage goods inventory information and the corresponding quantity of goods to obtain the goods cost relationship;

[0103] The analysis and prediction layer is used to perform multi-objective optimization based on the relationship between goods costs and the goods outbound rate, and output the predicted warehouse goods incoming situation;

[0104] The result output layer is used to obtain recommended storage shelves based on the predicted storage goods purchase situation;

[0105] In the data fitting layer, data fitting is performed based on the inventory information of the previous x secondary storage goods and the quantity of goods in the warehoused storage goods inventory information, and F(r)=f(k, t, m) is obtained; f(k, t, m) is the relationship between the cost of goods, k is the distance of a single transportation, t is the time of a single transportation, m is the quantity of goods in a single transportation, and F(r) is the cost of a single transportation;

[0106] In the analysis and prediction layer, the maximum profit of goods is defined as the goal, and the quantity of goods transported in a single trip is adjusted by constraints; the quantity of goods transported in a single trip fluctuates within β-γ to β+γ; the maximum number of iterations is set; simulation calculations are performed based on the relationship between goods costs and the rate of goods leaving the warehouse to obtain the fitness; when the maximum number of iterations is reached, the quantity of goods transported in a single trip whose fitness meets the goal is output, and the predicted warehouse goods purchase situation is obtained; β represents the baseline value of the quantity of goods transported in a single trip, and γ represents the fluctuation value of the constraint adjustment;

[0107] The profit of goods is affected by many factors. When the fitness reaches the maximum, the highest profit value of goods is simulated and calculated. At this time, there will be no backlog of inventory in the warehouse, and the storage space can be reasonably planned.

[0108] In the result output layer, the volume is estimated according to the predicted warehouse cargo incoming situation to obtain the predicted warehouse cargo volume valuation; the volume is fitted using digital twin technology based on the predicted warehouse cargo volume valuation and standard storage shelves to obtain the recommended storage shelves for the next warehousing; by estimating the volume of warehouse cargo incoming situation, the volume of the next warehousing can be predicted more accurately, which helps to avoid insufficient or excessive shelf storage and improve the efficiency of cargo storage; using digital twin technology for volume fitting can match the predicted cargo volume with the standard storage shelves, optimize space utilization, and ensure that each cargo can find a suitable storage location, thereby maximizing the use of storage space. By optimizing space utilization and reducing errors, the cost of warehousing management can be reduced, and additional costs caused by insufficient or excessive storage can be avoided, thereby improving management efficiency and economic benefits. Example

[0109] A warehouse logistics data processing system, see Figure 1 As shown, including:

[0110] The warehouse goods incoming quality inspection module is used to deploy the goods quality inspection model, obtain the logistics data of the warehouse goods to be entered, and obtain the logistics data of the warehouse goods to be entered, which includes the goods information, consignor information, transportation information, transportation status, and cost information; perform quantity verification and quality inspection on the warehouse goods to be entered, and obtain the loss situation of the warehouse goods to be entered; update the logistics data of the warehouse goods to be entered based on the loss situation of the warehouse goods to be entered, and obtain new logistics data of the warehouse goods to be entered; establish point cloud data for the warehouse goods to be entered, and obtain the point cloud of the warehouse goods to be entered;

[0111] The storage goods inbound allocation module is used to extract features from the point cloud of the storage goods to be inbound, obtain the volume features of the storage goods to be inbound, and determine the storage partition based on the goods information in the logistics data of the storage goods to be inbound; determine the storage shelves in the storage partition to which the storage goods to be inbound belong according to the goods information in the logistics data of the storage goods to be inbound, the recommended storage shelves and the volume features of the storage goods to be inbound; the recommended storage shelves are matched according to the last predicted storage goods purchase situation;

[0112] The storage goods logistics data storage module is used to integrate the storage partitions and the storage shelves in the storage partitions to which the storage goods to be stored belong, the point cloud of the storage goods to be stored, and the logistics data of the storage goods to be stored, so as to obtain the inventory information of the stored goods that have been stored; the inventory information of the stored goods that have been stored is combined with the inventory information of the stored goods that have been stored of the same storage goods, so as to obtain new inventory information of the stored goods that have been stored;

[0113] The warehouse goods outbound module is used to obtain outbound warehouse goods information, which includes the quantity and name of goods; match the corresponding warehouse goods with the warehoused warehouse goods inventory information based on the name of the goods; locate the position by using the warehouse code of the goods in the inventory information of the warehoused warehouse goods, and select the outbound warehouse goods from the first warehoused warehouse goods in the warehoused warehouse goods inventory information according to the quantity of goods; update the warehoused warehouse goods inventory information based on the outbound warehouse goods information to obtain new warehoused warehouse goods inventory information;

[0114] The storage cargo data analysis module is used to deploy the purchase prediction analysis model. At preset time intervals q, all the outbound storage cargo information from the previous period to this period and the changes in the cargo quantity in the current storage cargo inventory information are extracted to obtain a cargo fluctuation quantity table; data analysis is performed on the cargo fluctuation quantity table and the previous x storage cargo inventory information in the storage cargo inventory information to obtain the predicted storage cargo purchase situation; the idle storage shelves at the next time of storage are adjusted according to the predicted storage cargo purchase situation to obtain recommended storage shelves.

[0115] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A method for processing warehousing logistics data, characterized in that: include: S1. Obtain the logistics data of the warehoused goods to be entered, and obtain the logistics data of the warehoused goods to be entered, which includes the goods information, consignor information, transportation information, transportation status and cost information; perform quantity verification and quality inspection on the warehoused goods to be entered, and obtain the loss situation of the warehoused goods to be entered; update the logistics data of the warehoused goods to be entered based on the loss situation of the warehoused goods to be entered, and obtain new logistics data of the warehoused goods to be entered; establish point cloud data for the warehoused goods to be entered, and obtain the point cloud of the warehoused goods to be entered; S2. Extract features from the point cloud of the warehouse goods to be stored, obtain volume features of the warehouse goods to be stored, and determine the storage partition based on the cargo information in the new logistics data of the warehouse goods to be stored; determine the storage shelves in the storage partition to which the warehouse goods to be stored belong according to the cargo information in the new logistics data of the warehouse goods to be stored, the recommended storage shelves and the volume features of the warehouse goods to be stored; the recommended storage shelves are matched according to the last predicted purchase situation of the warehouse goods; S3. Integrate the storage partitions and the storage shelves in the storage partitions to which the stored goods to be stored belong, the point cloud of the stored goods to be stored, and the logistics data of the new stored goods to be stored to obtain the inventory information of the stored goods that have been stored; combine the inventory information of the stored goods that have been stored with the inventory information of the stored goods that have been stored for the same storage goods to obtain the new inventory information of the stored goods that have been stored; S4. Obtain outbound storage cargo information, which includes cargo quantity and cargo name; Matching the new warehoused stored goods inventory information of the corresponding warehoused goods based on the goods name; locating the position of the goods storage code in the new warehoused stored goods inventory information, and selecting the outbound stored goods from the first warehoused stored goods in the new warehoused stored goods inventory information according to the quantity of goods; updating the new warehoused stored goods inventory information based on the outbound stored goods information to obtain new warehoused stored goods inventory information; S5. Extract all the outbound storage goods information from the previous period to this period and the changes in the quantity of goods in the current new storage goods inventory information at preset time intervals q to obtain a goods fluctuation quantity table; perform data analysis on the goods fluctuation quantity table and the new storage goods inventory information of the previous x times in the new storage goods inventory information to obtain a predicted storage goods purchase situation; adjust the idle storage shelves at the next storage according to the predicted storage goods purchase situation to obtain recommended storage shelves; The specific steps of step S1 include: S11. Obtain the logistics data of the goods to be stored in the warehouse, and obtain the logistics data of the goods to be stored in the warehouse. The logistics data of the goods to be stored in the warehouse includes the goods information, the consignor information, the transportation information, the transportation status, and the cost information; S12. Use RFID technology to obtain the number of warehouse goods to be stored, and obtain the real number of warehouse goods to be stored; weigh the warehouse goods to be stored, and obtain the total weight of the warehouse goods to be stored; determine whether the total weight of the warehouse goods to be stored, the real number of the warehouse goods to be stored, and the unit weight of the goods are consistent; if they are not consistent, record the real number of the warehouse goods to be stored, and feedback the abnormal situation of the goods to the consignor; otherwise, no operation; S13. Take p images of the goods to be stored in the warehouse to obtain a set of images of the goods to be stored in the warehouse; based on the image set of the goods to be stored in the warehouse and the goods quality inspection model, obtain a list of unqualified stored goods; based on the actual number of goods to be stored in the warehouse and the list of unqualified stored goods, obtain the loss of goods to be stored in the warehouse; S14. Update the logistics data of the warehoused goods to be stored based on the loss of the warehoused goods to be stored, and obtain new logistics data of the warehoused goods to be stored; S15. Establish point cloud data for any of the warehoused goods to be stored, and obtain a point cloud of the warehoused goods to be stored; The cargo quality inspection model described in step S13 includes an image preprocessing layer, a feature recognition layer and a result output layer; The image preprocessing layer is used to preprocess the image set of warehouse goods to be stored in the warehouse, and obtain the preprocessed image set of warehouse goods to be inspected; The feature recognition layer is used to perform feature recognition on the pre-processed warehouse goods image set to be inspected and mark abnormal values ​​to obtain abnormal warehouse goods images; the abnormal warehouse goods images are marked with the locations of unqualified warehouse goods; The result output layer is used to output abnormal warehouse goods images. Based on the abnormal warehouse goods images, the abnormal warehouse goods are removed from the warehouse goods to be stored, and a list of unqualified warehouse goods is obtained; The YOLOv3 model is used as the basic model of the cargo inspection model. A number of labeled warehouse cargo images to be inspected are collected to obtain a training set of warehouse cargo images to be inspected. The training set of warehouse cargo images to be inspected is input into the initial model of cargo inspection to train the model with the labeled information as the target to obtain the labeling accuracy. If the labeling accuracy is within the error standard range, the trained initial model of cargo inspection is output to obtain the cargo inspection model. Otherwise, the training set of warehouse cargo images to be inspected is used to continue model training. The specific steps of step S5 include: S51. Extract all the outbound storage goods information from the previous period to this period and the goods quantity information in the new inbound storage goods inventory information at every preset time interval q to obtain a goods fluctuation quantity table; S52. Analyze the quantity of goods in the goods fluctuation quantity table and the new warehoused goods inventory information to obtain the goods outbound rate; S53. The cargo outbound rate and the new warehoused cargo inventory information of the previous x times are input into the purchase forecast analysis model for data analysis to obtain the predicted purchase of warehoused cargo; S54. Obtain the recommended storage shelves for the next warehousing according to the predicted storage goods purchase situation; The purchase forecast analysis model includes data processing layer, data fitting layer, analysis and forecast layer, and result output layer; The data processing layer is used to pre-process the new warehoused stored goods inventory information of the previous x times in the new warehoused stored goods inventory information, and extract the previous x times of secondary warehoused goods inventory information and the corresponding goods quantity in the new warehoused stored goods inventory information; The data fitting layer is used to perform data fitting based on the previous x secondary storage goods inventory information and the corresponding goods quantity in the new warehoused storage goods inventory information to obtain the goods cost relationship; The analysis and prediction layer is used to perform multi-objective optimization based on the relationship between goods costs and the goods outbound rate, and output the predicted warehouse goods incoming situation; The result output layer is used to obtain recommended storage shelves based on the predicted storage goods purchase situation; In the data fitting layer, data fitting is performed based on the secondary storage cargo inventory information and cargo quantity of the previous x times in the new warehoused storage cargo inventory information, and F(r)=f(k, t, m) is obtained; f(k, t, m) is the cargo cost relationship, k is the single transportation distance, t is the single transportation time, m is the single transportation cargo quantity, and F(r) is the single transportation cost; In the analysis and prediction layer, the maximum profit of goods is defined as the goal, and the quantity of goods transported in a single trip is adjusted by constraints; the quantity of goods transported in a single trip fluctuates within β-γ to β+γ; the maximum number of iterations is set; simulation calculations are performed based on the relationship between goods costs and the rate of goods leaving the warehouse to obtain the fitness; when the maximum number of iterations is reached, the quantity of goods transported in a single trip whose fitness meets the goal is output, and the predicted warehouse goods purchase situation is obtained; β represents the baseline value of the quantity of goods transported in a single trip, and γ represents the fluctuation value of the constraint adjustment; In the result output layer, the volume is estimated based on the predicted warehouse cargo purchase situation to obtain the predicted warehouse cargo volume estimate; the volume is fitted using digital twin technology based on the predicted warehouse cargo volume estimate and the standard warehouse shelves to obtain the recommended warehouse shelves for the next warehousing.

2. A method for processing warehousing logistics data according to claim 1, characterized in that: The specific steps of step S2 include: S21. Extracting features from the point cloud of the goods to be stored in the warehouse to obtain volume features of the goods to be stored in the warehouse; S22. Match the goods name in the goods information in the warehouse goods table to obtain the storage partition corresponding to the goods name; S23. Based on the recommended storage shelves, the quantity of goods in the goods information and the volume characteristics of the goods to be stored in the warehouse are fitted with the digital twin technology, and the storage shelves in the storage partition to which the goods to be stored in the warehouse belong are output; S24. Perform warehousing operations on the goods to be stored based on the storage partitions and the storage shelves in the storage partitions to which the goods to be stored belong.

3. A method for processing warehousing logistics data according to claim 2, characterized in that: The specific steps of step S23 include: S231. Obtain a point cloud image of a recommended storage shelf and obtain point cloud data of the recommended storage shelf; S232. Perform point cloud segmentation on the recommended storage point cloud data according to the volume characteristics of the storage cargo to be stored in the warehouse, and obtain a number of point cloud storage cargo blocks based on the volume characteristics of the storage cargo to be stored in the warehouse; S233. According to the quantity of goods in the goods information, a point cloud storage cargo block is selected from a number of point cloud storage cargo blocks based on the volume characteristics of the goods to be stored as the storage location of the goods to be stored, and according to the storage shelf to which the selected point cloud storage cargo block belongs, the storage shelf corresponding to the goods to be stored is obtained.

4. A method for processing warehousing logistics data according to claim 1, characterized in that: The specific steps of step S3 include: S31. Digitally encode the storage partitions and the storage shelves in the storage partitions to which the goods to be stored belong to obtain the goods storage code; S32. Compress the point cloud of the goods to be stored in the warehouse to obtain a compressed point cloud of the goods; S33. The new warehouse cargo logistics data to be stored is split to obtain the main warehouse cargo inventory information and the secondary warehouse cargo inventory information; the main warehouse cargo inventory information includes cargo information, cargo storage code and compressed cargo point cloud; the secondary warehouse cargo inventory information includes consignor information, transportation information, transportation status and cost information; S34. The main warehouse goods inventory information and the secondary warehouse goods inventory information are partitioned and stored to obtain the warehoused warehouse goods inventory information; the warehoused warehouse goods inventory information is combined with the warehoused warehouse goods inventory information of the same warehouse goods to obtain new warehoused warehouse goods inventory information.

5. A warehousing logistics data processing system, characterized in that: The system applies a storage logistics data processing method according to any one of claims 1 to 4, including: The warehouse goods incoming quality inspection module is used to deploy the goods quality inspection model, obtain the logistics data of the incoming warehouse goods, and obtain the logistics data of the incoming warehouse goods, which includes the goods information, consignor information, transportation information, transportation status and cost information; perform quantity verification and quality inspection on the incoming warehouse goods, and obtain the loss situation of the incoming warehouse goods; update the logistics data of the incoming warehouse goods based on the loss situation of the incoming warehouse goods, and obtain new logistics data of the incoming warehouse goods; establish point cloud data for the incoming warehouse goods, and obtain the point cloud of the incoming warehouse goods; The storage goods entry allocation module is used to extract features from the point cloud of the storage goods to be entered, obtain the volume features of the storage goods to be entered, and determine the storage partition based on the goods information in the new logistics data of the storage goods to be entered; determine the storage shelves in the storage partition to which the storage goods to be entered belong according to the goods information in the new logistics data of the storage goods to be entered, the recommended storage shelves and the volume features of the storage goods to be entered; the recommended storage shelves are matched according to the last predicted storage goods purchase situation; The storage goods logistics data storage module is used to integrate the storage partitions and the storage shelves in the storage partitions to which the storage goods to be stored belong, the point cloud of the storage goods to be stored, and the new logistics data of the storage goods to be stored to obtain the inventory information of the stored goods that have been stored; combine the inventory information of the stored goods that have been stored with the inventory information of the stored goods that have been stored of the same storage goods to obtain new inventory information of the stored goods that have been stored; The storage goods outbound module is used to obtain outbound storage goods information, which includes the quantity and name of goods; match the new warehoused storage goods inventory information of the corresponding storage goods based on the name of the goods; locate the position of the goods storage code in the new warehoused storage goods inventory information, and select outbound storage goods from the first warehoused storage goods in the new warehoused storage goods inventory information according to the quantity of goods; update the new warehoused storage goods inventory information based on the outbound storage goods information to obtain new warehoused storage goods inventory information; The storage cargo data analysis module is used to deploy the purchase prediction analysis model, extract all the outbound storage cargo information from the previous period to this period and the cargo quantity changes in the current new storage cargo inventory information at preset time intervals q, and obtain a cargo fluctuation quantity table; perform data analysis on the cargo fluctuation quantity table and the new storage cargo inventory information of the previous x times in the new storage cargo inventory information to obtain the predicted storage cargo purchase situation; adjust the idle storage shelves at the next warehousing according to the predicted storage cargo purchase situation to obtain recommended storage shelves.

Images