Intelligent warehouse management method and system based on Internet of Things

Abstract

The present invention discloses an intelligent warehouse management method and system based on the Internet of Things, the method comprising: based on the goods to be stored, using an RFID reader to scan the goods information before storage, and updating the storage goods information library; according to the goods attribute information of the goods to be stored, determining the storage location of the goods to be stored in the warehouse according to the pre-planned storage layout, the goods attribute information includes the name, volume, weight, and classification information of the goods; according to the storage location, using an automatic guided vehicle AGV to transport the goods to be stored to the storage location; based on the temperature sensors, humidity sensors, light sensors, and cameras pre-arranged in the warehouse, real-time collection of warehouse temperature and humidity data and images of materials stored in the warehouse, monitoring the warehouse temperature and humidity status and material status. The present invention realizes the automated storage, outbound and movement of goods, and realizes efficient and accurate goods management.

Description

Technical Field

[0001] The present invention relates to the field of intelligent warehousing technology, and in particular to an intelligent warehousing management method and system based on the Internet of Things. Background Art

[0002] With the rapid development of e-commerce and logistics industries, traditional warehousing and logistics models have gradually exposed many problems, such as low warehouse space utilization, low manual operation efficiency, high error rate, inaccurate inventory management, etc. These problems not only lead to increased operating costs, but also affect the customer service experience. Therefore, there is an urgent need for an intelligent solution that can effectively improve the efficiency and management level of warehousing and logistics. As one of the important applications of artificial intelligence in the field of logistics, the intelligent warehousing system has broad development prospects and huge potential. Summary of the invention

[0003] In view of the problems existing in the above-mentioned prior art, the present invention provides an intelligent warehouse management method and system based on the Internet of Things, which effectively improves warehouse operation efficiency and cargo storage density, and improves warehouse management efficiency. The technical solution is as follows:

[0004] In a first aspect, an intelligent warehouse management method based on the Internet of Things is provided, comprising:

[0005] Based on the goods to be stored, the RFID reader is used to scan the goods information before storage and update the warehouse goods information database;

[0006] Based on the cargo attribute information of the cargo to be stored, the storage location of the cargo to be stored in the warehouse is determined according to the pre-planned storage layout, wherein the cargo attribute information includes cargo name, volume, weight, and classification information, and the storage layout determines the cargo storage location according to the major category to which the cargo belongs and the group to which it belongs under the major category;

[0007] According to the storage location, an automatic guided vehicle (AGV) is used to transport the goods to be stored to the storage location;

[0008] Based on the temperature sensors, humidity sensors, light sensors and cameras pre-installed in the warehouse, the warehouse temperature and humidity data and images of materials stored in the warehouse are collected in real time to monitor the warehouse temperature and humidity status and material status.

[0009] In some embodiments, the pre-planned storage layout planning method includes:

[0010] Acquire attribute information of all goods, classify all goods into A categories according to application fields, and obtain the total amount of goods in each category in the A categories, wherein the A categories include at least one special category, and the special category is used to characterize goods with special storage requirements;

[0011] The required storage area size for each major category of goods is determined according to the total amount of each major category of goods, and A storage areas are set up in the warehouse to store each major category of goods respectively. The location of the storage area for each major category is determined based on the first in-and-out cost value of the major category of goods. The larger the first in-and-out cost value of the major category of goods, the smaller the distance between the location of the storage area of ​​the major category and the warehouse entrance; the first in-and-out cost value represents the AGV handling workload without considering the transfer distance when the goods are in or out of the warehouse.

[0012] In some embodiments, the first inbound and outbound cost value is determined based on the number of times the type of goods are handled, the inbound and outbound frequency, and the inbound and outbound weight within a preset time period.

[0013] In some implementations, the planning method further includes:

[0014] Grouping multiple different goods in a single major category to obtain different groups of goods, each group includes multiple different goods;

[0015] Note that a single major category of goods includes B groups of goods. The required storage area size of each group of goods is determined according to the total amount of each group of goods. B sub-storage areas are set in the single major category of goods storage area to store each group of goods respectively. The location of the storage area of ​​each group of goods is determined based on the second in-and-out warehouse cost value of the group of goods. The larger the second in-and-out warehouse cost value of the group of goods, the smaller the distance between the location of the storage area of ​​the group of goods and the warehouse entrance.

[0016] In some embodiments, the grouping of multiple different goods in a single category to obtain different groups of goods includes:

[0017] Based on the cargo type information on the outbound and inbound orders, the correlation between different cargoes is mined, and the correlation represents the correlation between the spatial and temporal combinations of different cargoes;

[0018] Calculate the in-and-out workload of different goods based on the in-and-out information of goods within a preset long period of time;

[0019] Based on the balance of the total in-and-out workload of different groups of goods within a preset long period of time and the maximum correlation of goods in the same group as the optimization goal, multiple different goods in a single category are iteratively grouped.

[0020] In some implementations, mining the association between different goods based on the goods category information on the outbound order and the inbound order includes:

[0021] Based on the material type information and supplier and purchaser information on a single outbound / inbound order, obtain the material type information required by a single supplier and purchaser on the single outbound / inbound order to form a single material combination example;

[0022] Based on the outbound / inbound orders at different times, obtain the time series of material combination examples in chronological order;

[0023] Based on the single material combination example in the material combination example time series, encoding is performed to obtain the encoding vector corresponding to the single material combination example, and then the encoding vector group corresponding to the material combination example time series is obtained;

[0024] Based on the coding vector group input spatiotemporal coding model, the spatial combination correlation relationship and temporal combination correlation relationship of different types of materials mined from the inbound / outbound orders at different times are obtained;

[0025] The results output by the spatiotemporal coding model are passed through a fully connected layer to obtain the correlation between different goods.

[0026] In some implementations, mining the association between different goods based on the goods category information on the outbound order and the inbound order further includes:

[0027] The names of suppliers and buyers are marked on the individual material combinations formed;

[0028] For the material combination time series, split it into material combination time series of different suppliers and buyers according to the supplier and buyer names;

[0029] Classify different suppliers and buyers based on the consistency of the types of purchased materials and the similarity of their business scopes to obtain at least one supplier and buyer cluster;

[0030] Based on the material combination example time series of the supplier purchaser in the same supplier purchaser cluster, a single material combination example in the material combination example time series is encoded to obtain a coding vector corresponding to the single material combination example, and then a coding vector group corresponding to the material combination example time series is obtained;

[0031] Based on the coding vector group input spatiotemporal coding model, the spatial combination correlation relationship and temporal combination correlation relationship of different types of materials mined from the inbound / outbound orders of different suppliers and buyers are obtained;

[0032] The results output by the spatiotemporal coding model are passed through a fully connected layer to obtain the correlation between different goods.

[0033] In some embodiments, the step of determining the storage location of the goods to be stored in the warehouse according to the property information of the goods to be stored and the pre-planned storage layout includes:

[0034] (1) Obtaining the property information of the goods to be stored, wherein the property information includes the category information to which the goods belong;

[0035] (2) Based on the cargo attribute information of the cargo to be stored, check whether a storage area corresponding to the cargo already exists in the existing storage layout. If so, proceed to step (3); otherwise, proceed to step (4);

[0036] (3) storing the goods to be stored according to the remaining space in the storage area corresponding to the goods to be stored in the existing storage layout;

[0037] (4) In the remaining free areas of the entire warehouse, the storage locations of the goods to be stored are determined based on the property information of the goods to be stored, the correlation between the goods to be stored and other goods, and the distance between different locations in the free areas and the storage entrance, thus forming a new warehouse layout.

[0038] In the second aspect, an intelligent warehouse management system based on the Internet of Things is provided, including:

[0039] The incoming scanning unit is used to scan the goods information with an RFID reader before the goods are put into storage, and update the storage goods information database;

[0040] A storage layout unit is used to determine the storage location of the goods to be stored in the warehouse according to the property information of the goods to be stored and the pre-planned storage layout. The property information of the goods includes the name, volume, weight and classification information of the goods. The storage layout determines the storage location of the goods according to the major category to which the goods belong and the group to which they belong under the major category;

[0041] A storage and handling unit is used to transport the goods to be stored to the storage location using an automatic guided vehicle (AGV) according to the storage location;

[0042] The warehouse monitoring unit is used to collect warehouse temperature and humidity data and images of materials stored in the warehouse in real time based on the temperature sensors, humidity sensors, light sensors, and cameras pre-installed in the warehouse, and monitor the temperature and humidity status of the warehouse and the status of materials.

[0043] In a second aspect, an intelligent warehouse management device based on the Internet of Things is provided, the device comprising:

[0044] processor;

[0045] a memory for storing processor-executable instructions;

[0046] Among them, the processor implements the intelligent warehousing management method based on the Internet of Things as described in the first aspect above by running the executable instructions.

[0047] The present invention provides an intelligent warehouse management method and system based on the Internet of Things, which has the following beneficial effects: by using an RFID reader to scan the cargo information of the goods to be stored to record the properties of the goods to be stored, determining the optimal storage location for the goods to be stored, automatically and intelligently transporting the goods into the warehouse, and real-time monitoring of the warehouse goods with multi-sensor information, the intelligent storage management of the goods is realized. The automated storage, outbound and movement of goods are realized, which reduces manual operations and improves work efficiency. Based on the Internet of Things sensors, data related to materials in the warehouse are collected, the warehouse environment and the condition of the goods are monitored in real time, and the storage safety of temperature-controlled goods is ensured. The storage layout planning of the present invention realizes the intelligent classification, positioning and optimal arrangement of goods, effectively improving the warehouse operation efficiency and the storage density of goods. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of an intelligent warehouse management method based on the Internet of Things in an embodiment of the present application;

[0049] Figure 2 This is a flow chart of a method for pre-planning a warehouse layout in an embodiment of the present application;

[0050] Figure 3 This is a flow chart of a method for grouping goods under a single category of a storage layout in an embodiment of the present application;

[0051] Figure 4 This is a flow chart of a method for analyzing the relevance of goods under a single category of a warehouse layout in an embodiment of the present application;

[0052] Figure 5 It is a structural diagram of an intelligent warehouse management system based on the Internet of Things in an embodiment of the present application. DETAILED DESCRIPTION

[0053] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0054] See also Figure 1 The present application embodiment provides an intelligent warehouse management method based on the Internet of Things, the method comprising the following steps:

[0055] Step 1: Based on the goods to be stored, use an RFID reader to scan the goods information before storage and update the storage goods information database;

[0056] Step 2: Determine the storage location of the goods to be stored in the warehouse according to the pre-planned storage layout based on the goods attribute information of the goods to be stored, wherein the goods attribute information includes the name, volume, weight, and classification information of the goods, and the storage layout determines the storage location of the goods according to the categories to which the goods belong and the groups under the categories;

[0057] Step 3, according to the storage location, using an automatic guided vehicle (AGV) to transport the goods to be stored to the storage location;

[0058] Step 4: Based on the temperature sensors, humidity sensors, light sensors, and cameras pre-installed in the warehouse, real-time data on warehouse temperature and humidity and images of materials stored in the warehouse are collected to monitor the temperature and humidity status of the warehouse and the status of materials.

[0059] In the embodiment of the present application, the intelligent storage management of goods is realized by using RFID reader to scan the goods information of the goods to be stored to record the properties of the goods to be stored, determine the optimal storage location of the goods to be stored, automatically and intelligently transport the goods to the warehouse, and monitor the goods in the warehouse in real time with multi-sensor information. The automatic storage, outbound and movement of goods are realized, which reduces manual operations and improves work efficiency.

[0060] In the embodiment of the present application, data related to materials in the warehouse are collected based on IoT sensors, and the sensors include temperature sensors, humidity sensors, light sensors, RFID readers, cameras, etc. These sensors are distributed in various key locations of the warehouse and can monitor the warehouse environment and the status of goods in real time. For example, temperature and humidity sensors are used to monitor the environmental conditions of the warehouse to ensure the storage safety of temperature-controlled goods; RFID readers and cameras are used to track the location and status of goods in real time to prevent the loss and damage of goods.

[0061] In one embodiment, see Figure 2 In the above step 2, the pre-planned storage layout planning method includes:

[0062] Step 21, obtaining attribute information of all goods, dividing all goods into A categories according to application fields and obtaining the total amount of goods in each category in the A categories, wherein the A categories include at least one special category, and the special category is used to characterize goods with special storage requirements;

[0063] Step 22, determine the required storage area size for each major category of goods based on the total amount of each major category of goods, set up A storage areas in the warehouse for storing each major category of goods, and the location of each major category of storage area is determined based on the first in-and-out cost value of the major category of goods. The larger the first in-and-out cost value of the major category of goods, the smaller the distance between the location of the major category of storage area and the warehouse entrance; the first in-and-out cost value represents the AGV handling workload when the goods are in or out of the warehouse without considering the transfer distance.

[0064] It is understandable that when planning the warehouse layout, the required storage area sizes for different categories of goods are different. When planning, it is necessary to consider the location and area size of the storage area for major categories of goods at the same time to form multiple storage areas of different sizes at different locations in the warehouse.

[0065] The first storage cost represents the AGV handling workload when the goods are stored or unstored without considering the transfer distance. For goods with a large handling workload, the location of the storage area is closer to the warehouse entrance to improve the efficiency of goods storage.

[0066] The warehouse layout planning in the embodiment of the present application realizes the intelligent classification, positioning and optimized arrangement of goods, effectively improving the warehouse operation efficiency and the goods storage density.

[0067] Specifically, the first in-and-out cost value is determined based on the number of handling times, in-and-out frequency, and in-and-out weight of the type of goods within a preset time period. In one embodiment, the first in-and-out cost value is determined based on the product of the number of handling times, in-and-out frequency, and in-and-out weight.

[0068] In the embodiment of the present application, in the above step 2, the pre-planned storage layout, after the above step 22, the planning method further includes:

[0069] Step 23, grouping the multiple different goods in a single category to obtain different groups of goods, each group including multiple different goods;

[0070] Step 24, record that a single major category of goods includes B groups of goods, determine the required storage area size of each group of goods according to the total amount of each group of goods, set B sub-storage areas in the single major category of goods storage area for storing each group of goods respectively, and the location of the storage area of ​​each group of goods is determined based on the second in-and-out cost value of the group of goods. The larger the second in-and-out cost value of the group of goods, the smaller the distance between the location of the storage area of ​​the group of goods and the warehouse entrance.

[0071] In one embodiment, see Figure 3 In the above step 23, grouping a plurality of different goods in a single category to obtain goods in different groups includes the following steps:

[0072] Step 231, based on the cargo type information on the outbound order and the inbound order, mining the correlation between different cargoes, wherein the correlation represents the correlation between the spatial combination and the temporal combination of different cargoes;

[0073] Step 232, calculating the in-and-out workload of different goods based on the in-and-out information of goods within a preset long period of time;

[0074] Step 233, based on the balance of the total in-and-out workload of goods of different groups within a preset long period of time and the maximum correlation of goods in the same group as the optimization goal, multiple different goods in a single category are iteratively grouped.

[0075] In the embodiment of the present application, a variety of different goods in a single category are grouped, and the in-and-out workload of different goods is calculated. The in-and-out workload is determined based on factors such as the number of handling times and the weight of a single handling within a preset long period of time, without considering the distance between the storage area location and the warehouse entrance. Specifically, a variety of different goods in a single category can be preliminarily grouped, the sum of the in-and-out workload of goods in a single group and the correlation of goods in a single group are calculated, and the objective function is calculated. The objective function includes the balance of the sum of the in-and-out workload of goods in different groups and the value of the goods correlation. The objective function is continuously optimized by adjusting the grouping results until the objective function is optimal, and the final classification results of the various different goods in a single category are determined.

[0076] Specifically, in the optimization objective function in step 233, the correlation degree of goods in the same group for: ,in, represents the kth group, Indicates the total number of groups. , Indicates different goods. represents the set of goods in the kth group, Indicates the degree of association, Indicates the goods in the kth group and goods The correlation degree of the goods in the same group is the largest, so max is used. calculate;

[0077] The total balance of the inbound and outbound workload of the different groups of goods for: , represents the total workload of the kth group of goods in and out of the warehouse, It represents the mean of the total in-and-out workload of m groups of goods. The total in-and-out workload of different groups of goods is optimally balanced, and min is used. calculate.

[0078] In one embodiment, see Figure 4 In the above step 231, based on the cargo type information on the outbound order and the inbound order, mining the correlation between different cargoes includes the following steps:

[0079] Step 23101, based on the material type information and supplier and purchaser information on the single outbound / inbound order, obtain the material type information required by a single supplier and purchaser on the single outbound / inbound order to form a single material combination example;

[0080] Step 23102, based on the outbound / inbound orders at different times, obtain a time series of material combination examples in chronological order;

[0081] Step 23103, encoding a single material combination example in the material combination example time series to obtain a coding vector corresponding to the single material combination example, and then obtaining a coding vector group corresponding to the material combination example time series;

[0082] Step 23104, inputting the spatiotemporal coding model based on the coding vector group, obtaining the spatial combination association relationship and the temporal combination association relationship representing different types of materials mined from the inbound / outbound warehouse orders at different times;

[0083] Step 23105, based on the result output by the spatiotemporal coding model, the correlation between different goods is obtained through a fully connected layer. This fully connected layer is similar to the last fully connected layer in conventional neural networks when solving classification or regression problems. It is used to integrate the output of the previous layer and map it to the output classification or regression result.

[0084] In the embodiment of the present application, when mining the correlation between different goods, the material information on the in / out order is mined. A single in / out order contains information on multiple material types, and corresponds to the purchase and supply material information of different suppliers and purchasers. In the embodiment of the present application, the types of materials required by a single supplier and purchaser on a single in / out order are formed into a material combination example. This material combination example affects the result of grouping multiple different goods in a single category. A material combination example may correspond to a group in a single category. In the embodiment of the present application, the correlation between different goods is mined based on the material combination example of the time series. On the one hand, the data contains time series information, and the correlation between different goods in the in / out time can be mined. For example, the in / out time of some goods is relatively close, and these goods may have a time sequence correlation in demand; on the other hand, the data contains different material combination examples, and the correlation between the in / out combination of different goods can be mined. In the embodiment of the present application, the spatial combination association relationship and the temporal combination association relationship of different types of materials are mined, and a spatiotemporal coding model is adopted. Before inputting the spatiotemporal coding model, the material combination example is first encoded, that is, different materials are encoded separately, and the material combination example corresponds to a coding vector; in the spatiotemporal coding model, the model can be implemented by a transform model that incorporates a self-attention mechanism, or by a convolutional network containing a temporal attention unit and a channel attention unit, or by other neural networks containing a temporal attention unit and a channel attention unit, and the spatiotemporal coding model is obtained by training the network model with a pre-prepared material combination example time series sample set carrying sample annotation data.

[0085] Through the embodiments of the present application, the combination association of different goods in a single major category when entering and leaving the warehouse, and the combination association in time sequence can be mined. In fact, further, the association between different goods in supply and demand can be mined according to the entry and exit quantities of different goods on the entry and exit orders. Some goods have large consumption and large supply and demand, while some goods have small consumption and small supply and demand. Through the combination association of different goods in a single major category when entering and leaving the warehouse, and the combination association in time sequence, different goods in a single major category are classified, and the final classification result of goods in a single major category is determined based on the optimization goal of balancing the total entry and exit workload of goods in different groups and maximizing the correlation of goods in the same group.

[0086] In one embodiment, the above step 231 mines the association between different goods based on the goods type information on the outbound order and the inbound order, and further includes:

[0087] Step 23111, based on the material type information and supplier and purchaser information on the single outbound / inbound order, obtain the material type information required by a single supplier and purchaser on the single outbound / inbound order to form a single material combination example; mark the supplier and purchaser name on the formed single material combination example;

[0088] Step 23112: For the material combination time series, split it into material combination time series of different suppliers and buyers according to the supplier and buyer names;

[0089] Step 23113, classify different suppliers and purchasers according to the consistency of the types of purchased materials and the similarity of their business scopes, and obtain at least one supplier and purchaser cluster;

[0090] Step 23114, based on the material combination example time series of the supplier purchaser in the same supplier purchaser cluster, encode a single material combination example in the material combination example time series to obtain a coding vector corresponding to the single material combination example, and then obtain a coding vector group corresponding to the material combination example time series;

[0091] Step 23115, based on the coding vector group, input the spatiotemporal coding model to obtain the spatial combination association relationship and the temporal combination association relationship representing different types of materials mined from the inbound / outbound orders of different suppliers and buyers;

[0092] Step 23116, based on the output of the spatiotemporal coding model, the correlation between different goods is obtained through a fully connected layer.

[0093] Furthermore, in the embodiment of the present application, when mining the correlation between different goods based on the goods type information on the outbound and inbound orders, the factors of suppliers and purchasers are also considered, and suppliers and purchasers with similar business material scopes are divided into a cluster for analysis and mining, and mining is performed on the purchase / supply material combination examples of similar suppliers and purchasers, and the material combination examples are mined in a refined manner to obtain the spatial combination and time combination correlation relationship of materials. The goods correlation mining results considering the classification of suppliers and purchasers provided in the embodiment of the present application can be combined with the goods correlation mining results of the previous embodiment, and the goods classification results in a single major category are finally determined based on the two goods mining results.

[0094] In one embodiment, the pre-planned storage layout in step 2 above, the planning method further includes:

[0095] Predict future inventory needs based on historical cargo in and out data;

[0096] The estimated storage space required for the goods is determined based on the predicted inventory demand of the goods, and the warehouse layout planning is performed according to the estimated storage space required for the goods.

[0097] In one embodiment, in the above step 2, based on the cargo attribute information of the cargo to be stored, the storage location of the cargo to be stored in the warehouse is determined according to the pre-planned storage layout, including:

[0098] (1) Obtaining the property information of the goods to be stored, wherein the property information includes the category information to which the goods belong;

[0099] (2) Based on the cargo attribute information of the cargo to be stored, check whether a storage area corresponding to the cargo already exists in the existing storage layout. If so, proceed to step (3); otherwise, proceed to step (4);

[0100] (3) storing the goods to be stored according to the remaining space in the storage area corresponding to the goods to be stored in the existing storage layout;

[0101] (4) In the remaining free areas of the entire warehouse, the storage locations of the goods to be stored are determined based on the property information of the goods to be stored, the correlation between the goods to be stored and other goods, and the distance between different locations in the free areas and the storage entrance, thus forming a new warehouse layout.

[0102] In this embodiment, the storage layout planning can be flexibly adjusted and optimized according to the actual situation of the warehouse to achieve efficient and accurate goods management. For the goods to be stored, it is necessary to detect whether it is the first time for the goods to enter the warehouse for storage. If it is the first time for the goods to enter the warehouse for storage, step (4) is executed to determine the size of the storage area according to the volume and weight of the goods, and analyze the correlation between the goods to be stored and other goods according to the name and classification information of the goods, combined with the in-and-out order information of other warehouses or other goods commodity association analysis database. The storage area planning is carried out for the newly appeared goods. The storage area planning for each type of goods takes into account the in-and-out value of the goods and the correlation of the goods in the adjacent storage areas. It can be understood that if the correlation between two goods is large, they are placed in the adjacent storage areas first to facilitate out-of-warehouse. It can be understood that for the newly stored goods, an initial storage location can be determined when they enter the warehouse for storage for the first time. With the subsequent multiple in-and-out order information, in-and-out frequency, in-and-out quantity and other information, the storage location will change adaptively.

[0103] See also Figure 5 , an intelligent warehouse management system based on the Internet of Things provided by the embodiment of the present application includes:

[0104] The incoming scanning unit is used to scan the goods information with an RFID reader before the goods are put into storage, and update the storage goods information database;

[0105] A storage layout unit is used to determine the storage location of the goods to be stored in the warehouse according to the property information of the goods to be stored and the pre-planned storage layout. The property information of the goods includes the name, volume, weight and classification information of the goods. The storage layout determines the storage location of the goods according to the major category to which the goods belong and the group to which they belong under the major category;

[0106] A storage and handling unit is used to transport the goods to be stored to the storage location using an automatic guided vehicle (AGV) according to the storage location;

[0107] The warehouse monitoring unit is used to collect warehouse temperature and humidity data and images of materials stored in the warehouse in real time based on the temperature sensors, humidity sensors, light sensors, and cameras pre-installed in the warehouse, and monitor the temperature and humidity status of the warehouse and the status of materials.

[0108] For the specific definition of the intelligent warehouse management system based on the Internet of Things, please refer to the definition of the intelligent warehouse management method based on the Internet of Things above, which will not be repeated here. Each unit in the above-mentioned intelligent warehouse management system based on the Internet of Things can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned units can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned units.

[0109] The embodiment of the present application provides an intelligent warehouse management device based on the Internet of Things, and the device includes: a processor; a memory for storing processor executable instructions; wherein the processor implements the intelligent warehouse management method based on the Internet of Things described in the above embodiment by running the executable instructions. Specifically, the device includes: at least one processor, a memory, a user interface, and at least one network interface. The various components in the electronic device are coupled together through a bus system. It can be understood that the bus system is used to achieve connection and communication between these components. The memory can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. The network interface is used to communicate with an external terminal through a network connection.

[0110] The present invention is not limited to the above-mentioned specific implementation modes. Various changes made by ordinary technicians in this field based on the above-mentioned concepts without creative work are all within the protection scope of the present invention.

Claims

1. An intelligent warehouse management method based on the Internet of Things, characterized in that: include: Based on the goods to be stored, the RFID reader is used to scan the goods information before storage and update the warehouse goods information database; Based on the cargo attribute information of the cargo to be stored, the storage location of the cargo to be stored in the warehouse is determined according to the pre-planned storage layout, wherein the cargo attribute information includes cargo name, volume, weight, and classification information, and the storage layout determines the cargo storage location according to the major category to which the cargo belongs and the group to which it belongs under the major category; According to the storage location, an automatic guided vehicle (AGV) is used to transport the goods to be stored to the storage location; Based on the temperature sensors, humidity sensors, light sensors, and cameras pre-installed in the warehouse, the warehouse temperature and humidity data and images of materials stored in the warehouse are collected in real time to monitor the warehouse temperature and humidity status and material status; The method of determining the group to which goods belong under a major category includes: Based on the cargo type information on the outbound and inbound orders, the correlation between different cargoes is mined, and the correlation represents the correlation between the spatial and temporal combinations of different cargoes; Calculate the in-and-out workload of different goods based on the in-and-out information of goods within a preset long period of time; Based on the balance of the total in-and-out workload of different groups of goods in a preset long period of time and the maximum correlation of goods in the same group as the optimization goal, multiple different goods in a single category are iteratively grouped; the correlation of goods in the same group for: ,in, represents the kth group, m represents the total number of groups, i and j represent different goods, represents the set of goods in the kth group, P represents the degree of association, Indicates the correlation between goods i and goods j in the kth group; the goods in the same group have the largest correlation, and max is used Calculate the balance of the total workload of the different groups of goods in and out of the warehouse for: , represents the total workload of the kth group of goods in and out of the warehouse, It represents the mean of the total in-and-out workload of m groups of goods. The total in-and-out workload of different groups of goods is optimally balanced, and min is used. calculate; The method of mining the correlation between different goods based on the goods type information on the outbound and inbound orders includes: Based on the material type information and supplier and purchaser information on a single outbound / inbound order, obtain the material type information required by a single supplier and purchaser on the single outbound / inbound order to form a single material combination example; Based on the outbound / inbound orders at different times, obtain the time series of material combination examples in chronological order; Based on the single material combination example in the material combination example time series, encoding is performed to obtain the encoding vector corresponding to the single material combination example, and then the encoding vector group corresponding to the material combination example time series is obtained; Based on the coding vector group input spatiotemporal coding model, the spatial combination correlation relationship and temporal combination correlation relationship of different types of materials mined from the inbound / outbound orders at different times are obtained; The results output by the spatiotemporal coding model are passed through a fully connected layer to obtain the correlation between different goods.

2. According to claim 1, the intelligent warehouse management method based on the Internet of Things is characterized in that: The pre-planned storage layout and planning method include: Acquire attribute information of all goods, classify all goods into A categories according to application fields, and obtain the total amount of goods in each category in the A categories, wherein the A categories include at least one special category, and the special category is used to characterize goods with special storage requirements; The required storage area size for each major category of goods is determined according to the total amount of each major category of goods, and A storage areas are set up in the warehouse to store each major category of goods respectively. The location of the storage area for each major category is determined based on the first in-and-out cost value of the major category of goods. The larger the first in-and-out cost value of the major category of goods, the smaller the distance between the location of the storage area of ​​the major category and the warehouse entrance; the first in-and-out cost value represents the AGV handling workload without considering the transfer distance when the goods are in or out of the warehouse.

3. According to claim 2, the intelligent warehouse management method based on the Internet of Things is characterized in that: The first in-and-out storage cost is determined based on the number of times the type of goods are handled, the in-and-out storage frequency, and the in-and-out storage weight within a preset length time period.

4. The intelligent warehouse management method based on the Internet of Things according to claim 2 is characterized in that: The planning method further includes: Grouping multiple different goods in a single major category to obtain different groups of goods, each group includes multiple different goods; Note that a single major category of goods includes B groups of goods. The required storage area size of each group of goods is determined according to the total amount of each group of goods. B sub-storage areas are set in the single major category of goods storage area to store each group of goods respectively. The location of the storage area of ​​each group of goods is determined based on the second in-and-out warehouse cost value of the group of goods. The larger the second in-and-out warehouse cost value of the group of goods, the smaller the distance between the location of the storage area of ​​the group of goods and the warehouse entrance.

5. The intelligent warehouse management method based on the Internet of Things according to claim 1 is characterized in that: The mining of the association between different goods based on the goods type information on the outbound order and the inbound order also includes: The names of suppliers and buyers are marked on the individual material combinations formed; For the material combination time series, split it into material combination time series of different suppliers and buyers according to the supplier and buyer names; Classify different suppliers and buyers based on the consistency of the types of purchased materials and the similarity of their business scopes to obtain at least one supplier and buyer cluster; Based on the material combination example time series of the supplier purchaser in the same supplier purchaser cluster, a single material combination example in the material combination example time series is encoded to obtain a coding vector corresponding to the single material combination example, and then a coding vector group corresponding to the material combination example time series is obtained; Based on the coding vector group input spatiotemporal coding model, the spatial combination correlation relationship and temporal combination correlation relationship of different types of materials mined from the inbound / outbound orders of different suppliers and buyers are obtained; The results output by the spatiotemporal coding model are passed through a fully connected layer to obtain the correlation between different goods.

6. The intelligent warehouse management method based on the Internet of Things according to claim 1 is characterized in that: The step of determining the storage location of the goods to be stored in the warehouse according to the pre-planned storage layout based on the goods attribute information of the goods to be stored includes: (1) Obtaining the property information of the goods to be stored, wherein the property information includes the category information to which the goods belong; (2) Based on the cargo attribute information of the cargo to be stored, check whether a storage area corresponding to the cargo already exists in the existing storage layout. If so, proceed to step (3); otherwise, proceed to step (4); (3) storing the goods to be stored according to the remaining space in the storage area corresponding to the goods to be stored in the existing storage layout; (4) In the remaining free areas of the entire warehouse, the storage locations of the goods to be stored are determined based on the property information of the goods to be stored, the correlation between the goods to be stored and other goods, and the distance between different locations in the free areas and the storage entrance, thus forming a new warehouse layout.

7. An intelligent warehouse management system based on the Internet of Things, characterized in that: include: The incoming scanning unit is used to scan the goods information with an RFID reader before the goods are put into storage, and update the storage goods information database; A storage layout unit is used to determine the storage location of the goods to be stored in the warehouse according to the property information of the goods to be stored and the pre-planned storage layout. The property information of the goods includes the name, volume, weight and classification information of the goods. The storage layout determines the storage location of the goods according to the categories to which the goods belong and the groups under the categories. A storage and handling unit is used to transport the goods to be stored to the storage location using an automatic guided vehicle (AGV) according to the storage location; The warehouse monitoring unit is used to collect warehouse temperature and humidity data and images of materials stored in the warehouse in real time based on the temperature sensors, humidity sensors, light sensors, and cameras pre-installed in the warehouse, and monitor the warehouse temperature and humidity status and material status; In the storage layout unit, a method for determining the group to which goods belong under a major category includes: mining the correlation between different goods based on the goods type information on the outbound and inbound orders, wherein the correlation represents the correlation between the spatial combination and the time combination of different goods; calculating the inbound and outbound workload of different goods based on the inbound and outbound information of goods in a preset long period of time; iteratively grouping a plurality of different goods in a single major category based on the balance of the total inbound and outbound workload of goods of different groups in a preset long period of time and the maximum correlation of goods in the same group as the optimization target; and calculating the correlation between goods in the same group based on the balance of the total inbound and outbound workload of goods of different groups in a preset long period of time. for: ,in, represents the kth group, m represents the total number of groups, i and j represent different goods, represents the set of goods in the kth group, P represents the degree of association, Indicates the correlation between goods i and goods j in the kth group; the goods in the same group have the largest correlation, and max is used Calculate the balance of the total workload of the different groups of goods in and out of the warehouse for: , represents the total workload of the kth group of goods in and out of the warehouse, It represents the mean of the total in-and-out workload of m groups of goods. The total in-and-out workload of different groups of goods is optimally balanced, and min is used. calculate; The method of mining the association between different goods based on the goods type information on the outbound and inbound orders includes: obtaining the required material type information of a single supplier and buyer on a single outbound / inbound order based on the material type information and supplier and buyer information on a single outbound / inbound order to form a single material combination example; obtaining a time series of material combination examples in chronological order based on outbound / inbound orders at different times; encoding a single material combination example in the material combination example time series to obtain a coding vector corresponding to the single material combination example, and then obtaining a coding vector group corresponding to the material combination example time series; inputting a spatiotemporal coding model based on the coding vector group to obtain a spatial combination association relationship and a temporal combination association relationship representing different types of materials mined from outbound / inbound orders at different times; and obtaining the association between different goods through a fully connected layer based on the result output by the spatiotemporal coding model.

8. An intelligent warehouse management device based on the Internet of Things, characterized in that: The device comprises: processor; a memory for storing processor-executable instructions; The processor implements the intelligent warehousing management method based on the Internet of Things as described in any one of claims 1 to 6 by running the executable instructions.

Images