Method, device and equipment for processing to-be-warehoused cargo data and storage medium

Abstract

The present application relates to the technical field of logistics, and provides a method, device and equipment for processing data of goods to be warehoused and a storage medium, which are used to improve the matching accuracy of a goods placement strategy of a goods placement storage rack based on the condition of high outbound loading efficiency. The method for processing data of goods to be warehoused comprises the following steps: classifying the goods to be warehoused according to the time to be assembled to obtain target classified goods; searching a preset rack knowledge graph according to the attribute feature information of the goods, the time to be assembled and the vehicle information to be assembled to obtain target rack information; classifying, priority sorting and extracting the target classified goods with the same priority by using a preset classification model and the goods information to obtain target warehoused goods; and generating target rack placement information corresponding to the target warehoused goods by using a preset loading model and the target rack information based on a preset target function and an assembly strategy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of logistics, and in particular to a method and device for processing data of goods to be stored in a warehouse, equipment and a storage medium. BACKGROUND

[0002] With the development of Internet technology, the logistics industry has also developed, and the goods to be delivered are also increasing. The warehouse management of goods has become one of the matters of concern in the industry. At present, the warehouse management of goods generally adopts a method of matching the corresponding goods storage shelves according to the volume data of the goods, the volume data and the vacancy information of the storage shelves.

[0003] However, the above-mentioned warehouse management of goods has low matching accuracy of the goods storage strategy of the goods placed in the storage shelves based on high outbound loading efficiency, because the matching factors of the goods placed in the storage shelves are single, and the auxiliary analysis of the goods storage strategy of the goods placed in the storage shelves is low. SUMMARY

[0004] The present application provides a method and device for processing data of goods to be stored in a warehouse, equipment and a storage medium, which can improve the matching accuracy of the goods storage strategy of the goods placed in the storage shelves based on high outbound loading efficiency.

[0005] The first aspect of the present application provides a method for processing data of goods to be stored in a warehouse, comprising:

[0006] Obtaining goods information and to-be-assembled information of goods to be stored in a warehouse, the goods information comprising goods attribute feature information of the goods to be stored in a warehouse, and the to-be-assembled information comprising to-be-assembled time and to-be-assembled vehicle information;

[0007] According to the to-be-assembled time, the goods to be stored in a warehouse are classified and processed to obtain target classified goods;

[0008] According to the goods attribute feature information, to-be-assembled time and to-be-assembled vehicle information corresponding to the target classified goods, a preset shelf knowledge graph is searched to obtain target shelf information, the target shelf information comprising position information, volume information and bearing information of the target shelf;

[0009] Through a preset classification model and the goods information, the target classified goods are classified, prioritized and extracted with equal priority to obtain target storage goods;

[0010] The target storage rack placement information corresponding to the target storage goods is generated based on a preset target function and an assembly strategy, the target function is a function constructed based on a greedy algorithm, and the target storage rack placement information includes a placement combination and a placement order of a storage rack corresponding to the target storage goods.

[0011] Optionally, in the first implementation manner of the first aspect, the classifying, priority sorting and same-priority extracting of the target classification goods based on the preset classification model and the goods information to obtain the target storage goods comprises:

[0012] A plurality of classifiers in the preset classification model are used to calculate a transportation category probability value of the target classification goods based on the goods information, to obtain a plurality of classification values corresponding to each transportation category respectively;

[0013] A mean value of the plurality of classification values corresponding to each transportation category respectively is calculated to obtain a target probability value corresponding to each transportation category respectively, and a target transportation category of the target classification goods is determined according to the target probability value corresponding to each transportation category respectively, and the target classification goods of which the target transportation category is determined are determined as the goods to be sorted;

[0014] The priority matching of the goods to be sorted is performed based on a preset priority decision tree to obtain initial storage goods;

[0015] The initial storage goods are classified according to the same priority to obtain the target storage goods.

[0016] Optionally, in the second implementation manner of the first aspect, the generating of the target storage rack placement information corresponding to the target storage goods based on the preset loading model and the target storage rack information comprises:

[0017] The target storage rack placement information corresponding to the target storage goods is generated based on the preset loading model and the target storage rack information, the target function is a function constructed based on a greedy algorithm, and the target storage rack placement information includes a placement combination and a placement order of a storage rack corresponding to the target storage goods.

[0018] The target storage rack placement information corresponding to the target storage goods is generated based on the preset loading model and the target storage rack information, the target function is a function constructed based on a greedy algorithm, and the target storage rack placement information includes a placement combination and a placement order of a storage rack corresponding to the target storage goods.

[0019] The placement order of the initial storage rack placement information is adjusted based on a preset assembly strategy to obtain candidate storage rack placement information;

[0020] The candidate storage rack placement information is verified to obtain the target storage rack placement information.

[0021] Optionally, in a third implementation form of the first aspect of the present application, the verifying the candidate shelf placement information to obtain target shelf placement information comprises:

[0022] The product information and the target shelf information are subjected to shelf placement information calculation and selection through a preset genetic algorithm, dynamic programming algorithm and assembly strategy to obtain comparative shelf placement information.

[0023] The similarity value between the candidate shelf placement information and the comparative shelf placement information is calculated through a preset similarity algorithm.

[0024] It is judged whether the similarity value is greater than a preset comparison value, and if the similarity value is greater than the preset comparison value, the candidate shelf placement information is determined as the target shelf placement information.

[0025] Optionally, in a fourth implementation form of the first aspect of the present application, the classifying the to-be-assembled goods according to the to-be-assembled time to obtain target classified goods comprises:

[0026] The to-be-assembled goods are sorted according to the time sequence of the to-be-assembled time to obtain a goods sequence.

[0027] The goods with the same to-be-assembled time in the goods sequence are classified into the same time set to obtain initial classified goods.

[0028] The goods assembly value information of the initial classified goods is obtained, and the goods assembly value information is compared and analyzed with preset assembly value information to obtain target assembly value information meeting the preset assembly value information.

[0029] The initial classified goods corresponding to the target assembly value information are determined as the target classified goods.

[0030] Optionally, in a fifth implementation form of the first aspect of the present application, the retrieving a preset shelf knowledge graph according to the product attribute feature information, to-be-assembled time and to-be-assembled vehicle information corresponding to the target classified goods to obtain target shelf information comprises:

[0031] The product attribute feature information, to-be-assembled time and to-be-assembled vehicle information corresponding to the target classified goods are subjected to keyword extraction and keyword entity pair generation in sequence to obtain entity information.

[0032] The preset shelf knowledge graph is subjected to random walk and information matching through the entity information to obtain the target shelf information.

[0033] Optionally, in a sixth implementation form of the first aspect of the present application, after the preset loading model and the target shelf information are used to generate the target shelf placement information corresponding to the target warehouse-in cargo based on a preset target function and an assembly strategy, the method further includes:

[0034] The warehouse-in shelf placement and warehouse-out loading of the warehouse-in cargo are performed based on the target shelf placement information through a preset simulation system to obtain simulation information, and the loading model is optimized according to the simulation information.

[0035] The second aspect of the present application provides a processing device for warehouse-in cargo data, which includes:

[0036] An acquisition module is configured to acquire cargo information and assembly information of a warehouse-in cargo, wherein the cargo information includes attribute characteristic information of the warehouse-in cargo, and the assembly information includes assembly time and assembly vehicle information;

[0037] A classification module is configured to perform classification processing on the warehouse-in cargo according to the assembly time to obtain target classified cargo.

[0038] A retrieval module is configured to perform retrieval on a preset shelf knowledge graph based on attribute characteristic information of the target classified cargo, assembly time and assembly vehicle information to obtain target shelf information, wherein the target shelf information includes position information, volume information and bearing information of a target shelf.

[0039] A classification extraction module is configured to perform classification, priority sorting and same-priority extraction on the target classified cargo based on a preset classification model and the cargo information to obtain target warehouse-in cargo.

[0040] A generation module is configured to generate target shelf placement information corresponding to the target warehouse-in cargo based on a preset target function and an assembly strategy through a preset loading model and the target shelf information, wherein the target function is a function constructed based on a greedy algorithm, and the target shelf placement information includes placement combination and placement order of a shelf corresponding to the target warehouse-in cargo.

[0041] Optionally, in a first implementation form of the second aspect of the present application, the classification extraction module is specifically configured to:

[0042] A plurality of classifiers in the preset classification model are used to calculate transport category probability values of the target classified cargo based on the cargo information to obtain a plurality of classification values corresponding to respective transport categories.

[0043] Calculate mean values of the classification values corresponding to each transport category respectively to obtain target probability values corresponding to each transport category respectively, and determine a target transport category of the target classified goods according to the target probability values corresponding to each transport category respectively, and determine the target classified goods with the target transport category as the goods to be sorted;

[0044] Match the priority of the goods to be sorted through the preset priority decision tree to obtain initial warehouse-in goods;

[0045] According to the same priority, the initial warehouse-in goods are classified to obtain target warehouse-in goods.

[0046] Optionally, in the second implementation manner of the second aspect of the present application, the generation module comprises:

[0047] The calculation unit is configured to calculate a rack load constraint value and a rack volume constraint value corresponding to the target rack information through a preset loading model and the target rack information;

[0048] The prediction unit is configured to predict a rack placement combination of the target warehouse-in goods based on a preset target function, the target rack information, the rack load constraint value and the rack volume constraint value to obtain initial rack placement information;

[0049] The adjustment unit is configured to adjust the placement order of the initial rack placement information through a preset assembly strategy to obtain candidate rack placement information;

[0050] The verification unit is configured to verify the candidate rack placement information to obtain target rack placement information.

[0051] Optionally, in the third implementation manner of the second aspect of the present application, the verification unit is specifically configured to:

[0052] Calculate and select the rack placement information of the goods information and the target rack information through a preset genetic algorithm, dynamic programming algorithm and assembly strategy to obtain comparative rack placement information;

[0053] Calculate a similarity value between the candidate rack placement information and the comparative rack placement information through a preset similarity algorithm;

[0054] Judge whether the similarity value is greater than a preset comparison value, and if the similarity value is greater than the preset comparison value, determine the candidate rack placement information as the target rack placement information.

[0055] Optionally, in the fourth implementation manner of the second aspect of the present application, the classification module is specifically configured to:

[0056] sequencing the to-be-warehoused goods according to the time sequence of the to-be-assembled time, to obtain a goods sequence;

[0057] classifying the goods with the same to-be-assembled time in the goods sequence into a same time set, to obtain initial classified goods;

[0058] obtaining goods assembly value information of the initial classified goods, comparing and analyzing the goods assembly value information with preset assembly value information, to obtain target assembly value information meeting the preset assembly value information;

[0059] determining the initial classified goods corresponding to the target assembly value information as target classified goods.

[0060] Optionally, in a fifth implementation manner of the second aspect of the present application, the searching module is specifically configured to:

[0061] performing keyword extraction and keyword entity pair generation on the goods attribute feature information, the to-be-assembled time and the to-be-assembled vehicle information corresponding to the target classified goods in sequence, to obtain entity information;

[0062] performing random walk and information matching on a preset goods shelf knowledge graph through the entity information, to obtain target goods shelf information.

[0063] Optionally, in a sixth implementation manner of the second aspect of the present application, the to-be-warehoused goods data processing device further includes:

[0064] a simulation optimization module configured to perform warehousing goods shelf placement and outbound loading on the to-be-warehoused goods based on the target goods shelf placement information through a preset simulation system, to obtain simulation information, and optimize the assembly and loading model according to the simulation information.

[0065] The third aspect of the present application provides a to-be-warehoused goods data processing device, which includes a memory and at least one processor, the memory has instructions stored therein; the at least one processor invokes the instructions in the memory, so that the to-be-warehoused goods data processing device executes the to-be-warehoused goods data processing method described above.

[0066] The fourth aspect of the present application provides a computer readable storage medium, which has instructions stored therein, when the instructions are run on a computer, the computer executes the to-be-warehoused goods data processing method described above.

[0067] The technical solution provided by this invention involves obtaining the goods information and assembly information of goods to be received into the warehouse. The goods information includes the attribute characteristics of the goods to be received into the warehouse, and the assembly information includes the assembly time and the vehicle information to be assembled. Based on the assembly time, the goods to be received into the warehouse are categorized to obtain target category goods. Based on the attribute characteristics, assembly time, and vehicle information corresponding to the target category goods, a pre-set shelf knowledge graph is retrieved to obtain target shelf information, which includes the location, volume, and load-bearing capacity of the target shelf. Using a pre-set classification model and goods information, the target category goods are classified, prioritized, and extracted with equal priority to obtain target goods to be received into the warehouse. Using a pre-set loading model and target shelf information, and based on a pre-set objective function and assembly strategy, target shelf placement information corresponding to the target goods to be received is generated. The objective function is a function constructed based on a greedy algorithm, and the target shelf placement information includes the placement combination and order of the shelves corresponding to the target goods to be received into the warehouse. In this embodiment of the invention, by classifying the goods to be received, and based on the product attribute characteristics, assembly time, and vehicle information corresponding to the target category of goods, a pre-set shelf knowledge graph is retrieved. Using a pre-set loading model and target shelf information, and based on a pre-set objective function and assembly strategy, target shelf placement information corresponding to the target goods is generated. Through classification and shelf knowledge graph retrieval, the matching accuracy of the target shelf is improved. By combining target shelf information and assembly strategies, the diversity of matching factors for the target shelf is enhanced, and the analysis accuracy of the target goods placement information is improved. This, in turn, improves the matching accuracy of goods placement strategies for warehouse shelves under the condition of high outbound loading efficiency. Attached Figure Description

[0068] Figure 1 This is a schematic diagram of an embodiment of the method for processing data of goods to be put into storage according to the present invention;

[0069] Figure 2 This is a schematic diagram of another embodiment of the method for processing data of goods to be put into storage in this invention;

[0070] Figure 3 This is a schematic diagram of an embodiment of the device for processing data of goods to be put into storage according to the present invention;

[0071] Figure 4 This is a schematic diagram of another embodiment of the device for processing data of goods to be put into storage in this invention;

[0072] Figure 5 This is a schematic diagram of one embodiment of the equipment for processing data of goods to be put into storage in this invention. Detailed Implementation

[0073] The embodiment of the present application provides a kind of to be put into warehouse goods data processing method, device, equipment and storage medium, improve the matching accuracy of goods placement strategy of goods placement warehouse rack based on high out-of-warehouse loading efficiency condition.

[0074] The terms "first", "second", "third", "fourth" and the like in the description, claims, and drawings of the present application, and the above-mentioned drawings (if any) are used to distinguish similar objects, and do not necessarily have to be described in a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0075] For ease of understanding, the specific flow of the embodiments of the present application is described below. Please refer to Figure 1 One embodiment of the method for processing data of goods to be put into warehouse in the embodiments of the present application includes:

[0076] 101, obtain the goods information and the to-be-assembled information of the goods to be put into warehouse, the goods information includes the goods attribute characteristic information of the goods to be put into warehouse, and the to-be-assembled information includes the to-be-assembled time and the to-be-assembled vehicle information.

[0077] It can be understood that the execution subject of the present application can be a data processing device for goods to be put into warehouse, and can also be a terminal or a server, and the specific place is not limited. The embodiments of the present application take the server as the execution subject for example.

[0078] In addition to the goods attribute characteristic information of the goods to be put into warehouse, the goods information can also include logistics information of the goods to be put into warehouse, which can include but is not limited to transportation information, address information of waybill and valuation data (valuation data includes valuation amount data, valuation type and valuation weight, etc.). In addition to the to-be-assembled time and the to-be-assembled vehicle information, the to-be-assembled information can also include to-be-assembled sequence, to-be-assembled batch, to-be-assembled vehicle information including type, rated volume data and rated load weight data of to-be-assembled vehicle, etc.

[0079] The server can obtain the product information and the assembly information of the to-be-warehoused goods by reading the product information and the assembly information of the to-be-warehoused goods stored in the logistics system. It should be noted that the server obtains the product information and the assembly information of the to-be-warehoused goods at a time before the to-be-warehoused goods are warehoused, that is, the server receives the positioning information of the vehicle carrying the to-be-warehoused goods in real time, calculates the arrival time of the vehicle by the positioning information and the position information of the warehouse and the speed of the vehicle carrying the to-be-warehoused goods, calls a preset timer, reads the product information and the assembly information of the to-be-warehoused goods from the preset database or the logistics system when the timer counts to the preset time of the arrival time of the vehicle, thereby enhancing the orderliness of the operation, reducing the calculation amount, and improving the reading efficiency of the product information and the assembly information of the to-be-warehoused goods.

[0080] 102. Classify the to-be-warehoused goods according to the assembly time to obtain target classified goods.

[0081] For example, the to-be-warehoused goods are A1-A12, the assembly time of A2, A3, A4, A8 and A9 is March 2, the assembly time of A7, A10 and A11 is the morning period of March 3, and the assembly time of A1 and A12 is the afternoon period of March 3. A2, A3, A4, A8 and A9 are classified as assembly goods B1 on March 2, A7, A10 and A11 are classified as assembly goods B2 in the morning of March 3, A1 and A12 are classified as assembly goods B3 in the afternoon of March 3, B1, B2 and B3 are sorted to obtain a classified goods sequence set B1, B2 and B3, and the classified goods sequence set B1, B2 and B3 is determined as the target classified goods.

[0082] 103. According to the product attribute feature information corresponding to the target classified goods, the assembly time and the assembly vehicle information, the preset shelf knowledge graph is searched to obtain target shelf information, and the target shelf information includes position information, volume information and bearing information of the target shelf.

[0083] The server pre-creates a shelf knowledge graph, which includes regional shelves corresponding to each type of goods and shelf information of the regional shelves. The shelf information includes, but is not limited to, position information, volume information, load-bearing information, item type, shelf number, shelf region information, and delivery convenience, etc. For example, the shelves of the warehouse are divided into regions to obtain regional shelves. The regions include regions that can be matched by a preset matching model and target shelf information, and regions that cannot be matched by the preset matching model and the target shelf information. Based on a preset target function and assembly strategy, target shelf placement information corresponding to the target warehouse-in goods is generated to plan and assemble the deployable regions. In addition, the regions that cannot be matched by the preset matching model and the target shelf information are based on the preset target function and the assembly strategy, and the target shelf placement information corresponding to the target warehouse-in goods is generated to plan and assemble the non-deployable regions (i.e. goods that need to be placed and matched, such as chemical goods). The shelves of each region (i.e. regional shelves) will indicate the types of goods that can be placed. In addition to the position information, volume information, and load-bearing information of the target shelf, the target shelf information can also include the item type, shelf number, shelf region information, and delivery convenience of the goods that can be placed on the target shelf, and the delivery convenience can correspond to the waiting assembly time. If the waiting assembly time is early or relatively urgent, the shelf with high delivery convenience is matched to improve the loading efficiency.

[0084] The server traverses and matches the node information of the preset shelf knowledge graph according to the goods attribute feature information, the waiting assembly time, and the waiting assembly vehicle information of the target classified goods, to obtain target shelf information that meets the waiting assembly vehicle information (such as the loading volume of the vehicle), the delivery convenience corresponding to the waiting assembly time, and the goods attribute feature information that meets the shelf placement conditions.

[0085] The server can generate a goods structured query language, a time structured query language, and a vehicle structured query language by respectively structuring the goods attribute feature information, the waiting assembly time, and the waiting assembly vehicle information. The goods structured query language, the time structured query language, and the vehicle structured query language are spliced to obtain a target structured query language. The target structured query language is used to query and match the information of the preset shelf knowledge graph, so as to obtain the target shelf information.

[0086] 104. The target warehouse-in goods are obtained by classifying, prioritizing, and extracting goods with the same priority through a preset classification model and goods information.

[0087] The server calls a preset classification model, which can be composed of a comprehensive framework connected by multiple network structures (such as gradient boosting decision tree network structure, random forest regression network structure, and gradient boosting tree network structure) in a preset connection mode. By the preset classification model and the goods information, the target classified goods are classified according to a preset classification category, and the classified goods are obtained. The preset classification category can include the types of loaded goods, such as frozen food, quick-frozen food, bulk goods, and box goods. The classified goods are prioritized and labeled according to a preset priority, and the prioritized goods are obtained. The priority includes the weight (such as light and heavy) and / or the delivery urgency (such as slow and fast) of the goods. The goods with the same priority in the prioritized goods are classified into the same level goods, thereby obtaining the target warehouse goods. For example, the prioritized goods are E1 (very urgent), E2 (very urgent), E3 (very urgent), E4 (relatively urgent), E5 (relatively urgent), E6 (not urgent), E7 (not urgent), and E8 (not urgent). E1, E2, and E3 are classified into the very urgent level goods F1, i.e., the target warehouse goods. E4 and E5 are classified into the relatively urgent level goods F2, i.e., the target warehouse goods. E6, E7, and E8 are classified into the not urgent level goods F3, i.e., the target warehouse goods.

[0088] 105. By the preset loading model and the target shelf information, the target shelf placement information corresponding to the target warehouse goods is generated based on a preset target function and a loading strategy. The target function is a function constructed based on a greedy algorithm, and the target shelf placement information includes the placement combination and the placement order of the target warehouse goods corresponding to the shelf.

[0089] The server constructs a loading model based on a preset greedy algorithm and a loading strategy in advance, and establishes a target function of the loading model. The loading strategy is, for example, according to the address of the shipping order, the goods unloaded first are placed on the left side of the shelf, and the goods unloaded later are placed on the rear side of the shelf. When the goods are placed up and down, according to the principle of heavy not pressing light and large not pressing small, the goods with light priority are placed on the goods with heavy priority, and the center of gravity of the goods is balanced after stacking.

[0090] The server calculates a rack average density value of the target rack according to the target rack information, calculates a cargo average density value of the target warehouse cargo according to the cargo information, obtains a rack load constraint value and a rack volume constraint value corresponding to the target rack information, and obtains rated load data, rated volume data and a cargo type of the to-be-assembled vehicle information in the to-be-assembled information, takes the rack average density value, the cargo average density value, the rack load constraint value and the rack volume constraint value corresponding to the target rack information, and the rated load data, the rated volume data and the cargo type of the to-be-assembled vehicle information in the to-be-assembled information as parameter factors, and performs prediction through the parameter factors and the target function to obtain prediction of rack placement combination of the target warehouse cargo, and obtain initial rack placement information; and the initial rack placement information is adjusted in placement order through the preset assembly strategy to obtain target rack placement information, and the target rack placement information includes placement combination and placement order of the rack corresponding to the target warehouse cargo.

[0091] Through the preset loading model and the target rack information, the target rack placement information corresponding to the target warehouse cargo is generated based on the preset target function and the assembly strategy, which avoids the problem that repeated multiple exhaustive iterations are time-consuming and the result cannot meet the expectation, improves the calculation efficiency, improves the matching accuracy of the target rack placement information, and further improves the matching accuracy of the cargo placement strategy of the warehouse rack.

[0092] In the embodiment of the present application, the to-be-warehoused cargo is classified and processed, the preset rack knowledge graph is searched according to the cargo attribute feature information corresponding to the target classified cargo, the to-be-assembled time and the to-be-assembled vehicle information, the target rack placement information corresponding to the target warehouse cargo is generated through the preset loading model and the target rack information based on the preset target function and the assembly strategy, the classification processing and the rack knowledge graph searching improve the matching accuracy of the target rack, the combination of the target rack information and the assembly strategy enhances the diversity of the matching factors of the target rack, and the analysis accuracy of the target cargo placement information is improved, thereby improving the matching accuracy of the cargo placement strategy of the warehouse rack under the condition of high warehouse loading efficiency.

[0093] Please refer to Figure 2 Another embodiment of the method for processing to-be-warehoused cargo data in the embodiment of the present application includes:

[0094] 201, obtain cargo information of to-be-warehoused cargo and to-be-assembled information, the cargo information includes cargo attribute feature information of the to-be-warehoused cargo, and the to-be-assembled information includes to-be-assembled time and to-be-assembled vehicle information.

[0095] The server can send a reading request to the preset logistics system, the reading request can include warehouse information and information type of reading, the logistics system receives the reading request, analyzes the reading request and performs keyword extraction to obtain keyword information, generates a key value of the keyword information, performs key value matching on a pre-stored goods information hash table through the key value, obtains corresponding goods information of the to-be-warehoused goods, the server traverses the assembly information decision tree stored in the preset database according to the goods information of the to-be-warehoused goods, and obtains corresponding to-be-assembled information.

[0096] 202. Classify the to-be-warehoused goods according to the to-be-assembled time to obtain target classified goods.

[0097] Specifically, the server sorts the to-be-warehoused goods according to the time sequence of the to-be-assembled time to obtain a goods sequence; classifies the goods with the same to-be-assembled time in the goods sequence into the same time set to obtain initial classified goods; obtains goods assembly value information of the initial classified goods, compares and analyzes the goods assembly value information with preset assembly value information to obtain target assembly value information meeting the preset assembly value information; and determines the initial classified goods corresponding to the target assembly value information as the target classified goods.

[0098] The preset assembly value information includes, but is not limited to, preset goods value data corresponding to the initial classified goods, user importance level data, and goods danger level number, etc. For example, the to-be-assembled time of the to-be-warehoused goods C1-C12, C2, C3, C4, C8 and C9 is March 2, the to-be-assembled time of C7, C10 and C11 is the morning period of March 3, and the to-be-assembled time of C1 and C12 is the afternoon period of March 3. According to the time sequence of the to-be-assembled time, C1-C12 is sorted to obtain a goods sequence C2, C3, C4, C8, C9, C7, C10, C11, C1 and C12. C2, C3, C4, C8 and C9 in the goods sequence C2, C3, C4, C8, C9, C7, C10, C11, C1 and C12 are classified as initial classified goods 1, C7, C10 and C11 in the goods sequence C2, C3, C4, C8, C9, C7, C10, C11, C1 and C12 are classified as initial classified goods 2, and C1 and C12 in the goods sequence C2, C3, C4, C8, C9, C7, C10, C11, C1 and C12 are classified as initial classified goods 3. The article danger level number (i.e., the goods assembly value information) of each goods in the initial classified goods 1 is obtained. The goods with an article danger level number meeting a preset goods danger level number (taking the preset assembly value information as the goods danger level number as an example) are filtered out from the initial classified goods 1 to obtain target classified goods 1. The filtered-out goods are processed according to a preset processing strategy. Similarly, the initial classified goods 2 and the initial classified goods 3 correspond to target classified goods 2 and target classified goods 3, respectively.

[0099] 203. According to the goods attribute feature information corresponding to the target classified goods, the to-be-assembled time and the to-be-assembled vehicle information, the preset goods shelf knowledge graph is searched to obtain target goods shelf information. The target goods shelf information includes position information, volume information and bearing information of the target goods shelf.

[0100] Specifically, the server sequentially performs keyword extraction and keyword entity pair generation on the goods attribute feature information corresponding to the target classified goods, the to-be-assembled time and the to-be-assembled vehicle information to obtain entity information. Through the entity information, random walk and information matching are performed on the preset goods shelf knowledge graph to obtain target goods shelf information.

[0101] The server respectively performs keyword extraction and keyword entity pair generation on the goods attribute feature information corresponding to the target classified goods, the to-be-assembled time and the to-be-assembled vehicle information in sequence, obtains entity information, and then calls a preset random walk algorithm to perform random walk on a preset shelf knowledge graph to obtain multiple initial sequences. The similarity between the entity information and the multiple initial sequences is calculated to obtain multiple sequence similarities. It is determined whether each sequence similarity is greater than a preset threshold. If yes, the initial sequence is determined as a candidate sequence. If no, the execution of the sequence is stopped. The number of candidate sequences is obtained, and it is determined whether the number is greater than 1. If yes, the candidate sequences are sorted in descending order of the values of the sequence similarities. The candidate sequence ranked first is determined as a target sequence. If no, the candidate sequence is determined as the target sequence. The shelf information corresponding to the target sequence is determined as target shelf information.

[0102] 204. The target classified goods are classified, prioritized and same-priority extracted by the preset classification model and the goods information to obtain a target warehouse-in goods.

[0103] Specifically, the server calculates the transportation category probability value of the target classified goods based on the goods information by using multiple classifiers in the preset classification model to obtain multiple classification values corresponding to each transportation category. The mean value of the multiple classification values corresponding to each transportation category is calculated to obtain a target probability value corresponding to each transportation category. The target transportation category of the target classified goods is determined according to the target probability value corresponding to each transportation category, and the target classified goods whose target transportation category is determined are determined as to-be-sequenced goods. The to-be-sequenced goods are prioritized by using a preset priority decision tree to obtain initial warehouse-in goods. The initial warehouse-in goods are classified according to the same priority to obtain the target warehouse-in goods.

[0104] For example, the server calculates the transportation category probability values of the target classified goods G based on the goods information through multiple classifiers in the preset classification model, obtains multiple classification values corresponding to transportation category 1 as G1, G2, G3, G4, G5 and G6, multiple classification values corresponding to transportation category 2 as H1, H2, H3, H4, H5 and H6, and multiple classification values corresponding to transportation category 3 as I1, I2, I3, I4, I5 and I6, calculates (G1+G2+G3+G4+G5+G6) / 6=L1 to obtain the target probability value L1 corresponding to transportation category 1, calculates (H1+H2+H3+H4+H5+H6) / 6=L2 to obtain the target probability value L2 corresponding to transportation category 2, and calculates (I1+I2+I3+I4+I5+I6) / 6=L3 to obtain the target probability value L3 corresponding to transportation category 3, sorts the target probability value L1, the target probability value L2 and the target probability value L3 in descending order of value to obtain the target probability value L3 ranked first, and determines whether the target probability value L3 is greater than a preset target value. If yes, the transportation type corresponding to the target probability value L3 is determined as the target transportation category of the target classified goods, and if no, the classification of the classification model is continued until the target transportation category of the target classified goods is obtained, and the target classified goods with the determined target transportation category are determined as the goods to be sorted; the priority matching of the goods to be sorted is performed through the preset priority decision tree to obtain the initial warehouse-in goods; and the same priority goods in the priority sorted goods are classified as the same level goods according to the same priority, so as to obtain the target warehouse-in goods.

[0105] 205、Through the preset loading model and the target shelf information, the target shelf placement information corresponding to the target warehouse-in goods is generated based on the preset target function and the assembly strategy. The target function is a function constructed based on a greedy algorithm, and the target shelf placement information includes a placement combination and a placement order of the target warehouse-in goods corresponding to the shelf.

[0106] Specifically, the server calculates the shelf load constraint value and the shelf volume constraint value corresponding to the target shelf information through the preset loading model and the target shelf information; predicts the shelf placement combination of the target warehouse-in goods based on the preset target function, the target shelf information, the shelf load constraint value and the shelf volume constraint value to obtain the initial shelf placement information; adjusts the placement order of the initial shelf placement information through the preset assembly strategy to obtain the candidate shelf placement information; and verifies the candidate shelf placement information to obtain the target shelf placement information.

[0107] The target shelf information includes rated load data and rated volume data of each target shelf. Each target shelf includes a plurality of interval grids, and the shelf load constraint value and the shelf volume constraint value corresponding to the target shelf information include a shelf load constraint value and a shelf volume constraint value corresponding to each interval grid of each target shelf; and the target warehouse-in cargo shelf placement combination includes a target warehouse-in cargo shelf placement combination corresponding to each interval grid of each target shelf.

[0108] The server obtains the types of the target warehouse-in cargos on the same target shelf, calls a preset loading model and the target shelf information, and calculates the shelf load constraint value corresponding to the target shelf information by The server calculates the shelf volume constraint value corresponding to the target shelf information by The server calculates the shelf average density value of each target shelf based on the total shelf mass data and the total shelf volume data of each target shelf in the target shelf information, and the factors corresponding to the shelf load constraint value and the shelf volume constraint value, by The server calculates the target warehouse-in cargo average density value of each target shelf based on the total cargo mass data and the total cargo volume data of each target shelf in the product information, and the factors corresponding to the shelf load constraint value and the shelf volume constraint value, by The server calculates the target warehouse-in cargo average density value of each target shelf based on the total cargo mass data and the total cargo volume data of each target shelf in the product information, and the factors corresponding to the shelf load constraint value and the shelf volume constraint value, by kj W represents the total shelf mass data of the jth target warehouse-in cargo loaded on the kth target shelf, W represents the volume data of the jth target warehouse-in cargo loaded on the kth target shelf, V represents the rated load data of the kth target shelf, and V represents the rated volume data of the kth target shelf. k W represents the total shelf mass data of the jth target warehouse-in cargo loaded on the kth target shelf, W represents the volume data of the jth target warehouse-in cargo loaded on the kth target shelf, V represents the rated load data of the kth target shelf, and V represents the rated volume data of the kth target shelf. kj W represents the total shelf mass data of the jth target warehouse-in cargo loaded on the kth target shelf, W represents the volume data of the jth target warehouse-in cargo loaded on the kth target shelf, V represents the rated load data of the kth target shelf, and V represents the rated volume data of the kth target shelf. k W represents the total shelf mass data of the jth target warehouse-in cargo loaded on the kth target shelf, W represents the volume data of the jth target warehouse-in cargo loaded on the kth target shelf, V represents the rated load data of the kth target shelf, and V represents the rated volume data of the kth target shelf.

[0109] Based on the preset target function The target shelf information, the shelf load constraint value and the shelf volume constraint value are used to predict the target warehouse-in cargo shelf placement combination to obtain first shelf placement information of each target shelf, wherein the target function K in the target function represents the total number of target shelves, M1 represents the shelf average density value of the k target shelves, and M2 represents the target warehouse-in cargo average density value of the k target shelves.

[0110] ​The server calculates the rack load constraint value and the rack volume constraint value corresponding to each interval grid of each target rack through the preset loading model and the target rack information. Based on the preset target function, the target rack information, the rack load constraint value and the rack volume constraint value corresponding to each interval grid of each target rack, the server predicts the rack placement combination of the target warehouse-in cargo corresponding to each interval grid of each target rack to obtain the second rack placement information of each target rack. The specific operation is as follows: the server obtains the cargo type of the target warehouse-in cargo on the same target rack, and calls the preset loading model and the target rack information to calculate the rack load constraint value corresponding to each interval grid of each target rack through The rack volume constraint value corresponding to each interval grid of each target rack is calculated through The rack volume constraint value corresponding to each interval grid of each target rack is calculated through The rack average density value corresponding to each interval grid of each target rack is calculated through The average density value of the target warehouse-in cargo corresponding to each interval grid of each target rack is calculated, wherein n represents the cargo type of the target warehouse-in cargo on the same interval grid of the same target rack, J represents the Jth target warehouse-in cargo on the same interval grid, p represents the pth interval grid of the same target rack, w pJ represents the total mass data of the Jth target warehouse-in cargo loaded on the pth interval grid, w p represents the rated load data of the pth interval grid, v pJ represents the volume data of the Jth target warehouse-in cargo loaded on the pth target rack, v p represents the rated volume data of the pth interval grid.

[0111] Based on the preset target function The target rack information, the rack load constraint value and the rack volume constraint value corresponding to each interval grid of each target rack are used to predict the rack placement combination of the target warehouse-in cargo corresponding to each interval grid of each target rack to obtain the second rack placement information of each target rack, wherein the target function in the target function P represents the total number of interval grids of the same target rack, m1 represents the rack average density value of the pth interval grid of the same target rack, and m2 represents the average density value of the target warehouse-in cargo of the pth interval grid of the same target rack.

[0112] The first shelf placement information of each target shelf and the second shelf placement information of each target shelf are determined as initial shelf placement information; the initial shelf placement information is adjusted in placement order according to a preset assembly strategy (for example, according to the address of the delivery order, the goods unloaded first are placed on the left side of the shelf, and the goods unloaded later are placed on the rear side of the shelf; when the goods are placed up and down, according to the principle of heavy not pressing light and large not pressing small, the goods with a light priority are placed on the goods with a heavy priority, and the center of gravity of the goods is balanced after stacking), to obtain candidate shelf placement information; the candidate shelf placement information is verified by a preset algorithm to obtain target shelf placement information.

[0113] Specifically, the server calculates and selects the shelf placement information of the goods information and the target shelf information by a preset genetic algorithm, a dynamic programming algorithm and an assembly strategy, to obtain comparative shelf placement information; the similarity value between the candidate shelf placement information and the comparative shelf placement information is calculated by a preset similarity algorithm; it is judged whether the similarity value is greater than a preset comparison value, if the similarity value is greater than the preset comparison value, the candidate shelf placement information is determined as the target shelf placement information.

[0114] The server can generate and select the initial genetic selection shelf placement information of the goods information and the target shelf information by a preset genetic algorithm, and adjust the initial genetic selection shelf placement information by a preset assembly strategy to obtain the target genetic selection shelf placement information; the initial planning shelf placement information is obtained by solving the structural characteristics of the optimal solution of the shelf placement information of the goods information and the target shelf information by a preset dynamic programming algorithm, and the initial planning shelf placement information is adjusted by a preset assembly strategy to obtain the target planning shelf placement information; the target genetic selection shelf placement information and the target planning shelf placement information are combined to obtain the comparative shelf placement information.

[0115] The server can also fuse the preset genetic algorithm and the dynamic programming algorithm to obtain a fusion algorithm, calculate the fusion shelf placement information by the fusion algorithm, and adjust the fusion shelf placement information by a preset assembly strategy to obtain the comparative shelf placement information, wherein the fusion can be to take the output of the genetic algorithm as the input of the dynamic programming algorithm, or to take the output of the dynamic programming algorithm as the input of the genetic algorithm, or to calculate the first shelf placement information of each target shelf by the genetic algorithm (or the dynamic programming algorithm), and to calculate the second shelf placement information of each target shelf by the dynamic programming algorithm (or the genetic algorithm).

[0116] The server calculates the similarity value between the candidate shelf placement information and the comparison shelf placement information through a preset similarity algorithm (such as a cosine similarity function); determines whether the similarity value is greater than a preset comparison value, if yes, determines the candidate shelf placement information as the target shelf placement information, if no, re-calculates the candidate shelf placement information until the target shelf placement information is obtained.

[0117] 206. The target shelf placement information is used to simulate the warehousing and loading of the target warehousing goods through a preset simulation system, to obtain simulation information, and the simulation information is used to optimize the loading model.

[0118] The server calls a preset simulation system (which is used to simulate the warehousing, loading and loading of the target warehousing goods based on the target shelf placement information), and uses the target shelf placement information to simulate the warehousing and loading of the target warehousing goods, and calculates the placement time of the warehousing and the loading time of the loading, and obtains the simulation shelf placement image after the warehousing and the simulation vehicle compartment image after the loading, calls a preset image processing model to segment the empty area, extract the features and perform convolution processing on the simulation shelf placement image to obtain the shelf empty information, calls a preset image processing model to segment the empty area, extract the features and perform convolution processing on the simulation vehicle compartment image to obtain the vehicle compartment empty information, and determines the placement time, the loading time, the shelf empty information and the vehicle compartment empty information as the simulation information, and uses the simulation information and a preset optimization algorithm to optimize the model parameters and network structure of the loading model.

[0119] By optimizing the loading model according to the simulation information, the accuracy of the loading model is improved, thereby improving the matching accuracy of the goods placement strategy of the goods placement storage shelf under the condition of high loading efficiency.

[0120] In the embodiment of the application, not only the matching accuracy of the target shelf is improved, the diversity of the matching factors of the target shelf is enhanced, and the analysis accuracy of the target goods placement information is improved, thereby improving the matching accuracy of the goods placement strategy of the goods placement storage shelf under the condition of high loading efficiency, but also the accuracy of the loading model is improved by optimizing the loading model according to the simulation information, thereby improving the matching accuracy of the goods placement strategy of the goods placement storage shelf under the condition of high loading efficiency.

[0121] The processing method of the warehousing goods data in the embodiment of the application is described above, and the processing device of the warehousing goods data in the embodiment of the application is described below, please refer to Figure 3 The processing device of the warehousing goods data in the embodiment of the application includes one embodiment:

[0122] The acquisition module 301 is configured to acquire goods information of the to-be-warehoused goods and to-be-assembled information, the goods information comprising goods attribute characteristic information of the to-be-warehoused goods, and the to-be-assembled information comprising to-be-assembled time and to-be-assembled vehicle information;

[0123] The classification module 302 is configured to perform classification processing on the to-be-warehoused goods according to the to-be-assembled time, to obtain target classified goods.

[0124] The retrieval module 303 is configured to perform retrieval on a preset goods shelf knowledge graph according to the goods attribute characteristic information corresponding to the target classified goods, the to-be-assembled time and the to-be-assembled vehicle information, to obtain target goods shelf information, the target goods shelf information comprising position information, volume information and load-bearing information of a target goods shelf.

[0125] The classification extraction module 304 is configured to perform classification, priority sorting and same-priority extraction on the target classified goods by using a preset classification model and the goods information, to obtain target warehoused goods.

[0126] The generation module 305 is configured to generate target goods shelf placement information corresponding to the target warehoused goods by using a preset loading model and the target goods shelf information, based on a preset target function and an assembly strategy, the target function being a function constructed based on a greedy algorithm, and the target goods shelf placement information comprising placement combination and placement order of a goods shelf corresponding to the target warehoused goods.

[0127] The functions of each module in the above-mentioned to-be-warehoused goods data processing apparatus correspond to the steps in the above-mentioned to-be-warehoused goods data processing method embodiment, and the functions and implementation processes will not be repeated here.

[0128] In the embodiment, the to-be-warehoused goods are classified, the preset goods shelf knowledge graph is retrieved according to the goods attribute characteristic information corresponding to the target classified goods, the to-be-assembled time and the to-be-assembled vehicle information, the preset loading model and the target goods shelf information are used, the preset target function and the assembly strategy are used to generate the target goods shelf placement information corresponding to the target warehoused goods, the classification processing and the goods shelf knowledge graph retrieval improve the matching accuracy of the target goods shelf, the combination of the target goods shelf information and the assembly strategy enhances the diversity of the matching factors of the target goods shelf, and the analysis accuracy of the target goods placement information is improved, thereby improving the matching accuracy of the goods placement strategy of the goods placement storage goods shelf under the condition of high out-of-warehouse loading efficiency.

[0129] Referring to Figure 4 Another embodiment of the to-be-warehoused goods data processing apparatus in the embodiment comprises:

[0130] The acquisition module 301 is configured to acquire goods information of the to-be-warehoused goods and to-be-assembled information, the goods information comprising goods attribute characteristic information of the to-be-warehoused goods, and the to-be-assembled information comprising to-be-assembled time and to-be-assembled vehicle information;

[0131] The classification module 302 is configured to perform classification processing on the to-be-warehoused goods according to the to-be-assembled time, to obtain target classified goods;

[0132] The retrieval module 303 is configured to perform retrieval on a preset goods shelf knowledge graph according to the goods attribute characteristic information corresponding to the target classified goods, the to-be-assembled time and the to-be-assembled vehicle information, to obtain target goods shelf information, the target goods shelf information comprising position information, volume information and load-bearing information of a target goods shelf;

[0133] The classification extraction module 304 is configured to perform classification, priority sorting and same-priority extraction on the target classified goods by using a preset classification model and the goods information, to obtain target warehoused goods;

[0134] The generation module 305 is configured to generate target goods shelf placement information corresponding to the target warehoused goods by using a preset loading model and the target goods shelf information, based on a preset target function and an assembly strategy, the target function being a function constructed based on a greedy algorithm, and the target goods shelf placement information comprising placement combination and placement order of a goods shelf corresponding to the target warehoused goods.

[0135] The simulation optimization module 306 is configured to perform warehousing shelf placement and outbound loading on the to-be-warehoused goods based on the target goods shelf placement information by using a preset simulation system, to obtain simulation information, and to optimize the loading model according to the simulation information.

[0136] Optionally, the classification extraction module 304 can be specifically configured to:

[0137] calculate, by using a plurality of classifiers in the preset classification model, a transportation category probability value of the target classified goods based on the goods information, to obtain a plurality of classification values corresponding to each transportation category respectively;

[0138] calculate a mean value of the plurality of classification values corresponding to each transportation category respectively, to obtain a target probability value corresponding to each transportation category respectively, and determine a target transportation category of the target classified goods according to the target probability value corresponding to each transportation category respectively, and determine the target classified goods of which the target transportation category is determined as to-be-sorted goods;

[0139] perform priority matching on the to-be-sorted goods by using a preset priority decision tree, to obtain initial warehoused goods;

[0140] perform classification processing on the initial warehoused goods according to the same priority, to obtain the target warehoused goods.

[0141] Optionally, the generating module 305 comprises:

[0142] a calculating unit 3051 configured to calculate a shelf load constraint value and a shelf volume constraint value corresponding to the target shelf information according to a preset loading model and the target shelf information;

[0143] a predicting unit 3052 configured to predict a shelf placement combination of the target warehouse entry goods based on a preset target function, the target shelf information, the shelf load constraint value and the shelf volume constraint value, to obtain initial shelf placement information;

[0144] an adjusting unit 3053 configured to adjust a placement order of the initial shelf placement information according to a preset assembly strategy, to obtain candidate shelf placement information;

[0145] a verifying unit 3054 configured to verify the candidate shelf placement information, to obtain target shelf placement information.

[0146] Optionally, the verifying unit 3054 can be specifically configured to:

[0147] calculate and select the shelf placement information of the goods information and the target shelf information according to a preset genetic algorithm, a dynamic programming algorithm and an assembly strategy, to obtain comparative shelf placement information;

[0148] calculate a similarity value between the candidate shelf placement information and the comparative shelf placement information according to a preset similarity algorithm;

[0149] determine whether the similarity value is greater than a preset comparison value, and if the similarity value is greater than the preset comparison value, determine the candidate shelf placement information as the target shelf placement information.

[0150] Optionally, the classifying module 302 can be specifically configured to:

[0151] sort the warehouse entry goods according to a time sequence of the assembly time, to obtain a goods sequence;

[0152] classify the goods with the same assembly time in the goods sequence into a same time set, to obtain initial classified goods;

[0153] obtain goods assembly value information of the initial classified goods, compare and analyze the goods assembly value information with preset assembly value information, to obtain target assembly value information meeting the preset assembly value information;

[0154] determine the initial classified goods corresponding to the target assembly value information as target classified goods.

[0155] Optionally, the searching module 303 can be specifically configured to:

[0156] The target classified goods corresponding to the goods attribute feature information, the to-be-assembled time and the to-be-assembled vehicle information are sequentially subjected to keyword extraction and keyword entity pair generation to obtain entity information.

[0157] The preset goods shelf knowledge graph is subjected to random walk and information matching through the entity information to obtain target goods shelf information.

[0158] The functions of the modules and units in the above-mentioned processing device for the to-be-warehoused goods data correspond to the steps in the above-mentioned processing method for the to-be-warehoused goods data, and the functions and implementation processes will not be repeated here.

[0159] In the embodiment of the application, not only the matching accuracy of the target goods shelf is improved, the diversity of the matching factors of the target goods shelf is enhanced, and the analysis accuracy of the target goods placement information is improved, but also the matching accuracy of the goods placement strategy of the goods placement warehouse shelf under the condition of high warehouse-out loading efficiency is improved, and the accuracy of the loading model is improved by optimizing the loading model according to the simulation information, thereby improving the matching accuracy of the goods placement strategy of the goods placement warehouse shelf under the condition of high warehouse-out loading efficiency.

[0160] The above Figure 3 and Figure 4 The processing device for the to-be-warehoused goods data in the embodiment of the application is described in detail from the perspective of modular functional entities, and the processing device for the to-be-warehoused goods data in the embodiment of the application is described in detail from the perspective of hardware processing.

[0161] Figure 5 is a structural schematic diagram of a processing device for to-be-warehoused goods data provided by the embodiment of the application. The processing device for to-be-warehoused goods data 500 can have great differences due to different configurations or performances, and can include one or more processors (central processing units, CPUs) 510 (for example, one or more processors) and a memory 520, one or more storage media 530 (for example, one or more mass storage devices) for storing application programs 533 or data 532. The memory 520 and the storage media 530 can be temporary storage or persistent storage. The programs stored in the storage media 530 can include one or more modules (not shown in the figure), and each module can include a series of instruction operations in the processing device for to-be-warehoused goods data 500. Furthermore, the processor 510 can be configured to communicate with the storage media 530 and execute a series of instruction operations in the storage media 530 on the processing device for to-be-warehoused goods data 500.

[0162] The processing device 500 of the to-be-warehoused cargo data can further include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input / output interfaces 560, and / or one or more operating systems 531, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art can understand that, Figure 5 The structure of the processing device of the to-be-warehoused cargo data shown is not a limitation on the processing device of the to-be-warehoused cargo data, and can include more or fewer components than shown, or combine certain components, or arrange different components.

[0163] The present application also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, and the computer-readable storage medium stores instructions, which, when executed on a computer, cause the computer to perform the steps of the processing method of the to-be-warehoused cargo data.

[0164] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0165] If the integrated units are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application or the whole or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method embodiments of the present application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0166] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for processing data on goods awaiting warehousing, characterized in that, The method for processing the data of goods to be received into the warehouse includes: Obtain the goods information and assembly information of the goods to be put into the warehouse. The goods information includes the goods attribute characteristics information of the goods to be put into the warehouse, and the assembly information includes the assembly time and the vehicle information to be assembled. Based on the assembly time, the goods to be put into storage are classified to obtain the target category of goods; Based on the product attribute characteristics, assembly time and vehicle information corresponding to the target category of goods, the pre-set shelf knowledge graph is retrieved to obtain the target shelf information, which includes the location, volume and load-bearing information of the target shelf. Using a pre-set classification model and the product information, the target category of goods is classified, prioritized, and extracted with equal priority to obtain the target goods to be put into storage. Based on the preset loading model and the target shelf information, and based on the preset objective function and assembly strategy, the target shelf placement information corresponding to the target inbound goods is generated. The objective function is a function constructed based on a greedy algorithm. The target shelf placement information includes the placement combination and placement order of the shelves corresponding to the target inbound goods. The step of classifying, prioritizing, and extracting goods with equal priority based on a pre-set classification model and the goods information to obtain target goods for warehousing includes: using multiple classifiers in the pre-set classification model, calculating the transportation category probability value of the target goods based on the goods information, obtaining multiple classification values ​​corresponding to each transportation category; calculating the mean of the multiple classification values ​​corresponding to each transportation category, obtaining the target probability value corresponding to each transportation category, and determining the target transportation category of the target goods based on the target probability value corresponding to each transportation category, and identifying the target goods with the determined target transportation category as goods to be sorted; using a pre-set priority decision tree, performing priority matching on the goods to be sorted to obtain initial goods for warehousing; and classifying the initial goods for warehousing according to equal priority to obtain target goods for warehousing. The step of generating target shelf placement information for the target inbound goods based on a preset loading model and target shelf information, and a preset objective function and assembly strategy, includes: calculating shelf load constraint values ​​and shelf volume constraint values ​​corresponding to the target shelf information using the preset loading model and target shelf information; predicting shelf placement combinations for the target inbound goods based on the preset objective function, target shelf information, shelf load constraint values, and shelf volume constraint values ​​to obtain initial shelf placement information; adjusting the placement order of the initial shelf placement information using a preset assembly strategy to obtain candidate shelf placement information; and verifying the candidate shelf placement information to obtain target shelf placement information.

2. The method for processing data of goods to be received into the warehouse according to claim 1, characterized in that, The step of verifying the candidate shelf placement information to obtain the target shelf placement information includes: By using a pre-set genetic algorithm, dynamic programming algorithm, and assembly strategy, the shelf placement information of the goods information and the target shelf information is calculated and selected to obtain comparative shelf placement information; The similarity value between the candidate shelf placement information and the comparison shelf placement information is calculated using a preset similarity algorithm. Determine whether the similarity value is greater than a preset comparison value. If the similarity value is greater than the preset comparison value, then determine the candidate shelf placement information as the target shelf placement information.

3. The method for processing data of goods to be received into the warehouse according to claim 1, characterized in that, The process of classifying the goods to be put into storage according to the assembly time to obtain target category goods includes: The goods to be put into storage are sorted according to the time sequence of the assembly time to obtain a goods sequence; The goods in the goods sequence that have the same assembly time are classified into the same time set to obtain the initial classified goods; Obtain the cargo loading value information of the initially classified goods, compare and analyze the cargo loading value information with the preset loading value information, and obtain the target loading value information that conforms to the preset loading value information; The initial category of goods corresponding to the target loading value information is determined as the target category of goods.

4. The method for processing data of goods to be received into the warehouse according to claim 1, characterized in that, The step involves retrieving the target shelf information from a pre-set shelf knowledge graph based on the product attribute characteristics, assembly time, and vehicle information corresponding to the target category of goods. This includes: The target category of goods is sequentially subjected to keyword extraction and keyword entity pair generation for the corresponding goods attribute feature information, assembly time and assembly vehicle information to obtain entity information. Using the entity information, a random walk and information matching are performed on the pre-set shelf knowledge graph to obtain the target shelf information.

5. The method for processing data of goods to be received into the warehouse according to any one of claims 1-4, characterized in that, After generating the target shelf placement information corresponding to the target inbound goods based on the preset loading model and the target shelf information, and according to the preset objective function and assembly strategy, the process further includes: Using a pre-set simulation system, based on the target shelf placement information, the goods to be stored are placed on the storage shelves and loaded out of the warehouse to obtain simulation information, and the loading model is optimized based on the simulation information.

6. A device for processing data on goods to be received into the warehouse, characterized in that, The device for processing the data of goods to be received includes: The acquisition module is used to acquire the goods information and assembly information of the goods to be put into the warehouse. The goods information includes the goods attribute feature information of the goods to be put into the warehouse, and the assembly information includes the assembly time and the vehicle information to be assembled. The classification module is used to classify the goods to be put into storage according to the assembly time to obtain the target category of goods; The retrieval module is used to retrieve the target shelf information by searching the pre-set shelf knowledge graph based on the product attribute feature information, assembly time and vehicle information corresponding to the target category of goods. The target shelf information includes the location information, volume information and load-bearing information of the target shelf. The classification and extraction module is used to classify, prioritize, and extract goods of equal priority according to the preset classification model and the goods information to obtain the target goods to be put into storage. The generation module is used to generate target shelf placement information corresponding to the target inbound goods based on a preset loading model and the target shelf information, and a preset objective function and assembly strategy. The objective function is a function constructed based on a greedy algorithm, and the target shelf placement information includes the placement combination and placement order of the shelves corresponding to the target inbound goods. The step of classifying, prioritizing, and extracting goods with equal priority based on a pre-set classification model and the goods information to obtain target goods for warehousing includes: using multiple classifiers in the pre-set classification model, calculating the transportation category probability value of the target goods based on the goods information, obtaining multiple classification values ​​corresponding to each transportation category; calculating the mean of the multiple classification values ​​corresponding to each transportation category, obtaining the target probability value corresponding to each transportation category, and determining the target transportation category of the target goods based on the target probability value corresponding to each transportation category, and identifying the target goods with the determined target transportation category as goods to be sorted; using a pre-set priority decision tree, performing priority matching on the goods to be sorted to obtain initial goods for warehousing; and classifying the initial goods for warehousing according to equal priority to obtain target goods for warehousing. The step of generating target shelf placement information for the target inbound goods based on a preset loading model and target shelf information, and a preset objective function and assembly strategy, includes: calculating shelf load constraint values ​​and shelf volume constraint values ​​corresponding to the target shelf information using the preset loading model and target shelf information; predicting shelf placement combinations for the target inbound goods based on the preset objective function, target shelf information, shelf load constraint values, and shelf volume constraint values ​​to obtain initial shelf placement information; adjusting the placement order of the initial shelf placement information using a preset assembly strategy to obtain candidate shelf placement information; and verifying the candidate shelf placement information to obtain target shelf placement information.

7. A device for processing data on goods to be received into the warehouse, characterized in that, The device for processing the data of goods to be put into storage includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the processing device for the goods to be received to execute the processing method for the goods to be received as described in any one of claims 1-5.

8. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements the method for processing data of goods to be put into storage as described in any one of claims 1-5.

Images