An intelligent warehouse batch sorting optimization method and system based on artificial intelligence

Abstract

The present invention relates to the technical field of warehouse management, and discloses an intelligent warehouse batch sorting optimization method and system based on artificial intelligence. The system comprises a batch sorting information acquisition module, a batch sorting commodity and customer information management module, and a batch sorting commodity storage and extraction module; the commodity information of the sorting process is acquired in real time by a laser scanner, and the commodity name and specification keywords of the scanned commodity information and warehouse sorting commodity data are identified by an intelligent search algorithm to scientifically confirm the storage and sorting commodity information, so as to ensure that the sorted commodity is consistent with the commodity information of the customer order; the confirmed batch sorting commodity objects are matched with the customer identity information and accurately marked on the packaging surface of the commodity to be sorted, so as to improve the service quality of batch warehouse sorting, and realize the automatic sorting operation according to the storage information and customer information in cooperation with the industrial handling robot, so as to improve the batch sorting effect and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to an intelligent warehouse batch sorting optimization method and system based on artificial intelligence. Background Art

[0002] Warehousing sorting is the operation of stacking items in different categories according to their types and the order of entry and exit. Sorting is the preparatory work to improve and support delivery, and it is the inevitable extension of different distribution companies competing and improving their own economic benefits during delivery. Sorting is an inevitable requirement for delivery to develop into a higher form; the logistics distribution center quickly and accurately sorts the goods from their storage locations or other locations according to the customer's order requirements or distribution plan, and classifies and concentrates them in a certain way. There are usually three ways of sorting: order sorting, batch sorting and composite sorting; Order sorting, definition: Order sorting is for each order, the sorting personnel sorts the goods from the storage area or sorting area according to the goods and quantities listed in the order, and then collects them together. Batch sorting, definition: Batch sorting is to collect multiple orders into a batch, add them up according to the types and categories of the goods, and then pick them, and then classify and concentrate them according to different customers or different orders. Composite sorting, definition: In order to overcome the shortcomings of order sorting and batch sorting, the distribution center can also adopt a composite sorting method that combines order sorting and batch sorting. Among them, batch sorting treats warehouse goods as centralized sorting according to the type of goods, which helps to improve sorting efficiency and quality. However, the existing warehouse batch sorting is sorted according to the type and specification of goods, and the sorting process is sorted and stored according to the category of goods, but it cannot guarantee the consistency of sorted goods and customer information. Although it is guaranteed that the type, specification and quantity of goods picked up by customers are consistent with those before warehousing, it is also impossible to guarantee that the goods picked up by customers are consistent with the source of goods before sorting and warehousing, resulting in the appearance of other people's problematic goods when customers pick up goods, which reduces the quality of warehousing sorting.

[0003] The Chinese invention patent with publication number CN112239043A discloses a logistics sorting method, device, warehousing system and storage medium, which uses the method of judging whether the outbound items corresponding to the outbound task of the item confluence and collection are stored in multiple storage logic areas; if so, generating at least one segmented logic area group based on the multiple storage logic areas; wherein the segmented logic area group includes multiple picking logic areas; generating and executing a zone group picking task corresponding to the segmented logic area group; wherein obtaining the area priorities corresponding to the multiple picking logic areas of the segmented logic area group, based on the area priorities Generate a regional sorting sequence; generate multiple logical area picking tasks corresponding to multiple picking logical areas according to the regional sorting sequence and control the multiple logical area picking tasks to be executed in sequence; after the execution of the zone group picking task is completed, transport at least one item picking device corresponding to the zone group picking task to the confluence area; if all the item picking devices corresponding to all the zone group picking tasks are transported to the confluence area, then transport all the item picking devices to the review buffer lane located in the review area to perform a review operation on the outbound items transported by all the item picking devices. The above scheme can perform segmented confluence processing on the confluence collection order, which can reduce the number of review buffer lanes in the review area and increase the storage space of the warehouse; it can reduce the waiting time required to use the review buffer lane to collect items, and can reduce storage costs. However, the logistics sorting process cannot guarantee that the goods and customer information are completely consistent in statistics, which leads to the inconsistency between the goods picked up by the customer and the goods before sorting, and the goods picked up contain other people's problematic goods. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In order to solve the problem that the existing warehousing batch sorting is sorted according to the type and specification of goods, the sorting process is sorted and stored according to the category of goods, but it cannot ensure the consistency of sorted goods and customer information. Although the type, specification and quantity of goods picked up by customers are guaranteed to be consistent with those before warehousing, it cannot ensure that the goods picked up by customers are from the same source as those before sorting and warehousing, resulting in the appearance of other people's problematic goods when picked up by customers, which reduces the quality of warehousing sorting. The above purpose is to ensure the unity of batch sorted goods and customer information, ensure that the goods picked up by customers are consistent with the source before sorting and warehousing, improve the batch sorting effect and accuracy, and provide batch sorting service quality.

[0006] (II) Technical solution

[0007] The present invention is implemented by the following technical solution: an intelligent warehouse batch sorting optimization method based on artificial intelligence, the method comprising the following steps:

[0008] S1. Collect customer warehousing and sorting order data;

[0009] S2. Using keyword search to identify warehouse sorting commodity data and warehouse sorting customer identity data based on the customer warehouse sorting order data;

[0010] S3, collecting the data of the commodity objects to be sorted and matching the commodity names and specifications with the warehouse sorted commodity data, and confirming the customer's batch sorted commodity objects according to the keyword matching results;

[0011] S4, combining the to-be-sorted commodity object data corresponding to the confirmed batch sorting commodity object with the warehousing sorting customer identity data to generate warehousing sorting commodity belonging to the customer identity identification data and marking it on the packaging surface of the batch sorting commodity object, and performing sorting operations on the marked batch sorting commodity object;

[0012] S5, generating specific storage location data of the warehouse for the sorted warehouse and sorted commodity objects according to the identity identification data of the customer to whom the warehouse and sorted commodity belongs;

[0013] S6, performing storage operations on the storage sorting commodity objects according to the storage location data in the specific storage location data of the commodity storage;

[0014] S7. Perform commodity extraction operations according to the commodity storage location data, commodity data, and customer identity data based on the commodity storage location data, the warehouse sorting commodity customer identity data, the warehouse sorting commodity data, and the warehouse sorting customer identity data.

[0015] Preferably, the steps of collecting customer warehousing sorting order data are as follows:

[0016] S11. Collect warehousing business orders online through the order management platform and generate customer warehousing sorting order data A.

[0017] Preferably, the operation steps of identifying the warehouse sorting commodity data and the warehouse sorting customer identity data by keyword search based on the customer warehouse sorting order data are as follows:

[0018] S21. Obtain the number of warehouse sorting orders from customers A;

[0019] S22, searching for warehousing and sorting commodity data from the customer warehousing and sorting order number A by using the KD tree nearest neighbor search algorithm combined with commodity keywords and generating a warehousing and sorting commodity data set A'=(a'1, ..., a' m ,…,a' φ ), m=1,2,3,…,φ;a' mrepresents the storage and sorting commodity data corresponding to the mth commodity, φ represents the maximum value of the storage and sorting commodity types, the commodity keywords include commodity name, commodity specification and commodity model; the storage and sorting commodity data includes at least one of commodity type, commodity name, commodity specification, commodity model, commodity unit price, commodity function and commodity usage method;

[0020] The customer identity data of the warehousing and sorting commodities are searched out from the customer warehousing and sorting order number A by using the KD tree nearest neighbor search algorithm in combination with the identity information keywords, and the warehousing and sorting customer identity data A is generated. The identity information keywords include name and title; the warehousing and sorting customer identity data includes at least one of the customer's name, title, ID number, and enterprise code.

[0021] Preferably, the collected data of the commodity objects to be sorted are matched with the warehouse sorted commodity data for commodity names and specifications keywords, and the operation steps of confirming the batch sorted commodity objects of the customer according to the keyword matching results are as follows:

[0022] S31, using a laser scanner to identify the barcode of the commodity object to be sorted, collecting commodity data of the commodity object to be sorted and generating commodity object data b to be sorted;

[0023] S32: compare the to-be-sorted commodity object data b with the warehouse sorting commodity data a' in the warehouse sorting commodity data set A' m Perform keyword matching retrieval of product names and specifications; use data retrieval algorithm to match the object data b of the product to be sorted with the warehouse sorting product data a' m The steps are as follows:

[0024] S321, initialize parameters, dimension φ in the warehouse sorting commodity data set A' space, commodity data identification manta ray population size F, maximum number of iterations T and update commodity data identification manta ray population position, where commodity data identification manta ray population position update formula in the warehouse sorting commodity data set A' space is Where i=1,2,3,…,F, It represents the position of manta ray individual i in the φ-dimensional warehouse sorting commodity data set A' space after t+1 algorithm iterations. represents the position of manta ray individual i in the φ-dimensional warehouse sorting commodity data set A' space after t algorithm iterations, ζ represents the somersault factor, ζ takes the value of 2, α and β are random numbers in [0,1], Represents commodity data identification manta ray individual in the φ-th dimension warehouse sorting commodity data set A' warehouse sorting commodity data a' m Identify the location with the maximum fitness value with the object data b of the commodity to be sorted;

[0025] S322, initialize commodity data to identify manta ray populations, use the initial commodity data to identify manta ray population fitness values ​​to obtain storage sorting commodity data a' that matches the commodity object data b to be sorted m The corresponding optimal fitness commodity data identifies manta ray individuals;

[0026] S323, generate a random number in [0,1], determine the numerical value of the random number and t / T, when the random number is less than t / T, go to step S324, t represents the current iteration number of the algorithm, t=t+1 means that when executing step S323, the current iteration number is equal to the previous current iteration number plus 1, t≤T; otherwise, re-execute steps S321 and S322 to generate new commodity data to identify manta ray individuals and obtain the fitness value, that is, search for new warehouse sorting commodity data a' that matches the identification of the commodity object data b to be sorted in the warehouse sorting commodity data set A' space. m And calculate b and a' m The fitness value of

[0027] S324, judging the fitness value of the manta ray individual identified by the new product data and the fitness value of the manta ray individual identified by the original product data, when the fitness value of the manta ray individual identified by the new product data is less than the fitness value of the manta ray individual identified by the original product data, using the new product data to identify the manta ray individual to replace the original product data to identify the manta ray individual; otherwise, retaining the original product data to identify the manta ray individual, judging the fitness value of the manta ray individual identified by the new product data and the fitness value of the manta ray individual identified by the global product data, when the fitness value of the manta ray individual identified by the new product data is less than the fitness value of the manta ray individual identified by the global product data, using the new product data to identify the manta ray individual to replace the original global product data to identify the manta ray optimally; otherwise, retaining the original global product data to identify the manta ray optimally;

[0028] S325. When the algorithm meets the maximum number of iterations, the storage and sorting commodity data a' with the largest matching suitability value with the to-be-sorted commodity object data b in the storage and sorting commodity data set A' space is output. m ;

[0029] S33, when b and a' m If the product name and specifications are matched successfully, it means that the product corresponding to the product object data b to be sorted is the product of the warehouse customer, and then the product to be sorted and the customer identity identification stage will be entered;

[0030] When b and a' m If the name and specification of the product are not matched successfully, it means that the product corresponding to the product object data b to be sorted is not the product of the warehouse customer, then the confirmation procedure will be continued for the next sorted product object according to step S32.

[0031] Preferably, the commodity object data to be sorted corresponding to the confirmed batch sorting commodity object is combined with the warehouse sorting customer identity data to generate warehouse sorting commodity belonging to the customer identity identification data and marked on the packaging surface of the batch sorting commodity object, and the operation steps of performing the sorting operation on the marked batch sorting commodity object are as follows:

[0032] S41, when b and a' m According to the successful matching of the name and specification of the commodity, the object data b of the commodity to be sorted and the identity data A of the warehouse sorting customer are obtained;

[0033] S42, combining the to-be-sorted commodity object data b and the storage and sorting customer identity data A” to generate storage and sorting commodity-attributed customer identity identification data B=[A”, b];

[0034] S43, converting the customer identification data B of the warehouse sorted goods into a barcode and marking it on the packaging surface of the goods to be sorted by a laser marking machine. After the goods to be sorted are marked, the goods to be sorted are sorted by sorting equipment according to the types of goods.

[0035] Preferably, the steps of generating the specific storage location data of the warehouse for the sorted warehouse sorted commodity objects according to the identity identification data of the customer to whom the warehouse sorted commodity belongs are as follows:

[0036] S51, using industrial sensors to collect idle workstation location data in the warehouse in order according to the numerical values ​​of the workstation numbers and generate a warehouse idle workstation location data set C, and using the KD tree nearest neighbor search algorithm to filter out the warehouse idle workstation location data set C with the smallest number. min ;

[0037] S52, the position data of the smallest storage idle station C min Combined with the customer identification data B of the warehouse sorted goods to generate the specific storage location data of the corresponding warehouse goods object after the sorting operation is performed

[0038] Preferably, the operation steps of performing storage operations on the storage sorting commodity objects according to the storage location data in the specific storage location data of the commodity storage are as follows:

[0039] S61. Obtaining specific storage location data of commodity warehouse

[0040] S62: Based on the specific storage location data of the commodity storage The industrial handling robot will store the sorted warehouse commodity objects in the designated storage location according to the warehouse station number.

[0041] Preferably, the steps of performing the commodity extraction operation according to the commodity storage location data, the commodity data, and the customer identity data according to the commodity storage location data, the commodity data, and the customer identity data are as follows:

[0042] S71, based on the specific storage location data of the commodity storage The warehouse sorting goods belong to the customer identity identification data B, the warehouse sorting goods data set A', and the warehouse sorting customer identity data A". The industrial handling robot goes to the storage warehouse according to the specific storage station number, product data, and customer identity data of the stored goods object to pick up the goods stored in the warehouse for the customer after batch sorting.

[0043] A system for implementing the intelligent warehousing batch sorting optimization method based on artificial intelligence, the system includes a batch sorting information collection module, a batch sorting commodity and customer information management module, and a batch sorting commodity storage and extraction module;

[0044] The batch sorting information acquisition module includes a customer warehousing sorting order acquisition unit, a customer warehousing sorting commodity information acquisition unit, and a warehousing sorting customer information acquisition unit;

[0045] The customer warehousing and sorting order collection unit is used for the order management platform to collect customer warehousing and sorting order data; the customer warehousing and sorting commodity information acquisition unit uses keyword search to identify warehousing and sorting commodity data based on the customer warehousing and sorting order data; the warehousing and sorting customer information acquisition unit uses keyword search to identify warehousing and sorting customer identity data based on the customer warehousing and sorting order data;

[0046] The batch sorting commodity and customer information management module includes a batch sorting commodity object information collection and identification unit, a batch sorting commodity object and customer data matching unit, a batch sorting commodity object and commodity customer information identification execution unit, and a batch sorting commodity sorting execution unit;

[0047] The batch sorting commodity object information collection and identification unit collects the commodity object data to be sorted through a laser scanner; the batch sorting commodity object and customer data matching unit matches the commodity object data to be sorted with the warehouse sorting commodity data by commodity name and specification keywords, and confirms the customer's batch sorting commodity object according to the keyword matching result; the batch sorting commodity object and commodity customer information identification execution unit combines the commodity object data to be sorted corresponding to the confirmed batch sorting commodity object with the warehouse sorting customer identity data to generate warehouse sorting commodity belonging to the customer identity identification data and marks it on the packaging surface of the batch sorting commodity object through a laser marking machine; the batch sorting commodity sorting execution unit performs sorting operations on the marked batch sorting commodity objects through a sorting device;

[0048] The batch sorting commodity storage and extraction module includes a batch sorting commodity storage location information generation unit, a batch sorting commodity storage execution unit, and a batch sorting commodity extraction execution unit;

[0049] The batch sorted commodity storage location information generating unit generates commodity storage specific storage location data for the sorted storage and sorting commodity objects according to the customer identity identification data to which the storage and sorting commodities belong; the batch sorted commodity storage execution unit executes storage operations on the storage and sorting commodity objects according to the storage location data in the commodity storage specific storage location data; the batch sorted commodity extraction execution unit performs commodity extraction operations according to the commodity storage specific storage location data, the customer identity identification data to which the storage and sorting commodities belong, the storage and sorting commodity data, and the storage and sorting customer identity data according to the commodity storage location data, commodity data, and customer identity data.

[0050] (III) Beneficial effects

[0051] The present invention provides an intelligent warehouse batch sorting optimization method and system based on artificial intelligence. It has the following beneficial effects:

[0052] 1. The customer warehousing and sorting order data can be conveniently and accurately collected through the order management platform by the customer warehousing and sorting order collection unit, thereby improving the efficiency of obtaining warehousing batch sorting goods and customer information; the customer warehousing and sorting product information acquisition unit and the warehousing and sorting customer information acquisition unit cooperate with each other to use keywords in the customer warehousing and sorting order data through the search algorithm to scientifically and efficiently search for warehousing and sorting product data and customer identity data, thereby realizing the scientific separation of batch sorting product data and customer identity data, providing data support for the subsequent batch sorting process product information verification and product and customer identity information verification, thereby improving the efficiency and quality of warehousing batch sorting.

[0053] 2. The batch sorting commodity object information collection and identification unit and the batch sorting commodity object and customer data matching unit cooperate with each other to obtain the commodity information of the sorting process in real time through the laser scanner, and the scanned commodity information and the warehouse sorting commodity data are identified by commodity name and specification keywords through the intelligent search algorithm to scientifically confirm the storage and sorting commodity information, ensuring that the sorted commodities are consistent with the customer order commodity information; the batch sorting commodity object and commodity customer information identification execution unit and the batch sorting commodity sorting execution unit match the confirmed batch sorting commodity objects with the customer identity information and accurately mark them on the packaging surface of the to-be-sorted commodities, ensuring that the goods picked up by customers are consistent with the source before sorting and warehousing, improving the service quality of batch warehousing sorting, and cooperating with industrial handling robots according to storage information and customer information to realize automated sorting operations, improving batch sorting effects and accuracy.

[0054] 3. Through the cooperation of the batch sorting commodity storage location information generation unit and the batch sorting commodity storage execution unit, the free workstations in the storage warehouse are accurately collected, and the search algorithm is combined to analyze the minimum number of storage workstation numbers and match the commodity and customer identity data to generate specific storage information for commodity storage, so as to realize the intelligent and reliable allocation of storage locations after batch sorting, ensure the completeness of the commodities before and after storage for customers, and ensure the quality of commodities for storage and sorting customers; the batch sorting commodity extraction execution unit combines the commodity storage location data, commodity data, and customer identity data to ensure that the commodities extracted by the customer are consistent with the commodities before storage and sorting, thereby improving the effect of warehousing batch sorting and improving the management efficiency and service quality of warehousing sorting. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of a module of an intelligent warehouse batch sorting optimization system based on artificial intelligence provided by the present invention;

[0056] Figure 2 A flowchart of an intelligent warehousing batch sorting optimization method based on artificial intelligence provided by the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] The embodiments of the intelligent warehouse batch sorting optimization method and system based on artificial intelligence are as follows:

[0059] See also Figure 2, an intelligent warehouse batch sorting optimization method based on artificial intelligence, the method comprises the following steps:

[0060] S1. Collect customer warehousing and sorting order data;

[0061] S2. Using keyword search to identify warehouse sorting commodity data and warehouse sorting customer identity data based on customer warehouse sorting order data;

[0062] S3. Collect the data of the commodity objects to be sorted and match the commodity names and specifications with the warehouse sorting commodity data, and confirm the customer's batch sorting commodity objects based on the keyword matching results;

[0063] S4, combining the warehouse sorting commodity data corresponding to the confirmed batch sorting commodity object with the warehouse sorting customer identity data to generate warehouse sorting commodity customer identity identification data and marking it on the packaging surface of the batch sorting commodity object, and performing sorting operations on the marked batch sorting commodity object;

[0064] S5. Generate specific storage location data of the warehoused sorted commodity for the sorted warehoused sorted commodity object based on the customer identity identification data of the warehoused sorted commodity;

[0065] S6. Performing storage operations on the storage sorting commodity objects according to the storage location data in the specific storage location data of the commodity storage;

[0066] S7. Perform commodity extraction operations according to the commodity storage location data, commodity data, and customer identity data based on the commodity storage location data, commodity data, and customer identity data.

[0067] For further information, see Figure 1 - Figure 2 The steps for collecting customer warehousing and sorting order data are as follows:

[0068] S11. Collect warehousing business orders online through the order management platform and generate customer warehousing sorting order data A.

[0069] The steps for identifying warehouse sorting commodity data and warehouse sorting customer identity data by keyword search based on customer warehouse sorting order data are as follows:

[0070] S21. Obtain the number of warehouse sorting orders from customers A;

[0071] S22, using the KD tree nearest neighbor search algorithm combined with the product keywords, search for the product data of the warehouse sorting from the customer warehouse sorting order number A and generate the warehouse sorting product data set A'=(a'1,...,a' m ,…,a'φ ), m=1,2,3,…,φ;a' m represents the storage and sorting commodity data corresponding to the mth commodity, φ represents the maximum value of the storage and sorting commodity types, commodity keywords include commodity name, commodity specification and commodity model; storage and sorting commodity data includes at least one of commodity type, commodity name, commodity specification, commodity model, commodity unit price, commodity function and commodity usage method;

[0072] The customer identity data of the warehousing and sorting goods is searched from the customer warehousing and sorting order number A through the KD tree nearest neighbor search algorithm combined with the identity information keywords, and the warehousing and sorting customer identity data A is generated. The identity information keywords include name and title; the warehousing and sorting customer identity data includes at least one of the customer's name, title, ID number, and enterprise code.

[0073] The customer warehousing and sorting order data can be conveniently and accurately collected through the order management platform through the customer warehousing and sorting order collection unit, thereby improving the efficiency of obtaining warehousing batch sorting goods and customer information; the customer warehousing and sorting goods information acquisition unit and the warehousing and sorting customer information acquisition unit cooperate with each other to use keywords in the customer warehousing and sorting order data through the search algorithm to scientifically and efficiently search for warehousing and sorting goods data and customer identity data, thereby realizing the scientific separation of batch sorting goods data and customer identity data, providing data support for the subsequent batch sorting process goods information verification and goods and customer identity information verification, thereby improving the efficiency and quality of warehousing batch sorting.

[0074] Furthermore, the data of the commodity objects to be sorted are collected and matched with the warehouse sorted commodity data for commodity name and specification keywords. The steps for confirming the customer's batch sorted commodity objects based on the keyword matching results are as follows:

[0075] S31, using a laser scanner to identify the barcode of the commodity object to be sorted, collecting commodity data of the commodity object to be sorted and generating commodity object data b to be sorted;

[0076] S32: compare the to-be-sorted commodity object data b with the warehouse sorting commodity data a' in the warehouse sorting commodity data set A' m Perform keyword matching retrieval of product names and specifications; use data retrieval algorithm to match the object data b of the product to be sorted with the warehouse sorting product data a' m The steps are as follows:

[0077] S321, initialize parameters, dimension φ in the warehouse sorting commodity data set A' space, commodity data identification manta ray population size F, maximum number of iterations T and update commodity data identification manta ray population position, where commodity data identification manta ray population position update formula in the warehouse sorting commodity data set A' space is Where i = 1, 2, 3, ..., F, It represents the position of manta ray individual i in the φ-dimensional warehouse sorting commodity data set A' space after t+1 algorithm iterations. represents the position of manta ray individual i in the φ-dimensional warehouse sorting commodity data set A' space after t algorithm iterations, ζ represents the somersault factor, ζ takes the value of 2, α and β are random numbers in [0,1], Represents commodity data identification manta ray individual in the φ-th dimension warehouse sorting commodity data set A' warehouse sorting commodity data a' m Identify the location with the maximum fitness value with the object data b of the commodity to be sorted;

[0078] S322, initialize commodity data to identify manta ray populations, use the initial commodity data to identify manta ray population fitness values ​​to obtain storage sorting commodity data a' that matches the commodity object data b to be sorted m The corresponding optimal fitness commodity data identifies manta ray individuals;

[0079] S323, generate a random number in [0,1], determine the numerical value of the random number and t / T, when the random number is less than t / T, go to step S324, t represents the current iteration number of the algorithm, t=t+1 means that when executing step S323, the current iteration number is equal to the previous current iteration number plus 1, t≤T; otherwise, re-execute steps S321 and S322 to generate new commodity data to identify manta ray individuals and obtain the fitness value, that is, search for new warehouse sorting commodity data a' that matches the identification of the commodity object data b to be sorted in the warehouse sorting commodity data set A' space. m And calculate b and a' m The fitness value of

[0080] S324, judging the fitness value of the manta ray individual identified by the new product data and the fitness value of the manta ray individual identified by the original product data, when the fitness value of the manta ray individual identified by the new product data is less than the fitness value of the manta ray individual identified by the original product data, using the new product data to identify the manta ray individual to replace the original product data to identify the manta ray individual; otherwise, retaining the original product data to identify the manta ray individual, judging the fitness value of the manta ray individual identified by the new product data and the fitness value of the manta ray individual identified by the global product data, when the fitness value of the manta ray individual identified by the new product data is less than the fitness value of the manta ray individual identified by the global product data, using the new product data to identify the manta ray individual to replace the original global product data to identify the manta ray optimally; otherwise, retaining the original global product data to identify the manta ray optimally;

[0081] S325. When the algorithm meets the maximum number of iterations, the storage and sorting commodity data a' with the largest matching suitability value with the to-be-sorted commodity object data b in the storage and sorting commodity data set A' space is output. m;

[0082] S33, when b and a' m If the product name and specifications are matched successfully, it means that the product corresponding to the product object data b to be sorted is the product of the warehouse customer, and then the product to be sorted and the customer identity identification stage will be entered;

[0083] When b and a' m If the name and specification of the product are not matched successfully, it means that the product corresponding to the product object data b to be sorted is not the product of the warehouse customer, then the confirmation procedure will be continued for the next sorted product object according to step S32.

[0084] The warehouse sorting commodity data corresponding to the confirmed batch sorting commodity object and the warehouse sorting customer identity data are combined to generate the warehouse sorting commodity belonging to the customer identity identification data and marked on the packaging surface of the batch sorting commodity object. The operation steps for performing the sorting operation on the marked batch sorting commodity object are as follows:

[0085] S41, when b and a' m According to the successful matching of the name and specification of the goods, the object data b of the goods to be sorted and the identity data A of the warehouse sorting customer are obtained";

[0086] S42, combining the to-be-sorted commodity object data b and the storage and sorting customer identity data A” to generate storage and sorting commodity-attributed customer identity identification data B=[A”, b];

[0087] S43, converting the customer identification data B of the warehouse sorted goods into a barcode and marking it on the packaging surface of the goods to be sorted by a laser marking machine. After the goods to be sorted are marked, the goods to be sorted are sorted by sorting equipment according to the types of goods.

[0088] The batch sorting commodity object information collection and identification unit and the batch sorting commodity object and customer data matching unit cooperate with each other to obtain commodity information of the sorting process in real time through a laser scanner, and the scanned commodity information and the warehouse sorting commodity data are identified with commodity names and specification keywords through an intelligent search algorithm to scientifically confirm the storage and sorting commodity information, ensuring that the sorted commodities are consistent with the commodity information of the customer order; the batch sorting commodity object and commodity customer information identification execution unit and the batch sorting commodity sorting execution unit match the confirmed batch sorting commodity objects with the customer identity information and accurately mark them on the packaging surface of the commodities to be sorted, ensuring that the commodities picked up by customers are consistent with the sources before sorting and warehousing, improving the service quality of batch warehousing sorting, and realizing automated sorting operations in cooperation with industrial handling robots based on storage information and customer information, improving batch sorting effects and accuracy.

[0089] For further information, see Figure 1- Figure 2 The steps for generating the specific storage location data of the warehouse for the warehouse sorting commodity object after sorting according to the customer identity identification data of the warehouse sorting commodity are as follows:

[0090] S51, using industrial sensors to collect idle workstation location data in the warehouse in order according to the numerical value of the workstation number and generate a warehouse idle workstation location data set C, and using the KD tree nearest neighbor search algorithm to filter out the warehouse idle workstation location data set C with the smallest number. min ;

[0091] S52, the position data of the smallest storage idle station C min Combined with the customer identification data B of the warehouse sorting goods to generate the specific storage location data of the corresponding warehouse goods object after the sorting operation is executed

[0092] The steps for performing storage operations on storage sorting commodity objects according to the storage location data in the specific storage location data of the commodity warehouse are as follows:

[0093] S61. Obtaining specific storage location data of commodity warehouse

[0094] S62, based on the specific storage location data of the commodity warehouse The industrial handling robot will store the sorted warehouse commodity objects in the designated storage location according to the warehouse station number.

[0095] The steps for extracting goods according to the specific storage location data of the goods warehouse, the identification data of the customer to whom the goods are sorted, the goods data of the warehouse sorting, and the customer identity data of the warehouse sorting are as follows:

[0096] S71. Based on the specific storage location data of the commodity warehouse The warehouse sorting goods belong to the customer identity identification data B, the warehouse sorting goods data set A', and the warehouse sorting customer identity data A". The industrial handling robot goes to the storage warehouse according to the specific storage station number, product data, and customer identity data of the stored goods object to pick up the goods stored in the warehouse for the customer after batch sorting.

[0097] The batch sorting commodity storage location information generation unit and the batch sorting commodity storage execution unit cooperate with each other to accurately collect the vacant workstations in the storage warehouse and combine the search algorithm to analyze the minimum number of storage workstation numbers and match the commodity and customer identity data to generate specific storage information for commodity storage, thereby realizing intelligent and reliable allocation of storage locations after batch sorting, ensuring that the customer's commodities are complete before and after storage, and ensuring the quality of commodities for storage and sorting customers; the batch sorting commodity extraction execution unit combines commodity storage location data, commodity data, and customer identity data to ensure that the commodities extracted by the customer are consistent with the commodities before storage and sorting, thereby improving the effect of warehousing batch sorting, and improving the management efficiency and service quality of warehousing sorting.

[0098] A system for realizing an intelligent warehouse batch sorting optimization method based on artificial intelligence, please refer to Figure 1 ,The system includes a batch sorting information collection module, a batch sorting commodity and customer information management module, and a batch sorting commodity storage and extraction module;

[0099] The batch sorting information collection module includes a customer warehousing sorting order collection unit, a customer warehousing sorting commodity information acquisition unit, and a warehousing sorting customer information acquisition unit;

[0100] The customer warehousing and sorting order collection unit is used for the order management platform to collect customer warehousing and sorting order data; the customer warehousing and sorting commodity information acquisition unit uses keyword search to identify warehousing and sorting commodity data based on customer warehousing and sorting order data; the warehousing and sorting customer information acquisition unit uses keyword search to identify warehousing and sorting customer identity data based on customer warehousing and sorting order data;

[0101] The batch sorting commodity and customer information management module includes a batch sorting commodity object information collection and identification unit, a batch sorting commodity object and customer data matching unit, a batch sorting commodity object and commodity customer information identification execution unit, and a batch sorting commodity sorting execution unit;

[0102] The batch sorting commodity object information collection and identification unit collects the data of the commodity objects to be sorted through a laser scanner; the batch sorting commodity object and customer data matching unit matches the commodity object data to be sorted with the warehouse sorting commodity data by commodity name and specification keywords, and confirms the customer's batch sorting commodity objects according to the keyword matching results; the batch sorting commodity object and commodity customer information identification execution unit combines the warehouse sorting commodity data corresponding to the confirmed batch sorting commodity objects with the warehouse sorting customer identity data to generate the warehouse sorting commodity belonging to the customer identity identification data and marks it on the packaging surface of the batch sorting commodity objects through a laser marking machine; the batch sorting commodity sorting execution unit performs sorting operations on the marked batch sorting commodity objects through sorting equipment;

[0103] The batch sorting commodity storage and extraction module includes a batch sorting commodity storage location information generation unit, a batch sorting commodity storage execution unit, and a batch sorting commodity extraction execution unit;

[0104] The batch sorting commodity storage location information generation unit generates commodity storage specific storage location data for the sorted storage sorting commodity objects according to the customer identity identification data to which the storage sorting commodities belong; the batch sorting commodity storage execution unit executes storage operations for the storage sorting commodity objects according to the storage location data in the commodity storage specific storage location data; the batch sorting commodity extraction execution unit performs commodity extraction operations according to the commodity storage specific storage location data, the customer identity identification data to which the storage sorting commodities belong, the storage sorting commodity data, and the storage sorting customer identity data according to the commodity storage location data, commodity data, and customer identity data.

[0105] It should be noted that in the technical solution of the present invention, the collection and processing of the customer identity information involved are in compliance with the provisions of relevant laws and regulations and are used for legal and legitimate purposes.

[0106] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent warehouse batch sorting optimization method based on artificial intelligence, characterized in that: The method comprises the following steps: S1. Collect customer warehousing and sorting order data A; S2: identifying the warehouse sorting commodity data and the warehouse sorting customer identity data by keyword search based on the customer warehouse sorting order data, wherein S2 comprises the following steps: S21, obtaining the customer warehousing sorting order data A; S22, searching for warehousing and sorting commodity data from the customer warehousing and sorting order number A by using the KD tree nearest neighbor search algorithm combined with commodity keywords and generating a warehousing and sorting commodity data set A'=(a'1, ..., a' m ,…,a' φ ), m=1,2,3,…,φ;a' m represents the storage and sorting commodity data corresponding to the mth commodity, φ represents the maximum value of the storage and sorting commodity types, the commodity keywords include commodity name, commodity specification and commodity model; the storage and sorting commodity data includes at least one of commodity type, commodity name, commodity specification, commodity model, commodity unit price, commodity function and commodity usage method; The customer identity data of the warehousing and sorting commodities is searched from the customer warehousing and sorting order number A by using the KD tree nearest neighbor search algorithm combined with the identity information keywords, and the warehousing and sorting customer identity data A" is generated, wherein the identity information keywords include name and title; the warehousing and sorting customer identity data includes at least one of the customer's name, title, ID number, and enterprise code; S3, collecting the data of the commodity objects to be sorted and matching the commodity names and specifications with the warehouse sorted commodity data, and confirming the batch sorted commodity objects of the customer according to the keyword matching results, said S3 includes the following steps: S31, using a laser scanner to identify the barcode of the commodity object to be sorted, collecting commodity data of the commodity object to be sorted and generating commodity object data b to be sorted; S32, compare the to-be-sorted commodity object data b with the warehouse sorting commodity data a′ in the warehouse sorting commodity data set A′ m Perform keyword matching search on product names and specifications; S33, when b and a′ m If the name and specification of the product are matched successfully, it means that the product corresponding to the object data b of the product to be sorted is the product of the warehouse customer, and then the process goes to step S4 of identifying the product to be sorted and the customer; When b and a′ m If the name and specification of the product are not matched successfully, it means that the product corresponding to the product object data b to be sorted is not the product of the warehouse customer, then the confirmation procedure is continued for the next sorted product object according to step S32; S4, combining the to-be-sorted commodity object data corresponding to the confirmed batch sorting commodity object with the warehousing sorting customer identity data to generate warehousing sorting commodity belonging to the customer identity identification data and marking it on the packaging surface of the batch sorting commodity object, and performing sorting operations on the marked batch sorting commodity object; S5, generating specific storage location data of the warehouse for the sorted warehouse and sorted commodity objects according to the identity identification data of the customer to whom the warehouse and sorted commodity belongs; S6, performing storage operations on the storage sorting commodity objects according to the storage location data in the specific storage location data of the commodity storage; S7. Perform commodity extraction operations according to the commodity storage location data, commodity data, and customer identity data based on the commodity storage location data, the warehouse sorting commodity customer identity data, the warehouse sorting commodity data, and the warehouse sorting customer identity data.

2. According to claim 1, an intelligent warehouse batch sorting optimization method based on artificial intelligence is characterized by: The S1 comprises the following steps: S11. Collect warehousing business orders online through the order management platform and generate customer warehousing sorting order data A.

3. According to the artificial intelligence-based intelligent warehousing batch sorting optimization method of claim 1, it is characterized by: In step S32, a data retrieval algorithm is used to match the object data b of the commodity to be sorted with the warehouse sorted commodity data a′. m The steps are as follows: S321, initializing parameters, dimension φ in the storage and sorting commodity data set A' space, commodity data identification manta ray population size F, maximum number of iterations T, and updating commodity data identification manta ray population location; S322, initialize commodity data to identify manta ray populations, use the initial commodity data to identify manta ray population fitness values ​​to obtain storage sorting commodity data a′ that matches the commodity object data b to be sorted m The corresponding optimal fitness commodity data identifies manta ray individuals; S323, generate a random number in [0,1], determine the numerical value of the random number and t / T, when the random number is less than t / T, go to step S324, t represents the current iteration number of the algorithm, t=t+1 means that when executing step S323, the current iteration number is equal to the previous current iteration number plus 1, t≤T; otherwise, re-execute steps S321 and S322 to generate new commodity data to identify manta ray individuals and obtain the fitness value, that is, search for new warehouse sorting commodity data a′ that matches the identification of the commodity object data b to be sorted in the warehouse sorting commodity data set A' space m And calculate b and a′ m The fitness value of S324, judging the fitness value of the manta ray individual identified by the new product data and the fitness value of the manta ray individual identified by the original product data, when the fitness value of the manta ray individual identified by the new product data is less than the fitness value of the manta ray individual identified by the original product data, using the new product data to identify the manta ray individual to replace the original product data to identify the manta ray individual; otherwise, retaining the original product data to identify the manta ray individual, judging the fitness value of the manta ray individual identified by the new product data and the fitness value of the manta ray individual identified by the global product data, when the fitness value of the manta ray individual identified by the new product data is less than the fitness value of the manta ray individual identified by the global product data, using the new product data to identify the manta ray individual to replace the original global product data to identify the manta ray optimally; otherwise, retaining the original global product data to identify the manta ray optimally; S325. When the algorithm meets the maximum number of iterations, the storage and sorting commodity data a′ with the largest matching fitness value with the to-be-sorted commodity object data b in the storage and sorting commodity data set A′ is output. m .

4. The method for optimizing intelligent warehouse batch sorting based on artificial intelligence according to claim 3 is characterized in that: The S4 comprises the following steps: S41, when b and a′ m According to the successful matching of the name and specification of the commodity, the object data b of the commodity to be sorted and the identity data A of the warehouse sorting customer are obtained; S42, combining the to-be-sorted commodity object data b and the storage and sorting customer identity data A” to generate storage and sorting commodity-attributed customer identity identification data B=[A”, b]; S43, converting the customer identification data B of the warehouse sorted goods into a barcode and marking it on the packaging surface of the goods to be sorted by a laser marking machine. After the goods to be sorted are marked, the goods to be sorted are sorted by sorting equipment according to the types of goods.

5. The method for optimizing intelligent warehouse batch sorting based on artificial intelligence according to claim 4 is characterized in that: The S5 The following steps are involved: S51, using industrial sensors to collect idle workstation location data in the warehouse in order according to the numerical values ​​of the workstation numbers and generate a warehouse idle workstation location data set C, and using the KD tree nearest neighbor search algorithm to filter out the warehouse idle workstation location data set C with the smallest number. min ; S52, the minimum number of storage idle station location data C min Combined with the customer identification data B of the warehouse sorted goods to generate the specific storage location data of the corresponding warehouse goods object after the sorting operation is performed 6. The method for optimizing intelligent warehouse batch sorting based on artificial intelligence according to claim 5, characterized in that: The S6 comprises the following steps: S61. Obtaining specific storage location data of commodity warehouse S62: Based on the specific storage location data of the commodity storage The industrial handling robot will store the sorted warehouse commodity objects in the designated storage location according to the warehouse station number.

7. The method for optimizing intelligent warehouse batch sorting based on artificial intelligence according to claim 6, characterized in that: The S7 comprises the following steps: S71, based on the specific storage location data of the commodity storage The warehouse sorting goods belong to the customer identity identification data B, the warehouse sorting goods data set A', and the warehouse sorting customer identity data A". The industrial handling robot goes to the storage warehouse according to the specific storage station number, product data, and customer identity data of the stored goods object to pick up the goods stored in the warehouse for the customer after batch sorting.

8. A system for implementing the artificial intelligence-based intelligent warehousing batch sorting optimization method as described in any one of claims 1 to 7.

Images