Warehouse selection method, device, computer equipment and storage medium

Abstract

The present application relates to a warehouse product selection method, device, computer equipment and storage medium. By obtaining warehouse product order data and warehouse cost data; determining the first warehouse product to be selected according to the warehouse product order data; determining the second warehouse product to be selected according to the warehouse product order data and the warehouse cost data through a multi-objective optimization algorithm based on the optimization target, the optimization target is to reduce the order delivery split rate and the warehouse product holding cost, and meet the warehouse's own constraints; warehouse selection is performed according to the first warehouse product to be selected and the second warehouse product to be selected. The present application first determines the first warehouse product to be selected, and then further screens the second warehouse product to be selected based on multi-objective planning, and performs warehouse selection. It is possible to simultaneously balance the reduction of the order delivery split rate and the warehouse product holding cost, and meet the warehouse's own constraints when selecting warehouse products, and can flexibly give warehouse selection results while meeting multiple constraints.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a warehouse product selection method, device, computer equipment and storage medium. Background Art

[0002] With the rapid development of computer technology and e-commerce, the segmentation of the consumer market has become more and more obvious, and the concept of customized demand has gradually become popular. Therefore, the number of SKUs (Stock Keeping Units) managed in the company's supply chain has exploded. Taking the auto parts industry as an example, the supplier can provide customers with more than 100,000 SKUs, but due to the large number of long-tail sales products and the volume restrictions of regional warehouses / forward warehouses, it is difficult to place all categories in the regional warehouse. Generally, the warehouse selection optimization method can be used to determine which SKUs to place in the warehouse to meet customer needs.

[0003] Currently, warehouse product selection is generally optimized through a heuristic method based on optimizing the order splitting rate. This method can ensure that the selected SKUs have a high correlation, but this method cannot meet the requirements of the corresponding holding cost and instant satisfaction rate of the SKU. Summary of the invention

[0004] Based on this, it is necessary to provide a warehouse selection method, device, computer equipment and storage medium that can simultaneously meet the holding cost and instant satisfaction rate in response to the above technical problems.

[0005] A warehouse product selection method, the method comprising:

[0006] Obtain warehouse product order data and warehouse cost data;

[0007] Determining a first stored product to be selected according to the stored product order data;

[0008] According to the warehouse product order data and the warehouse cost data, a second warehouse product to be selected is determined by a multi-objective optimization algorithm based on an optimization goal, wherein the optimization goal is to reduce the order delivery split rate and the warehouse product holding cost, and to meet the warehouse's own constraints;

[0009] The warehouse selects products according to the first warehouse products to be selected and the second warehouse products to be selected.

[0010] In one embodiment, determining the first to-be-selected stored product according to the stored product order data comprises:

[0011] Acquire storage product classification demand data, and determine a storage product classification method according to the storage product classification demand data;

[0012] Obtaining classification data corresponding to the stored product classification method in the stored product order data;

[0013] A first stored product to be selected is determined from among various types of stored products in the stored product order data according to the classification data.

[0014] In one embodiment, determining the second selected stored product according to the stored product order data and the storage cost data by a multi-objective optimization algorithm based on the optimization objective includes:

[0015] Initialize the population of the selected storage products according to the optimization objective to obtain an initial solution and an objective function;

[0016] Determining the Pareto rank of the initial solution according to the objective function result corresponding to the initial solution through fast non-dominated sorting;

[0017] Iterate and cycle the initial solution through the genetic algorithm, and obtain the cycle processing result after a preset number of cycles;

[0018] A Pareto optimal solution set is determined from the loop processing result, and a second storage product to be selected is obtained according to the Pareto optimal solution set.

[0019] In one embodiment, determining the Pareto level of the initial solution according to the objective function result corresponding to the initial solution by fast non-dominated sorting includes:

[0020] By fast non-dominated sorting, when an individual in the initial solution does not satisfy a constraint condition, a Pareto rank corresponding to the violation value of the constraint condition is assigned to the individual.

[0021] In one embodiment, the iterative loop processing of the initial solution by the genetic algorithm, after a preset number of loop rounds, obtaining the loop processing result comprises:

[0022] Obtaining a sub-generation solution according to the initial solution;

[0023] The initial solution and the child solution are combined to obtain a combined solution, and an iterative loop process begins;

[0024] Obtaining an elite solution from the merged solution by fast non-dominated sorting and congestion calculation;

[0025] Taking the elite solution as the initial solution, returning to the step of obtaining the child solution according to the initial solution, and accumulating the number of iterative cycles;

[0026] When the accumulated number of loop rounds reaches the preset number of rounds, the elite solution after the last loop processing is used as the loop processing result.

[0027] In one embodiment, determining the second storage product to be selected according to the Pareto optimal solution set includes:

[0028] Obtaining the order splitting rate target corresponding to each solution in the Pareto optimal solution set;

[0029] The solution with the smallest order splitting rate target is used as the second storage product to be selected corresponding to the Pareto optimal solution set.

[0030] A warehouse product selection device, comprising:

[0031] Data acquisition module, used to obtain warehouse product order data and warehouse cost data;

[0032] A storage product search module, used to determine a first storage product to be selected according to the storage product order data;

[0033] A product selection and allocation module is used to determine the second to-be-selected stored product according to the stored product order data and the storage cost data through a multi-objective optimization algorithm based on an optimization goal, wherein the optimization goal is to reduce the order delivery split rate and the storage product holding cost, and to meet the constraints of the warehouse itself;

[0034] The warehouse product selection module is used to select warehouse products according to the first warehouse products to be selected and the second warehouse products to be selected.

[0035] In one embodiment, the stored product search module is specifically used to:

[0036] Acquire storage product classification demand data, and determine a storage product classification method according to the storage product classification demand data;

[0037] Obtaining classification data corresponding to the stored product classification method in the stored product order data;

[0038] A first stored product to be selected is determined from among various types of stored products in the stored product order data according to the classification data.

[0039] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0040] Obtain warehouse product order data and warehouse cost data;

[0041] Determining a first stored product to be selected according to the stored product order data;

[0042] According to the warehouse product order data and the warehouse cost data, a second warehouse product to be selected is determined by a multi-objective optimization algorithm based on an optimization goal, wherein the optimization goal is to reduce the order delivery split rate and the warehouse product holding cost, and to meet the warehouse's own constraints;

[0043] The warehouse selects products according to the first warehouse products to be selected and the second warehouse products to be selected.

[0044] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0045] Obtain warehouse product order data and warehouse cost data;

[0046] Determining a first stored product to be selected according to the stored product order data;

[0047] According to the warehouse product order data and the warehouse cost data, a second warehouse product to be selected is determined by a multi-objective optimization algorithm based on an optimization goal, wherein the optimization goal is to reduce the order delivery split rate and the warehouse product holding cost, and to meet the warehouse's own constraints;

[0048] The warehouse selects products according to the first warehouse products to be selected and the second warehouse products to be selected.

[0049] The above-mentioned warehouse selection method, device, computer equipment and storage medium obtain the warehouse product order data and the warehouse cost data; determine the first warehouse product to be selected according to the warehouse product order data; determine the second warehouse product to be selected according to the warehouse product order data and the warehouse cost data through a multi-objective optimization algorithm based on the optimization target, wherein the optimization target is to reduce the order delivery split rate and the warehouse product holding cost, and meet the warehouse's own constraints; select warehouse products according to the first warehouse product to be selected and the second warehouse product to be selected. This application first determines the first warehouse product to be selected, and then further screens the second warehouse product to be selected based on multi-objective planning, and selects warehouse products. When selecting warehouse products, it is possible to simultaneously balance the reduction of the order delivery split rate and the warehouse product holding cost, and meet the warehouse's own constraints, and flexibly give the warehouse selection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is an application environment diagram of a warehouse product selection method in one embodiment;

[0051] Figure 2 A schematic diagram of a process of selecting products for a warehouse in one embodiment;

[0052] Figure 3 In one embodiment Figure 2 Schematic diagram of the sub-process of step 203;

[0053] Figure 4 In one embodiment Figure 2 Schematic diagram of the sub-process of step 205;

[0054] Figure 5 is a structural block diagram of a warehouse product selection device in one embodiment;

[0055] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0057] The warehouse selection method provided in this application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the warehouse selection server 104 through the network. The terminal 102 can send the warehouse selection corresponding to the warehouse selection of storage product order data and storage cost data to the warehouse selection server 104, so as to select the warehouse through the warehouse selection server 104, and the warehouse selection server 104 obtains the storage product order data and storage cost data; determines the first storage product to be selected according to the storage product order data; determines the second storage product to be selected according to the storage product order data and the storage cost data through a multi-objective optimization algorithm based on the optimization target, and the optimization target is to reduce the order delivery split rate and the storage product holding cost, and meet the warehouse's own constraints; select the warehouse according to the first storage product to be selected and the second storage product to be selected. Among them, the terminal 102 can be but not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices, and the warehouse selection server 104 can be implemented with an independent server or a server cluster composed of multiple servers. In another embodiment, the warehouse selection method of the present application is specifically used in the field of logistics and transportation to identify illegal actions of logistics personnel. The terminal 102 may be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, and portable wearable devices. The warehouse selection server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0058] In one embodiment, Figure 2 As shown, a warehouse selection method is provided, which is applied to Figure 1 Taking the warehouse selection server 104 in the example as an example, the following steps are included:

[0059] Step 201, obtaining warehouse product order data and warehouse cost data.

[0060] Among them, when a warehouse is newly built or expanded, it is necessary to consider the selection of products to be stored in the warehouse, that is, warehouse selection. Usually, the categories of goods stored in a warehouse are determined according to the business, that is, the business plan determines which category or categories of goods can be stored, such as food, clothing or electronic products, etc. The storage products in the warehouse can be stored in the warehouse in the form of SKU units. The storage product order data specifically refers to the data of existing orders corresponding to the storage products in the warehouse. These existing orders contain the data such as the quantity of storage products corresponding to the single storage product to be supplied by the warehouse that performs warehouse selection, the supply time of the storage products, etc. The storage cost data specifically refers to the holding cost per unit time of all storage products. The holding cost is also called inventory cost, which refers to the various expenses incurred in the process of ordering, purchasing and storing inventory, as well as the cost loss caused by inventory shortages, etc. The storage product order data and storage cost data corresponding to the warehouse selection can specifically refer to the storage product order data and storage cost data that have been determined in the future.

[0061] Specifically, when the staff of the terminal 102 wants to select warehouse products for a certain warehouse or a certain warehouse cluster, the warehouse selection can be implemented through the warehouse selection server 104. Specifically, the staff can collect the warehouse product order data and the storage cost data corresponding to these warehouse products in the future period, and then the warehouse selection server 104 implements the warehouse selection. In another embodiment, the staff of the terminal 102 can also select warehouse products based on historical data. In this case, the staff can provide the warehouse product order data and storage cost data in the past period to the warehouse selection server 104, and control the warehouse selection based on the historical data.

[0062] Step 203: determining the first stored product to be selected according to the stored product order data.

[0063] Among them, the first warehouse product to be selected may specifically refer to hot-selling goods.

[0064] Specifically, the warehouse selection server 104 can determine which SKUs are hot-selling products based on the order data of the warehouse products, and then determine the first warehouse products that must be selected as warehouse products based on these hot-selling SKUs, and then perform subsequent warehouse selection.

[0065] Step 205, according to the warehouse product order data and the warehouse cost data, the second warehouse product to be selected is determined by a multi-objective optimization algorithm based on the optimization goal. The optimization goal is to reduce the order delivery split rate and the warehouse product holding cost, and meet the warehouse's own constraints.

[0066] Step 207, selecting warehouse products according to the first warehouse product to be selected and the second warehouse product to be selected.

[0067] Among them, the multi-objective optimization algorithm specifically refers to the simultaneous existence of multiple optimization objectives, and the optimization problem based on multiple optimization objectives is to achieve the best possible optimization problem in a given area. In general, the various sub-objectives under multi-objective optimization are contradictory. The improvement of one sub-objective may cause the performance of another or several other sub-objectives to decline. Therefore, it is impossible to achieve the optimal solution for multiple sub-objectives at the same time, and they can only be coordinated and compromised. Splitting an order refers to the supply of the same order by different warehouses, and reducing the splitting rate is one of the main requirements for warehouse selection. This application uses the reduction of the order delivery splitting rate and the storage cost of the warehoused goods, and meets the warehouse's own constraints as the optimization objectives for multi-objective optimization. It can be determined which warehoused goods need to be selected into the warehouse in addition to the first candidate warehoused goods. These selected warehoused goods are the second warehoused goods to be selected. Then, the warehouse is selected according to the first warehoused goods to be selected and the second warehoused goods to be selected, and the warehoused goods are stored in the warehouse according to the result data corresponding to the warehouse selection.

[0068] Specifically, in a specific embodiment, the present application constructs an objective function by reducing the order delivery split rate and the storage product holding cost, and establishes a constraint function by satisfying the warehouse's own constraints as the optimization target, and then solves the multi-objective optimization problem of warehouse selection through a genetic algorithm. After obtaining the final Pareto optimal solution set, the second storage product to be selected in the selection allocation result is determined from the Pareto optimal solution set through the preset optimization constraints, and then the selection can be optimized based on the selection allocation result to determine the best warehouse selection plan, which can simultaneously ensure the split rate and the holding cost requirements.

[0069] The above-mentioned warehouse selection method obtains the warehouse product order data and the warehouse cost data; determines the first warehouse product to be selected according to the warehouse product order data; determines the second warehouse product to be selected according to the warehouse product order data and the warehouse cost data through a multi-objective optimization algorithm based on the optimization target, and the optimization target is to reduce the order delivery split rate and the warehouse product holding cost, and meet the warehouse's own constraints; warehouse selection is performed based on the first warehouse product to be selected and the second warehouse product to be selected. This application first determines the first warehouse product to be selected, and then further screens the second warehouse product to be selected based on multi-objective planning, and performs warehouse selection. It can simultaneously balance the reduction of the order delivery split rate and the warehouse product holding cost, and meet the warehouse's own constraints when selecting warehouse products, and can flexibly give warehouse selection results while meeting multiple constraints.

[0070] In one embodiment, Figure 3 As shown, step 203 includes:

[0071] Step 302, obtaining storage product classification demand data, and determining a storage product classification method according to the storage product classification demand data.

[0072] Step 304, obtaining classification data corresponding to the storage product classification method in the storage product order data.

[0073] Step 306: determine the first stored product to be selected from each type of stored product in the stored product order data according to the classification data.

[0074] Among them, the storage product classification demand data specifically refers to the data required to determine how to classify the selected storage products. The storage product classification demand data is specifically used to determine the storage product classification method. The storage product classification method includes the ABC (Activity Based Classification) inventory classification management method. The ABC inventory classification management method refers to dividing the inventory items into three levels according to the variety and the amount of funds occupied, namely, particularly important inventory (Class A), generally important inventory (Class B) and unimportant inventory (Class C), and then managing and controlling the different levels separately. Such a classification management method can achieve the following effects: compressing the total inventory, releasing the occupied funds, rationalizing the inventory and saving the management investment. And the XYZ classification method. The XYZ classification rule is classified according to the accuracy of the annual sales forecast: the X-class finished product forecast has the highest accuracy, and the inventory can be appropriately reduced. The Y-class in the middle can store a certain amount of inventory, while the Z-class is the least accurate, and it is necessary to consider a relatively high inventory. The principle of managing the XYZ classification method is to reduce the total inventory by reducing the finished product inventory that is easier to predict, but not to the point of being out of stock. In addition, it also includes ABC / XYZ analysis that combines the two, and analyzes the types of warehouse products through a matrix. The warehouse product classification demand data is determined based on the selected classification method, generally sales or sales volume.

[0075] Specifically, the warehouse selection server 104 needs to first obtain the user's classification requirements, so as to determine which method to use to determine the warehouse product categories corresponding to each type of SKU. Then, the classification data required by these classification methods are searched from the order data. For example, the ABC classification method can be used to determine which warehouse products in the warehouse product order data belong to Class A SKU, and then use it as the first warehouse product to be selected for warehouse selection. In this embodiment, by first determining the warehouse product classification requirement data and then determining the classification method based on the requirements, and extracting the data required for classification from the order, classification is performed and the first warehouse product to be selected is determined, so that the selected warehouse products can be classified according to the requirements and the accuracy of the classification can be guaranteed.

[0076] In one embodiment, if Figure 4 As shown, step 205 includes:

[0077] Step 401, initialize the population of the storage products to be selected according to the optimization target, and obtain the initial solution and the target function.

[0078] Step 403, determining the Pareto rank of the initial solution according to the objective function result corresponding to the initial solution through fast non-dominated sorting.

[0079] Step 405, iteratively loop processing is performed on the initial solution through a genetic algorithm, and after a preset number of loop rounds, a loop processing result is obtained.

[0080] Step 407, determine the Pareto optimal solution set from the loop processing result, and obtain the second storage product to be selected according to the Pareto optimal solution set.

[0081] Among them, the present application can solve the multi-objective optimization problem by means of a fast non-dominated sorting genetic algorithm. The goal of the fast non-dominated sorting is to divide the Pareto levels according to the objective function results of the initial solution. Then, through the iterative processing of the genetic algorithm, the optimized Pareto optimal solution set is obtained from the subsequent loop processing results, thereby determining the second storage product to be selected.

[0082] Specifically, after determining the first candidate stored product that must be selected into the warehouse in the stored product order data, it is also necessary to select the second candidate stored product that needs to be placed in the warehouse from the remaining stored products. At this time, the fast non-dominated sorting genetic algorithm can be used to construct and solve the multi-objective optimization problem.

[0083] We can first construct input parameters and decision variables based on the warehouse product order data and warehouse cost data: S represents the number of alternative SKUs, and s represents the specific SKU. H represents the holding cost per unit time of all SKUs, H = (h1, h2, ..., hs). O: the number of orders, o represents the specific order. D: the average demand per unit time of all SKUs, D = (D1, D2, ..., Ds). cap: the maximum number of categories that can be allocated to the regional warehouse, that is, the category constraint of the regional warehouse. r: the immediate demand satisfaction rate provided by the SKUs stored in the warehouse, that is, how much total demand can be covered by the selected SKU on average. The decision variable is X = (x1, x2, ..., xs), where xi = 1 means that the warehouse selects SKUi, otherwise xi = 0.

[0084] At this time, the objective function constructed to reduce the order delivery split rate and the storage product holding cost is as follows:

[0085] 1) Objective function of minimizing the order splitting rate. The objective of minimizing the order splitting rate is transformed into maximizing the order volume of SKUs that appear in the same order. This indicator can be understood as the correlation of the selected SKUs. Design the SKU co-occurrence matrix C, C ij Indicates that SKU is also included i and SKU j The order quantity, C ii Indicates that SKU is included i In addition, assume that the order SKU matrix RO×S , where r os = 1 if and only if s∈o, otherwise r os = 0. Then, the SKU co-occurrence matrix C = R T R. Maximizing the co-occurrence frequency of SKUs is equivalent to minimizing the inverse of the co-occurrence frequency. Based on the decision variables, the minimization objective function can be expressed as:

[0086]

[0087] 2) Minimize the holding cost objective function

[0088] The SKU holding cost uses the unit time holding cost of the average inventory. Based on the input parameters and decision variables, the objective function can be expressed as:

[0089]

[0090] At the same time, it is also necessary to meet the constraints of the warehouse itself. The constraints for product selection generally include the warehouse's volume / category upper limit and the immediate demand fulfillment rate of the SKUs provided by the warehouse. The two constraints can be expressed as follows.

[0091] SKU category upper limit constraints

[0092]

[0093] SKU instant gratification rate constraint

[0094]

[0095] According to the population size N required by the genetic algorithm, N initial solutions (X1, X2, ..., XN) are created, and each initial solution satisfies the constraints. At the same time, the corresponding two objective function values ​​are calculated for each initial solution. Then, the corresponding Pareto level is calculated through fast non-dominated sorting. And the initial solution is iteratively looped through steps such as population merging, crowding calculation, and elite retention strategy. And after looping the preset rounds, the final loop result is obtained. The set of the last generation of solutions of the loop, in which each individual is a feasible result. Individuals with a Pareto level of 1 can be selected from them, that is, the Pareto optimal solution set. Based on the Pareto optimal solution, the second warehouse product to be selected can be selected from the remaining warehouse products. In this embodiment, the warehouse selection problem of multi-objective planning is solved by a fast non-dominated sorting genetic algorithm, which can ensure the efficiency and accuracy of multi-objective planning solution, thereby more effectively selecting the second warehouse product to be selected.

[0096] In one embodiment, step 403 includes: when an individual in the initial solution does not satisfy the constraint condition, assigning the individual a Pareto level corresponding to the violation value of the constraint condition by fast non-dominated sorting.

[0097] Specifically, the goal of fast non-dominated sorting is to divide the Pareto levels according to the objective function results of the initial solution. The Pareto optimal solution of the multi-objective problem is defined as: for the minimization multi-objective optimization problem, if there are two objective components f1(X) and f2(X), if any decision variable X a and X b , the following conditions are met, then X a DominateX b :①f1(X a )≤f1(X b ) and f2(X a )≤f2(X b );②f1(X a ) <f1(X b ) or f2(X a ) <f2(X b ).

[0098] If there is no other decision variable that can dominate a decision variable, then the decision variable is called a non-dominated solution. The Pareto rank is defined as: in a set of solutions, the Pareto rank of the non-dominated solution is defined as 1. The non-dominated solution is deleted from the solution set, and the Pareto rank of the remaining solution is defined as 2. By analogy, the Pareto rank of all solutions in the solution set can be obtained. Therefore, in the product selection problem, due to the existence of constraints, it is defined that when an individual X does not meet the constraints, its Pareto rank is the violation value of the constraints.

[0099] In one embodiment, step 405 includes: obtaining a child solution based on the initial solution; merging the initial solution with the child solution to obtain a merged solution, and starting iterative loop processing; obtaining an elite solution in the merged solution through fast non-dominated sorting and congestion calculation; using the elite solution as the initial solution, returning to the step of obtaining the child solution based on the initial solution, and accumulating iterative loop rounds; when the accumulated loop rounds reach a preset round, using the elite solution after the last loop processing as the loop processing result.

[0100] Specifically, when the genetic algorithm is used for iterative solution, for the initial solution, a set of offspring solutions equal to the number of the initial solution population can be obtained from the initial solution by standard selection, crossover and mutation operators. Then, if the iterative generation of the genetic algorithm reaches the predicted round. If not, the iterative loop processing of the genetic algorithm is entered. During the iterative processing, the population merging is to merge the current parent and offspring populations into an overall population with a population size of 2*N, and the crowding degree calculation is used to determine which individual has a better fitness when the Pareto level of two individuals is the same. We believe that individuals with greater crowding have better fitness. After fast non-dominated sorting and crowding degree calculation, individuals with high Pareto level (1 is the highest) and high crowding degree can be preferentially selected as elites to enter the next generation. Individuals sorted according to these two indicators are added to the next generation one by one until the population size reaches N, and the solution set at this time is an elite solution. After the preset rounds of iteration, the elite solution after the last loop processing can be used as the loop processing result. That is, the result obtained by the last loop processing. The number of rounds of cyclic processing can be determined according to the accuracy required for product selection, warehouse capacity, category constraints of product selection, etc. In this embodiment, the Pareto optimal solution in the fast non-dominated sorting process is obtained by genetic algorithm, which can effectively improve the accuracy of the solution and ensure the accuracy of warehouse product selection.

[0101] In one embodiment, step 405 includes: taking the solution with the smallest order splitting rate target as the second storage product to be selected corresponding to the Pareto optimal solution set.

[0102] Specifically, after obtaining the set of elite solutions of the last generation, each individual in the set is a feasible result. Individuals with a Pareto level of 1, namely the Pareto optimal solution set, can be selected from them. At the same time, based on business experience or business needs, the solution with the smallest order splitting rate target in the solution set can generally be selected as the final product selection result Xopt. In this embodiment, by using the solution with the smallest order splitting rate target as the second storage product to be selected corresponding to the Pareto optimal solution set, it is more in line with the business needs of actual warehouse product selection and ensures the degree of fit between warehouse product selection and business processes.

[0103] It should be understood that although Figure 2-4 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2-4At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0104] In one embodiment, Figure 5 As shown, a warehouse product selection device is provided, comprising:

[0105] Data acquisition module 502, used to acquire warehouse product order data and warehouse cost data;

[0106] A storage product search module 504, used to determine the first storage product to be selected according to the storage product order data;

[0107] The product selection and allocation module 506 is used to determine the second storage product to be selected according to the storage product order data and storage cost data through a multi-objective optimization algorithm based on the optimization goal, wherein the optimization goal is to reduce the order delivery split rate and the storage product holding cost, and meet the constraints of the warehouse itself;

[0108] The warehouse product selection module 508 is used to select warehouse products according to the first warehouse products to be selected and the second warehouse products to be selected.

[0109] In one of the embodiments, the stored product search module 504 is specifically used to: obtain stored product classification demand data, and determine the stored product classification method based on the stored product classification demand data; obtain classification data corresponding to the stored product classification method in the stored product order data; and determine the first stored product to be selected from each type of stored product in the stored product order data based on the classification data.

[0110] In one of the embodiments, the product selection and allocation module 506 is specifically used to: initialize the population of the storage products to be selected according to the optimization goal to obtain the initial solution and the objective function; determine the Pareto level of the initial solution according to the objective function result corresponding to the initial solution through fast non-dominated sorting; iteratively loop process the initial solution through a genetic algorithm, and obtain the loop processing result after a preset number of rounds; determine the Pareto optimal solution set from the loop processing result, and obtain the second storage product to be selected according to the Pareto optimal solution set.

[0111] In one embodiment, the product selection allocation module 506 is specifically used to: through fast non-dominated sorting, when an individual in the initial solution does not meet the constraint condition, assign a Pareto level corresponding to the violation value of the constraint condition to the individual.

[0112] In one embodiment, the product selection and allocation module 506 is specifically used to: obtain child solutions based on the initial solution; merge the initial solution with the child solution to obtain a merged solution, and start iterative loop processing; obtain the elite solution in the merged solution through fast non-dominated sorting and congestion calculation; use the elite solution as the initial solution, return to the step of obtaining the child solution based on the initial solution, and accumulate iterative loop rounds; when the accumulated loop rounds reach the preset rounds, use the elite solution after the last loop processing as the loop processing result.

[0113] In one embodiment, the product selection and allocation module 506 is specifically used to: obtain the order splitting rate target corresponding to each solution in the Pareto optimal solution set; and use the solution with the smallest order splitting rate target as the second storage product to be selected corresponding to the Pareto optimal solution set.

[0114] For the specific definition of the warehouse selection device, please refer to the definition of the warehouse selection method above, which will not be repeated here. Each module in the above-mentioned warehouse selection device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0115] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store warehouse selection data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a warehouse selection method is implemented.

[0116] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0117] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0118] Obtain warehouse product order data and warehouse cost data;

[0119] Determine the first storage product to be selected according to the storage product order data;

[0120] According to the warehouse product order data and storage cost data, the second candidate warehouse product is determined through a multi-objective optimization algorithm based on the optimization goal. The optimization goal is to reduce the order delivery split rate and the storage product holding cost, and meet the warehouse's own constraints.

[0121] The warehouse is selected based on the first warehouse product to be selected and the second warehouse product to be selected.

[0122] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining warehouse product classification demand data, and determining a warehouse product classification method based on the warehouse product classification demand data; obtaining classification data corresponding to the warehouse product classification method in the warehouse product order data; and determining a first warehouse product to be selected from each category of warehouse products in the warehouse product order data based on the classification data.

[0123] In one embodiment, the processor also implements the following steps when executing the computer program: initialize the population of the storage products to be selected according to the optimization goal to obtain the initial solution and the objective function; determine the Pareto level of the initial solution according to the objective function result corresponding to the initial solution through fast non-dominated sorting; iteratively loop process the initial solution through a genetic algorithm, and obtain the loop processing result after a preset number of rounds; determine the Pareto optimal solution set from the loop processing result, and obtain the second storage product to be selected according to the Pareto optimal solution set.

[0124] In one embodiment, when the processor executes the computer program, the following steps are further implemented: when an individual in the initial solution does not satisfy the constraint condition, a Pareto level corresponding to the violation value of the constraint condition is assigned to the individual through fast non-dominated sorting.

[0125] In one embodiment, the processor further implements the following steps when executing the computer program: obtaining a child solution based on the initial solution; merging the initial solution with the child solution to obtain a merged solution, and starting iterative loop processing; obtaining an elite solution from the merged solution through fast non-dominated sorting and congestion calculation; using the elite solution as the initial solution, returning to the step of obtaining the child solution based on the initial solution, and accumulating iterative loop rounds; when the accumulated loop rounds reach a preset round, using the elite solution after the last loop processing as the loop processing result.

[0126] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the order splitting rate target corresponding to each solution in the Pareto optimal solution set; and taking the solution with the smallest order splitting rate target as the second selected storage product corresponding to the Pareto optimal solution set.

[0127] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0128] Obtain warehouse product order data and warehouse cost data;

[0129] Determine the first storage product to be selected according to the storage product order data;

[0130] According to the warehouse product order data and storage cost data, the second candidate warehouse product is determined through a multi-objective optimization algorithm based on the optimization goal. The optimization goal is to reduce the order delivery split rate and the storage product holding cost, and meet the warehouse's own constraints.

[0131] The warehouse is selected based on the first warehouse product to be selected and the second warehouse product to be selected.

[0132] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining storage product classification demand data, and determining a storage product classification method based on the storage product classification demand data; obtaining classification data corresponding to the storage product classification method in the storage product order data; and determining a first storage product to be selected from each type of storage product in the storage product order data based on the classification data.

[0133] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: the population of storage products to be selected is initialized according to the optimization objective to obtain an initial solution and an objective function; the Pareto level of the initial solution is determined according to the objective function result corresponding to the initial solution through fast non-dominated sorting; the initial solution is iteratively looped through a genetic algorithm, and after a preset number of loops, a loop processing result is obtained; the Pareto optimal solution set is determined from the loop processing result, and the second storage product to be selected is obtained according to the Pareto optimal solution set.

[0134] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: when an individual in the initial solution does not satisfy the constraint condition, a Pareto level corresponding to the violation value of the constraint condition is assigned to the individual through fast non-dominated sorting.

[0135] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining a child solution based on the initial solution; merging the initial solution with the child solution to obtain a merged solution, and starting iterative loop processing; obtaining an elite solution from the merged solution through fast non-dominated sorting and congestion calculation; using the elite solution as the initial solution, returning to the step of obtaining the child solution based on the initial solution, and accumulating iterative loop rounds; when the accumulated loop rounds reach a preset round, using the elite solution after the last loop processing as the loop processing result.

[0136] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the order splitting rate target corresponding to each solution in the Pareto optimal solution set; and taking the solution with the smallest order splitting rate target as the second selected storage product corresponding to the Pareto optimal solution set.

[0137] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0138] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0139] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A warehouse product selection method, the method comprising: Obtain warehouse product order data and warehouse cost data; Acquire storage product classification demand data, and determine a storage product classification method according to the storage product classification demand data; Obtaining classification data corresponding to the stored product classification method in the stored product order data; Determine a first stored product to be selected from each type of stored products in the stored product order data according to the classification data; Constructing input parameters and decision variables based on the warehouse product order data and the warehouse cost data; According to the input parameters and the decision variables, an objective function of minimizing the order delivery split rate and an objective function of minimizing the storage product holding cost are respectively constructed, and the constraints of the warehouse itself are satisfied; Initialize the population based on the constraints of the warehouse itself and create an initial solution for the number of populations required by the genetic algorithm; Determining the Pareto rank of the initial solution according to the objective function result corresponding to the initial solution through fast non-dominated sorting; Iterate and cycle the initial solution through the genetic algorithm, and obtain the cycle processing result after a preset number of cycles; Determining a Pareto optimal solution set from the loop processing results; Obtaining the order splitting rate target corresponding to each solution in the Pareto optimal solution set; The solution with the minimum order splitting rate target is used as the second storage product to be selected corresponding to the Pareto optimal solution set; The warehouse selects products according to the first warehouse products to be selected and the second warehouse products to be selected.

2. The method according to claim 1, characterized in that The warehouse product classification methods include the ABC inventory classification management method and the XYZ classification method.

3. The method according to claim 1, characterized in that Determining the Pareto level of the initial solution according to the objective function result corresponding to the initial solution through fast non-dominated sorting includes: By fast non-dominated sorting, when an individual in the initial solution does not satisfy a constraint condition, a Pareto rank corresponding to the violation value of the constraint condition is assigned to the individual.

4. The method according to claim 1, characterized in that: The iterative loop processing of the initial solution by the genetic algorithm, after a preset number of loop rounds, obtaining the loop processing result includes: Obtaining a sub-generation solution according to the initial solution; The initial solution and the child solution are combined to obtain a combined solution, and an iterative loop process begins; Obtaining an elite solution from the merged solution by fast non-dominated sorting and congestion calculation; Taking the elite solution as the initial solution, returning to the step of obtaining the child solution according to the initial solution, and accumulating the number of iterative cycles; When the accumulated number of loop rounds reaches the preset number of rounds, the elite solution after the last loop processing is used as the loop processing result.

5. A warehouse product selection device, characterized in that: The device comprises: Data acquisition module, used to obtain warehouse product order data and warehouse cost data; A storage product search module is used to obtain storage product classification demand data, determine a storage product classification method according to the storage product classification demand data; obtain classification data corresponding to the storage product classification method in the storage product order data; and determine a first storage product to be selected from each type of storage product in the storage product order data according to the classification data; A product selection and allocation module is used to construct input parameters and decision variables based on the storage product order data and the storage cost data; construct an objective function of minimizing the order delivery splitting rate and an objective function of minimizing the storage product holding cost respectively according to the input parameters and the decision variables, and satisfy the warehouse's own constraints; initialize the population based on the warehouse's own constraints, and create an initial solution of the population quantity required by the genetic algorithm; determine the Pareto level of the initial solution according to the objective function result corresponding to the initial solution through fast non-dominated sorting; iterate and loop the initial solution through the genetic algorithm, and obtain the loop processing result after a preset number of loops; determine the Pareto optimal solution set from the loop processing result; obtain the splitting rate target corresponding to each solution in the Pareto optimal solution set; and use the solution with the smallest splitting rate target as the second storage product to be selected corresponding to the Pareto optimal solution set; The warehouse product selection module is used to select warehouse products according to the first warehouse products to be selected and the second warehouse products to be selected.

6. The device according to claim 5, characterized in that The warehouse product classification methods include the ABC inventory classification management method and the XYZ classification method.

7. The device according to claim 5, characterized in that The product selection and allocation module is specifically used for: By fast non-dominated sorting, when an individual in the initial solution does not satisfy a constraint condition, a Pareto rank corresponding to the violation value of the constraint condition is assigned to the individual.

8. The device according to claim 5, characterized in that The product selection and allocation module is also used for: Obtaining a sub-generation solution according to the initial solution; The initial solution and the child solution are combined to obtain a combined solution, and an iterative loop process begins; Obtaining an elite solution from the merged solution by fast non-dominated sorting and congestion calculation; Taking the elite solution as the initial solution, returning to the step of obtaining the child solution according to the initial solution, and accumulating the number of iterative cycles; When the accumulated number of loop rounds reaches the preset number of rounds, the elite solution after the last loop processing is used as the loop processing result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images