Finished product selection methods, equipment and systems

Abstract

The present application discloses a method, device and system for selecting finished products, the method comprising: obtaining selection targets for each set of finished products, and selecting corresponding first-class finished products from each set of finished products according to the selection targets, constructing first-class finished product sets corresponding to all finished product sets according to the first-class finished products corresponding to each set of finished products, and generating a first-class raw material list according to the first-class finished product set, calculating the similarity between the first-class raw material list and the required raw material list of non-first-class finished products, and inferring second-class finished products and third-class finished products from non-first-class finished products according to the similarity; because the present application selects finished products based on the raw material relationship of each type of finished product, the accuracy of finished product selection is improved, the raw materials of important finished products are effectively guaranteed to be complete, and the rapid delivery of superior finished products is achieved.

Description

Technical Field

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

[0002] At present, the market and customer demands are constantly changing, while the operating resources of the supply chain are limited and relatively stable. How to call on limited and relatively stable resources to continuously meet unknown market fluctuations is a test of the flexible management of the supply chain. These tests include unknown demand and limited production capacity, fast delivery and reasonable inventory, etc. The industry usually divides finished products into three categories (Category A: finished product delivery with inventory; Category B: finished product production on order, corresponding raw material inventory; Category C: order procurement, finished products / raw materials are not in inventory) for delivery management. However, the relevant technology has a single selection angle when selecting finished products (that is, dividing finished products into three categories of A / B / C), resulting in inaccurate finished product selection results. Summary of the invention

[0003] The main purpose of this application is to provide a finished product selection method, equipment and system, aiming to solve the technical problem that the related technology has a single selection angle when selecting finished products, resulting in inaccurate finished product selection results.

[0004] To achieve the above objectives, the present application provides a finished product selection method, the finished product selection method comprising:

[0005] Acquire the selection target of each finished product set, and select the corresponding first-category finished product from each finished product set according to the selection target, wherein the first-category finished product is the stocked finished product;

[0006] Constructing a first-category finished product set corresponding to all finished product sets according to the first-category finished products corresponding to each finished product set, and generating a first-category raw material list according to the first-category finished product set;

[0007] Calculate the similarity between the first category raw material list and the required raw material list of non-first category finished products, and reversely infer the second category finished products and the third category finished products from the non-first category finished products based on the similarity, wherein the non-first category finished products are the finished products other than the first category finished products in the set of all finished products, the second category finished products are the finished products that are not in stock but have the corresponding raw materials, and the third category finished products are the finished products that are not in stock and do not have the corresponding raw materials.

[0008] Optionally, obtaining the selection target of each finished product set and selecting the corresponding first-category finished product from each finished product set according to the selection target includes:

[0009] Obtaining the selection target and the selection target weight of each finished product set, and constructing the optimization target of each finished product set according to the selection target and the selection target weight;

[0010] Obtaining constraints, and building a finished product selection model for each finished product set according to the optimization objectives of each finished product set and the constraints, wherein the constraints include that the proportion of the number of first-category finished products in the finished product set is not greater than a preset first threshold, and the raw material similarity is not less than a preset second threshold, and the raw material similarity is the degree of similarity between the first-category raw material list and the second-category raw material list corresponding to the second-category finished product, and the first-category raw material list includes preset first-category raw materials and raw materials required for the first-category finished product;

[0011] A finished product selection model based on each finished product set selects a corresponding first-category finished product from each finished product set.

[0012] Optionally, the finished product selection model based on each finished product set selects a corresponding first-category finished product from each finished product set, including:

[0013] Traverse each finished product set, and use the traversed finished product set as the current finished product set;

[0014] Generate a first type of finished product initial solution of the finished product selection model according to the current finished product set and the constraint conditions;

[0015] Improving the initial solution of the first type of finished products by a meta-heuristic algorithm to obtain an improved solution of the first type of finished products, wherein the meta-heuristic algorithm includes a neighborhood search algorithm and / or a genetic algorithm;

[0016] Inversely infer the current second-category finished product based on the improved solution of the first-category finished product;

[0017] Calculate the fitness improvement of the first type of finished product improved solution according to the optimization target, the first type of finished product improved solution and the current second type of finished product;

[0018] If the fitness improvement is less than a preset third threshold, or the number of solution improvements reaches a maximum number of iterations, the first-category finished product improvement solution is used as the first-category finished product corresponding to the current finished product set;

[0019] After the traversal is completed, the first category of finished products corresponding to each finished product set is obtained.

[0020] Optionally, improving the initial solution of the first type of finished products by a meta-heuristic algorithm to obtain an improved solution of the first type of finished products includes:

[0021] Calculating the raw material similarity between the finished products in the current finished product set by comparing the bill of materials attributes of the finished products in the current finished product set, wherein the bill of materials attributes include the type, quantity and supplier of the raw materials;

[0022] Combining the finished products in the current finished product set according to the raw material similarity to obtain a target finished product combination, and calculating the fitness of the target finished product combination, wherein the raw material similarity between the finished products in the target finished product combination is greater than a preset fourth threshold;

[0023] The initial solution of the first type of finished products is adjusted according to the fitness of the target finished product combination to obtain an improved solution of the first type of finished products.

[0024] Optionally, the step of inferring the current second-category finished product based on the improved solution of the first-category finished product includes:

[0025] Obtain a list of raw materials corresponding to the first type of finished product improvement solution;

[0026] Calculate the similarity between the raw material list corresponding to the improved solution of the first-class finished product and the required raw material list of the non-first-class finished product;

[0027] According to the similarity between the raw material list corresponding to the improved solution of the first category finished product and the required raw material list of the non-first category finished product, the current second category finished product is inferred from the non-first category finished product.

[0028] Optionally, after calculating the fitness improvement amount of the improved solution for the first type of finished product according to the optimization target, the improved solution for the first type of finished product and the current second type of finished product, the method further includes:

[0029] If the fitness improvement is not less than the preset third threshold and the number of improvements of the solution has not reached the maximum number of iterations, the improved solution of the first type of finished product is used as the initial solution of the first type of finished product, and the steps of improving the initial solution of the first type of finished product through the metaheuristic algorithm to obtain the improved solution of the first type of finished product are repeated until the fitness improvement is not less than the preset third threshold, or the number of improvements of the solution reaches the maximum number of iterations.

[0030] Optionally, generating the first type of finished product initial solution of the finished product selection model according to the current finished product set and the constraint condition includes:

[0031] Selecting a finished product combination from the current finished product set according to the constraint condition;

[0032] Calculating the fitness of the finished product combination according to the selection target;

[0033] An initial solution of a first type of finished products is selected from the finished product combination according to the fitness of the finished product combination.

[0034] Optionally, before obtaining the selection target of each finished product set and selecting the corresponding first-category finished product from each finished product set according to the selection target, the method further includes:

[0035] Predicting the future demand for each finished product through a preset prediction model, and constructing an evaluation index matrix based on the future demand, wherein the preset prediction model includes a gradient boosting decision tree machine learning algorithm, an exponential smoothing algorithm, and a time series prediction algorithm based on deep learning;

[0036] Obtaining first-category finished product selection constraints for each finished product set, and selecting corresponding first-category finished product candidate sets from each finished product set according to the first-category finished product selection constraints;

[0037] Accordingly, the step of obtaining the selection target of each finished product set and selecting the corresponding first-category finished product from each finished product set according to the selection target includes:

[0038] The selection target of each finished product set is determined according to the evaluation index matrix, and the first category of finished products is selected from the first category of finished product candidate set according to the selection target.

[0039] In addition, to achieve the above-mentioned purpose, the present application also proposes a finished product selection device, which includes a memory, a processor, and a finished product selection program stored in the memory and executable on the processor, and the finished product selection program is configured to implement the finished product selection method described above.

[0040] In addition, to achieve the above-mentioned purpose, the present application also proposes a finished product selection system, which includes: a sensor, a finished product selection device as described above, and a terminal device;

[0041] The sensor is used to collect finished product information in response to a finished product selection request, and send the finished product information to the finished product selection device;

[0042] The finished product selection device is used to analyze the finished product information, obtain the selection target of each finished product set, and execute the finished product selection method as described above;

[0043] The terminal device is used to display the finished product selection results on a display screen, wherein the finished product selection results include first-category finished products, second-category finished products, and third-category finished products.

[0044] One or more technical solutions proposed in this application have at least the following technical effects:

[0045] In the present application, the method of obtaining the selection target of each finished product set is disclosed, and the corresponding first-class finished products are selected from each finished product set according to the selection target, wherein the first-class finished products are the finished products in stock, the first-class finished product sets corresponding to all finished product sets are constructed according to the first-class finished products corresponding to each finished product set, and the first-class raw material list is generated according to the first-class finished product set, the similarity between the first-class raw material list and the required raw material list of non-first-class finished products is calculated, and the second-class finished products and the third-class finished products are inferred from the non-first-class finished products according to the similarity, wherein the non-first-class finished products are the other finished products in all finished product sets except the first-class finished products, the second-class finished products are the finished products that are not in stock but have the corresponding raw materials, and the third-class finished products are the finished products that are not in stock and do not have the corresponding raw materials; since the present application first generates the first-class raw material list according to the selection target, and then infers the second-class finished products and the third-class finished products according to the similarity between the first-class raw material list and the required raw material list of non-first-class finished products, the finished products can be selected based on the raw material relationship of each type of finished products, thereby improving the accuracy of finished product selection, effectively ensuring that the raw materials of important finished products are complete, and realizing the rapid delivery of superior finished products. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0048] Figure 1 This is a flow chart of the first embodiment of the finished product selection method of the present application;

[0049] Figure 2 This is a flow chart of the second embodiment of the finished product selection method of the present application;

[0050] Figure 3 This is a flow chart of the third embodiment of the finished product selection method of the present application;

[0051] Figure 4 This is a schematic diagram of the calculation of the finished product selection model of an embodiment of the finished product selection method of the present application;

[0052] Figure 5 This is a flow chart of the third embodiment of the finished product selection method of the present application;

[0053] Figure 6 A schematic diagram of the device structure of the hardware operating environment involved in the finished product selection method in the embodiment of the present application;

[0054] Figure 7 This is a schematic diagram of the module structure of the finished product selection system of the embodiment of the present application.

[0055] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0056] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0057] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0058] At present, the market and customer demands are constantly changing, while the operating resources of the supply chain are limited and relatively stable. How to call on limited and relatively stable resources to continuously meet unknown market fluctuations tests the flexible management of the supply chain. These tests include unknown demand and limited production capacity, fast delivery and reasonable inventory, etc. The industry usually divides finished products into three categories (Category A: finished product stock delivery; Category B: finished product order production, corresponding raw material stock; Category C: order purchase, finished products / raw materials are not stocked) for delivery management. Related technology 1 discloses an intelligent chemical system based on material selection, which first obtains several materials that meet the requirements according to user needs, and then outputs the best material based on the two-dimensional matrix information of material performance, comprehensively considers the performance gain of candidate materials compared with basic materials, and outputs the best material. Related technology 2 discloses a material selection and verification system for transmission and transformation engineering equipment based on big data. First, the actual laying distance of the cable and the actual required transmission volume are obtained according to actual needs, and the cable demand level is divided based on these two information. Then, the performance coefficients of all cables are calculated, and the cable performance levels are divided based on the cable performance coefficients. Then, the cable demand levels and cable performance levels are matched one by one. After the matching is completed, the electrical data and mechanical data are verified through the verification module. Related technology 3 discloses a method for recommending alternative materials. First, matching target components are found through the obtained performance requirements, and then the scores of target components and candidate alternative components are calculated according to the set selection goals and the pre-processed knowledge graph, and a list of recommended alternative materials is generated for recommendation. In summary, the related technology has a single selection angle when selecting finished products, so there is a defect of inaccurate finished product selection results.

[0059] Therefore, in order to overcome the above-mentioned defects, the present application provides a solution, which includes: obtaining the selection target of each finished product set, and selecting the corresponding first-class finished products from each finished product set according to the selection target, wherein the first-class finished products are the finished products in stock, constructing the first-class finished product sets corresponding to all finished product sets according to the first-class finished products corresponding to each finished product set, and generating a first-class raw material list according to the first-class finished product set, calculating the similarity between the first-class raw material list and the required raw material list of non-first-class finished products, and inferring the second-class finished products and the third-class finished products from the non-first-class finished products according to the similarity, wherein the non-first-class finished products are the other finished products in all finished product sets except the first-class finished products, the second-class finished products are the finished products that are not in stock but have the corresponding raw materials, and the third-class finished products are the finished products that are not in stock and do not have the corresponding raw materials.

[0060] Since this application first generates a list of first-category raw materials based on the selection objectives, and then reversely infers the second-category finished products and the third-category finished products based on the similarity between the list of first-category raw materials and the required list of raw materials for non-first-category finished products, it is possible to select finished products based on the relationship between the raw materials of each type of finished products, thereby improving the accuracy of finished product selection, effectively ensuring that the raw materials of important finished products are complete, and achieving rapid delivery of superior finished products.

[0061] Based on this, the present application embodiment provides a finished product selection method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the finished product selection method of the present application.

[0062] In the first embodiment, the finished product selection method includes:

[0063] Step S10: obtaining the selection target of each finished product set, and selecting the corresponding first-category finished product from each finished product set according to the selection target;

[0064] Among them, the first category of finished products are finished products in stock.

[0065] It should be noted that the execution subject of the embodiment of the present application can be a finished product selected device with data processing, network communication and program running functions, such as a cloud server, terminal equipment, etc., or other electronic devices that can achieve the same or similar functions. This embodiment does not impose any restrictions on this.

[0066] For ease of understanding, the following examples are given, but the present application is not limited. As an example, assuming that the finished product selection device is a cloud server, the user can enter the target file on the terminal device, and the terminal device uploads the target file to the cloud server. Among them, the target file can be an Excel spreadsheet, a word document, and a finished product image, etc., which is not limited in this embodiment. The cloud server obtains the selection target of each finished product set from the target file, and executes subsequent finished product selection steps to obtain the finished product selection results, wherein the finished product selection results include a first category of finished products, a second category of finished products, and a third category of finished products. The cloud server sends the finished product selection results to the terminal device, and the terminal device displays the finished product selection results on the display screen in a preset manner. Among them, the preset method can be a file display, a picture display, and a video display, etc., which is not limited in this embodiment.

[0067] As another example, assuming that the finished product selection device is a cloud server, in response to the finished product selection request, the sensor obtains the finished product information, and the sensor uploads the finished product information to the cloud server. Among them, the finished product selection request may refer to a request to screen and select finished products according to specific requirements (such as customer orders, inventory management, etc.) during the production or sales process. The sensor can be a camera, a barcode scanner, and other devices, which are not limited in this embodiment. The finished product information may refer to finished product images, finished product videos, and finished product label information, etc., which are not limited in this embodiment. The cloud server analyzes the finished product information, obtains the selection targets of each finished product set, and executes subsequent finished product selection steps to obtain the finished product selection results. The cloud server sends the finished product selection results to the terminal device, and the terminal device displays the finished product selection results in a preset manner on the display screen.

[0068] As another example, assuming that the finished product selection device is a terminal device, the user can directly input the target file on the terminal device, and the terminal device obtains the selection target of each finished product set from the target file, and executes the subsequent finished product selection steps to obtain the finished product selection result. The terminal device displays the finished product selection result in a preset manner on the display screen.

[0069] As another example, assuming that the finished product selection device is a terminal device, in response to a finished product selection request, the sensor obtains finished product information, and the sensor sends the finished product information to the terminal device. The terminal device analyzes the finished product information, obtains the selection target of each finished product set, and executes subsequent finished product selection steps to obtain finished product selection results. The terminal device displays the finished product selection results in a preset manner on the display screen.

[0070] It should be understood that in order to accurately screen the first category of finished products and ensure the supply of key finished products, in this embodiment, the corresponding first category of finished products are selected from each finished product set according to the selection target of each finished product set. Among them, the finished product selection can refer to the classification of finished products, specifically, the finished products can be divided into the first category of finished products, the second category of finished products and the third category of finished products, wherein the first category of finished products are finished products with stock, the second category of finished products are finished products that are not in stock but have corresponding raw materials, and the third category of finished products are finished products that are not in stock and do not have corresponding raw materials. The finished product set can refer to a set of finished products of a certain type, for example, the finished product set can be a camera finished product set, and the camera finished product set includes multiple models of camera finished products; the finished product set can also be a display finished product set, and the display finished product set includes multiple models of display finished products, which is not limited in this embodiment. The selection target can refer to the screening criteria or goals set for a specific finished product set based on market demand, corporate strategy, cost-effectiveness and other factors, which are used to select finished products suitable for a specific strategy from the finished product set.

[0071] In a specific implementation, the first category of finished products may be category A finished products, the second category of finished products may be category B finished products, and the third category of finished products may be category C finished products. The selection objectives include, but are not limited to, the proportion of future orders of category A finished products, the proportion of future orders of category A+B finished products, the proportion of future order items of category A finished products, the proportion of future order items of category A+B finished products, and the future turnover days of category A finished products, which are not limited in this embodiment. Selecting the corresponding first category of finished products from each set of finished products according to the selection objective may be selecting the corresponding first category of finished products from each set of finished products by making the selection objective the best.

[0072] For ease of understanding, the following examples are provided, but are not intended to limit the present application. As an example, assume that the finished product set includes ,in, Represents the nth finished product set, with the first finished product set For example, suppose the finished product set The camera finished product collection has the following selection targets: the proportion of future orders of Class A finished products, the proportion of future orders of Class A+B finished products, the proportion of future order items of Class A finished products, and the proportion of future order items of Class A+B finished products. The first type of finished product can be selected from the finished product set to optimize the selection target. Select the corresponding first category finished product .

[0073] Step S20: construct a first category of finished product sets corresponding to all finished product sets based on the first category of finished products corresponding to each finished product set, and generate a first category of raw material list based on the first category of finished product sets.

[0074] It is understandable that, in order to improve the comprehensiveness of the first-category raw material list, generating the first-category raw material list according to the first-category finished product set may be to analyze the raw materials required for each first-category finished product in the first-category finished product set to obtain the first-category raw material list; or it may be to first analyze the raw materials required for each first-category finished product in the first-category finished product set to obtain the raw materials required for the first-category finished product set, then obtain the preset first-category raw materials, add the raw materials required for the first-category finished product set to the preset first-category raw materials, and obtain the first-category raw material list, which is not limited in this embodiment. Among them, the preset first-category raw materials may be preset default first-category raw materials.

[0075] For ease of understanding, the following examples are provided, but are not intended to limit the present application. As an example, assume that the finished product set includes ,in, Represents the nth finished product set, the finished product set The corresponding first category finished products are ,in, represents the first type of finished product corresponding to the nth finished product set, then the first type of finished product set corresponding to all finished product sets can be obtained ,in, Represents the first type of finished product set. Analyze the raw materials required for each first-class finished product to obtain the raw materials required for the first-class finished product collection , and then obtain the preset first type of raw materials , according to the raw materials required for the first category of finished product collection and preset the first type of raw materials Calculate the first category raw material listCalculate the category A raw material list , the calculation formula is .

[0076] Step S30: calculating the similarity between the first category raw material list and the required raw material list of the non-first category finished product, and inferring the second category finished product and the third category finished product from the non-first category finished product according to the similarity;

[0077] Among them, the non-first category finished products are the other finished products in the set of all finished products except the first category finished products, the second category finished products are the finished products that are not in stock but have corresponding raw materials, and the third category finished products are the finished products that are not in stock and do not have corresponding raw materials.

[0078] It should be understood that in order to accurately select the second category finished products and the third category finished products, in this embodiment, the second category finished products and the third category finished products are inferred by the similarity between the first category raw material list and the required raw material list of non-first category finished products. In a specific implementation, step S30 can specifically be to traverse all non-first category finished products and calculate the similarity between the required raw material list of non-first category finished products and the first category raw material list. According to the similarity threshold, non-first category finished products are divided into two categories: those with a similarity not lower than the similarity threshold are regarded as second category finished products (i.e., finished products that are not in stock but have corresponding raw materials), and those with a similarity lower than the similarity threshold are regarded as third category finished products (i.e., finished products that are not in stock and do not have corresponding raw materials), wherein the similarity threshold can be preset.

[0079] For ease of understanding, the following examples are provided, but are not intended to limit the present application. As an example, assume that the finished product set includes ,in, Indicates the nth set of finished products. The first type of finished products are A-type finished products, the second type of finished products are B-type finished products, and the third type of finished products are C-type finished products. The specific steps for finished product selection are as follows:

[0080] Step 1: Select the finished product set according to the selection target of the finished product set Select the corresponding first-class finished product and A-class finished product .

[0081] Step 2: Assemble the finished products Corresponding Class A finished products Get the Class A finished product collection corresponding to all finished product collections , analyze the raw materials required for each Class A finished product in the Class A finished product set, and obtain the raw materials required for the Class A finished product set , and then obtain the preset Class A raw materials , according to the raw materials required for Class A finished product collection and preset Class A raw materials Calculate the Class A Raw Materials List , the calculation formula is .

[0082] Step 3: Traverse all non-Class A finished products. If a non-Class A finished product requires raw materials and a Class A raw material list If the similarity is not lower than the similarity threshold, the non-Class A finished product is considered to be a Class B finished product; otherwise, the non-Class A finished product is considered to be a Class C finished product.

[0083] Step 4: After the above steps, the A, B, and C selection results of all finished product sets are obtained.

[0084] It should be understood that in order to facilitate users to view the finished product selection results, this embodiment can also output two reports after step S30 and display them on the business front-end page: the two reports are the delivery strategy finished product ABC selection result report and the delivery strategy finished product difference raw material report.

[0085] The delivery strategy finished product ABC selection result report records the selection results of all finished product materials, historical order information, Class A selection constraint satisfaction, material comprehensive sorting and selection result explanation fields. Only some important field information is displayed here. The delivery strategy finished product ABC selection result report is shown in Table 1:

[0086] Table 1

[0087]

[0088] The delivery strategy finished product difference raw material report records which raw materials shortages of Class C finished products may cause delayed delivery of orders, helping to remind planners to communicate in advance and formulate corresponding countermeasures. The delivery strategy finished product difference raw material report is shown in Table 2:

[0089] Table 2

[0090]

[0091] The finished product ABC selection results output by this application realize the hierarchical management of finished product materials, help establish a standardized product delivery strategy, and promote the company's synergy of the "R&D-Purchase-Production-Sales" strategy:

[0092] 1. R&D - finished product prototype identification and differential material analysis: Output the most similar prototype and differential raw materials for each finished product material, and encourage the R&D department to reduce the use of differential raw materials when designing products.

[0093] 2. Procurement - Reserve Class A raw materials required for stocking finished products: Output all Class A raw materials required for stocking finished products (Class A finished products), and promote the procurement department to implement standardized stocking of important raw materials.

[0094] 3. Production - Standardized stocking of finished products for storage: Output a collection of finished products for storage (Class A finished products) to encourage the planning department to carry out standardized stocking.

[0095] 4. Sales - Standard delivery period calculation: The coefficient outputs the ABC selection results of finished products. Combined with the ABC selection results of semi-finished products and raw materials, the standard delivery period of all materials (raw material procurement cycle + finished product and semi-finished product production cycle) can be calculated as a basis for guiding customers to place orders.

[0096] This embodiment first generates a list of first-category raw materials according to the selection objectives, and then reversely infers the second-category finished products and the third-category finished products according to the similarity between the list of first-category raw materials and the required list of raw materials for non-first-category finished products. This enables the finished products to be selected based on the relationship between the raw materials of each type of finished products, thereby improving the accuracy of finished product selection, effectively ensuring that the raw materials of important finished products are complete, and achieving rapid delivery of superior finished products.

[0097] Reference Figure 2 , Figure 2 This is a flow chart of the second embodiment of the finished product selection method of this application, based on the above Figure 1 The first embodiment shown is a second embodiment of the finished product selection method of the present application.

[0098] In the second embodiment, the step S10 includes:

[0099] Step S101: obtaining the selection target and the selection target weight of each finished product set, and constructing the optimization target of each finished product set according to the selection target and the selection target weight.

[0100] It should be understood that in order to make the selected finished products meet the optimization requirements as a whole, in this embodiment, the optimization target can also be constructed according to the selection target and the selection target weight. Among them, the selection target weight can refer to the relative importance or priority of each selection target in the decision-making process, which is used to weight each selection target when calculating the comprehensive optimization target. The optimization target can refer to a comprehensive objective function constructed according to the selection target and the selection target weight, which is used to guide the solution of the finished product selection model so that the selected finished product set meets the optimization requirements as a whole.

[0101] For ease of understanding, the following examples are provided, but are not intended to limit the present application. As an example, assume that the finished product set includes , with finished product collection For example, Represents the first finished product set. Assume that the finished product set The selection goal is , finished product collection The selection target weight , then the finished product set is constructed according to the selection target and the selection target weight through the following formula The optimization goal is:

[0102]

[0103] In the formula, Represents a finished product collection The optimization goal is Represents a finished product collection The i-th selection target, Represents a finished product collection The selection target weight of the i-th selection target.

[0104] In the specific implementation, for example, assuming that the finished product collection is a camera finished product collection. There are four selection targets for the camera finished product collection, namely: the proportion of future orders of Class A finished products, the proportion of future orders of Class A+B finished products, the proportion of future order items of Class A finished products, and the proportion of future order items of Class A+B finished products. The weights of the four selection targets are 0.25, 0.25, 0.25 and 0.25 respectively. Then the finished product collection The optimization goal is , where Represents a finished product collection The i-th selection target, Represents a finished product collection The selection target weight of the i-th selection target. Similarly, the optimization targets of other finished product sets can be constructed, which will not be repeated here.

[0105] Step S102: Obtain constraint conditions, and construct a finished product selection model for each finished product set according to the optimization target of each finished product set and the constraint conditions;

[0106] Among them, the constraints include that the proportion of the number of first-category finished products in the finished product set is not greater than a preset first threshold, and the raw material similarity is not less than a preset second threshold. The raw material similarity is the degree of similarity between the first-category raw material list and the second-category raw material list corresponding to the second-category finished products. The first-category raw material list includes the preset first-category raw materials and the raw materials required for the first-category finished products.

[0107] It is understandable that in order to ensure that the selected finished products meet the actual needs, in this embodiment, constraints are also constructed, and the finished product selection model of each finished product set is constructed according to the optimization objectives and constraints of each finished product set. Among them, the constraints may refer to the limiting conditions that must be met in the finished product selection process, which are used to ensure that the selected finished products meet the actual needs.

[0108] For ease of understanding, the following examples are provided, but are not intended to limit the present application. As an example, assume that the finished product set includes , with finished product collection For example, suppose the finished product set The selection goal is , finished product collection The selection target weight , then according to the finished product collection The optimization objectives and constraints are used to construct an integer programming problem (i.e., the finished product selection model) as shown below:

[0109]

[0110] In the formula, Represents a finished product collection The optimization goal is Represents a finished product collection The i-th selection target, Represents a finished product collection The selection target weight of the i-th selection target. The planning goal of the integer programming problem is: the weighted selection target is optimal. The decision variable of the integer programming problem is: determine which finished products in the finished product set should be selected as Class A finished products. This planning problem is a 0-1 integer programming problem. The constraints of the integer programming problem are: Condition 1, the proportion of Class A finished products ≤ the given threshold, the default is 20%; Condition 2, the similarity between the raw materials required for Class B finished products and the Class A raw materials list ≥ the given threshold, the default is 99%. Among them, the Class A raw materials list It consists of two parts, one of which is the raw materials required for the Class A finished product collection. , the other part is preset Class A raw materials Since the finished products of type A are decision variables, they are constantly changing during the whole process of solving the integer programming problem. Therefore, the raw materials required for set A are and Class A Raw Materials List The set of Class B products launched by Class A products is also constantly changing. Therefore, the selection algorithm disclosed in this embodiment hopes to find prototypes and baseline materials with relatively large sales as Class A materials and the prototype bottom layer Materials with higher similarity are classified as Class B materials.

[0111] Step S103: Select corresponding first-category finished products from each finished product set based on the finished product selection model of each finished product set.

[0112] It should be understood that selecting the corresponding first-class finished products from each finished product set based on the finished product selection model may be performed by solving the finished product selection model of each finished product set to obtain the corresponding first-class finished products selected from each finished product set. Specifically, the integer programming problem of each finished product set may be solved by a neighborhood search algorithm and / or a genetic algorithm to obtain the corresponding first-class finished products selected from each finished product set.

[0113] This embodiment first obtains the selection objectives and selection objective weights of each finished product set, then constructs the optimization objectives based on the selection objectives and the selection objective weights, then constructs the finished product selection model of each finished product set based on the optimization objectives and constraints of each finished product set, and then selects the corresponding first-class finished products from each finished product set based on the finished product selection model, so that the selected finished products can meet the optimization requirements as a whole and ensure that the selected finished products meet actual needs.

[0114] Reference Figure 3 , Figure 3 This is a flow chart of the third embodiment of the finished product selection method of this application, based on the above Figure 2 The embodiment shown is a third embodiment of the finished product selection method of the present application.

[0115] In the third embodiment, the step S103 includes:

[0116] Step S1031: traverse each finished product set, and use the traversed finished product set as the current finished product set.

[0117] It should be understood that in order to ensure that each finished product set is considered and processed and provide a basis for subsequent steps, in this embodiment, each finished product set is processed one by one by traversing. Among them, the finished product set can refer to a set of a certain type of finished products, for example, the finished product set can be a camera finished product set, and the camera finished product set includes multiple models of camera finished products; the finished product set can also be a display finished product set, and the display finished product set includes multiple models of display finished products, which is not limited in this embodiment. The current finished product set can refer to the finished product set that is currently being processed during the traversal process.

[0118] For ease of understanding, the following examples are provided, but are not intended to limit the present application. As an example, assume that the finished product set includes ,in, Represents the i-th finished product set, from Start to traverse each finished product set one by one in order, record the currently traversed finished product set as the current finished product set, and obtain the first type of finished product corresponding to the current finished product set through subsequent analysis steps. After the traversal is completed (that is, all finished product sets After traversing all of them), obtain the first type of finished products corresponding to each finished product set.

[0119] Step S1032: Generate an initial solution for the first type of finished products of the finished product selection model according to the current finished product set and the constraint conditions.

[0120] It is understandable that in order to generate a feasible initial solution and provide a starting point for subsequent optimization and improvement, in this embodiment, the first type of finished product initial solution of the finished product selection model is first generated based on the current finished product set and the constraint conditions. Among them, the constraint conditions may refer to the conditions or restrictions that need to be met when generating the first type of finished product initial solution of the finished product selection model. The first type of finished product initial solution may refer to the initial finished product combination selected from the current finished product set under the constraint conditions.

[0121] In a specific implementation, the first type of finished product initial solution of the finished product selection model generated according to the current finished product set and the constraint conditions can be a combination of finished products randomly selected from the current finished product set based on the constraint conditions as the first type of finished product initial solution, or a combination of finished products with higher fitness can be selected from the current finished product set based on the constraint conditions as the first type of finished product initial solution, which is not limited in this embodiment. Among them, fitness can refer to an indicator used to measure the degree of excellence of a solution in meeting the optimization goal,

[0122] Furthermore, in order to improve the feasibility of the initial solution of the first type of finished products, the step S1032 includes: selecting a finished product combination from the current finished product set according to the constraint conditions; calculating the fitness of the finished product combination according to the selection target; and selecting the initial solution of the first type of finished products from the finished product combination according to the fitness of the finished product combination.

[0123] In a specific implementation, for example, firstly, according to the finished products and constraints in the current finished product set, a finished product combination that satisfies the constraints is selected, and then the fitness of each finished product combination is calculated according to the optimization goal, and the finished product combination with higher fitness is selected as the initial solution of the first category of finished products. Optionally, for each finished product set, the initial solution of the first category of finished products is generated in the above manner, and the initial solution of the first category of finished products includes multiple first category finished products.

[0124] Step S1033: improving the initial solution of the first type of finished products by a meta-heuristic algorithm to obtain an improved solution of the first type of finished products, wherein the meta-heuristic algorithm includes a neighborhood search algorithm and / or a genetic algorithm.

[0125] It should be understood that in order to improve the quality of the solution and make it closer to the optimal solution, in this embodiment, a metaheuristic algorithm can be used to improve the initial solution of the first type of finished product to obtain an improved solution of the first type of finished product. Among them, the improved solution of the first type of finished product can refer to a set of finished products obtained after a series of optimization and improvement operations are performed on the initial solution of the first type of finished product, aiming to better meet the selection objectives and constraints. In a specific implementation, improving the initial solution of the first type of finished product to obtain an improved solution of the first type of finished product can be to use a neighborhood search algorithm and / or a genetic algorithm to improve the initial solution of the first type of finished product to obtain an improved solution of the first type of finished product.

[0126] Furthermore, in order to improve the improvement effect of the initial solution, in this embodiment, during the neighborhood search, the initial solution is also improved with reference to the raw material similarity between each finished product in the current finished product set, and the step S1033 includes: calculating the raw material similarity between each finished product in the current finished product set by comparing the bill of materials attributes of each finished product in the current finished product set, wherein the bill of materials attributes include the type, quantity and supplier of raw materials; combining the finished products in the current finished product set according to the raw material similarity to obtain a target finished product combination, and calculating the fitness of the target finished product combination, wherein the raw material similarity between the finished products in the target finished product combination is greater than a preset fourth threshold; adjusting the initial solution of the first category of finished products according to the fitness of the target finished product combination to obtain an improved solution for the first category of finished products.

[0127] In a specific implementation, during the neighborhood search, the specific steps of improving the initial solution by referring to the raw material similarity between the finished products in the current finished product set include:

[0128] 1. Calculate the similarity of raw materials: For each finished product in the current finished product set, calculate the similarity between them based on raw materials. Specifically, the similarity of raw materials between the finished products in the current finished product set can be calculated by comparing the BOM (Bill of Materials) attributes of each finished product in the current finished product set, where the BOM attributes include but are not limited to the type, quantity and supplier of raw materials, etc., which is not limited in this embodiment.

[0129] 2. Obtaining a target finished product combination: According to the raw material similarity between any two finished products in the current finished product set, the finished products in the current finished product set are combined. If the raw material similarity between the two finished products is greater than a preset fourth threshold, the two finished products can form a finished product combination. After traversing all the finished products in the current finished product set, this embodiment can obtain multiple finished product combinations, wherein the number of finished products included in the finished product combination is at least 1.

[0130] 3. Calculate the fitness of the target finished product combination: Calculate the sum of the fitness of all finished products in the finished product combination as the fitness of the current finished product combination.

[0131] 4. Determine the neighborhood search range: Select a combination of finished products with higher fitness as the range of the neighborhood search. Although some finished products have higher order prediction values, the similarity between the raw materials they need and other finished products is not high. Although some finished products have medium order prediction values, the similarity between the raw materials they need and other finished products is very high. During the search process, this embodiment prefers to select the second type of finished product as Class A finished product in order to increase the possibility of finding a better solution.

[0132] 5. Perform neighborhood search: within the determined search range, adjust the initial solution of the first category of finished products according to the information of the target finished product combination with higher fitness. Specifically, randomly remove a finished product with lower fitness from the initial solution of the first category of finished products, and then select a finished product with higher order prediction value from the combination of finished products with higher fitness, and add it to the current initial solution of the first category of finished products to generate an improved solution. Optionally, standard genetic algorithms (such as crossover, mutation and other operations) and neighborhood search strategies (such as exchange, insertion and other operations) can also be used to improve the candidate solutions and generate improved solutions.

[0133] Step S1034: reverse infer the current second category finished products based on the improved solution of the first category finished products.

[0134] It can be understood that in order to provide necessary inputs for subsequent fitness calculations and verify the validity of the improved solution, in this embodiment, the current second-class finished product is also inferred based on the improved solution of the first-class finished product. Among them, the current second-class finished product can refer to the raw materials required for the improved solution of the first-class finished product, and the set of finished products that are obtained by inversion and have a high similarity with the raw materials of the first-class finished product or meet specific conditions. In a specific implementation, inverting the current second-class finished product based on the improved solution of the first-class finished product can be to obtain the raw material list corresponding to the improved solution of the first-class finished product; calculate the similarity between the raw material list corresponding to the improved solution of the first-class finished product and the required raw material list of the non-first-class finished product; invert the current second-class finished product from the non-first-class finished product according to the similarity between the raw material list corresponding to the improved solution of the first-class finished product and the required raw material list of the non-first-class finished product. Among them, the raw material list corresponding to the improved solution of the first-class finished product can include the raw materials required for the finished product corresponding to the improved solution of the first-class finished product, and can also include the raw materials required for the finished product corresponding to the preset first-class raw materials and the first-class finished product improved solution, and this embodiment is not limited to this.

[0135] In a specific implementation, according to the similarity between the raw material list corresponding to the improved solution of the first-class finished product and the required raw material list of the non-first-class finished product, the current second-class finished product can be inferred from the non-first-class finished product by taking the non-first-class finished product with a similarity not less than a similarity threshold as the current second-class finished product. The similarity threshold can be preset.

[0136] Step S1035: Calculate the fitness improvement of the improved solution for the first category of finished products according to the optimization target, the improved solution for the first category of finished products, and the current second category of finished products.

[0137] It should be understood that, in order to facilitate the subsequent screening of improved solutions, in this embodiment, the fitness improvement amount of the improved solution of the first type of finished product is also calculated based on the optimization target, the improved solution of the first type of finished product, and the current second type of finished product. The fitness improvement amount may refer to the improvement amount of the improved solution in fitness relative to the initial solution, which is used to measure the quality of the solution. In a specific implementation, for example, the fitness improvement amount of the improved solution relative to the initial solution is calculated based on the optimization target.

[0138] Furthermore, in order to gradually approach the global optimal solution and improve the fitness of the solution through continuous iteration and screening, the step S1035 also includes: if the fitness improvement is not less than the preset third threshold, and the number of solution improvements has not reached the maximum number of iterations, then the improved solution of the first category of finished products is used as the initial solution of the first category of finished products, and the steps of improving the initial solution of the first category of finished products by the metaheuristic algorithm to obtain the improved solution of the first category of finished products are repeated until the fitness improvement is not less than the preset third threshold, or the number of solution improvements reaches the maximum number of iterations.

[0139] It can be understood that if the fitness improvement of the improved solution is large enough (not less than the preset third threshold) and the number of iterations has not reached the maximum number of iterations, the improved solution will be used as the initial solution for the next iteration, and the improvement process will be repeated until the fitness improvement of the improved solution is no longer significant or the maximum number of iterations is reached.

[0140] Step S1036: If the fitness improvement is less than the preset third threshold, or the number of solution improvements reaches the maximum number of iterations, the first-category finished product improved solution is used as the first-category finished product corresponding to the current finished product set.

[0141] It is understandable that in order to avoid over-optimization and waste of computing resources while ensuring a reasonable solution, in this embodiment, when the fitness improvement is no longer significant or reaches the maximum number of iterations, the improvement of the current finished product set is stopped and the current improved solution is used as the final result.

[0142] Step S3037: After the traversal is completed, the first category of finished products corresponding to each finished product set is obtained.

[0143] It should be understood that in order to provide comprehensive information for the final selection decision, in this embodiment, after traversing all finished product sets, the first category of finished products corresponding to each finished product set is obtained.

[0144] For ease of understanding, refer to Figure 4 As an example, based on the finished product selection model, the corresponding first-class finished product process is selected from each finished product set. Figure 4 As shown, the specific steps include:

[0145] 1. Generate the initial solution of Class A finished products: randomly generate the initial solution from the finished product set according to the constraints, or select the finished product material combination with higher fitness (relative to the fitness of the selection target) from the finished product set according to the constraints to obtain the initial solution.

[0146] 2. Meta-heuristic algorithm solution: Adaptive Large Neighborhood Search (ALNS) and Genetic Algorithm (GA) are used to improve the existing solution through a series of operations, and select some solutions from the current solution set to enter the next iteration. In terms of search strategy, in addition to mutation and crossover operations, this embodiment designs some local search strategies to speed up the convergence speed, such as: based on BOM similarity preprocessing to obtain a set of finished products with similar BOMs, and give priority to selecting materials in the set of finished products with higher fitness as Class A finished products.

[0147] 3. Update Class A raw material collection , traverse non-Class A materials and obtain Class B finished products: For the improved solution obtained by the metaheuristic algorithm in step 2, based on the set of Class A finished products of the improved solution, update the set of Class A raw materials, and obtain the Class B finished products launched by the latest Class A raw materials.

[0148] 4. Fitness calculation: For the improved solution obtained by the metaheuristic algorithm and the Class B finished product information it introduced, calculate the fitness of the improved solution (relative to the fitness of the selection target), and retain some solutions with higher fitness in the group to enter the next iteration.

[0149] 5. Repeat steps 2-4 until the termination conditions are met, such as reaching the maximum number of iterations, the improvement of the solution is less than a certain threshold, etc., to obtain Class A finished products corresponding to each finished product set.

[0150] In this embodiment, by generating an initial solution for the first category of finished products of the finished product selection model and iteratively improving the initial solution for the first category of finished products to obtain the final first category of finished products, a set of finished products that better meets the selection objectives and constraints can be obtained, thereby improving the accuracy and reliability of the selection results.

[0151] Reference Figure 5 , Figure 5 This is a flow chart of the fourth embodiment of the finished product selection method of the present application. Based on the above embodiments, the fourth embodiment of the finished product selection method of the present application is proposed.

[0152] In the fourth embodiment, before step S10, the method further includes:

[0153] Step S01: Predict the future demand for each finished product through a preset prediction model, and construct an evaluation index matrix based on the future demand, wherein the preset prediction model includes a gradient boosting decision tree machine learning algorithm, an exponential smoothing algorithm, and a time series prediction algorithm based on deep learning.

[0154] It should be understood that in order to provide a decision-making basis for the selection algorithm, in this embodiment, the future demand for each finished product is predicted, and an evaluation index matrix is ​​constructed based on the future demand. Among them, the future demand can refer to the market demand forecast for each finished product in the future period of time, which can be based on historical sales data, market trend analysis, customer orders and other factors. The evaluation index matrix can be a matrix containing multiple evaluation indicators, which is used to evaluate the performance of each finished product in future demand. These indicators can be highly correlated with the core performance goals of the supply chain plan (such as line items, number of units shipped, inventory turnover, etc.).

[0155] For ease of understanding, the following examples are given, but this application is not limited. As an example, the selection of finished materials in the field of supply chain planning will be used to guide subsequent actual business execution, such as: the range of finished product stock materials, the range of raw material stock materials, etc. Therefore, the finished material selection method is best associated with the core performance goal of the supply chain plan "order fulfillment rate × number of units shipped × inventory turnover", which requires that the finished material selection method needs to calculate the evaluation indicators that are highly correlated with the three goals "line items × number of units × turnover" as the decision basis for the selection algorithm. Among them, the predicted value of order line items in the future period and the predicted value of the number of orders in the future period are the most important evaluation indicators and selection decision basis.

[0156] The specific steps are as follows:

[0157] Step 1: Forecast the future demand for all finished products, and obtain the forecast values ​​of future order line items and future order quantity of all finished products. Optionally, the finished product demand forecast data for a future period manually formulated in the system can be directly read. Optionally, a preset forecasting model can be used to predict the finished product demand for all finished products in a future period, where the preset forecasting model includes a gradient boosting decision tree machine learning algorithm (such as the LightGBM method (LightGradient Boosting Machine)), an exponential smoothing algorithm, and a time series forecasting algorithm based on deep learning (such as the N-BEATS method (Neural Basis Expansion Analysis for Time Series Forecasting)).

[0158] Step 2: Calculate the evaluation index matrix as the decision basis for the selection algorithm. The finished material selection method requires the calculation of evaluation indicators that are highly correlated with the performance target "line item × number of units × turnover". Some of the indicators included in the evaluation index matrix are shown in Table 3:

[0159] Table 3

[0160]

[0161] Step S02: obtaining first-category finished product selection constraints of each finished product set, and selecting corresponding first-category finished product candidate sets from each finished product set according to the first-category finished product selection constraints.

[0162] It is understandable that in order to introduce expert experience, avoid mistakenly selecting finished products that do not meet expectations, and ensure the accuracy and feasibility of the selection results, in this embodiment, the first-class finished product selection constraints of each finished product set are first obtained, and the corresponding first-class finished product candidate set is selected from each finished product set according to the first-class finished product selection constraints, and then the first-class finished product is selected from the first-class finished product candidate set based on the selection target. Among them, the first-class finished product selection constraints can refer to the hard conditions set for the first-class finished product (used to screen finished products that meet specific requirements as candidate sets.

[0163] For ease of understanding, the following examples are provided, but the present application is not limited. As an example, the selection constraints and selection targets of the material selection method are read through the system interface. The selection constraints and selection targets of each selection dimension (material group or material category) are manually maintained on the front-end page. When the material selection method is running, it will automatically read the data maintained manually on the front-end page, and obtain the candidate set of Class A finished products based on the selection constraint information. The role of manually maintaining the selection constraints of Class A finished products is to introduce expert experience, avoid the material selection system outputting Class A finished products that do not meet expectations, and avoid the impact of misselection on subsequent business execution.

[0164] The specific steps are as follows:

[0165] Step 1: Read all selection dimensions maintained by the front end through the system interface ,in Indicates the i-th selected material set, specifically representing a material group or a material category. At the same time, the selection constraints of Class A finished products in each dimension are read through the system interface as hard conditions that Class A finished products must meet. i-th selection dimension Examples of constraints for the selection of Class A finished products are shown in the following table:

[0166] Table 4

[0167]

[0168] Step 2: Based on the selection constraints of Class A finished products in each selection dimension, obtain the Class A candidate set of each selection dimension. For example, suppose there are 100 materials in the selection material set of the i-th selection dimension, of which only 30 materials meet the selection constraints of Class A finished products in this selection dimension, then these 30 materials are called the Class A finished product candidate set of this selection dimension.

[0169] Accordingly, the step S10 includes:

[0170] Step S10': determining the selection target of each finished product set according to the evaluation index matrix, and selecting the first category of finished products from the first category of finished product candidate set according to the selection target.

[0171] This embodiment predicts the future demand for each finished product and constructs an evaluation index matrix based on the future demand, so as to provide a decision-making basis for the selection algorithm. In addition, this embodiment first obtains the first-class finished product selection constraints of each finished product set, and selects the corresponding first-class finished product candidate set from the each finished product set according to the first-class finished product selection constraints, and then selects the first-class finished product from the first-class finished product candidate set based on the selection goal, so as to introduce expert experience, avoid misselection of finished products that do not meet expectations, and ensure the accuracy and feasibility of the selection results.

[0172] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the finished product selection method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0173] The present application provides a finished product selection device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the finished product selection method in the above-mentioned embodiment one.

[0174] Reference below Figure 6 , which shows a schematic diagram of the structure of a finished product selection device suitable for implementing the embodiment of the present application. The finished product selection device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6The finished product selection device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0175] like Figure 6 As shown, the finished product selection device may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to the program stored in the ROM (Read Only Memory) 1002 or the program loaded from the storage device 1003 to the RAM (Random Access Memory) 1004. In the RAM 1004, various programs and data required for the operation of the finished product selection device are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, an LCD (Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the finished product selection device to communicate with other devices wirelessly or wired to exchange data. Although the finished product selection device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0176] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0177] The finished product selection device provided by the present application adopts the finished product selection method in the above embodiment, which can solve the technical problem that the selection angle of the related technology is single when selecting finished products, resulting in inaccurate finished product selection results. Compared with the related technology, the beneficial effects of the finished product selection device provided by the present application are the same as the beneficial effects of the finished product selection method provided by the above embodiment, and the other technical features in the finished product selection device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0178] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0179] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0180] This application also provides a finished product selection system, please refer to Figure 7 , the finished product selection system comprises: a sensor 10, a finished product selection device 20 as described above, and a terminal device 30;

[0181] The sensor 10 is used to collect finished product information in response to a finished product selection request and send the finished product information to the finished product selection device;

[0182] The finished product selection device 20 is used to analyze the finished product information, obtain the selection target of each finished product set, and execute the finished product selection method as described above;

[0183] The terminal device 30 is used to display the finished product selection results on a display screen, wherein the finished product selection results include first-category finished products, second-category finished products, and third-category finished products.

[0184] It should be understood that the sensor 10 responds to the finished product selection request, obtains the finished product information, and uploads the finished product information to the cloud server. Among them, the finished product selection request may refer to a request to screen and select finished products according to specific requirements (such as customer orders, inventory management, etc.) during the production or sales process. The sensor may be a camera, a barcode scanner, and other devices, which are not limited in this embodiment. The finished product information may refer to finished product images, finished product videos, and finished product label information, etc., which are not limited in this embodiment. The cloud server analyzes the finished product information, obtains the selection target of each finished product set, and executes subsequent finished product selection steps to obtain the finished product selection result. The cloud server sends the finished product selection result to the terminal device, and the terminal device displays the finished product selection result on the display screen in a preset manner. Among them, the preset method may be file display, picture display, and video display, etc., which are not limited in this embodiment.

[0185] The finished product selection system provided by the present application adopts the finished product selection method in the above embodiment, which can solve the technical problem that the related technology has a single selection angle when selecting finished products, resulting in inaccurate finished product selection results. Compared with the related technology, the beneficial effects of the finished product selection system provided by the present application are the same as the beneficial effects of the finished product selection method provided by the above embodiment, and other technical features of the finished product selection system are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0186] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0187] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0188] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned finished product selection method, and can solve the technical problem that the related art has a single selection angle when selecting finished products, resulting in inaccurate finished product selection results. Compared with the related art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the finished product selection method provided by the above-mentioned embodiment, and will not be repeated here.

[0189] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A finished product selection method, characterized in that: The finished product selection method comprises: Acquire the selection target of each finished product set, and select the corresponding first-category finished product from each finished product set according to the selection target, wherein the first-category finished product is the stocked finished product; Constructing a first-category finished product set corresponding to all finished product sets according to the first-category finished products corresponding to each finished product set, and generating a first-category raw material list according to the first-category finished product set; Calculate the similarity between the first category raw material list and the required raw material list of non-first category finished products, and infer the second category finished products and the third category finished products from the non-first category finished products based on the similarity, wherein the non-first category finished products are all finished products except the first category finished products in the set of all finished products, the second category finished products are finished products that are not in stock but have corresponding raw materials, and the third category finished products are finished products that are not in stock and do not have corresponding raw materials; The step of obtaining the selection target of each set of finished products and selecting the corresponding first-category finished products from each set of finished products according to the selection target includes: Obtaining the selection target and the selection target weight of each finished product set, and constructing the optimization target of each finished product set according to the selection target and the selection target weight; Obtaining constraints, and building a finished product selection model for each finished product set according to the optimization objectives of each finished product set and the constraints, wherein the constraints include that the proportion of the number of first-category finished products in the finished product set is not greater than a preset first threshold, and the raw material similarity is not less than a preset second threshold, and the raw material similarity is the degree of similarity between the first-category raw material list and the second-category raw material list corresponding to the second-category finished product, and the first-category raw material list includes preset first-category raw materials and raw materials required for the first-category finished product; A finished product selection model based on each finished product set selects a corresponding first-category finished product from each finished product set.

2. The finished product selection method according to claim 1, characterized in that: The finished product selection model based on each finished product set selects a corresponding first-category finished product from each finished product set, including: Traverse each finished product set, and use the traversed finished product set as the current finished product set; Generate a first type of finished product initial solution of the finished product selection model according to the current finished product set and the constraint conditions; Improving the initial solution of the first type of finished products by a meta-heuristic algorithm to obtain an improved solution of the first type of finished products, wherein the meta-heuristic algorithm includes a neighborhood search algorithm and / or a genetic algorithm; Inversely infer the current second-category finished product based on the improved solution of the first-category finished product; Calculate the fitness improvement of the first type of finished product improved solution according to the optimization target, the first type of finished product improved solution and the current second type of finished product; If the fitness improvement is less than a preset third threshold, or the number of solution improvements reaches a maximum number of iterations, the first-category finished product improvement solution is used as the first-category finished product corresponding to the current finished product set; After the traversal is completed, the first category of finished products corresponding to each finished product set is obtained.

3. The finished product selection method according to claim 2, characterized in that: The step of improving the initial solution of the first type of finished products by a meta-heuristic algorithm to obtain an improved solution of the first type of finished products includes: Calculating the raw material similarity between the finished products in the current finished product set by comparing the bill of materials attributes of the finished products in the current finished product set, wherein the bill of materials attributes include the type, quantity and supplier of the raw materials; Combining the finished products in the current finished product set according to the raw material similarity to obtain a target finished product combination, and calculating the fitness of the target finished product combination, wherein the raw material similarity between the finished products in the target finished product combination is greater than a preset fourth threshold; The initial solution of the first type of finished products is adjusted according to the fitness of the target finished product combination to obtain an improved solution of the first type of finished products.

4. The finished product selection method according to claim 2, characterized in that: The method of inferring the current second-category finished product based on the improved solution of the first-category finished product comprises: Obtain a list of raw materials corresponding to the first type of finished product improvement solution; Calculate the similarity between the raw material list corresponding to the improved solution of the first-class finished product and the required raw material list of the non-first-class finished product; According to the similarity between the raw material list corresponding to the improved solution of the first category finished product and the required raw material list of the non-first category finished product, the current second category finished product is inferred from the non-first category finished product.

5. The finished product selection method according to claim 2, characterized in that: After calculating the fitness improvement amount of the first type of finished product improved solution according to the optimization target, the first type of finished product improved solution and the current second type of finished product, the method further includes: If the fitness improvement is not less than the preset third threshold and the number of improvements of the solution has not reached the maximum number of iterations, the improved solution of the first type of finished product is used as the initial solution of the first type of finished product, and the steps of improving the initial solution of the first type of finished product through the metaheuristic algorithm to obtain the improved solution of the first type of finished product are repeated until the fitness improvement is not less than the preset third threshold, or the number of improvements of the solution reaches the maximum number of iterations.

6. The finished product selection method according to claim 2, characterized in that: The generating of the first type of finished product initial solution of the finished product selection model according to the current finished product set and the constraint condition comprises: Selecting a finished product combination from the current finished product set according to the constraint condition; Calculating the fitness of the finished product combination according to the selection target; An initial solution of a first type of finished products is selected from the finished product combination according to the fitness of the finished product combination.

7. The finished product selection method according to any one of claims 1 to 6, characterized in that: Before obtaining the selection target of each finished product set and selecting the corresponding first-category finished product from each finished product set according to the selection target, the method further includes: Predicting the future demand for each finished product through a preset prediction model, and constructing an evaluation index matrix based on the future demand, wherein the preset prediction model includes a gradient boosting decision tree machine learning algorithm, an exponential smoothing algorithm, and a time series prediction algorithm based on deep learning; Obtaining first-category finished product selection constraints for each finished product set, and selecting corresponding first-category finished product candidate sets from each finished product set according to the first-category finished product selection constraints; Accordingly, the step of obtaining the selection target of each finished product set and selecting the corresponding first-category finished product from each finished product set according to the selection target includes: The selection target of each finished product set is determined according to the evaluation index matrix, and the first category of finished products is selected from the first category of finished product candidate set according to the selection target.

8. A finished product selection device, characterized in that: The finished product selection device includes: a memory, a processor, and a finished product selection program stored in the memory and executable on the processor. When the finished product selection program is executed by the processor, the finished product selection method according to any one of claims 1 to 7 is implemented.

9. A finished product selection system, characterized in that: The finished product selection system comprises: a sensor, a finished product selection device as claimed in claim 8, and a terminal device; The sensor is used to collect finished product information in response to a finished product selection request, and send the finished product information to the finished product selection device; The finished product selection device is used to analyze the finished product information, obtain the selection target of each finished product set, and execute the finished product selection method according to any one of claims 1 to 7; The terminal device is used to display the finished product selection results on a display screen, wherein the finished product selection results include first-category finished products, second-category finished products, and third-category finished products.

Images