Storehouse network planning method and device, electronic equipment and storage medium

Abstract

The invention provides a warehouse network planning method and device, electronic equipment and a storage medium, and the method comprises the steps: determining a constraint condition and an optimization direction of a warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information and warehouse network configuration information, decoding a code, namely a candidate solution, of a candidate warehouse network planning decision, and carrying out the decoding of the candidate solution. And obtaining a candidate warehouse network planning decision comprising an opening decision of the warehouse and a service decision of the warehouse network transportation path, performing iterative optimization on the current warehouse network planning decision based on the constraint condition, the optimization direction and the candidate warehouse network planning decision, and when a termination condition is reached, performing optimization on the current warehouse network planning decision. And determining the current warehouse network planning decision after iterative optimization as a target warehouse network planning decision. Therefore, iterative optimization of the current warehouse network planning decision can be realized by adopting a heuristic iterative optimization mechanism without depending on artificial experience, so that a reasonable target warehouse network planning decision conforming to the optimization direction is obtained, the actual demand of warehouse network planning is met, and loss is avoided.

Description

Technical Field

[0001] This application relates to the field of logistics technology, specifically to a warehouse network planning method, apparatus, electronic device, and storage medium. Background Technology

[0002] In general, during the transit and transportation of products in warehouses, enterprise users need to plan warehouse networks (i.e., warehouse networks) to ensure smooth flow of related needs. In the past, warehouse network planning methods usually relied on statistical analysis of downstream, i.e., customer product demand and a deep understanding of the supply chain network, making warehouse network planning decisions based on manual experience or direct numerical ranking.

[0003] However, since warehouse network planning needs to meet relatively complex specific conditions, methods based on manual experience or direct numerical sorting may ignore some key factors, making it impossible to obtain reasonable and efficient planning decisions. In some cases, the planning scheme may fail to meet the actual needs of warehouse network planning, resulting in direct or indirect losses in operation, warehousing and transportation. Summary of the Invention

[0004] Based on the aforementioned defects and shortcomings of the existing technology, this application proposes a warehouse network planning method, device, electronic device, and storage medium. It can iteratively optimize the current warehouse network planning decision based on constraints and optimization directions such as transportation timeliness, warehouse capacity and quantity, and candidate warehouse network planning decisions. It then determines the iteratively optimized current warehouse network planning decision as the target parameter network planning decision, thereby obtaining a reasonable target warehouse network planning decision that conforms to the optimization direction, meets the actual needs of warehouse network planning, and avoids losses.

[0005] According to a first aspect of the embodiments of this application, a warehouse network planning method is provided, comprising:

[0006] Based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information, the constraints and optimization directions of the warehouse network planning problem are determined. The warehouse network configuration information is determined based on the needs of enterprise users who provide products to customers. The constraints include transportation timeliness, warehouse capacity, and number of warehouses.

[0007] Decode the candidate solutions to obtain candidate warehouse network planning decisions, which include warehouse opening decisions and warehouse network transportation route service decisions. The candidate solutions are the codes of the candidate warehouse network planning decisions.

[0008] Based on the constraints, the optimization direction, and the candidate warehouse network planning decisions, the current warehouse network planning decision is iteratively optimized until the termination condition is met. Then, the iteratively optimized current warehouse network planning decision is determined as the target warehouse network planning decision.

[0009] According to a second aspect of the embodiments of this application, a warehouse network planning device is provided, comprising:

[0010] The determination module is used to determine the constraints and optimization direction of the warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information. The warehouse network configuration information is determined based on the needs of enterprise users who provide products to customers. The constraints include transportation timeliness constraints, warehouse capacity constraints, and warehouse quantity constraints.

[0011] The decoding module is used to decode the candidate solutions to obtain candidate warehouse network planning decisions. The candidate warehouse network planning decisions include warehouse opening decisions and warehouse network transportation path service decisions. The candidate solutions are the encodings of the candidate warehouse network planning decisions.

[0012] The iterative optimization module is used to iteratively optimize the current warehouse network planning decision based on the constraints, the optimization direction, and the candidate warehouse network planning decisions, until the termination condition is met, and then determine the iteratively optimized current warehouse network planning decision as the target warehouse network planning decision.

[0013] According to a third aspect of the embodiments of this application, an electronic device is provided, including a memory and a processor;

[0014] The memory is connected to the processor and is used to store programs;

[0015] The processor is used to implement the warehouse network planning method as described in the first aspect by running the program in the memory.

[0016] According to a fourth aspect of the embodiments of this application, a storage medium is provided, on which a computer program is stored, and when the computer program is run by a processor, it implements the warehouse network planning method as described in the first aspect.

[0017] The aforementioned warehouse network planning methods, devices, electronic equipment, and storage media can determine constraints such as transportation timeliness, warehouse capacity, and number of warehouses, as well as optimization directions, based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information. The encoded candidate solutions for candidate warehouse network planning decisions are decoded to obtain candidate warehouse network planning decisions. Based on the constraints, optimization directions, and candidate warehouse network planning decisions, the current warehouse network planning decision is iteratively optimized until a termination condition is met. The iteratively optimized current warehouse network planning decision is then determined as the target warehouse network planning decision. In this way, without relying on human experience, a heuristic iterative optimization mechanism can be used to iteratively optimize the current warehouse network planning decision based on the optimization direction, constraints, and candidate warehouse network decisions, resulting in a reasonable target warehouse network planning decision that conforms to the optimization direction, meeting the actual needs of warehouse network planning and avoiding losses. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0019] Figure 1 A flowchart illustrating a warehouse network planning method provided in an embodiment of this application;

[0020] Figure 2 This is a schematic diagram illustrating an iterative solution process for a warehouse network planning problem, as provided in an embodiment of this application.

[0021] Figure 3 This is a schematic diagram of the structure of a warehouse network planning device provided in an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of the structure of an electronic device proposed in an embodiment of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0024] Overview

[0025] As described in the background section, traditional warehouse network planning methods typically rely on statistical analysis of downstream (customer) product demand and a deep understanding of the supply chain, making warehouse network planning decisions based on manual experience or direct numerical ranking. However, because warehouse network planning needs to meet relatively complex and specific conditions, methods based on manual experience or direct numerical ranking may overlook certain key factors, failing to obtain reasonable and efficient planning decisions, and even leading to planning schemes that do not meet the actual needs of warehouse network planning, resulting in direct or indirect losses in operations, warehousing, and transportation.

[0026] Building upon this foundation, the inventors further discovered that by determining warehouse network configuration information based on warehouse information, factory information, customer demand information, warehouse network transportation information, and the needs of enterprise users providing products to customers, constraints such as transportation timeliness, warehouse capacity, and number of warehouses, as well as optimization directions, are identified. The encoding of candidate warehouse network planning decisions (candidate solutions) is then decoded to obtain candidate warehouse network planning decisions. Based on the constraints, optimization directions, and candidate warehouse network planning decisions, the current warehouse network planning decision is iteratively optimized until a termination condition is met. The iteratively optimized current warehouse network planning decision is then determined as the target warehouse network planning decision. In this way, without relying on human experience, a heuristic iterative optimization mechanism, based on optimization directions, constraints, and candidate warehouse network decisions, can iteratively optimize the current warehouse network planning decision to obtain a reasonable target warehouse network planning decision that conforms to the optimization direction, meeting the actual needs of warehouse network planning and avoiding losses.

[0027] Based on the above concept, this specification provides a warehouse network planning method, which will be described exemplarily below with reference to the accompanying drawings.

[0028] Exemplary methods

[0029] Please see Figure 1 In one exemplary embodiment, a warehouse network planning method is provided, applicable to any electronic device. For example... Figure 1 As shown, the warehouse network planning method includes steps S101-S103:

[0030] S101: Based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information, determine the constraints and optimization direction of the warehouse network planning problem.

[0031] Specifically, warehouse information refers to the basic information of candidate warehouses, including warehouse name (i.e., warehouse identifier), warehouse location, warehouse capacity, warehouse type, warehouse level, and stored item identifiers.

[0032] The warehouse location includes the province and city where the warehouse is located, as well as the specific address of the warehouse.

[0033] In addition, warehouse levels are based on the role of warehouses in the logistics system. Generally, warehouse levels can be divided into, for example, central warehouses, regional warehouses, and forward warehouses.

[0034] For example, warehouse information includes warehouse identifier such as C0001, province and city information of the warehouse such as Hangzhou City, Zhejiang Province, specific address of the warehouse such as No. x, xx Road, Yuhang District, Hangzhou City, Zhejiang Province, warehouse level such as regional warehouse / central warehouse, and stored item identifiers such as P0001 and P0002.

[0035] Specifically, factory information refers to the factory's basic information and production information. Generally, basic factory information includes the factory name (i.e., warehouse identification), factory location, and product information, while production information includes the factory's maximum production capacity.

[0036] The factory location includes the province and city where the factory is located, as well as the factory's specific address.

[0037] For example, factory information includes factory identifier such as F0001, province and city where the factory is located such as Beijing, specific address of the factory such as No. x, xx Road, Daxing District, Beijing, and product information including product identifier such as P0001, maximum production capacity of 10,000 pieces P0001 / day, etc.

[0038] Specifically, customer demand information includes basic customer information and required product information. Generally, basic customer information includes customer name (i.e., customer identifier) ​​and customer location, while demand information includes product name (i.e., product identifier), product quantity, product weight, product volume, and product delivery time.

[0039] The customer's location includes the province and city where the customer is located, as well as the customer's specific address.

[0040] For example, customer demand information includes customer identifier such as K0001, customer's province and city such as Dongguan City, Guangdong Province, customer's specific address such as Room xxx, No. x, xx Road, Nancheng District, Dongguan City, Guangdong Province, product identifier such as P0001, product quantity such as 10,000 pieces, product weight such as 40,000 kilograms, product volume such as 100 cubic meters, and product delivery time such as 48 hours.

[0041] Specifically, warehouse network transportation information refers to transportation information between factories, warehouses, and customers, including information such as origin, destination, distance, transportation mode, transportation time, and transportation cost.

[0042] Transportation costs can be calculated based on the number of products transported, their weight, volume, and transportation distance.

[0043] For example, the warehouse network transportation information includes the origin, such as warehouse C0001, the destination, such as customer K0001, the transportation method, such as container truck, the transportation time, such as 24 hours, and the transportation cost, such as 2 yuan / (kg*km).

[0044] Specifically, the warehouse network configuration information is determined based on the needs of enterprise users who provide products to customers.

[0045] Among them, the needs of enterprise users who provide products to customers can be determined by enterprise users based on the current scenario, customer needs, and enterprise development needs.

[0046] More specifically, warehouse network configuration information may include the number of activated warehouses and transportation time, such as a transportation time of 24 / 48 hours from an activated warehouse to the customer. Understandably, by configuring transportation time, warehouses that are too far from the customer's location in the warehouse network transportation information can be filtered out.

[0047] More specifically, the warehouse network configuration information also includes mandatory warehouses, i.e., warehouses that must be enabled. Of course, depending on the actual needs, the warehouse network configuration information may also include other configuration items.

[0048] Specifically, based on customer demand information, the decision-making factors are summarized, namely the decision to activate the warehouse and the service decision of the warehouse network transportation route. Based on the warehouse network configuration information, combined with factory information, warehouse information, transportation network information, etc., the constraints and optimization directions of the warehouse network planning problem are generated.

[0049] The constraints of the warehouse network planning problem include transportation timeliness, warehouse capacity, and number of warehouses.

[0050] Generally, the number of warehouses includes the total number of all warehouses in operation.

[0051] Furthermore, the number of warehouses also includes the number of warehouses of each type that are in use, and the number of warehouses at each level that are in use.

[0052] In addition, constraints may include mandatory warehouse selection, etc.

[0053] For example, constraints include warehouse capacity limitations. For instance, if warehouse 1 is started and needs to provide 10,000 products to 10 customers, but warehouse 1 is limited by its capacity and may only be able to provide 6,000 products, then warehouse 1 cannot provide all 10,000 products. In this case, based on the constraints, a trade-off needs to be made, such as abandoning some customers, even if the cost of these customers going to warehouse 1 is the lowest compared to other warehouses.

[0054] Alternatively, for example, constraints may include time limits for delivery, such as when the product required by the customer is a frozen item, in which case the time from the warehouse to the user must not exceed 24 / 48 hours, otherwise the quality of the product delivered to the customer cannot be guaranteed.

[0055] Furthermore, the optimization direction of the warehouse network planning problem refers to the desired improvement in key performance indicators (KPIs) during the iterative optimization of warehouse network planning decisions. For example, the optimization direction could be to achieve optimal cost, optimal number of warehouses (i.e., the number of warehouses closest to the constraints or the minimum number of warehouses), and optimal timeliness.

[0056] S102: Decode the candidate solutions to obtain the candidate warehouse network planning decision.

[0057] Among them, the candidate warehouse network planning decision includes the decision to open the warehouse and the service decision of the warehouse network transportation route.

[0058] Specifically, the warehouse opening decision represents which warehouses need to be opened, and the warehouse network transportation route service decision represents the transportation route from the warehouse to the customer.

[0059] Furthermore, candidate solutions are the encodings of candidate warehouse network planning decisions. By executing various search mechanisms within the algorithm, the existing encodings (i.e., the encodings of the current solution) are searched to find improved solutions that can enhance the current solution. Specifically, the encoding of the current solution is the encoding of the current warehouse network planning decision, which is the warehouse network planning decision that best fits the current optimization direction.

[0060] The current warehouse network planning decision can be either an initial warehouse network planning decision or a candidate warehouse network planning decision.

[0061] S103: Based on constraints, optimization direction and candidate warehouse network planning decisions, iteratively optimize the current warehouse network planning decision until the termination condition is met, and then determine the iteratively optimized current warehouse network planning decision as the target warehouse network planning decision.

[0062] Specifically, during the iterative optimization of the current warehouse network planning decision, if a candidate warehouse network planning decision is more in line with the optimization direction than the current warehouse network planning decision, then the current warehouse network planning decision is optimized and updated to the candidate warehouse network planning decision. Conversely, if the current warehouse network planning decision is more in line with the optimization direction than the candidate warehouse network planning decision, then the current warehouse network planning decision is not updated and the candidate warehouse network planning decision is discarded.

[0063] The termination conditions are typically the maximum number of iterations, the maximum execution time, and the maximum number of iterations without improvement.

[0064] For example, the maximum number of iterations can range from 500 to 4000, the maximum execution time can range from 10 to 30 minutes, and the maximum number of iterations without improvement can range from 30 to 50.

[0065] In this embodiment, based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information, constraints such as transportation timeliness, warehouse capacity, and number of warehouses, as well as optimization directions, are determined. The encoding of candidate warehouse network planning decisions (candidate solutions) is decoded to obtain candidate warehouse network planning decisions. Based on the constraints, optimization directions, and candidate warehouse network planning decisions, the current warehouse network planning decision is iteratively optimized until a termination condition is met. The iteratively optimized current warehouse network planning decision is then determined as the target warehouse network planning decision. Thus, without relying on human experience, a heuristic iterative optimization mechanism, based on optimization directions, constraints, and candidate warehouse network decisions, can iteratively optimize the current warehouse network planning decision to obtain a reasonable target warehouse network planning decision that conforms to the optimization direction, meeting the actual needs of warehouse network planning and avoiding losses.

[0066] Since the products required by customers need to be transported from the warehouse to the customers to meet their needs, warehouse and customer configuration is required when planning the warehouse network. In other words, the warehouse network configuration information must include warehouse configuration items and customer configuration items. These are the key configuration items in the warehouse network configuration information. If the key information items are missing from the warehouse network configuration information, warehouse network planning may not be possible, or the warehouse network planning effect may be poor.

[0067] Therefore, in some embodiments, before determining the constraints and optimization direction of the warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information, it is verified whether the warehouse network configuration information is missing key information items. Based on the verification results, relevant personnel are reminded or the next operation is performed, that is, the operation of determining the constraints and optimization direction of the warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information.

[0068] Specifically, the system verifies whether the warehouse network configuration information is missing key information items, namely warehouse configuration items and customer configuration items. If so, the verification result indicates that the warehouse network configuration information is missing key information items, and a prompt message is generated to remind users of the missing key information items. If not, the verification result indicates that the warehouse network configuration information is not missing key information items, and the system executes operations to determine the constraints and optimization direction of the warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information.

[0069] Depending on the actual working conditions, the warehouse network configuration information may vary, and other configuration items may also be included in the key information items of the warehouse network configuration information.

[0070] Understandably, in some scenarios involving central warehouses, it may not be necessary to consider factories or multi-level warehouses. Furthermore, flexible warehouse network configuration can be achieved by configuring warehouses through key information items in the warehouse network configuration information, such as warehouse configuration items.

[0071] In this embodiment, the system checks whether the acquired warehouse network configuration information is missing any key information items. If key information items are missing, a prompt message is generated to remind relevant personnel of the missing key information items in the warehouse network configuration information, so that relevant personnel can promptly supplement and input the key information items. If no key information items are missing, the system performs operations to determine the constraints and optimization direction of the warehouse network planning problem. This avoids situations where warehouse network planning cannot be carried out or reasonable warehouse network planning decisions cannot be obtained due to the lack of key information items, ensuring the smooth progress of warehouse network planning and the rationality of the final warehouse network planning decision.

[0072] In some embodiments, before decoding the candidate solutions to obtain the candidate warehouse network planning decision, the current solution is searched based on various search mechanisms to obtain candidate solutions.

[0073] Among them, the search mechanism is usually referred to as an operator. Common operators include neighborhood search operator, variable neighborhood search operator, cross search operator, insertion search operator, and destruction repair search operator.

[0074] The neighborhood search operator's search mechanism is as follows: In combinatorial optimization problems, a neighborhood is generally defined as the set of nodes in the given problem domain obtained by transforming each node in the given problem domain according to a given transformation rule. Neighborhood search involves performing an operation, namely a neighborhood action, on the current solution to obtain a set of all corresponding candidate solutions.

[0075] The variable neighborhood search operator is an improved local search algorithm that uses neighborhood structures formed by different actions to perform alternating searches, achieving a better balance between concentration and dispersion, and obtaining corresponding candidate solutions.

[0076] In genetic algorithms, the crossover operator refers to the operator that recombines the genes of two individuals to produce a new individual. The crossover operation, based on the natural phenomenon of gene recombination, is used to improve the fitness of individuals and increase evolutionary diversity.

[0077] The insertion search operator randomly selects a node from one solution and inserts it into another solution to obtain a new candidate solution.

[0078] The destruction-repair search operator uses a combination of removal and insertion operations to first remove some nodes from the sequence and then (according to certain rules) insert them into the original sequence to obtain new candidate solutions.

[0079] Specifically, before decoding the candidate solutions to obtain the candidate warehouse network planning decision, a preset search operator and the corresponding preset usage frequency are first obtained, and a candidate solution is obtained by searching the current solution based on the preset search operator and the preset usage frequency.

[0080] Among them, there is at least one preset search operator (or at least one preset search operator), and correspondingly, the preset usage frequency corresponds to the preset search operator.

[0081] More specifically, at least one preset search operator and the preset usage frequency of each preset search operator are obtained, and candidate solutions are obtained by searching the current solution based on the at least one search operator and the preset usage frequency of each preset search operator.

[0082] Among them, at least one of the above-mentioned preset search operators and the corresponding preset usage frequency are determined based on actual working conditions.

[0083] In this embodiment, based on at least one preset search operator and the preset usage frequency of each preset search operator, a search is performed on the current solution to obtain candidate solutions, so as to ensure high search efficiency and obtain solutions that are more in line with the optimization direction.

[0084] In some embodiments, candidate solutions are decoded to obtain warehouse opening decisions in candidate warehouse network planning decisions. Then, based on warehouse opening decisions and optimization directions, the levels of warehouses through which products are transported to customers are determined, thus obtaining service decisions for warehouse network transportation routes.

[0085] Among them, the candidate solution is the encoding of the candidate warehouse network planning decision.

[0086] For example, the candidate solution is encoded using the common 0-1 encoding.

[0087] Based on the encoded information, i.e., candidate solutions, the warehouse opening decision is mapped to downstream customers, or customer demand points, and also to upstream source factories. The mapping logic can be determined based on optimization directions such as minimum cost or optimal timeliness. Based on this mapping logic and the decoding method of the candidate solutions, the warehouse opening decision and the service decision of the warehouse network transportation path are obtained in the candidate warehouse network planning decision.

[0088] In other words, specifically, based on the warehouse opening decision and the optimization direction, the service decision of the transportation route of the warehouse network is determined step by step to determine the level of warehouses and even factories of the customers.

[0089] For example, taking the transportation of products from a factory and a central warehouse to customer 1 as an example, based on customer 1's customer demand information and the warehouse opening decision, the warehouse a that best fits the optimization direction (e.g., lowest cost or shortest delivery time) among the open central warehouses is first selected as the central warehouse in customer 1's warehouse network transportation path. Then, based on the warehouse information of warehouse a and the factory information, the factory b that best fits the optimization direction (e.g., lowest cost or shortest delivery time) is selected as the factory in customer 1's warehouse network transportation path, thereby obtaining the service decision for customer 1's warehouse network transportation path, namely the warehouse network transportation path of factory b-warehouse a-customer 1.

[0090] For example, the warehouse opening decision is encoded using a 0-1 encoding method. To simplify the logic, only the matching of a single-level warehouse (i.e., a warehouse at a certain level) with the customer is considered. Taking 10 candidate warehouses as an example, encoding the warehouse opening decision in a feasible solution yields a feasible encoding as shown in Table 1 below. Warehouses W1, W3, W4, W8, and W10, with an encoding of 1, are enabled / opened warehouses. Correspondingly, warehouses W2, W5, W6, W7, and W9, with an encoding of 0, are not enabled / not opened warehouses.

[0091] Table 1

[0092] storehouse W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 coding 1 0 1 1 0 0 0 1 0 1

[0093] Specifically, after decoding the candidate solutions and obtaining the warehouse opening decision, based on the optimization direction (e.g., minimum / lowest cost) and the cost of transporting products from the opened warehouse to the customer, the warehouse with the lowest cost of transporting products to the customer is matched with that customer, and the warehouse is mapped to the corresponding customer. If the obtained warehouse network configuration information also includes factories, the factory with the lowest cost of providing products to the corresponding warehouse for transportation to the customer can be mapped to that customer, thereby obtaining the service decision for the warehouse network transportation path from the factory to the warehouse and then to the customer.

[0094] For example, based on the coding of the 10 candidate warehouses in the above example, and considering 10 customers C1-C10, the warehouse selected by each customer is determined based on the minimum transportation cost from these 10 candidate warehouses to the customers. The product demand (kg), transportation time requirements (hours), and transportation costs (ten thousand yuan) and transportation time (hours) from the warehouse selected by each customer to the corresponding customer are shown in Table 2 below. Customers C1, C2, and C5 all chose warehouse W1, with product demands of 10,000, 8,000, and 5,000 kg respectively, delivery time requirements of 24, 36, and 36 hours respectively, transportation costs of 10,000, 20,000, and 15,000 yuan respectively, and transportation times of 15, 20, and 30 hours respectively; Customers C3 and C7 both chose warehouse W3, with product demands of 7,000 and 9,000 kg respectively, delivery time requirements of 48 and 24 hours respectively, transportation costs of 30,000 and 22,000 yuan respectively, and transportation times of 33 and 36 hours respectively; Customers C4 and C8 both chose warehouse W4, with product demands of 10,000, 8,000, and 5,000 kg respectively, delivery time requirements of 24, 36, and 36 .... Customers C6 and C9 both choose warehouse W8, with product demands of 8,000 and 10,000 kg respectively, delivery time requirements of 24 and 36 hours respectively, transportation costs of 20,000 and 37,000 yuan respectively, and transportation times of 40 and 12 hours respectively; Customer C10 chooses warehouse W10, with product demands of 10,000 kg, delivery time requirements of 24 hours, transportation costs of 20,000 yuan respectively, and transportation times of 30 hours respectively.

[0095] Table 2

[0096]

[0097] In this embodiment, based on the warehouse opening decision in the candidate warehouse network planning decision obtained from the candidate decoding, and combined with the optimization direction, the warehouses at all levels through which the product is transported to the customer are determined, and the service decision of the warehouse network transportation path is obtained, that is, the warehouse network transportation path is obtained, thereby obtaining the warehouse network transportation path that conforms to the optimization direction, meeting the needs of warehouse network planning in the optimization direction, and avoiding losses.

[0098] In some embodiments, when iteratively optimizing the current warehouse network planning decision based on constraints, optimization direction, and candidate warehouse network planning decisions, the evaluation function of the warehouse network planning decision is first obtained, and then the evaluation function is used to evaluate the candidate warehouse network planning decisions to obtain the implementation cost of the candidate warehouse network planning decisions. Based on the implementation cost of the candidate warehouse network planning decisions, the implementation cost of the current warehouse network planning decision, and the optimization direction, the current warehouse network planning decision is iteratively optimized.

[0099] The evaluation function is used to evaluate the implementation cost of warehouse network planning decisions.

[0100] Specifically, the evaluation function can be obtained directly from local storage or it can be constructed in real time based on the current operating conditions.

[0101] More specifically, when constructing the evaluation function in real time based on the current working conditions, the evaluation function can be constructed based on the constraints and optimization directions of the above-mentioned warehouse network planning problem.

[0102] Specifically, when evaluating candidate warehouse network planning decisions using an evaluation function, the evaluation is based on warehouse information, factory information, customer demand information, and warehouse network transportation information to obtain the implementation cost of the candidate warehouse network planning decisions.

[0103] Similarly, the implementation cost of current warehouse network planning decisions is also evaluated using an evaluation function, which evaluates the corresponding warehouse network planning decisions based on warehouse information, factory information, customer demand information, and warehouse network transportation information.

[0104] It is understandable that the current warehouse network planning decision is the candidate warehouse network planning decision that best fits the optimization direction up to the current moment in the iterative optimization process.

[0105] Specifically, based on the implementation costs of candidate warehouse network planning decisions, the implementation costs of the current warehouse network planning decisions, and the optimization direction, when iteratively optimizing the current warehouse network planning decisions, the optimization direction is to minimize the implementation cost. First, the implementation costs of candidate warehouse network planning decisions and the current warehouse network planning decisions are compared to determine whether the implementation cost of candidate warehouse network planning decisions is lower than that of the current warehouse network planning decisions. Then, based on the comparison results, the current warehouse network planning decisions are iteratively updated.

[0106] If the implementation cost of the candidate warehouse network planning decision is lower than that of the current warehouse network planning decision, then the current warehouse network planning decision is updated to the candidate warehouse network planning decision. If not, that is, the implementation cost of the candidate warehouse network planning decision is not lower than that of the current warehouse network planning decision, then the current warehouse network planning decision is not updated, that is, the current warehouse network planning decision remains unchanged, and the candidate warehouse network planning decision is discarded.

[0107] In this embodiment, an evaluation function is constructed based on the constraints and optimization directions of the warehouse network planning problem. This results in an evaluation function that can comprehensively evaluate warehouse network planning decisions based on constraints and optimization directions. Using this evaluation function, based on warehouse information, factory information, customer demand information, and warehouse network transportation information, an accurate evaluation of candidate warehouse network planning decisions can be achieved. The implementation cost of candidate warehouse network planning decisions can be determined more accurately. Therefore, based on the implementation costs and optimization directions of candidate and current warehouse network planning decisions, the current warehouse network planning decision can be iteratively optimized to ultimately obtain a reasonable target warehouse network planning decision.

[0108] It is understandable that this application employs a heuristic iterative optimization mechanism to solve the warehouse network planning problem, that is, it uses a heuristic algorithm for computation to solve the warehouse network planning problem. Heuristic algorithms, as opposed to optimization algorithms, can be considered algorithms constructed based on intuition or experience. Within an acceptable cost range (i.e., computation time and space), they provide a feasible solution for each example of the combinatorial optimization problem to be solved. The deviation of this feasible solution from the optimal solution is generally unpredictable. Currently, heuristic algorithms are mainly natural body-inspired algorithms, including ant colony optimization, simulated annealing, and neural networks. By encoding and representing decision information (i.e., warehouse decisions and service decisions for warehouse network transportation routes), and designing an effective decoding strategy, the encoding is parsed into complete candidate warehouse network planning decisions and evaluated. By designing a reasonable evolutionary strategy and iterative mechanism, the optimization result—the target warehouse network planning decision—adapting to the constraints and optimization direction of the current scenario can be obtained.

[0109] Because warehouse network planning is quite complex, in some cases, although the constraints are violated, the overall implementation cost is relatively low and the impact is small, which is more in line with actual needs. In order to guide the optimization process, determine a solution that is more in line with the actual situation, optimize warehouse network planning decisions and reduce implementation costs while avoiding violations of constraints as much as possible, in some embodiments, it is also necessary to evaluate the penalty cost of warehouse network planning decisions.

[0110] Optionally, the evaluation function includes a penalty cost evaluation function, and the implementation cost includes penalty costs.

[0111] Among them, the penalty cost is linearly related to the degree of violation of the constraints in the candidate warehouse network planning. In this case, the penalty cost evaluation function can be constructed based on the unit penalty cost.

[0112] An evaluation function is used to evaluate candidate warehouse network planning decisions based on warehouse information, factory information, customer demand information, and warehouse network transportation information. When obtaining the implementation cost of candidate warehouse network planning decisions, the degree to which the candidate warehouse network planning decisions violate the constraints is first determined, and then the penalty cost is determined based on the penalty cost evaluation function and the degree to which the candidate warehouse network planning decisions violate the constraints.

[0113] Specifically, based on warehouse information, factory information, customer demand information, and warehouse opening decisions in candidate warehouse network planning, the degree of violation of constraints in candidate warehouse network planning decisions is determined, and a penalty cost evaluation function is used to determine the penalty cost based on the degree of violation of constraints in candidate warehouse network planning decisions and the unit penalty cost.

[0114] For example, taking Table 2 above as an example, by comparing the transportation time from the warehouse to the customer with the timeliness requirements in the constraints, it can be determined that the timeliness requirement for customer C4 is 24 hours, but the transportation time is 40 hours; the timeliness requirement for customer C7 is 24 hours, but the transportation time is 36 hours; and the timeliness requirement for customer C10 is 24 hours, but the transportation time is 30 hours. That is, the transportation time for customers C4, C7, and C10 exceeds the timeliness requirements. If the unit penalty cost for violating the timeliness requirement in the penalty cost evaluation function is 100 yuan / hour, then the timeliness violation (i.e., the violation of the timeliness requirement / penalty cost) is (40-24+36-24+30-24)×100=3400 yuan.

[0115] For example, taking Table 2 above as an example, if the number of activated warehouses configured by the system does not exceed 4, and the candidate warehouse network planning decision corresponding to Table 2 includes 5 activated warehouses. If the unit penalty cost for violating the number of warehouses in the penalty cost evaluation function is 10,000 yuan / warehouse, then at this time, the penalty cost for violating the number of warehouses (i.e., exceeding the number of warehouses) is (5-4)×10,000=10,000 yuan.

[0116] For example, the warehouse has capacity limitations in dimensions such as items / weight / volume, and these capacity limitations are also included in the constraints. Taking the warehouse having a weight limit as an example, in Table 2 above, comparing the warehouse weight limit with the customer's demand (i.e., the demanded weight), it can be determined that the capacity limit of warehouse W3, 15,000 kg, is less than the customer's demand of 16,000 kg, meaning that warehouse W3 has exceeded its capacity limit. If the unit penalty cost for violating the warehouse capacity limit in the penalty cost evaluation function is 10 yuan / kg, then the warehouse capacity violation (i.e., the penalty cost for exceeding the warehouse capacity) is (16,000 - 15,000) × 10 = 10,000 yuan.

[0117] In this embodiment, the degree to which candidate warehouse network planning decisions violate constraints is evaluated based on the unit penalty cost and the penalty cost evaluation function, which yields a more accurate penalty cost. Based on this penalty cost, a more accurate implementation cost is determined, which facilitates better iterative optimization of warehouse network planning decisions and obtains a reasonable target warehouse network planning decision that meets actual needs.

[0118] In warehouse network planning decisions, in addition to the penalty costs incurred due to violations of demand during actual product transportation, there are also optimization costs that meet the demand. In some embodiments, the evaluation function also includes an optimization cost function, and the implementation cost also includes the optimization cost. Therefore, by using the evaluation function, based on warehouse information, factory information, customer demand information, and warehouse network transportation information, candidate warehouse network planning decisions are evaluated. When obtaining the implementation cost of the candidate warehouse network planning decisions, it is also necessary to determine the optimization cost of the candidate warehouse network planning decisions.

[0119] Specifically, first determine the construction cost of the warehouse and the transportation cost of each route in the warehouse network transportation path, and then determine the optimized cost based on the construction cost and the transportation cost of all transportation routes.

[0120] More specifically, the warehouse information also includes warehouse construction costs. Based on this information and the warehouse activation decisions in the candidate warehouse network planning decisions, the construction cost of each warehouse activated in that activation decision can be determined. Furthermore, based on the warehouse network transportation information, the transportation cost of each path in the service decisions of the warehouse network transportation routes in the candidate warehouse network planning decisions can be determined. Then, based on the construction costs of all activated warehouses in the warehouse activation decisions and the transportation costs of all paths in the service decisions of the warehouse network transportation routes, the optimized cost of the candidate warehouse network planning decision can be determined.

[0121] Among them, the optimized cost is the sum of the construction cost of all activated warehouses in the activation decision and the transportation cost of all paths in the service decision.

[0122] For example, the capacity (kg) and construction cost (or commissioning cost) (ten thousand yuan) of warehouses W1-W10 given in Table 2 above are shown in Table 3 below. The capacities of warehouses W1-W10 are 30,000, 40,000, 15,000, 30,000, 30,000, 40,000, 30,000, 40,000, 40,000, and 10,000 kg, respectively, and the construction costs of warehouses W1-W10 are 100,000, 120,000, 80,000, 90,000, 90,000, 100,000, 130,000, 60,000, 110,000, and 100,000 yuan, respectively.

[0123] At this point, in the warehouse opening decisions shown in Table 2 above, warehouses W1, W3, W4, W8, and W10 were opened. The total construction cost of these five warehouses is (10+8+9+6+10=43) million yuan. Allocating these five warehouses to customers C1-C10 incurs a total transportation cost of (1+2+3+2+1.5+3.5+2.2+3.7+4+2=24.9) million yuan. Therefore, the optimized cost of the warehouse network planning decision shown in Table 2 can be determined to be (43+24.9)=67.9 million yuan.

[0124] Table 3

[0125]

[0126] In other words, implementation costs consist of penalty costs and optimization costs.

[0127] For example, the composition of implementation costs can be shown in Table 4 below. Implementation costs include penalty costs and optimization (direction) costs. Penalty costs include multiple cost items such as timeliness violations, warehouse quantity violations, and warehouse capacity violations, while optimization costs include multiple cost items such as warehouse construction costs and transportation costs. The values ​​for timeliness violations, warehouse quantity violations, and warehouse capacity violations in the penalty costs are 0.34, 1, and 1 (ten thousand yuan), respectively, while the values ​​for warehouse construction costs and transportation costs in the optimization costs are 43 and 24.9 (ten thousand yuan), respectively.

[0128] Table 4

[0129]

[0130] In this embodiment, an optimized cost evaluation function is used to determine the construction cost of each (activated) warehouse in the warehouse activation decision based on warehouse information and candidate warehouse network planning decisions. Based on warehouse network transportation information, the transportation cost of each path in the service decision of warehouse network transportation paths in candidate warehouse network planning decisions is determined. Then, based on the construction costs of all activated warehouses and the transportation costs of all paths determined above, the optimal cost of candidate warehouse network planning decisions is determined more accurately, and the inherent cost of warehouse construction and transportation in the accurate warehouse path, i.e., the optimized cost, is obtained, thereby more accurately determining the implementation cost of candidate warehouse network planning decisions.

[0131] For example, based on Tables 1 and 2 above, the solution search process is described using the neighborhood search operator as an example. The feasible codes shown in Table 1 are used as the codes for the current solution, i.e., the current warehouse network planning decision. By randomly closing warehouse W10 through neighborhood search, the feasible codes are obtained as shown in Table 5, i.e., the code corresponding to W10 is adjusted from 1 in Table 1 to 0.

[0132] Table 5

[0133] storehouse W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 coding 1 0 1 1 0 0 0 1 0 0

[0134] For example, based on the feasible codes shown in Table 5, the warehouses selected by each customer are re-determined, as shown in Table 6 below. Compared to Table 2, only the warehouse selected by customer C10 has changed, from warehouse W10 to warehouse W4. At this time, the transportation cost and transportation time also change accordingly, with the transportation cost becoming 24,000 yuan and the transportation time becoming 20 hours.

[0135] At this point, based on the candidate warehouse network planning decision shown in Table 6, the implementation cost is recalculated as shown in Table 7 below. Compared to Table 4, the value of timeliness violation in the penalty cost becomes 0.28 (ten thousand yuan), the value of warehouse quantity violation becomes 0, the warehouse construction cost in the optimization cost becomes 33 (ten thousand yuan), and the transportation cost becomes 253,000 yuan.

[0136] Based on the examples in Tables 1-2 and 4-7, it can be seen that by using the local search operator, i.e., the neighborhood search operator, to search for solutions to the current solution and obtain candidate solutions, the original timeliness violation can be reduced. Furthermore, after reducing the number of warehouses, the warehouse quantity violation becomes 0. At this point, the transportation cost increases slightly, but the warehouse construction cost decreases significantly. The reduction in both timeliness and warehouse quantity violations leads to a decrease in penalty costs. At this point, the candidate solution shows a significant improvement compared to the current solution represented in Table 1-2 before applying the search operator, indicating that the local search operator brings improvement. Therefore, this candidate solution will be retained as the new current solution and participate in subsequent algorithm execution for iterative optimization.

[0137] Table 6

[0138]

[0139] Table 7

[0140]

[0141] For example, the iterative solution process for the above warehouse network planning problem can be as follows: Figure 2 As shown. First, data is acquired, including warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information. Then, based on the acquired data, constraints and optimization directions are determined for the warehouse network planning problem, generating an initial code (or initial solution). This initial code is then decoded and evaluated to determine the implementation cost of the corresponding warehouse network planning decision. This initial code can be generated randomly or based on certain rules, such as sorting and selecting the warehouses to transport products to customers at different levels based on warehouse location and construction costs. Next, if no termination condition is triggered, the initial code is used as the current solution, and iterative optimization is performed. In each iteration, a search operator is invoked to generate multiple candidate codes, which are then decoded and evaluated sequentially. The currently optimal code is retained as the current code, which becomes the initial code for the next iteration. If a termination condition is triggered, for example, after a certain iteration, the algorithm stops running and outputs the current code (i.e., the currently optimal code) and the corresponding warehouse network planning decision (i.e., the target warehouse network planning decision mentioned above). Commonly used termination conditions include the maximum number of iterations, the maximum execution time, and the maximum number of iterations without improvement.

[0142] For example, a certain enterprise user has an average monthly order volume of 113,000 orders during peak season (billing weight is 506 tons of product weight). By implementing the above warehouse network planning method, the number of planned warehouses is increased from 10 to 12, and two new warehouses, A and B, are added. This improves the transportation timeliness of the regions where A and B are located. After adding warehouses in A and B, it is expected that the proportion of the transportation timeliness in the regions where A and B are located within 48 hours will increase by 7%, while the transportation cost will decrease by 2%.

[0143] For example, a corporate user, based on its business development and existing factory locations, requires fixed warehouses to be opened in 5 cities (AE) in its warehouse distribution plan. The estimated monthly product demand is 800,000 units, with a 48-hour delivery rate exceeding 90%. The analysis provides an optimal warehouse network planning scheme. In this case, according to the aforementioned warehouse network planning method, the system's algorithm calculates that at least 7 more warehouses need to be opened / built to meet the requirement of a 48-hour delivery rate exceeding 90%, and the total cost is optimal with 12 warehouses (warehouse construction cost + transportation cost + other related operating costs, i.e., penalty costs).

[0144] Exemplary device

[0145] like Figure 3 As shown in the figure, this application embodiment also provides a warehouse network planning device, including a determination module 301, a decoding module 302, and an iterative optimization module 303.

[0146] in,

[0147] The determination module 301 is used to determine the constraints and optimization direction of the warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information. The warehouse network configuration information is determined based on the needs of enterprise users who provide products to customers. The constraints include transportation timeliness constraints, warehouse capacity constraints, and warehouse quantity constraints.

[0148] Decoding module 302 is used to decode candidate solutions to obtain candidate warehouse network planning decisions. The candidate warehouse network planning decisions include warehouse opening decisions and warehouse network transportation path service decisions. The candidate solutions are the encodings of the candidate warehouse network planning decisions.

[0149] The iterative optimization module 303 is used to iteratively optimize the current warehouse network planning decision based on the constraints, the optimization direction and the candidate warehouse network planning decisions, until the termination condition is met, and then determine the iteratively optimized current warehouse network planning decision as the target warehouse network planning decision.

[0150] The warehouse network planning device provided in this embodiment belongs to the same concept as the warehouse network planning method provided in the above embodiments of this application. It can execute the method provided in any of the above embodiments of this application and has the corresponding functional modules and beneficial effects of the method. Technical details not described in detail in this embodiment can be found in the specific processing content of the warehouse network planning method provided in the above embodiments of this application, and will not be repeated here.

[0151] The functions implemented by the determination module 301, the decoding module 302, and the iterative optimization module 303 can be implemented by the same or different processors calling software, and this application embodiment does not limit this.

[0152] Exemplary electronic devices

[0153] Another embodiment of this application also provides an electronic device, see [link to relevant documentation] Figure 4 As shown, the electronic device includes a memory 400 and a processor 410.

[0154] The memory 400 is connected to the processor 410 and is used to store programs;

[0155] The processor 410 is used to implement the warehouse network planning method disclosed in any of the above embodiments by running the program stored in the memory 400.

[0156] Specifically, the electronic device may also include: a bus, a communication interface 420, an input device 430, and an output device 440.

[0157] The processor 410, memory 400, communication interface 420, input device 430, and output device 440 are interconnected via a bus. Among them:

[0158] A bus can include a pathway for transmitting information between various components of a computer system.

[0159] The processor 410 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0160] Processor 410 may include a main processor, as well as a baseband chip, modem, etc.

[0161] The memory 400 stores a program for executing the technical solution of this application, and may also store an operating system and other critical business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory 400 may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0162] Input device 430 may include a device for receiving user input data and information, such as a keyboard, mouse, camera, scanner, light pen, voice input device, touch screen, pedometer, or gravity sensor.

[0163] Output device 440 may include devices that allow information to be output to a user, such as a display screen, printer, speaker, etc.

[0164] The communication interface 420 may include a device that uses any transceiver to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0165] The processor 410 executes the program stored in the memory 400 and calls other devices, which can be used to implement each step of any of the warehouse network planning methods provided in the above embodiments of this application.

[0166] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0167] This application also proposes a chip, which includes a processor and a data interface. The processor reads and runs a program stored in a memory through the data interface to execute the warehouse network planning method described in any of the above embodiments. For details of the processing and its beneficial effects, please refer to the embodiments of the warehouse network planning method described above.

[0168] In addition to the methods and apparatus described above, embodiments of this application provide a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps of the warehouse network planning methods according to various embodiments of this application as described in the "Exemplary Methods" section of this specification.

[0169] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0170] Furthermore, embodiments of this application also propose a storage medium storing a computer program thereon, the computer program being executed by a processor of the steps in the warehouse network planning methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0171] The basic principles of the present invention have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in the present invention are merely examples and not limitations, and should not be considered as essential features of each embodiment of the present invention. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the present invention to the necessity of employing the aforementioned specific details.

[0172] The block diagrams of devices, apparatuses, devices, and systems involved in this invention are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0173] It should also be noted that in the apparatus, device, and method of the present invention, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of the present invention.

[0174] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the invention. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the invention. Therefore, the invention is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0175] It should be understood that the qualifying terms "first", "second", "third", "fourth", "fifth" and "sixth" used in the description of the embodiments of the present invention are only used to more clearly illustrate the technical solutions and are not intended to limit the scope of protection of the present invention.

[0176] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the invention to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A warehouse network planning method, characterized in that, The method includes: Based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information, the constraints and optimization directions of the warehouse network planning problem are determined. The warehouse network configuration information is determined based on the needs of enterprise users who provide products to customers. The constraints include transportation timeliness, warehouse capacity, and number of warehouses. Decode the candidate solutions to obtain candidate warehouse network planning decisions, which include warehouse opening decisions and warehouse network transportation route service decisions. The candidate solutions are the codes of the candidate warehouse network planning decisions. Based on the constraints, the optimization direction, and the candidate warehouse network planning decisions, the current warehouse network planning decision is iteratively optimized until the termination condition is met. Then, the iteratively optimized current warehouse network planning decision is determined as the target warehouse network planning decision.

2. The warehouse network planning method according to claim 1, characterized in that, Before determining the constraints and optimization direction of the warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information, the method further includes: Verify whether the warehouse network configuration information is missing any key information items, including warehouse configuration items and customer configuration items; If so, a prompt message is generated, which is used to remind that the key information item is missing; If not, then perform the operation described above to determine the constraints and optimization direction of the warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information.

3. The warehouse network planning method according to claim 1, characterized in that, Before decoding the candidate solutions to obtain the candidate warehouse network planning decision, the method further includes: Obtain at least one preset search operator and the preset usage frequency of each preset search operator; The candidate solution is obtained by searching the current solution based on at least one preset search operator and the preset usage frequency of each preset search operator.

4. The warehouse network planning method according to claim 1, characterized in that, Decoding the candidate solutions to obtain the candidate warehouse network planning decision includes: The candidate solutions are decoded to obtain the decision to open the warehouse; Based on the warehouse opening decision and the optimization direction, the service decision for the warehouse network transportation route is obtained by determining the level of warehouses through which the product will be transported to the customer.

5. The warehouse network planning method according to any one of claims 1-4, characterized in that, Based on the constraints, the optimization direction, and the candidate warehouse network planning decisions, the current warehouse network planning decision is iteratively optimized, including: An evaluation function is constructed based on the constraints and the optimization direction; Using the evaluation function, the candidate warehouse network planning decisions are evaluated based on the warehouse information, factory information, customer demand information, and warehouse network transportation information to obtain the implementation cost of the candidate warehouse network planning decisions; Based on the implementation cost of the candidate warehouse network planning decision, the implementation cost of the current warehouse network planning decision, and the optimization direction, the current warehouse network planning decision is iteratively optimized.

6. The warehouse network planning method according to claim 5, characterized in that, The evaluation function includes a penalty cost evaluation function, and the implementation cost includes a penalty cost. The evaluation function is used to evaluate the candidate warehouse network planning decisions based on the warehouse information, factory information, customer demand information, and warehouse network transportation information, to obtain the implementation cost of the candidate warehouse network planning decisions, including: Based on the warehouse information, the factory information, the customer demand information, and the warehouse opening decision in the candidate warehouse network planning decision, determine the degree to which the candidate warehouse network planning decision violates the constraints; The penalty cost is determined using the penalty cost evaluation function, based on the degree to which the candidate warehouse network planning decision violates the constraints and the unit penalty cost.

7. The warehouse network planning method according to claim 6, characterized in that, The evaluation function further includes an optimization cost function, and the implementation cost further includes optimization cost. The step of using the evaluation function to evaluate the candidate warehouse network planning decisions based on the warehouse information, factory information, customer demand information, and warehouse network transportation information to obtain the implementation cost of the candidate warehouse network planning decisions also includes: Based on the warehouse information and the candidate warehouse network planning decision, the warehouse opening decision is made, and the construction cost of each warehouse in the opening decision is determined. Based on the warehouse network transportation information, determine the transportation cost of each path in the service decision of the warehouse network transportation path in the candidate warehouse network planning decision; The optimized cost is determined based on the construction cost of all warehouses in the warehouse opening decision and the transportation cost of all paths in the warehouse network transportation path service decision.

8. A warehouse network planning device, characterized in that, The device includes: The determination module is used to determine the constraints and optimization direction of the warehouse network planning problem based on warehouse information, factory information, customer demand information, warehouse network transportation information, and warehouse network configuration information. The warehouse network configuration information is determined based on the needs of enterprise users who provide products to customers. The constraints include transportation timeliness constraints, warehouse capacity constraints, and warehouse quantity constraints. The decoding module is used to decode the candidate solutions to obtain candidate warehouse network planning decisions. The candidate warehouse network planning decisions include warehouse opening decisions and warehouse network transportation path service decisions. The candidate solutions are the encodings of the candidate warehouse network planning decisions. The iterative optimization module is used to iteratively optimize the current warehouse network planning decision based on the constraints, the optimization direction, and the candidate warehouse network planning decisions, until the termination condition is met, and then determine the iteratively optimized current warehouse network planning decision as the target warehouse network planning decision.

9. An electronic device, characterized in that, Including memory and processor; The memory is connected to the processor and is used to store programs; The processor is used to implement the warehouse network planning method as described in any one of claims 1 to 7 by running the program in the memory.

10. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the warehouse network planning method as described in any one of claims 1 to 7.

Images