A cable distribution method and device based on genetic algorithm

Abstract

The present application provides a cable distribution method and device based on a genetic algorithm, wherein the method includes: obtaining user order information and inventory data; the user order information includes the required lengths of various cable models and the distribution strategy of the user order; the inventory data includes the inventory lengths of each model and the number of cables corresponding to each inventory length; for each model, a preset number of gene sequences are created based on the distribution strategy, required length, and inventory data to form an initial gene population; the initial gene population is iterated using a genetic algorithm until the end condition is met to obtain a target gene population; the gene sequences with the highest fitness in the target gene population are used to determine the cable distribution results corresponding to each model. Therefore, the cable distribution method can adapt to various specific needs of the cable manufacturing industry and improve the work efficiency and accuracy of sales management personnel.

Description

Technical Field

[0001] The present application relates to the field of supply chain management and optimization technology, and in particular to a cable distribution method and device based on genetic algorithm. Background Art

[0002] With the development of industrialization and urbanization in modern society, the scale of the wire and cable industry has grown steadily. This trend has not only promoted the innovation of related technologies, but also put higher requirements on material management and order processing in the cable manufacturing industry.

[0003] In some current cable manufacturing industries, the majority of orders are under 1 kilometer in length. This characteristic often leads to material waste in traditional, manually controlled material ordering processes. Due to the large volume and diverse requirements of short-length orders, traditional material management and allocation methods struggle to precisely match the specific needs of each order. This results in inefficient use of excess material, resulting in wasted resources.

[0004] Furthermore, with the continued steady growth of the cable segment's sales business, sales personnel are facing increasing pressure during peak business periods. The growing order volume demands not only improved efficiency but also increased accuracy. However, traditional material management and order processing systems for the cable manufacturing industry typically rely on manual operations and traditional management methods. These methods often involve manual recording of order requirements, material allocation, and inventory management, which is not only inefficient but also prone to errors.

[0005] In this context, traditional management methods are no longer able to meet the modern cable manufacturing industry's demand for efficient and accurate material management and order processing. Therefore, it is urgent to introduce more advanced technologies and management strategies to optimize this process and improve overall work efficiency and accuracy. Summary of the Invention

[0006] The embodiments of the present application provide a cable distribution method and device based on a genetic algorithm, which can adapt to various specific needs of the cable manufacturing industry and improve the work efficiency and accuracy of sales management personnel.

[0007] In a first aspect, an embodiment of the present application provides a cable distribution method based on a genetic algorithm, the method comprising:

[0008] Obtain user order information and inventory data; user order information includes the required lengths of various cable models and the allocation strategy for user orders; inventory data includes the inventory lengths of each model and the number of cables of each inventory length;

[0009] For each model, a preset number of gene sequences are created based on the allocation strategy, demand length, and inventory data to form the initial gene population;

[0010] The initial gene population is iterated using a genetic algorithm until the end condition is met to obtain a target gene population; the population iteration includes using a fitness function to calculate the fitness of each gene sequence in the gene population, and performing genetic operator operations on the gene sequences in the gene population according to the fitness to obtain a new gene population; the gene sequences with the highest fitness in the target gene population are used to determine the cable distribution results corresponding to each model.

[0011] This allows us to automatically query matching inventory and calculate available quantities based on the requested model, requested length, and allocation strategy specified in the user's order. On this basis, we further apply a genetic algorithm to find the optimal allocation solution.

[0012] In a second aspect, an embodiment of the present application provides a cable distribution device based on a genetic algorithm, the device comprising:

[0013] The acquisition module is used to obtain user order information and inventory data; user order information includes the required lengths of various cable models and the allocation strategy of the user order; inventory data includes the inventory lengths of each model and the number of cables of each inventory length;

[0014] A processing module is used to create a preset number of gene sequences for each model according to the allocation strategy, demand length and inventory data to form an initial gene population;

[0015] The processing module is also used to use a genetic algorithm to iterate the initial gene population until the end condition is met to obtain a target gene population; the population iteration includes using a fitness function to calculate the fitness of each gene sequence in the gene population, and performing genetic operator operations on the gene sequences in the gene population according to the fitness to obtain a new gene population; the gene sequences with the highest fitness in the target gene population are used to determine the cable distribution results corresponding to each model.

[0016] In a third aspect, an embodiment of the present application provides a computer storage medium on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute the method described in the first aspect or any possible implementation of the first aspect.

[0017] In a fourth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the method described in the first aspect or any possible implementation of the first aspect.

[0018] It can be understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A flowchart of an embodiment of the present application using a genetic algorithm to find an optimal distribution plan;

[0021] Figure 2 A flow chart of a cable distribution method based on a genetic algorithm provided in an embodiment of the present application;

[0022] Figure 3 A flow chart of another cable distribution method based on a genetic algorithm provided in an embodiment of the present application;

[0023] Figure 4 A schematic diagram of a cable distribution device based on a genetic algorithm provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings.

[0025] In the description of the embodiments of this application, any embodiment or design scheme using "exemplary," "for example," or "for example" should not be understood as being more preferred or advantageous than other embodiments or designs. Rather, the use of words such as "exemplary," "for example," or "for example" is intended to present the relevant concepts in a concrete manner.

[0026] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly identifying the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. The terms "include," "comprising," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized. "Multiple" can mean one or more, where "multiple" means two or more.

[0027] To address the issues mentioned in the background technology, the present application proposes a solution that automatically queries matching inventory and calculates available quantities based on the required model, length, and allocation strategy specified in the user's order. Furthermore, a genetic algorithm is applied to find the optimal allocation plan. This method automates material billing, reducing the workload of sales personnel. Furthermore, the allocation process for inventory cables can be optimized for specific needs, improving inventory utilization and reducing resource waste.

[0028] For example, Figure 1 A flowchart of using a genetic algorithm to find an optimal distribution plan provided in an embodiment of the present application is shown in FIG.

[0029] Genetic algorithm is a search heuristic algorithm that simulates natural selection and genetic mechanism. In this solution, it is used to solve the cable distribution optimization problem. Figure 1 As shown in Figure 2, the process of using genetic algorithm to find the optimal distribution plan includes the following steps:

[0030] Step S101: Establishing a set of actual problem parameters.

[0031] It is necessary to transform the actual problem of cable distribution into a set of parameters to obtain the actual problem parameter set.

[0032] Step S102: Convert to genetic code.

[0033] These parameters are converted into genetically encoded form so that they can be processed by the genetic algorithm.

[0034] Step S103: establishing an initialization population T.

[0035] Initialize a population T, where each individual in the population represents a possible solution.

[0036] Step S104: Calculate individual fitness.

[0037] Evaluate the fitness of each individual in the population. The fitness function is defined according to the specific requirements of the actual problem and can indicate the quality of the solution.

[0038] Step S105: genetic operator simulation.

[0039] Genetic operators (including selection, crossover, and mutation) are used to simulate the natural selection process and generate new populations. The functions of these operators are:

[0040] Selection operator: select individuals with higher fitness for reproduction.

[0041] Crossover operator: combines the genes of two individuals to produce a new individual.

[0042] Mutation operator: randomly changes the genes of certain individuals to increase the diversity of the population.

[0043] Step S106: Generate a new population T+1.

[0044] Step S107: Check whether the population iteration requirement is met.

[0045] Check whether the new population meets the preset conditions (such as reaching the maximum number of iterations, finding a satisfactory solution, etc.).

[0046] If the requirements are met, proceed to the next step, entering step S108.

[0047] If the requirements are not met, the new population is used as the current population, and the process returns to step S104 to continue iteration.

[0048] Step S108: performing gene decoding.

[0049] Decode the genes of the top-ranked individuals in the final population and convert them back into solutions to practical problems.

[0050] Step S109: Obtain an optimal solution.

[0051] The optimal individual is regarded as the optimal solution to the practical problem.

[0052] thus, Figure 1 The whole process of genetic algorithm from practical problem definition to solution is demonstrated, including gradually approaching the optimal solution of the problem through iterative optimization.

[0053] For example, Figure 2 A flow chart of a cable distribution method based on a genetic algorithm provided in an embodiment of the present application is shown.

[0054] like Figure 2 As shown, the cable distribution method is mainly implemented by the following modules: distribution management module 20, order module 21, inventory management module 22, data management module 23, algorithm module 24, and security encryption module 25. Each module interacts with each other to complete the cable distribution process based on the genetic algorithm.

[0055] Step S201: Send user order information.

[0056] For example, the distribution management module 20 provides an order input interface through which sales personnel can enter user order information. When completing this form, they must clearly specify the desired cable model, required length, and desired distribution strategy. This information directly influences the selection of the subsequent genetic algorithm and the application of the fitness function. After obtaining the user order information, the distribution management module 20 sends it to the order module 21.

[0057] In one implementation, the distribution management module 20 can be Microsoft Dynamics 365, an integrated business application platform that integrates customer relationship management and enterprise resource planning functions. It can provide a comprehensive solution for the entire cable sales business. For example, it can manage business processes in sales, customer service, marketing, finance, supply chain, human resources, and other areas.

[0058] Step S202: Request to search for a corresponding cable.

[0059] For example, the order module 21 groups the required cable models and required lengths, and fills the requirements of the same model into an order request message for sequential processing. Each order request message is sent to the inventory management module 22 to request to find the corresponding cable.

[0060] Step S203: Return the cable search result.

[0061] Exemplarily, the inventory management module 22 provides a web application for receiving order request information. Upon receiving this information, the inventory management module 22 first searches the system's available inventory table for inventory models that match the order request. It then statistically classifies all eligible inventory items by length, determines the "available number" of cables of each length specification, i.e., the number of cables of a specific length specification currently in inventory, and returns the statistical classification results as cable search results to the order module 21.

[0062] In addition, the inventory management module 22 is further configured to update the status (eg, cable inventory length) and quantity of the cable inventory after receiving the cable distribution result in step S212 .

[0063] Step S204: request the public key.

[0064] Step S205, returning the public key.

[0065] Step S206: Send the encrypted data.

[0066] Exemplarily, after receiving the cable search result, the order module 21 organizes the user order information and the cable search result into a structured data format.

[0067] Subsequently, for these structured data, the order module 21 requests the public key from the security encryption module 25 in real time, and receives the public key returned by the security encryption module 25.

[0068] Next, the order module 21 uses the public key to encrypt the structured data, obtains the ciphertext data, and sends it to the algorithm module 24.

[0069] In one implementation, during the inventory allocation calculation process, order module 21 homomorphically encrypts sensitive user order and inventory data before sending it to algorithm module 24. This unique encryption method allows algorithm module 24 to perform calculations directly on the encrypted data without first decrypting it, thereby ensuring data security and privacy even in third-party computing environments. Given that homomorphic encryption may increase data volume, order module 21 can implement an efficient compression algorithm to reduce the size of the encrypted data, thereby optimizing transmission efficiency.

[0070] During this process, the security encryption module 25 is used to ensure the security of data during transmission, storage, and calculation. By encrypting data before transmission and processing, the data remains encrypted during transmission and when the algorithm module 24 performs optimization calculations, thereby preventing the leakage of sensitive information. Even during the calculation process, the data exists in ciphertext form, ensuring the privacy and security of the data. At the same time, this module is also responsible for decrypting the results after the calculation is completed to obtain a usable plaintext output, which can protect data security without affecting the use of the final results.

[0071] Specifically, after receiving the public key request information, the security encryption module 25 uses a random number generation algorithm to generate a public key and private key pair for homomorphic encryption in real time, and returns the public key to the order module 21 for encryption processing of user order data and inventory data.

[0072] Furthermore, the security encryption module 25 receives a private key request from the distribution management module 20 in step S209, and returns the private key corresponding to the public key carried in the processed user order message to the distribution management module 20 for decryption of the calculation result. To enhance security, the security encryption key is updated and proactively distributed to relevant modules.

[0073] Optionally, the security encryption module 25 can also implement fine-grained access control policies to ensure that users with different roles can only access resources within their scope of authority. For example, ordinary users can only submit tasks, while administrators can monitor the operating status of the entire system and adjust the configuration of computing models.

[0074] Step S207, returning the calculation result.

[0075] For example, algorithm module 24 is used to perform calculations on encrypted data. It is the core functional module of the system and calculates and optimizes the cutting plan through data modeling. This module uses a genetic algorithm for optimization calculations, including three basic operators: selection, crossover, and mutation, to find a global or near-optimal solution. Genetic algorithms simulate natural selection and heredity mechanisms, iteratively generating new populations. Individuals are selected for reproduction and mutation based on fitness, gradually approaching the optimal solution.

[0076] Specifically, after receiving the encrypted data, the algorithm module 24 first creates an initial genetic population of possible solutions (i.e., cutting solutions for each available stock) based on a pre-set population size. This process involves generating a certain number of initial solution sets or "genetic populations" for each possible solution length.

[0077] Next, the genetic algorithm is executed, which involves:

[0078] (1) Fitness evaluation: The fitness value of each individual is calculated using the defined fitness function to measure the quality of its solution.

[0079] (2) Selection mechanism: The tournament selection method is used to select individuals with the highest fitness scores as outstanding representatives in the next generation population.

[0080] (3) Crossover operation: Perform a crossover operation on the two selected individuals to exchange genetic information between the selected positions and produce new offspring individuals.

[0081] (4) Mutation operation: In order to increase population diversity, a few gene fragments are randomly selected for modification or recombination to create new feature combinations that are different from the parent generation.

[0082] After a preset number of iterations, when a predetermined condition is met, the algorithm stops and returns the top individuals in the current population. These individuals are then decoded and converted into structured data as calculation results. This data is then sent to the algorithm module 24, which provides a specific solution for the cutting situation on each stock.

[0083] Step S208: forward the calculation result.

[0084] Exemplarily, the calculation result is forwarded by the algorithm module 24 to the distribution management module 20 .

[0085] Step S209: Request private key.

[0086] Step S210, return the private key.

[0087] Exemplarily, after receiving the calculation result, the distribution management module 20 sends a request to the security encryption module 25 for the calculation result to obtain the private key corresponding to the public key carried by the user order message processed this time, and receives the private key returned by the security encryption module 25.

[0088] Subsequently, the distribution management module 20 decrypts the calculation result using the private key to obtain the decrypted calculation result, which is the optimal solution for cable distribution for the user's order.

[0089] Next, in order to meet the special requirements put forward by the sales management department, such as avoiding the impact of specific excess material lengths on sales or supporting alternative materials for distribution, the distribution management module 20 will perform additional optimization and processing on the preferred solution to obtain the final cable distribution result for the user's order, so as to better serve the actual needs of the sales business.

[0090] Step S211: Send the decrypted calculation result and the cable distribution result.

[0091] Step S212: forwarding the cable distribution result.

[0092] Step S213: forward the user order information, the decrypted calculation result, and the cable distribution result.

[0093] For example, the order module 21 sends the cable allocation results sent by the allocation management module 20 to the inventory management module 22 for inventory update. It also forwards the user order information, decrypted calculation results, and cable allocation results to the data management module 23, which records the user order, calculation results, and cable allocation results for subsequent results review, problem troubleshooting, and genetic algorithm optimization.

[0094] Therefore, according to the demand model, demand length and distribution strategy specified in the user order, matching inventory can be automatically found and available quantity can be counted, and then the distribution plan with the highest or close to the highest material utilization rate can be calculated, thereby achieving efficient processing of sales orders and optimized distribution.

[0095] Based on the above content, another cable distribution method based on genetic algorithm proposed in this application is introduced in detail.

[0096] For example, Figure 3 A flowchart of another method for cable distribution based on a genetic algorithm provided in an embodiment of the present application is shown. This method can be implemented by any computing unit, server, device, device cluster, etc. with computing and processing capabilities, and mainly includes the following execution steps:

[0097] Step S301: Obtain user order information and inventory data. The user order information includes the required lengths of various cable models and the order allocation strategy. The inventory data includes the inventory lengths for each model and the number of cables in stock for each length.

[0098] In one embodiment, first, a user can place an order for different types of cables simultaneously in one user order through the distribution management module 20. Second, sales management personnel can select the distribution strategy corresponding to the user order through the distribution management module 20. For example, the minimum margin strategy aims to reduce waste generated by cutting cables, or the safety stock strategy aims to maintain a certain amount of safety stock to cope with demand fluctuations and supply chain uncertainties. These strategies can be used individually or in combination to adapt to cost, time, resource utilization or other business needs. By implementing an effective cable distribution strategy, companies can improve customer satisfaction, reduce costs, and improve overall operational efficiency.

[0099] Subsequently, the distribution management module 20 is used to group the cable requirements of various models in the user order according to the model, and the required lengths of multiple cables under each model are obtained, including the distribution strategy of the user order.

[0100] Next, use Figure 1 The inventory management module 22 shown searches for cables of various models according to their requirements, and obtains a plurality of different inventory lengths for each model, as well as the number of cables corresponding to each inventory length.

[0101] Further, using Figure 1 The order module shown obtains user order information from the distribution management module 20 and obtains inventory data from the inventory management module 22.

[0102] In the following content, the requirements of the same model will be used Figure 1 The genetic algorithm shown is processed in sequence to obtain the cable distribution results corresponding to each model.

[0103] Step S302: For each model, a preset number of gene sequences are created according to the distribution strategy, demand length and classification data to form an initial gene population.

[0104] For example, when applying a genetic algorithm to a cable allocation scenario, each allocation strategy has a corresponding pricing strategy. This pricing strategy determines the cost-effectiveness of different allocation options. The pricing strategy includes a pricing model that sets prices based on factors such as cable model, length, and material. This model can be linear or nonlinear, depending on the cost structure and market conditions.

[0105] In the genetic algorithm, pricing strategies are implemented through a price dictionary. This price dictionary, a key data structure, contains pricing information for each cable model. During the algorithm's iterations, this pricing information is used to estimate the costs of different stocking options, guiding the algorithm to find the lowest-cost stocking solution. In this way, pricing strategies directly influence the genetic algorithm's optimization direction and the final stocking decision.

[0106] After obtaining the user order information and inventory data, first, based on the price strategy corresponding to the allocation strategy and multiple different inventory lengths, the price corresponding to each inventory length can be obtained to form a price dictionary.

[0107] In one example, a linear pricing model is used for the pricing strategy of a specific cable model. This model sets the price using a linear ratio that reflects the relationship between cable length and price. Cable inventory management may include multiple lengths of this cable model. To reflect the importance of length in the pricing strategy, a linear ratio can be used to adjust the price of these different cable lengths.

[0108] Specifically, this process might involve adjusting the price of a cable using a linear ratio, scaling it up or down based on its actual length. If the cable is longer, the price will increase proportionally; if it is shorter, the price will decrease proportionally. In this way, a linear pricing model ensures fair pricing for cables of varying lengths.

[0109] For example, if the allocation strategy for a user's order is the minimum margin strategy and the corresponding pricing strategy includes a linear ratio of 1:10, then the price dictionary PRICE can be a simple key-value pair set, where the key is the different cable lengths and the value is the corresponding price. The following is an example of a simple price dictionary:

[0110] PRICE = {8: 0.8, 19: 1.9, 20: 2, 22: 2.2, 49: 4.9, 89: 8.9, 180: 18,580: 58, 780: 78, 940: 94, 1940: 194,2330: 233, 2580: 258, 3290: 329}

[0111] As shown in the price dictionary above, each length of stock cable has its own specific price. For example, for a cable with a length of 8, the price after shrinking using a linear ratio of 1:10 is 0.8.

[0112] Next, the price dictionary, demand length, and inventory data are structured to generate a problem dataset, which serves as input for the genetic algorithm model. The problem dataset includes a demand length dataset, an inventory length dataset, a price dataset consisting of pairs of inventory lengths and their corresponding prices, and a number of inventory items dataset consisting of pairs of inventory lengths and their corresponding number of items.

[0113] Continuing with the above example, we will create a problem dataset in the form of an array data structure.

[0114] For example, create a length dataset:

[0115] PIECE = [70, 15, 214, 88, 95, 111, 200, 34, 39, 64, 30, 180, 79, 133,254, 28, 37, 489],

[0116] The length dataset includes all required cable lengths for this model.

[0117] Create a stock length dataset:

[0118] STOCK= (8, 19, 20, 22, 49, 89, 180, 180, 580, 780, 940, 1940, 2330,2580, 3290),

[0119] The stock length data set includes the lengths of all stock cables that meet the model requirements.

[0120] The price dictionary PRICE in the above example is directly used as the price dataset.

[0121] Create a data set of numbers:

[0122] QUANTITY = {8: 1, 19: 1, 20: 1, 22: 1, 49: 1, 89: 1, 180: 2, 580: 1,780: 1, 940: 1, 1940: 1, 2330: 1, 2580: 1,3290: 1},

[0123] The Count dataset contains the "Available Count" of each length of stock cable in inventory.

[0124] Next, a greedy algorithm is used to perform a preset number of rounds (e.g., T rounds) of initial cable allocation based on the problem dataset, generating T gene sequences that constitute the initial gene population T. Each gene sequence consists of multiple vectors, corresponding to the results of an initial cable allocation. Each vector represents a set of at least one required length from the required length dataset, representing the allocation of at least one required length based on a single cable in stock. In a greedy algorithm, each selection step takes the best or optimal (i.e., most favorable) choice under the current state, hoping to achieve the best or optimal algorithmic strategy. In other words, the locally optimal solution can determine the globally optimal solution. Greedy algorithms involve a state space and a selection strategy. The state space describes the set of all possible states of the problem, while the selection strategy determines the greedy selection criteria for each step.

[0125] For each round of initial cable allocation, the problem dataset for that round is used as the state space of a greedy algorithm, and a benefit function is set as the greedy algorithm's selection strategy. The benefit function is based on the allocation strategy and must meet the cost, time, resource utilization, or other business requirements of the allocation strategy. Based on this, the greedy algorithm's benefit function is used to evaluate the problem dataset for that round, determining multiple vectors of gene sequences corresponding to that round as the results of each round of initial cable allocation. The order of the multiple demand lengths in the demand length dataset is randomly adjusted to obtain the problem dataset for the next round of initial cable allocation.

[0126] When the benefit function of the greedy algorithm is used to evaluate the problem data set of this round, the problem data set of this round is evaluated and iterated multiple times until multiple demand lengths in the demand length data set are allocated.

[0127] For each evaluation iteration, the inventory length dataset is traversed based on the item count dataset for the unassigned demand lengths in the demand length dataset. The first optimal inventory length corresponding to the maximum benefit is selected from the inventory length dataset using the benefit function. This also generates the first vector of the gene sequence for the current round. The number of items corresponding to the first optimal inventory length is non-zero. The number of items corresponding to the first optimal inventory length in the item count dataset is decremented by 1. The demand lengths included in the first vector are marked as assigned in the demand length dataset for the next evaluation iteration, until all demand lengths in the demand length dataset have been assigned.

[0128] Assuming that the allocation strategy is the minimum margin strategy, the benefit E of the first vector can be calculated using the following benefit function:

[0129] E=(∑Length i )÷Price1 (1)

[0130] Among them, Length iis the i-th required length included in the first vector, and Price1 is the price corresponding to the first optimal inventory length. The optimal inventory length is selected based on the benefit value E. A higher benefit value E indicates that more required lengths can be shipped per unit price, meaning that a single in-stock cable can accommodate more required lengths. Therefore, the selected inventory length is more optimal.

[0131] Continuing the above example, for the initial cable distribution in round m, we perform the jth evaluation iteration.

[0132] Step 1: Determine the unallocated demand length list in the demand length dataset: [70, 15, 214, ...],

[0133] Traverse all inventory lengths with non-zero inventory items, calculate the benefit value E using formula (1), and find the solution with the highest benefit E: select 180m of inventory (priced at 18 yuan) and load it with the required length of 180m (the index value in the required length dataset is 11, the inventory cables are just full, the cable margin is 0, and E is 10).

[0134] Add the index value of the required length in the PIECE array (integer ≥ 0) to the same vector to obtain the first vector:

[11] , which is added to the gene sequence obtained in the last evaluation (it can be understood that the gene sequence is initialized to an empty set when the first evaluation is performed).

[0135] Get the current gene sequence for the jth evaluation: […,

[11] ] (assuming the 11th in the PIECE array is 180m),

[0136] Step 2: Subtract 1 from the number of items in QUANTITY that have an inventory length of 180m.

[0137] Step 3: Mark the demand length ranked 11 in the demand length dataset as allocated.

[0138] Continue with the (j+1)th evaluation iteration,

[0139] Step 1: Determine the list of unallocated demand lengths in the demand length dataset: [70, 15, 214, ...] (excluding the 11th demand length compared to round j).

[0140] Traverse all inventory lengths with non-zero inventory items again and find the solution with the highest benefit E: select 580m of inventory (priced at 58 yuan) to meet the demand for more, such as 214+200+... (the total length does not exceed 580m).

[0141] Get the first vector: [2,6,...], add it to the gene sequence obtained in the last evaluation,

[0142] Get the (j+1)th evaluation of the current gene sequence: […,

[11] , [2,6,...]],

[0143] Step 2: ...,

[0144] Step 3: ...,

[0145] This process is repeated until all the required lengths in the required length dataset are allocated. The result of the initial cable allocation in the mth round is obtained, that is, the mth gene sequence in the initial population T. This gene sequence is the gene sequence obtained in the last evaluation iteration during the mth round of multiple evaluation iterations. In the above example, it can be represented by a vector array as follows: [

[0147]

[11] , # Group 1: Required length 11 (length 180)

[0148] [2, 6], # Group 2: Required length 2 (length 214) and 6 (length 200) → Total length 414 → Possible inventory length of 580 meters

[0149] [0, 1, 7, 8], # Group 3: Required lengths 0 (length 70), 1 (length 15), 7 (length 34), and 8 (length 39) → Total length 158 → Possible stock length of 180 meters ....]

[0151] The vector array above is the mth gene sequence. The process of dividing the required length into different vectors is the process of converting the problem data set (i.e. the input parameter of the genetic algorithm model) into a gene sequence.

[0152] To enhance the randomness of the genetic algorithm and prevent premature convergence to a local optimum, after completing the mth round of initial cable allocation, the order of the multiple required lengths in the length dataset PIECE is randomly adjusted (while maintaining the index value corresponding to each required length unchanged), thereby generating a new sequence of required lengths. This new sequence serves as the length dataset PIECE' for the next allocation round, guiding the initial cable allocation for the next round. It is understood that for the first round of initial cable allocation, the original order of required lengths in the length dataset PIECE is used directly.

[0153] As a result, the genetic algorithm can consider the required lengths in a different order in each iteration, which helps the algorithm to escape from the local optimum and improve the possibility of finding the global optimal solution.

[0154] For example, before the problem dataset consisting of the price dictionary, demand length, and inventory data is sent to algorithm module 24 for processing, it is encrypted using the public key provided by security encryption module 25 to obtain encrypted data. Furthermore, algorithm module 24 creates a preset number of gene sequences based on the encrypted data to form an initial gene population.

[0155] This ensures that data security and privacy are protected even in third-party computing environments.

[0156] In step S303, a genetic algorithm is used to iterate the initial gene population until the termination condition is met, thereby obtaining a target gene population. This population iteration involves calculating the fitness of each gene sequence in the gene population using a fitness function. Based on the fitness, genetic operators are then applied to the gene sequences in the gene population to obtain a new gene population. Gene sequences with the highest fitness in the target gene population are used to determine the cable allocation results for each model.

[0157] For example, in a genetic algorithm, a fitness function is used to score each individual in the population. The higher the score, the more adapted the individual is to the environment, that is, the better the solution is.

[0158] The initial gene population T is obtained using a greedy algorithm, which doesn't always yield a globally optimal solution because it only considers the local optimum at each step, without considering the overall situation. Therefore, when using the fitness function to calculate the fitness of each gene sequence in the initial gene population T, the following operations are performed on each gene sequence and each vector it contains, in order to re-determine the second-best inventory for each demand length included in each vector from the inventory length dataset.

[0159] For the current vector, the operation includes: traversing the inventory length dataset based on the number of items dataset, and using the benefit function of the greedy algorithm (such as formula (1)) to select the second optimal inventory length corresponding to the maximum benefit from the target data subset of the inventory length dataset. It can be understood that the number of items corresponding to the second optimal inventory length is not zero. After determining the second optimal inventory length of the current vector, the number of items corresponding to the second optimal inventory length in the number of items dataset is subtracted by 1 to determine the second optimal inventory of the next vector. The inventory length in the target data subset conforms to the global strategy, which is determined based on experience and represents the distribution strategy that the cable segment sales business usually needs to comply with. For example, the inventory length minus the demand length is non-negative, etc.

[0160] Finally, the prices corresponding to the second optimal inventory length of each vector are summed to obtain the fitness of the gene sequence.

[0161] Among them, the fitness function F is expressed by formula (2):

[0162] F=∑Price 2(j) (2)

[0163] Price 2(j) is the price corresponding to the second optimal inventory length of the j-th vector in the gene sequence.

[0164] After performing the above operations, the fitness of each gene sequence in the initial gene population T can be obtained.

[0165] Next, enter Figure 1 The genetic operator simulation stages are shown.

[0166] For example, three basic operators of a genetic algorithm are used for iterative population optimization: a selection operator (e.g., random selection) retains individuals with high fitness, a crossover operator (e.g., single-point or multi-point crossover), and a mutation operator (e.g., random perturbation) introduce diversity, generating a new population (population T+1). This iterative process continues until a termination condition is met (e.g., reaching a maximum number of iterations or fitness convergence).

[0167] First, in the selection phase, a selection operator is used to randomly select at least two gene sequences from the gene population, for example, 20 from 50 gene sequences. Fitness is used to determine which individuals will be selected for reproduction. Generally, individuals with higher fitness have a higher probability of being selected. Therefore, the first and second gene sequences with the highest and second highest fitness are then selected from the at least two gene sequences.

[0168] Secondly, in the crossover stage, the crossover operator is used to randomly select crossover points to re-splice partial vectors of the first gene sequence and the second gene sequence into offspring, that is, a new gene sequence (the first child gene sequence).

[0169] For example, the intersection of parent 1: intercept item_chromosome1[0:2] → [

[11] , [2,6]],

[0170] Intersection of parent 2: intercept item_chromosome2[1:3] → [[2,5], [0,7]],

[0171] The child inherits the first half of parent 1 + the second half of parent 2, so the child is [

[11] ,[0,7]].

[0172] The first child gene sequence replaces the vector represented by item_chromosome1[0:2] in parent 1 to obtain the third gene sequence.

[0173] Finally, in the mutation phase, a random determination is made as to whether to perform the mutation operation on the third gene sequence. If so, the mutation operation is used to obtain the fourth gene sequence, which is then added to the gene population to form a new gene population. The mutation operation is used to simulate the genetic mutation process in nature.

[0174] The random judgment process includes generating a random number and performing a mutation operation if the random number is greater than the preset mutation rate. Since the mutation operation is performed randomly, it can increase the exploration ability of the genetic algorithm.

[0175] If not executed, the third gene sequence is added to the gene population to obtain a new gene population.

[0176] The mutation operation includes using a mutation operator to randomly select a portion of the vectors of the third gene sequence, disrupting them, and then reassembling them into a new gene sequence (the fourth gene sequence).

[0177] Specifically, a partial vector of the third gene sequence is randomly selected, and the required length component included in the partial vector of the third gene sequence is used to form the required length dataset. The partial required length dataset is used to replace the required length dataset to obtain a partial problem dataset. An initial cable allocation is performed based on the partial problem dataset using a greedy algorithm to obtain a second sub-gene sequence. The second sub-gene sequence is used to replace the partial vector of the third gene sequence to obtain a fourth gene sequence.

[0178] For example, if the third gene sequence is [

[11] , [2,3,4], [0,7,8], [5,6]], randomly select two vectors [2,3,4] and [5,6], decompose the two vectors, and perform the initial cable matching again according to the method in step S302 to obtain the possible variant sequence [

[11] , [2], [5,6,3], [4], [0,7,8]], which is the second sub-gene sequence. The second sub-gene sequence is used to replace the two vectors [2,3,4] and [5,6] in the third gene sequence to obtain the fourth gene sequence.

[0179] The fourth gene sequence is added to the initial gene population T to obtain a new gene population T+1.

[0180] Thus, the genetic algorithm continuously optimizes the distribution plan by simulating the iterative process of natural selection, including operations such as selection, crossover, and mutation, gradually approaching the goal of cost optimization. This method not only helps reduce distribution costs but also improves inventory management efficiency, ensuring the maximum economic benefits of the cable distribution process.

[0181] In this way, the genetic algorithm is used to iterate the initial gene population until the termination condition is met, resulting in a target gene population. The multiple vectors included in the gene sequences with the highest fitness in the target gene population and the second-best inventory corresponding to each vector are used as the multiple optimal cable allocation results for each model.

[0182] Exemplarily, if the genetic algorithm creates a preset number of gene sequences based on encrypted data, the gene sequences with the highest fitness in the target gene population are also decrypted, and the decrypted data is used to determine multiple preferred cable distribution results corresponding to each model.

[0183] Optionally, based on special requirements raised by the sales management department, such as avoiding the impact of specific excess material lengths on sales or supporting alternative materials for distribution, the optimal solution is further optimized and processed to obtain the final cable distribution result for the user's order.

[0184] For example, a cable distribution result can be displayed as:

[0185] running GA_stock_cutting...

[0186] stock_length_8: []

[0187] stock_length_19: [[15, 4]]

[0188] stock_length_20: []

[0189] stock_length_22: []

[0190] stock_length_49: []

[0191] stock_length_89: [[88, 1]]

[0192] stock_length_180: [[180, 0], [111, 64, 5]]

[0193] stock_length_580: [[214, 200, 133, 30, 3]]

[0194] stock_length_780: [[489, 254, 37, 0]]

[0195] stock_length_940: [[39, 901]]

[0196] stock_length_1940: [[95, 79, 70, 34, 28, 1634]]

[0197] stock_length_2330: []

[0198] stock_length_2580: []

[0199] stock_length_3290: []

[0200] stock_length: 8 left qty: 1

[0201] stock_length: 19 left qty: 0

[0202] stock_length: 20 left qty: 1

[0203] stock_length: 22 left qty: 1

[0204] stock_length: 49 left qty: 1

[0205] stock_length: 89 left qty: 0

[0206] stock_length: 180 left qty: 0

[0207] stock_length: 580 left qty: 0

[0208] stock_length: 780 left qty: 0

[0209] stock_length: 940 left qty: 0

[0210] stock_length: 1940 left qty: 0

[0211] stock_length: 2330 left qty: 1

[0212] stock_length: 2580 left qty: 1

[0213] stock_length: 3290 left qty: 1

[0214] cost: 470.8

[0215] ---GA runtime: 0.04809141159057617 seconds ---

[0216] The array in the form of stock_length_ represents the cutting situation of the inventory cables, such as stock_length_180:[[180, 0], [111, 64, 5]], which includes two vectors, indicating that two 180-meter-long inventory cables are allocated. The first 180-meter-long inventory cable is used in its entirety without cutting, and the second cable needs to be cut twice to obtain 111 meters and 64 meters, with 5 meters of surplus material remaining.

[0217] stock_length: 180 left qty: 0, indicating that the remaining quantity of 180-meter cables available for sale is 0.

[0218] cost is the fitness of this cable distribution result.

[0219] In summary, the embodiments of this application propose a solution that automatically queries matching inventory and calculates available quantities based on the requested model, length, and allocation strategy specified in the user's order. Furthermore, a genetic algorithm is applied to find the optimal allocation plan. This method automates material billing, reducing the workload of sales personnel. Furthermore, the allocation process for inventory cables can be optimized for specific needs, improving inventory utilization and reducing resource waste.

[0220] It is understandable that the size of the sequence number of each step in the above embodiment does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application. In addition, in some possible implementations, the steps in the above embodiment can be selectively executed according to actual conditions, and can be partially executed or fully executed, which is not limited here. In addition, all or part of any features in the above embodiment can be freely and arbitrarily combined without contradiction. The combined technical solution is also within the scope of this application.

[0221] For example, Figure 4 Schematic diagram of a cable distribution device based on genetic algorithm provided in an embodiment of the present application is shown in FIG.

[0222] like Figure 4 As shown, the cable distribution device 400 includes:

[0223] Acquisition module 410 is used to obtain user order information and inventory data; user orders may include requirements for multiple cable models, including the models and required lengths of multiple cables of the same model, and the corresponding allocation strategy for the user order; inventory data includes multiple different inventory lengths for each model and the corresponding number of cables of each inventory length;

[0224] Processing module 420 is used to create a preset number of gene sequences for each model according to the allocation strategy, demand length and inventory data to form an initial gene population;

[0225] Processing module 420 is also used to use a genetic algorithm to perform population iteration on the initial gene population until the end condition is met to obtain a target gene population; the population iteration includes using a fitness function to calculate the fitness of each gene sequence in the gene population, and performing genetic operator operations on the gene sequences in the gene population according to the fitness to obtain a new gene population; the gene sequences with the highest fitness in the target gene population are used to determine the cable distribution results corresponding to each model.

[0226] Based on the method in the above embodiment, an embodiment of the present application provides an electronic device. The electronic device may include: at least one memory for storing a program; and at least one processor for executing the program stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method described in the above embodiment. Exemplarily, the electronic device may be a mobile phone, a tablet computer, a desktop computer, a laptop computer, a handheld computer, a notebook computer, a server, an ultra-mobile personal computer (UMPC), a netbook, a cellular phone, a personal digital assistant (PDA), or an artificial intelligence (AI) device. The embodiment of the present application does not impose any special restrictions on the specific type of the electronic device.

[0227] The above embodiments can be implemented in whole or in part through software, hardware, firmware, or any combination thereof. When implemented using software, they can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, the computer program instructions fully or partially generate the processes or functions described in accordance with the embodiments of the present invention. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disk, hard disk, tape), optical media (e.g., DVD), or semiconductor media (e.g., solid-state drive (SSD)).

[0228] It is understood that the various numerical numbers involved in the embodiments of the present application are only for the convenience of description and are not intended to limit the scope of the embodiments of the present application. It should be understood that in the embodiments of the present application, the order of the sequence numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0229] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of this application in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of this application. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of this application should be included in the scope of protection of this application.

Claims

1. A cable distribution method based on genetic algorithm, characterized in that: The method comprises: Obtain user order information and inventory data; the user order information includes the required lengths of various cable models and the allocation strategy for the user order; the inventory data includes the inventory lengths of each model and the number of cables of each inventory length; For each model, a preset number of gene sequences are created based on the allocation strategy, the demand length, and the inventory data to form an initial gene population; the creating a preset number of gene sequences based on the allocation strategy, the demand length, and the inventory data includes: According to the price strategy corresponding to the allocation strategy, prices corresponding to each inventory length are obtained to form a price dictionary; The price dictionary, the demand length, and the inventory data are structured to obtain a problem data set; the problem data set includes a demand length data set, an inventory length data set, a price data set consisting of pairs of values consisting of each inventory length and its corresponding price, and a number data set consisting of pairs of values consisting of each inventory length and its corresponding number of items; Performing the preset number of rounds of initial cable allocation based on the problem dataset using a greedy algorithm to obtain the preset number of gene sequences; each gene sequence includes multiple vectors corresponding to a result of the initial cable allocation, each vector being a set of at least one required length in the required length dataset, the set indicating that the at least one required length is allocated based on a stocked cable; The initial gene population is iterated using a genetic algorithm until an end condition is met to obtain a target gene population; the population iteration includes calculating the fitness of each gene sequence in the gene population using a fitness function, and performing genetic operator operations on the gene sequences in the gene population according to the fitness to obtain a new gene population; the gene sequences with the highest fitness in the target gene population are used to determine the cable distribution results corresponding to each model.

2. The method according to claim 1, characterized in that The method of performing a preset number of rounds of initial cable allocation based on the problem data set using a greedy algorithm includes: For each round of initial cable distribution, the problem data set of the round is evaluated using the benefit function of the greedy algorithm to determine multiple vectors of gene sequences corresponding to the round; the benefit function is set based on the distribution strategy; The order of multiple required lengths in the required length data set is randomly adjusted to obtain a question data set for the next round.

3. The method according to claim 2, characterized in that The use of the benefit function of the greedy algorithm to evaluate the problem data set of this round includes: For each round of initial cable allocation, multiple evaluation iterations are performed on the problem dataset of the round until multiple required lengths in the required length dataset are allocated; In each evaluation iteration, for the unallocated demand lengths in the demand length dataset, the inventory length dataset is traversed based on the item count dataset, and the first optimal inventory length corresponding to the maximum benefit is selected from the inventory length dataset using the benefit function, and a first vector of the gene sequence of the current round is generated; the number of items currently corresponding to the first optimal inventory length is not zero; and Subtract 1 from the number of items corresponding to the first optimal inventory length in the item number data set; and The demand length included in the first vector is marked as allocated in the demand length dataset for the next evaluation iteration.

4. The method according to claim 3, characterized in that The distribution strategy is the minimum margin strategy; The benefit E of the first vector is calculated using the following benefit function: E=(∑Length i )÷Price1, Among them, Length i is the i-th demand length included in the first vector, and Price1 is the price corresponding to the first optimal inventory length.

5. The method according to claim 1, wherein the step of calculating the fitness of each gene sequence in the gene population using a fitness function comprises: For each gene sequence, the following operations are performed in sequence for each vector included therein: based on the number of entries dataset, the inventory length dataset is traversed, and the benefit function of the greedy algorithm is used to select the second optimal inventory length corresponding to the maximum benefit from the target data subset of the inventory length dataset, where the number of entries currently corresponding to the second optimal inventory length is not zero; as well as Subtract 1 from the number of items corresponding to the second optimal inventory length in the item count data set to perform the next vector operation; The inventory length in the target data subset complies with a global policy, wherein the global policy is determined based on experience; Sum the prices corresponding to the second optimal inventory length of each vector to obtain the fitness of the gene sequence; The fitness function F is expressed as follows: F=∑Price 2(j) , Price 2(j) is the price corresponding to the second optimal inventory length of the j-th vector in the gene sequence.

6. The method according to claim 1, wherein the genetic operator operation is performed on the gene sequence in the gene population according to the fitness to obtain a new gene population, comprising: randomly selecting at least two gene sequences from the gene population; Selecting a first gene sequence and a second gene sequence with the highest and second highest fitness from at least two gene sequences; Randomly selecting intersection points to rejoin the partial vectors of the first gene sequence and the second gene sequence to obtain a first sub-gene sequence; Substituting the first sub-gene sequence for a portion of the vector of the first gene sequence to obtain a third gene sequence; randomly determining whether to perform a mutation operation on the third gene sequence; if so, obtaining a fourth gene sequence according to the mutation operation, and adding the fourth gene sequence to the gene population to obtain the new gene population; wherein the mutation operation is used to simulate a genetic mutation process in nature; If not executed, the third gene sequence is added to the gene population to obtain the new gene population.

7. The method according to claim 6, wherein obtaining a fourth gene sequence according to the mutation operation comprises: Randomly selecting a partial vector of the third gene sequence; Utilize the required length component required length data set included in the partial vector of the third gene sequence; Replacing the required length dataset with the partial required length dataset to obtain a partial problem dataset; Using a greedy algorithm to perform initial cable matching based on part of the problem dataset, we obtain the second sub-gene sequence; The second sub-gene sequence replaces a portion of the vector of the third gene sequence to obtain the fourth gene sequence.

8. The method according to claim 5, further comprising: A plurality of vectors including gene sequences with the highest fitness in the target gene population and the second best inventory corresponding to each vector are used as a plurality of optimal cable allocation results for each model.

9. The method according to claim 1, wherein the step of creating a preset number of gene sequences based on the distribution strategy, the demand length, and the inventory data comprises: According to the price strategy corresponding to the allocation strategy, prices corresponding to each inventory length are obtained to form a price dictionary; Encrypting the price dictionary, the demand length, and the inventory data to obtain encrypted data; Creating a preset number of gene sequences based on the encrypted data to form an initial gene population; The gene sequences with the highest fitness in the target gene population are used to determine the cable distribution results corresponding to each model, including: decrypted data of the gene sequences with the highest fitness in the target gene population are used to determine the cable distribution results corresponding to each model.

10. A cable distribution device based on genetic algorithm, characterized in that: The device comprises: An acquisition module is used to obtain user order information and inventory data; the user order information includes the required lengths of various cable models and the allocation strategy for the user order; the inventory data includes the inventory lengths of each model and the number of cables of each inventory length; A processing module is configured to create a preset number of gene sequences for each model based on the allocation strategy, the required length, and the inventory data to form an initial gene population; the creating a preset number of gene sequences based on the allocation strategy, the required length, and the inventory data includes: According to the price strategy corresponding to the allocation strategy, prices corresponding to each inventory length are obtained to form a price dictionary; The price dictionary, the demand length, and the inventory data are structured to obtain a problem data set; the problem data set includes a demand length data set, an inventory length data set, a price data set consisting of pairs of values consisting of each inventory length and its corresponding price, and a number data set consisting of pairs of values consisting of each inventory length and its corresponding number of items; Performing the preset number of rounds of initial cable allocation based on the problem dataset using a greedy algorithm to obtain the preset number of gene sequences; each gene sequence includes multiple vectors corresponding to a result of the initial cable allocation, each vector being a set of at least one required length in the required length dataset, the set indicating that the at least one required length is allocated based on a stocked cable; The processing module is further configured to perform population iteration on the initial gene population using a genetic algorithm until an end condition is met, thereby obtaining a target gene population; the population iteration includes calculating the fitness of each gene sequence in the gene population using a fitness function, and performing genetic operator operations on the gene sequences in the gene population based on the fitness to obtain a new gene population; the gene sequences with the highest fitness in the target gene population are used to determine the cable distribution results corresponding to each model.

Images