Configuration method and storage medium for production logistics distribution resources

Abstract

The present application relates to a configuration method and storage medium for production logistics distribution resources, the method comprising: determining revised configuration information according to received distribution demand change information; when it is determined that the distribution carrier configuration mode in the revised configuration information is the target configuration mode, based on material target information and alternative cloud vehicle supply information, performing gene coding and population initialization to generate multiple initial structure codes; using a swarm intelligence global optimization algorithm, performing swarm optimization operations on the multiple initial structure codes to generate multiple first structure codes, and based on the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the swarm intelligence global optimization algorithm, performing population update optimization on the multiple first structure codes to generate multiple candidate structure codes; among the multiple candidate structure codes, detecting the target structure code, and providing the distribution configuration scheme corresponding to the target structure as the configuration result of the production logistics distribution resources.

Description

Technical Field

[0001] The present application relates to the field of smart logistics technology, and in particular to a configuration method and storage medium for production logistics distribution resources. Background Art

[0002] In order to meet the dynamic, diversified and personalized needs of customers, manufacturing companies need to constantly adjust their resource structure, build a flexible production logistics system, and formulate highly adaptable production logistics plans. As an important part of the production logistics system, the optimization decision of the distribution link can effectively ensure the smooth implementation of the production logistics plan.

[0003] In the related technologies, faced with the multi-variety and highly dynamic distribution demands faced by manufacturing enterprises, it is difficult to meet the corresponding distribution demands by using the distribution resources owned by the enterprise itself (including but not limited to distribution vehicles) for material distribution. Therefore, the related enterprises adopt the method of leasing cloud vehicles from the cloud platform to meet the transportation needs. However, in the related technologies, based on leasing cloud vehicles from the cloud platform and combining the distribution resources owned by the enterprise for distribution planning, the corresponding material distribution service is prone to interruption of the distribution service of production materials and reduced production efficiency due to the differences in the service quality of the leased cloud vehicles and the unreasonable configuration planning scheme. At the same time, because the configuration planning of production logistics distribution vehicles involves many factors, is large-scale and mostly nonlinear planning, the optimization algorithm used in the related technologies has poor robustness, poor global search capability and is prone to fall into local optimality. Generating the corresponding distribution plan requires more computing power, the distribution efficiency of the distribution plan is low and the implementation of the distribution plan requires high cost.

[0004] Currently, there is no effective solution to the problem of low distribution efficiency corresponding to the method of production logistics distribution vehicle configuration planning in related technologies and the distribution efficiency of the distribution plan. Summary of the invention

[0005] The embodiments of the present application provide a method and storage medium for configuring production logistics distribution resources, so as to at least solve the problems of low planning efficiency and distribution efficiency corresponding to the distribution plan of the production logistics distribution vehicle configuration planning method in the related art.

[0006] In a first aspect, an embodiment of the present application provides a method for configuring production logistics distribution resources, comprising: in a process of using planned logistics distribution planning information for transportation and distribution, determining revised configuration information based on received distribution demand change information, wherein the distribution demand change information includes material target information of the material to be distributed, and the revised configuration information includes a distribution carrier configuration mode; judging whether the distribution carrier configuration mode is a target configuration mode, and when it is judged that the distribution carrier configuration mode is the target configuration mode, based on the material target information and the alternative cloud vehicle supply information, performing gene encoding and population initialization to generate multiple initial structure codes, wherein the target configuration mode represents a configuration mode based on rented cloud resources for on-demand rental, and each of the initial structure codes represents a distribution configuration plan for the material to be distributed; utilizing a preset swarm intelligence global optimization algorithm to perform gene encoding and population initialization on the multiple initial structure codes. A group optimization operation is performed on the initial structure code to generate multiple first structure codes, and based on the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the group intelligence global optimization algorithm, the multiple first structure codes are population updated and optimized to generate multiple candidate structure codes, wherein the group intelligence global optimization algorithm is an improved TLBO that uses the hybrid teaching strategy and the hybrid learning strategy as the teaching strategy and learning strategy of the teaching and learning algorithm TLBO respectively, the hybrid teaching strategy includes a cooperative learning strategy and a random individual insertion strategy, the hybrid learning strategy includes a self-learning strategy, a group learning strategy and a random insertion strategy, and the distribution cost fitness is determined according to a preset optimization configuration objective function; among the multiple candidate structure codes, the target structure code is detected, and the distribution configuration scheme corresponding to the target structure code is provided as the configuration result of the production logistics distribution resources.

[0007] In a second aspect, an embodiment of the present application provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for configuring production logistics and distribution resources as described in the first aspect above.

[0008] Compared with the related art, the configuration method and storage medium for production logistics distribution resources provided in the embodiment of the present application adopts, in the process of using the planned logistics distribution planning information for transportation and distribution, to determine the modified configuration information according to the received distribution demand change information, wherein the distribution demand change information includes the material target information of the material to be distributed, and the modified configuration information includes the vehicle configuration mode; it is judged whether the vehicle configuration mode is the target configuration mode, and when it is judged that the vehicle configuration mode is the target configuration mode, based on the material target information and the alternative cloud vehicle supply information, gene encoding and population initialization are performed to generate multiple initial structure codes; a preset swarm intelligence global optimization algorithm is used to perform swarm optimization operations on the multiple initial structure codes to generate multiple first structure codes, and based on the first structure codes corresponding to the The distribution cost fitness of the distribution configuration scheme and the swarm intelligence global optimization algorithm are used to perform population update optimization on multiple first structure codes to generate multiple candidate structure codes; among the multiple candidate structure codes, the target structure code is detected, and the distribution configuration scheme corresponding to the target structure code is provided as the configuration result of the production logistics distribution resources, and an improved teaching and learning algorithm is adopted, and the teaching strategy is adjusted and a diversified learning strategy is adopted to reduce the dependence on the initial solution and avoid falling into the local optimum. Information sharing and learning between individuals are adopted to promote the global search ability of the optimization process, so that the configuration results of the production logistics distribution resources generated by the configuration can effectively improve the generation efficiency, save material distribution costs, and solve the problems of low planning efficiency and distribution efficiency corresponding to the distribution scheme of the production logistics distribution vehicle configuration planning method in the related technology.

[0009] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0011] Figure 1 It is a hardware structure block diagram of a terminal of the method for configuring production logistics distribution resources according to an embodiment of the present application;

[0012] Figure 2 is a flow chart of a method for configuring production logistics distribution resources according to an embodiment of the present application;

[0013] Figure 3 It is a structural block diagram of a device for configuring production logistics and distribution resources according to an embodiment of the present application. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the contents disclosed in the present application, some changes such as design, manufacturing or production based on the technical contents disclosed in the present application are only conventional technical means, and should not be understood as insufficient contents disclosed in the present application.

[0015] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0016] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation and may represent the singular or plural. The terms "including", "comprising", "having" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include unlisted steps or units, or may also include other steps or units inherent to these processes, methods, products or devices. The "multiple links" involved in this application refer to links greater than or equal to two. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" may represent: A exists alone, A and B exist at the same time, and B exists alone. The terms "first", "second", "third" and the like involved in this application are only used to distinguish similar objects and do not represent a specific ordering of objects.

[0017] The teaching and learning algorithm involved in this application is described as follows:

[0018] The teaching and learning based optimization (TLBO) algorithm is a global optimization algorithm based on swarm intelligence. It divides individuals in the population into two types: teachers and students. It improves the optimization results through the interaction between teachers and students, and optimizes the population through mutual learning between teachers and information transmission to students, as well as learning between individual students.

[0019] Principles of teaching and learning algorithms

[0020] 1. Initialize the population

[0021] Assuming the population size is NP and the individual variable dimension is m, the random initial population is:

[0022]

[0023] The initial students are generated as follows:

[0024]

[0025] in, For Students , the upper and lower limits of each dimension of the student's decision variables are and , Generates a random number between 0 and 1.

[0026] 2. Teaching stage

[0027] The teaching stage simulates the process of teachers guiding students to learn. Students learn through the average value of their fitness and the difference of teachers. Specifically, the average gap between teachers and students is calculated according to the following formula:

[0028]

[0029]

[0030]

[0031] in, for The average value of students represents the value of the teacher, that is, the value of the student with the best fitness, To adaptively adjust the teaching factor, represents the maximum value of the teaching factor, represents the minimum value of the teaching factor, Indicates the current iteration number, Indicates the maximum number of iterations;

[0032] Students learn through the following formula and the above gap

[0033]

[0034] in, for The value of each student before teaching, After teaching The teaching value of each student.

[0035] 3. Learning stage

[0036] The learning stage mainly involves mutual learning through discussion and communication among students, and by learning from members with higher values ​​to improve their own values, it is beneficial for students , randomly select students who are different from Conduct learning exchanges and compare the fitness values ​​between students to adjust their learning. The formula is as follows:

[0037]

[0038] The specific steps of the existing teaching and learning algorithm include:

[0039] Step 1: Initialize the population: randomly distribute a certain number of individuals in the solution space as the initial population.

[0040] Step 2: Divide the population into teachers and students: In each generation, individuals are divided into two categories: teachers and students according to their fitness values. Individuals with high fitness values ​​are divided into teachers, and individuals with low fitness values ​​are divided into students.

[0041] Step 3: Interaction between individual teachers: Individual teachers communicate and learn from each other to improve their own solutions and exchange information with individual students.

[0042] Step 4: Individual student interaction: Individual students communicate and learn from each other, and obtain information from individual teachers to improve their own solutions.

[0043] Step 5: Update the population: Update the solution of the entire population based on the learning and communication between teachers and individual students.

[0044] Step 6: Termination condition: When the predetermined stopping criteria are reached, the algorithm execution is stopped and the optimal solution is output.

[0045] The specific embodiments of the present application are described below:

[0046] The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. Taking running on a terminal as an example, Figure 1 1 is a hardware structure block diagram of a terminal of the method for configuring production logistics distribution resources according to an embodiment of the present application. Figure 1 As shown, the terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Optionally, the terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the above terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0047] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the configuration method of production logistics distribution resources in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0048] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by the communication provider of the terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0049] This embodiment provides a method for configuring production logistics distribution resources running on the above terminal. Figure 2 is a flow chart of a method for configuring production logistics and distribution resources according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:

[0050] Step S201, in the process of using the planned logistics distribution planning information for transportation and distribution, determine the modified configuration information according to the received distribution demand change information, wherein the distribution demand change information includes the material target information of the material to be distributed, and the modified configuration information includes the distribution carrier configuration method.

[0051] In this embodiment, the configuration of production logistics distribution resources refers to the arrangement of distribution carriers, that is, the production materials are distributed to the corresponding processing and production areas (for example, production workshops) by arranging distribution carriers, wherein the distribution carriers include but are not limited to one of the following: distribution vehicles (for example, flatbed trailers, box trailers, trucks), transfer tools (for example, electric forklifts, AVG carts); in this embodiment and the following text, the distribution resources are explained in terms of distribution vehicles. It should be understood that the distribution vehicles are a type or a kind of configuration resources, but do not constitute a limitation on the distribution resources of this application.

[0052] In this embodiment, before executing the configuration method of production logistics distribution resources of the present application, the scheme configured before the current one is executed. The configured scheme can be a scheme generated by periodic pre-configuration based on the distribution resources owned by the relevant enterprise itself (for example, its own distribution vehicles) (for example, distribution vehicle arrangement, transportation lane arrangement, number of employees involved in transportation and distribution), or it can be a scheme generated by modifying the configuration based on demand change information before the current one. The scheme generated by the modified configuration meets the dynamic transportation demand changes generated by the current history. In this embodiment, when new transportation demand is generated in the production process (corresponding to the received distribution demand), the current one is automatically generated. Demand change information), at this time, it is necessary to determine whether the transportation plan corresponding to the currently executed logistics distribution planning information has redundant transportation capacity, and to determine whether the redundant transportation capacity can meet the newly generated transportation demand changes; in this embodiment, according to the received configuration demand change information, determining to correct the configuration information means determining what configuration planning method to adopt based on the newly generated configuration demand change information (including but not limited to the addition of new distribution materials, and the increase in the distribution volume of existing materials), for example: adopting the material distribution arrangement planning based on the company's own distribution vehicles, and for example: adopting the material distribution arrangement planning based on the introduction of new distribution resources (for example: cloud vehicle leasing).

[0053] Step S202, determine whether the distribution carrier configuration mode is the target configuration mode, and when it is determined that the distribution carrier configuration mode is the target configuration mode, perform gene encoding and population initialization based on the material target information and the alternative cloud vehicle supply information to generate multiple initial structure codes, wherein the target configuration mode represents a configuration mode based on renting cloud resources as needed, and each initial structure code represents a distribution configuration plan for distributing the material to be distributed.

[0054] In this embodiment, when it is determined that the distribution carrier configuration method represented by the modified configuration information is a configuration method based on leasing cloud resources for on-demand rental, dynamic configuration is started, that is, entering the modified configuration stage. In this stage, the distribution tasks corresponding to the newly added distribution demand changes and the current distribution tasks that have not yet been delivered are taken as the total distribution demand, and the company's own distribution vehicles and the appropriate number of cloud vehicles are selected to plan the material distribution arrangement; in this embodiment, the distribution carrier configuration corresponding to the material distribution arrangement is converted into a problem of optimizing the population individuals. Therefore, after determining that the distribution carrier configuration method is the target configuration method, genetic encoding and population initialization will be performed. That is, a plan is initially generated to allocate corresponding delivery vehicles to the total delivery demand. It can be understood that each plan formed by the allocation of tasks corresponds to a delivery cost. Seeking the optimal delivery cost (for example: minimum cloud vehicle rental cost, minimum maintenance cost, minimum power cost, and minimum overtime cost) is also the original intention of implementing the method for configuring production logistics and distribution resources in the embodiment of the present application. Therefore, when generating multiple initial structure codes, the initial structure codes will be subjected to population evolution operations to obtain the optimal population individuals, that is, the distribution plan corresponding to the corresponding structure code meets the current changing distribution demand, and at the same time, the corresponding distribution cost is also optimal.

[0055] In this embodiment, the initial structure code is generated using a set coding method. Specifically, the following coding method is used: 1. Cloud vehicle supplier coding: an integer variable C1 is used to represent the vehicle supplier selected for household material 1, with a value range of 0 to 7; an integer variable C2 is used to represent the vehicle supplier selected for household material 2, with a value range of 0 to 7; an integer variable C3 is used to represent the vehicle supplier selected for commercial material 3, with a value range of 0 to 7; 2. Rental quantity coding: an integer variable Q1 is used to represent the rental quantity of household material 1, with a value range of 0 to 10; an integer variable Q2 is used to represent the rental quantity of household material 2, with a value range of 0 to 10. The integer variable Q3 represents the rental quantity of commercial material 3, with a value range of 0 to 10; 3. Transportation mode coding: A Boolean variable T represents the transportation mode of household material 1, where T=True represents flatbed trailer transportation, and T=False represents truck transportation. For example, the initial structure encoding corresponding to period 1 is [C1=2, C2=5, C3=1, Q1=7, Q2=3, Q3=5, T=True], which means that household material 1 selects the third vehicle supplier (C1=2), household material 2 selects the sixth vehicle supplier (C2=5), and commercial material 3 selects Select the second vehicle supplier (C3=1), the number of household material 1 leased is 7 (Q1=7), the number of household material 2 leased is 3 (Q2=3), the number of commercial material 3 leased is 5 (Q3=5), and select flatbed trailer transportation (T=True); It can be understood that the above encoding method is used to solve the problem that there are many parameters in the production logistics distribution resource optimization configuration model, the number of feasible solutions in the problem may be too large, and the binary string encoding may cause the algorithm to run slowly due to its large length, and it is difficult to find a suitable string length to balance the representation capacity and operation efficiency of feasible solutions.

[0056] Step S203, using a preset swarm intelligence global optimization algorithm, perform swarm optimization operations on multiple initial structure codes to generate multiple first structure codes, and based on the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the swarm intelligence global optimization algorithm, perform population update optimization on the multiple first structure codes to generate multiple candidate structure codes, wherein the swarm intelligence global optimization algorithm is an improved TLBO that uses a hybrid teaching strategy and a hybrid learning strategy as the teaching strategy and learning strategy of the teaching and learning algorithm TLBO respectively, the hybrid teaching strategy includes a cooperative learning strategy and a random individual insertion strategy, the hybrid learning strategy includes a self-learning strategy, a group learning strategy and a random insertion strategy, and the distribution cost fitness is determined according to a preset optimization configuration objective function.

[0057] In this embodiment, the teaching strategy improvement includes: adjusting the teaching strategy by analyzing the student performance and the evolutionary process in the algorithm to improve the overall performance of the population; the diversity and convergence speed of the population can be improved by introducing diversity and adaptability; in this embodiment, the diverse teaching strategies adopted by the teaching strategy include personalized teaching, cooperative learning, practical teaching and random individual insertion strategies; in this embodiment, the learning strategy improvement includes: adopting more diverse learning strategies to enable students to make full use of their own information and learn more effectively, specifically including: self-learning, group learning, and hybrid learning.

[0058] In this embodiment, when calculating the distribution cost fitness, it is based on the distribution configuration scheme corresponding to each first structure code, and is calculated according to the preset overall optimization configuration objective function. It can be understood that the calculation of the distribution cost fitness based on the corresponding objective function is feasible and does not constitute an unclear limitation on the embodiments of the present application.

[0059] In this embodiment, the improved TLBO is used to perform group optimization operations on multiple initial structure codes, and based on the improved TLBO and the distribution cost fitness of the distribution configuration plan corresponding to the first structure code, the multiple first structure codes are population updated and optimized to generate corresponding distribution cost fitness that meets the set threshold, that is, the corresponding candidate structure code with the optimal distribution cost, to achieve the solution of population optimization, that is, to form a distribution plan that meets the distribution demand and has a distribution cost within an acceptable range.

[0060] Step S204 , detecting the target structure code among multiple candidate structure codes, and providing the distribution configuration scheme corresponding to the target structure code as the configuration result of the production logistics distribution resources.

[0061] In this embodiment, when generating a candidate structure code whose distribution cost fitness satisfies a set threshold, that is, the corresponding distribution cost is optimal, a candidate structure code is selected as a target structure code by selecting the best, so that a distribution configuration scheme and a configuration result of the corresponding production logistics distribution resources are obtained. Of course, it can be understood that after selecting the target structure code, the corresponding distribution configuration scheme can be obtained through the target structure code. For example, the target structure code is [C1=3, C2=5, C3=1, ..., Cn=6, Q1=7, Q2=3, Q3=5, ..., Qn= 7, T=True], the corresponding distribution configuration plan is: household material 1 selects the 4th vehicle supplier (C1=3), household material 2 selects the 6th vehicle supplier (C2=5), commercial material 3 selects the 2nd vehicle supplier (C3=1), and commercial material n selects the 7th vehicle supplier (Cn=6). The rental quantity of household material 1 is 7 vehicles (Q1=7), the rental quantity of household material 2 is 3 vehicles (Q2=3), the rental quantity of commercial material 3 is 5 vehicles (Q3=5), and the rental quantity of commercial material n is 7 vehicles (Qn=7), and flatbed trailer transportation is selected (T=True).

[0062] Through the above steps S201 to S204, in the process of using the planned logistics distribution planning information for transportation and distribution, according to the received distribution demand change information, the modified configuration information is determined, wherein the distribution demand change information includes the material target information of the material to be distributed, and the modified configuration information includes the distribution carrier configuration mode; it is determined whether the distribution carrier configuration mode is the target configuration mode, and when it is determined that the distribution carrier configuration mode is the target configuration mode, based on the material target information and the alternative cloud vehicle supply information, gene encoding and population initialization are performed to generate multiple initial structure codes; using the preset swarm intelligence global optimization algorithm, a swarm optimization operation is performed on the multiple initial structure codes to generate multiple first structure codes, and the distribution of the distribution configuration scheme corresponding to the first structure code is performed. The cost fitness and swarm intelligence global optimization algorithm performs population update optimization on multiple first structure codes to generate multiple candidate structure codes; among the multiple candidate structure codes, the target structure code is detected, and the distribution configuration plan corresponding to the target structure code is provided as the configuration result of the production logistics distribution resources. An improved teaching and learning algorithm is adopted, and the teaching strategy is adjusted and a diversified learning strategy is adopted to reduce the dependence on the initial solution and avoid falling into the local optimum. Information sharing and learning between individuals are adopted to promote the global search ability of the optimization process, so that the configuration results of the production logistics distribution resources generated by the configuration can effectively improve production efficiency, save material distribution costs, and solve the problems of low distribution efficiency corresponding to the production logistics distribution vehicle configuration planning method, planning efficiency and distribution plan in related technologies.

[0063] In some embodiments, the step S203 of performing a group optimization operation on a plurality of initial structure codes using a preset group intelligence global optimization algorithm to generate a plurality of first structure codes is implemented by the following steps:

[0064] Step 21 : Select a preset number of first teacher structure codes from a plurality of initial structure codes based on the distribution cost fitness corresponding to each initial structure code.

[0065] In this embodiment, teachers and students are divided into individuals according to the calculated delivery cost fitness.

[0066] Step 22, using a hybrid teaching strategy, performs teaching updates on all first teacher structure codes to generate a second teacher structure code, and based on the second teacher structure code and the hybrid teaching strategy, performs population individual teaching on the first student structure codes in multiple initial structure codes to generate a second student structure code, wherein the first student structure code includes all first structure codes except the first teacher structure code.

[0067] In this embodiment, the divided teacher individuals are different from the existing method of selecting teacher individuals using the tournament selection method, and a cooperative learning method is also adopted for the teacher individuals. In an optional implementation method of this embodiment, the two first teacher structure codes with the best fitness are selected according to the distribution cost fitness, and then the two first teacher structure codes are updated according to the hybrid teaching strategy to generate new teacher individuals, that is, generate the second teacher structure code.

[0068] In this embodiment, after the second teacher structure code is generated, the second teacher structure code is used as the target teacher individual to teach the first student structure code as the student individual to generate a corresponding student individual. The student individual will generate a population individual after the current population update after the hybrid learning strategy is performed; in this embodiment, in the process of teaching the first student structure code, the following teaching is performed based on a variety of teaching strategies:

[0069] 1. Personalized teaching

[0070] According to students' learning characteristics and performance, targeted teaching methods are adopted, such as exercises of different difficulty levels, interactive courseware, etc., to improve students' learning effects. The specific method is as follows:

[0071]

[0072] in, Indicates students exist Performance scores on learning tasks, Indicates students In the In the task performance in terms of for The weight of the aspect performance;

[0073] 2. Collaborative learning

[0074] Encourage students to cooperate with each other and complete tasks together to improve their collaboration and problem-solving abilities. The specific formula is as follows:

[0075]

[0076] in, Indicates students and The cooperation score between Indicates students In the score in each task;

[0077] 3. Practical teaching

[0078] Guide students to participate in solving practical problems and improve their practical and innovative abilities. The specific formula is as follows:

[0079]

[0080] Where E is the mean absolute value of the prediction error, n is the number of samples, and m is the number of predictor variables. For sample No. The true value of the predictor variable, For sample No. The predicted value of the predictor variable;

[0081] 4. Random individual insertion strategy

[0082] In each iteration, a certain proportion (for example, 10%) of random individuals are inserted into the population. This can increase the diversity of the population and thus improve the search ability of the population. The specific formula is as follows:

[0083]

[0084] in, For the previous generation population, For the current population, 10% of the individuals are randomly selected. is a set of randomly selected individuals.

[0085] Step 23, using a hybrid learning strategy, performs population individual information learning and updating on the second student structure code to generate a plurality of first candidate student structure codes, wherein the first structure code includes the second teacher structure code and the first candidate student structure code.

[0086] In this embodiment, the newly generated teacher individual is used to teach the student individual, thereby generating a new population individual; in this embodiment, the following learning strategy is used to learn and update the population individual information of the second student structure code:

[0087] 1. Self-learning

[0088] Students choose some of their own attributes to study in order to make full use of their own information, according to the following formula: ;

[0089] 2. Group learning

[0090] Individual students can choose other individuals to learn from, or they can choose the entire group to learn from for a more comprehensive learning experience, as follows: ;

[0091] 3. Blended Learning

[0092] Individual students can learn more flexibly by using both self-learning and group learning, as shown in the following formula: ,in, represents the best individual in the current population, The weight of self-learning, Learn weights for the population, , is the weight of hybrid learning.

[0093] Through steps 21 to 23, a group optimization operation is implemented based on the initial structure code to generate the corresponding first structure code; at the same time, the improved teaching strategy is used to generate new teacher individuals and teach individual students, thereby improving the diversity and convergence speed of the population, enhancing the global search capability, and using the improved learning strategy to learn individual students, so that individual students can make full use of their own information and learn more effectively, thereby making the generated distribution configuration plan tend to be optimal, improving the material distribution efficiency of the entire enterprise's processing and production, and saving material distribution costs.

[0094] In some embodiments, the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the swarm intelligence global optimization algorithm in step S203 perform population update optimization on multiple first structure codes to generate multiple candidate structure codes, which is implemented by the following steps:

[0095] Step 31, determine all the first structure codes to be updated and optimized for the Nth population, obtain the candidate structure codes, and determine the first difference between the distribution cost fitness corresponding to each candidate structure code and the corresponding historical structure code, where N=1, the historical structure code includes the initial structure code, N>1, the historical structure code includes the first structure code for N-1 population update optimization, and N is a positive integer.

[0096] In this embodiment, after using the improved TLBO to perform a population optimization operation on the initial structure code to generate the first structure code, the corresponding structure code obtained is not the target structure code, and multiple iterative updates are required to make the corresponding distribution cost fitness meet the set threshold or the number of iterations reaches the preset maximum number of times, then stop the iteration and obtain the candidate structure code for selecting the target structure code; it can be understood that the number of times the population update optimization is to be performed can be the first update after the population optimization operation is performed on the initial structure code to generate the first structure code, or it can be one of the times after the first update; in this embodiment When judging whether it is necessary to perform another population update optimization on the current corresponding structure code, the difference between the distribution cost fitness corresponding to the current corresponding structure code and the previously generated structure code is used to judge. If the corresponding difference is not greater than the set difference threshold, it is necessary to continue the population update optimization. If the corresponding difference is greater than the set difference threshold, it means that the distribution cost fitness of the current corresponding structure code is within the range corresponding to the objective function, and the population update optimization process for generating the current corresponding structure code is an effective population update optimization. At this time, the current corresponding structure code can be used as a candidate structure code for selecting the target structure code.

[0097] Step 32, determine whether the first difference is greater than a preset threshold, and if it is determined that the first difference is not greater than the preset threshold, select a third teacher structure code from multiple candidate structure codes according to the distribution cost fitness corresponding to the candidate structure code.

[0098] In this embodiment, when the first difference is not greater than the preset threshold, it indicates that the population update optimization needs to continue. At this time, it is necessary to generate corresponding teacher individuals based on the selected teacher individuals to realize subsequent teaching based on the newly generated teacher individuals.

[0099] Step 33, using a hybrid teaching strategy, updating all third teacher structure codes to generate a fourth teacher structure code, and based on the fourth teacher structure code and the hybrid teaching strategy, performing population individual teaching on the third student structure code among multiple alternative structure codes to generate a fourth student structure code, wherein the third student structure code includes all alternative structure codes except the third teacher structure code.

[0100] Step 34, using a hybrid learning strategy, performs population individual information learning and updating on the fourth student structure code to generate multiple second candidate student structure codes, wherein the candidate structure codes corresponding to the Nth population update optimization include the fourth teacher structure code and the second candidate student structure code.

[0101] Through the above steps 31 to 34, it is achieved to generate candidate structure codes for selecting target structure codes based on improved hybrid teaching strategy and improved hybrid learning strategy, generate new teacher individuals and teach student individuals by using improved teaching strategy, improve the diversity and convergence speed of population, enhance the global search capability, and use improved learning strategy to learn student individuals, so that student individuals can make full use of their own information and learn more effectively, thereby making the generated distribution configuration plan tend to be optimal, improving the material distribution efficiency of the entire enterprise's processing and production, and saving material distribution costs.

[0102] In some of the embodiments, based on the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the swarm intelligence global optimization algorithm, a plurality of first structure codes are subjected to population update optimization to generate a plurality of candidate structure codes, which is achieved by the following steps:

[0103] Step 41, determine all the first structure codes to be updated and optimized for the Nth population, obtain the candidate structure codes, and determine the first difference between the distribution cost fitness corresponding to each candidate structure code and the corresponding historical structure code, where N=1, the historical structure code includes the initial structure code, N>1, the historical structure code includes the first structure code for N-1 population update optimization, and N is a positive integer.

[0104] Step 42, determine whether the first difference is greater than a preset threshold, and if it is determined that the first difference is greater than the preset threshold, obtain the fifth teacher structure code and the third candidate student structure code generated by N-1 population update optimizations.

[0105] Step 43, determining that the candidate structure codes include the fifth teacher structure code and the third candidate student structure code.

[0106] In this embodiment, when judging whether it is necessary to perform another population update optimization on the current corresponding structure code, the difference between the distribution cost fitness corresponding to the current corresponding structure code and the previously generated structure code is judged. If the corresponding difference is greater than the set difference threshold, it means that the distribution cost fitness of the current corresponding structure code is within the range corresponding to the objective function, and the population update optimization process for generating the current corresponding structure code is an effective population update optimization. At this time, the current corresponding structure code can be used as a candidate structure code for selecting the target structure code, that is, the structure code composed of the fifth teacher structure code and the third candidate student structure code is used as the candidate structure code.

[0107] In some embodiments, teaching and updating the structure code includes the following steps:

[0108] Step 51, determine the current candidate structure code to be updated for teaching, wherein the current candidate structure code includes one of the following: the sixth teacher structure code selected based on the distribution cost fitness, the fifth student structure code to be used for population individual teaching, each current candidate structure code includes multiple first sub-codes, and the gene value of the first sub-code is used to represent the distribution configuration information for distributing a distribution material.

[0109] Step 52, using a preset collaborative algorithm and the first fitness corresponding to the first sub-code, all first sub-codes corresponding to each current candidate structure code are processed, and a preset number of first random sub-codes are randomly inserted into the processed first sub-code to generate a corresponding second sub-code, wherein the gene value of the first random sub-code is used to characterize the distribution configuration information of a randomly generated distribution material, and the first fitness is used to characterize the distribution efficiency of the distribution configuration information corresponding to the corresponding sub-code.

[0110] Step 53, based on the second sub-code, generate the first target structure code for completing the current teaching, and determine the distribution cost fitness corresponding to the first target structure code according to the first fitness corresponding to the second sub-code, wherein the first target structure code includes one of the following structure codes for completing the teaching update: the seventh teacher structure code corresponding to the sixth teacher structure code, and the sixth student structure code corresponding to the fifth student structure code.

[0111] In this embodiment, cooperative learning and random individual insertion strategies are used for optimization. When the cooperative learning strategy is used, the problem is first decomposed into sub-problems, and a learner is assigned to each sub-problem. Assume that the problem is decomposed into Sub-question, The decision variables of the subproblem are , , , , each learner can only see the sub-problem assigned to him, but can exchange information and cooperate to solve the problem; in this embodiment, to calculate the contribution of each sub-problem, a collaboration function is defined , which can be expressed as: ,in, Representing learners In the In the task performance in terms of , , , Represents the weight of the corresponding decision variable. In addition, a cooperation function is defined , which is used to integrate the solutions of all subproblems and calculate the delivery cost fitness: .

[0112] Through the above steps 51 to 53, the structure code is taught based on the hybrid teaching strategy to update the teaching of the structure code.

[0113] In some of the optional implementations, the teaching is carried out as follows:

[0114] Step 1: Generate an initial solution. Generate an initial solution as the starting point for learners.

[0115] Step 2: Decompose the problem. Decompose the distribution carrier configuration problem into multiple sub-problems. Each sub-problem corresponds to a learner, where the learners include corresponding teacher individuals and student individuals.

[0116] Step 3: Assign subproblems to different learners and execute the teaching algorithm on each learner.

[0117] Step 4: Collaborative learning. During the learning process, learners need to exchange information and collaborate to solve problems. The collaborative algorithm is executed on each learner to calculate the contribution of each sub-problem.

[0118] In this embodiment, the collaborative algorithm is executed to calculate the share of each sub-problem using the following formula: ,in, Representing learners In the In the task performance in terms of , , , represents the weight of the corresponding decision variable.

[0119] Step 5: Individual insertion: During the collaborative learning process, some learners are randomly selected for individual insertion, and some random solutions are selected and inserted into the learners’ solution set.

[0120] Step 6: Solution merging: merge all learners’ solutions into a global solution and calculate the delivery cost fitness.

[0121] Step 7: Iteration process: Repeat steps 3 to 6.

[0122] In some embodiments, a hybrid learning strategy is used to process the corresponding student structure code, including the following steps:

[0123] Step 61, determining a plurality of current student structure codes to be updated for population individual information learning, wherein each current student structure code includes a plurality of third sub-codes, and the gene value of the third sub-code is used to represent the distribution configuration information for distributing a distribution material.

[0124] Step 62, using the self-learning strategy and the group learning strategy in turn, to process all the third sub-codes and generate the fourth sub-code, wherein the self-learning strategy is used to guide the third sub-codes, and search and update the best sub-code among all the third sub-codes corresponding to their respective corresponding gene values ​​and each current student structure code, and the group learning strategy is used to guide the candidate sub-codes generated by the self-learning, and search and update them based on all the candidate sub-codes corresponding to multiple current student structure codes.

[0125] In the self-learning stage, each corresponding sub-code is searched and updated based on its own experience and the optimal solution of the sub-code, which is calculated according to the following formula:

[0126]

[0127] in, Indicates the third subcode exist The position vector of the generation, Indicates that the third subcode The optimal position vector in time, represents the position vector of the teacher subcode at the current moment, rand(n)∈[0,1] It means to generate a random number between 0 and 1. The random insertion probability for the random subcode, is the learning rate set according to the experiment;

[0128] In the group learning stage, the information of all sub-codes in the corresponding structure code is searched and updated. This stage is expressed as follows:

[0129]

[0130] in, Represents the average position vector of all sub-codes in the structure code, Encode the insertion probability for the structure.

[0131] Step 63, randomly inserting a preset number of second random sub-codes into the fourth sub-code to generate a corresponding target sub-code, wherein the corresponding candidate student structure code includes the target sub-code.

[0132] In this embodiment, n second random sub-codes are randomly generated: , … , The value of each new sub-code is randomly generated in the feasible solution space of the problem and satisfies the constraints of the problem. Specifically, for each structure code, the new sub-code corresponding to , calculated as follows: .

[0133] Through the above steps, multiple current student structure codes to be updated for population individual information learning are determined; all third sub-codes are processed in turn using self-learning strategy and group learning strategy to generate a fourth sub-code; a preset number of second random sub-codes are randomly inserted into the fourth sub-code to generate a corresponding target sub-code, wherein the corresponding candidate student structure code includes the target sub-code, thereby realizing the use of an improved hybrid learning strategy to generate candidate structure codes.

[0134] In some of the embodiments, detecting a target structure encoding among multiple candidate structure encodings is achieved by the following steps:

[0135] Step 71, determining the distribution cost fitness corresponding to each candidate structure code.

[0136] Step 72 : Select the candidate structure code with the highest fitness from among multiple candidate structure codes to obtain the target structure code.

[0137] By determining the distribution cost fitness corresponding to each candidate structure code in the above steps; among multiple candidate structure codes, the candidate structure code with the highest fitness is selected to obtain the target structure code, thereby realizing the selection of the target structure code to obtain the configuration result of the production logistics distribution resources.

[0138] In some embodiments, obtaining the candidate cloud vehicle supply information includes the following steps:

[0139] Step 81, obtaining a distribution vehicle parameter table, wherein the distribution vehicle parameter table includes material target information, distribution vehicle target information, and correspondence information between the material target information and the distribution vehicle target information.

[0140] Step 82, in the distribution vehicle parameter table, query the distribution vehicle target information corresponding to the material target information.

[0141] Step 83, in the obtained evaluation parameter table, query the evaluation parameters corresponding to the delivery vehicle target information, and use the delivery vehicle target information and the evaluation parameters corresponding to the delivery vehicle target information as cloud vehicle supply information, wherein the evaluation parameter table includes the delivery vehicle target information, the evaluation parameters and the correspondence information between the delivery vehicle target information and the evaluation parameters, and the evaluation parameters are obtained by processing the delivery vehicle target information using the fuzzy evaluation method to obtain parameters for evaluating the service quality of the corresponding delivery vehicle.

[0142] Through the above steps 81 to 83, it is possible to select a suitable cloud vehicle based on the service quality of the cloud vehicle and reduce the overall material distribution cost.

[0143] In some of the embodiments, when it is determined that the delivery carrier configuration does not include the target configuration, the following steps are further implemented:

[0144] Obtaining current delivery vehicle configuration information from the planned logistics delivery planning information, wherein the delivery vehicle configuration information includes the configured target delivery vehicle for each material to be delivered;

[0145] The number of distribution times implemented by the target distribution vehicle corresponding to each material to be distributed is modified to generate target distribution configuration information, wherein the configuration result of the production logistics distribution resources includes the target distribution configuration information.

[0146] This embodiment also provides a configuration method device for production logistics distribution resources, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the terms "module", "unit", "subunit", etc. can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0147] Figure 3 is a structural block diagram of a configuration device for production logistics distribution resources according to an embodiment of the present application, such as Figure 3 As shown, the device includes a determination module 31, an encoding module 32, a processing module 33 and a detection module 34, wherein:

[0148] The determination module 31 is used to determine the modified configuration information according to the received distribution demand change information during the transportation and distribution process using the planned logistics distribution planning information, wherein the distribution demand change information includes the material target information of the material to be distributed, and the modified configuration information includes the distribution carrier configuration method.

[0149] The encoding module 32 is coupled to the determination module 31, and is used to determine whether the distribution carrier configuration mode is the target configuration mode. When it is determined that the distribution carrier configuration mode is the target configuration mode, based on the material target information and the alternative cloud vehicle supply information, gene encoding and population initialization are performed to generate multiple initial structure codes, wherein the target configuration mode represents a configuration mode based on renting cloud resources as needed, and each initial structure code represents a distribution configuration plan for distributing materials to be distributed.

[0150] The processing module 33 is coupled to the encoding module 32, and is used to use a preset swarm intelligence global optimization algorithm to perform swarm optimization operations on multiple initial structure codes to generate multiple first structure codes, and based on the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the swarm intelligence global optimization algorithm, perform population update optimization on the multiple first structure codes to generate multiple candidate structure codes, wherein the swarm intelligence global optimization algorithm is an improved TLBO that uses a hybrid teaching strategy and a hybrid learning strategy as the teaching strategy and learning strategy of the teaching and learning algorithm TLBO respectively, the hybrid teaching strategy includes a cooperative learning strategy and a random individual insertion strategy, the hybrid learning strategy includes a self-learning strategy, a group learning strategy and a random insertion strategy, and the distribution cost fitness is determined according to a preset optimization configuration objective function.

[0151] The detection module 34 is coupled to the processing module 33 and is used to detect the target structure code among multiple candidate structure codes, and provide a distribution configuration scheme corresponding to the target structure code as a configuration result of production logistics distribution resources.

[0152] In some embodiments, the processing module 33 further includes:

[0153] A first selection unit is used to select a preset number of first teacher structure codes from a plurality of initial structure codes based on the distribution cost fitness corresponding to each initial structure code;

[0154] The first teaching unit is coupled to the first selection unit and is used to use a hybrid teaching strategy to perform teaching updates on all first teacher structure codes to generate a second teacher structure code, and based on the second teacher structure code and the hybrid teaching strategy, perform population individual teaching on the first student structure code in the plurality of initial structure codes to generate a second student structure code, wherein the first student structure code includes all first structure codes except the first teacher structure code;

[0155] The first learning unit is coupled to the first teaching unit and is used to utilize a hybrid learning strategy to perform population individual information learning and update on the second student structure code to generate a plurality of first candidate student structure codes, wherein the first structure code includes a second teacher structure code and a first candidate student structure code.

[0156] In some embodiments, the processing module 33 further includes:

[0157] A first determination unit is used to determine all first structure codes to be subjected to the Nth population update optimization, obtain candidate structure codes, and determine a first difference between the distribution cost fitness corresponding to each candidate structure code and the corresponding historical structure code, wherein N=1, the historical structure code includes the initial structure code, N>1, the historical structure code includes the first structure code subjected to the N-1th population update optimization, and N is a positive integer;

[0158] The first judging unit and the first determining unit are used to judge whether the first difference is greater than a preset threshold value, and when it is judged that the first difference is not greater than the preset threshold value, select a third teacher structure code from a plurality of candidate structure codes according to the distribution cost fitness corresponding to the candidate structure code;

[0159] The second teaching unit is coupled to the first judgment unit and is used to use a hybrid teaching strategy to perform teaching updates on all third teacher structure codes to generate a fourth teacher structure code, and based on the fourth teacher structure code and the hybrid teaching strategy, perform population individual teaching on the third student structure code in a plurality of candidate structure codes to generate a fourth student structure code, wherein the third student structure code includes all candidate structure codes except the third teacher structure code;

[0160] The second learning unit is coupled to the second teaching unit, and is used to utilize a hybrid learning strategy to perform population individual information learning and update on the fourth student structure code, and generate multiple second candidate student structure codes, wherein the candidate structure codes corresponding to the Nth population update optimization include the fourth teacher structure code and the second candidate student structure code.

[0161] In some of the embodiments, when the first judgment unit determines that the first difference is greater than a preset threshold, the processing module 33 is also used to obtain the fifth teacher structure code and the third candidate student structure code generated by N-1 population update optimizations; and determine that the candidate structure codes include the fifth teacher structure code and the third candidate student structure code.

[0162] In some of the embodiments, the configuration device of the production logistics distribution resources is also used to: determine the current candidate structure code to be updated for teaching, wherein the current candidate structure code includes one of the following: the sixth teacher structure code selected based on the distribution cost fitness, the fifth student structure code to be taught to the population individual, each current candidate structure code includes a plurality of first sub-codes, and the gene value of the first sub-code is used to characterize the distribution configuration information for distributing a distribution material; using a preset collaborative algorithm and the first fitness corresponding to the first sub-code, all the first sub-codes corresponding to each current candidate structure code are processed, and a preset number of first sub-codes are randomly inserted into the processed first sub-codes. The first random sub-code is used to generate the corresponding second sub-code, wherein the gene value of the first random sub-code is used to characterize the distribution configuration information of a randomly generated distribution material, and the first fitness is used to characterize the distribution efficiency of the distribution configuration information corresponding to the corresponding sub-code; based on the second sub-code, the first target structure code for completing the current teaching is generated, and according to the first fitness corresponding to the second sub-code, the distribution cost fitness corresponding to the first target structure code is determined, wherein the first target structure code includes one of the following structure codes for completing the teaching update: the seventh teacher structure code corresponding to the sixth teacher structure code, and the sixth student structure code corresponding to the fifth student structure code.

[0163] In some of the embodiments, the configuration device for production logistics distribution resources is also used to: determine multiple current student structure codes to be updated for population individual information learning, wherein each current student structure code includes multiple third sub-codes, and the gene value of the third sub-code is used to characterize the distribution configuration information for distributing a distribution material; use self-learning strategy and group learning strategy in turn to process all third sub-codes to generate a fourth sub-code, wherein the self-learning strategy is used to guide the third sub-codes to search and update based on the respective corresponding gene values ​​and the optimal sub-code among all the third sub-codes corresponding to each current student structure code, and the group learning strategy is used to guide the candidate sub-codes generated by self-learning to search and update based on all the candidate sub-codes corresponding to multiple current student structure codes; randomly insert a preset number of second random sub-codes into the fourth sub-code to generate a corresponding target sub-code, wherein the corresponding candidate student structure code includes the target sub-code.

[0164] In some embodiments, the detection module 34 further includes:

[0165] The second determination unit is used to determine the delivery cost fitness corresponding to each candidate structure code;

[0166] The second selection unit is coupled to the second determination unit and is used to select a candidate structure code with the highest fitness among multiple candidate structure codes to obtain a target structure code.

[0167] In some of the embodiments, the encoding module 32 is also used to obtain a delivery vehicle parameter table, wherein the delivery vehicle parameter table includes material target information, delivery vehicle target information, and correspondence information between the material target information and the delivery vehicle target information; in the delivery vehicle parameter table, the delivery vehicle target information corresponding to the material target information is queried; in the obtained evaluation parameter table, the evaluation parameters corresponding to the delivery vehicle target information are queried, and the delivery vehicle target information and the evaluation parameters corresponding to the delivery vehicle target information are used as cloud vehicle supply information, wherein the evaluation parameter table includes delivery vehicle target information, evaluation parameters, and correspondence information between the delivery vehicle target information and the evaluation parameters, and the evaluation parameters are obtained by processing the delivery vehicle target information using a fuzzy evaluation method to obtain parameters for evaluating the service quality of the corresponding delivery vehicle.

[0168] In some of the embodiments, when it is determined that the delivery carrier configuration mode does not include the target configuration mode, the configuration device for the production logistics distribution resources is also used to: obtain the current delivery carrier configuration information from the planned logistics distribution planning information, wherein the delivery carrier configuration information includes the target delivery vehicle that has been configured for each material to be delivered; modify the number of delivery times implemented by the target delivery vehicle corresponding to each material to be delivered, and generate the target delivery configuration information, wherein the configuration result of the production logistics distribution resources includes the target delivery configuration information.

[0169] This embodiment further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0170] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0171] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0172] S1, in the process of using the planned logistics distribution planning information for transportation and distribution, according to the received distribution demand change information, determine the modified configuration information.

[0173] S2, judging whether the distribution carrier configuration mode is the target configuration mode, and if it is judged that the distribution carrier configuration mode is the target configuration mode, performing gene encoding and population initialization based on material target information and alternative cloud vehicle supply information to generate multiple initial structure codes.

[0174] S3, using a preset swarm intelligence global optimization algorithm, performs swarm optimization operations on multiple initial structure codes to generate multiple first structure codes, and based on the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the swarm intelligence global optimization algorithm, performs population update optimization on the multiple first structure codes to generate multiple candidate structure codes.

[0175] S4, detecting the target structure code among multiple candidate structure codes, and providing the distribution configuration scheme corresponding to the target structure code as the configuration result of the production logistics distribution resources.

[0176] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0177] In addition, in combination with the configuration method of production logistics distribution resources in the above embodiments, the present application embodiment can provide a storage medium for implementation. The storage medium stores a computer program; when the computer program is executed by a processor, any configuration method of production logistics distribution resources in the above embodiments is implemented.

[0178] Those skilled in the art should understand that the technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0179] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for configuring production logistics distribution resources, characterized in that: include: In the process of transport and delivery using the planned logistics and delivery planning information, the revised configuration information is determined according to the received delivery demand change information, wherein the delivery demand change information includes material target information of the material to be delivered, and the revised configuration information includes the delivery carrier configuration mode; Determine whether the distribution carrier configuration mode is a target configuration mode, and if it is determined that the distribution carrier configuration mode is a target configuration mode, perform gene encoding and population initialization based on the material target information and the alternative cloud vehicle supply information to generate multiple initial structure codes, wherein the target configuration mode represents a configuration mode based on renting cloud resources as needed, and each of the initial structure codes represents a distribution configuration scheme for distributing the material to be distributed; Using a preset swarm intelligence global optimization algorithm, a swarm optimization operation is performed on the multiple initial structure codes to generate multiple first structure codes, and based on the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the swarm intelligence global optimization algorithm, a population update optimization is performed on the multiple first structure codes to generate multiple candidate structure codes, wherein the swarm intelligence global optimization algorithm is an improved TLBO that uses a hybrid teaching strategy and a hybrid learning strategy as the teaching strategy and learning strategy of the teaching and learning algorithm TLBO respectively, the hybrid teaching strategy includes a cooperative learning strategy and a random individual insertion strategy, the hybrid learning strategy includes a self-learning strategy, a swarm learning strategy and a random insertion strategy, and the distribution cost fitness is determined according to a preset optimization configuration objective function; Among the plurality of candidate structure codes, the target structure code is detected, and the distribution configuration scheme corresponding to the target structure code is provided as the configuration result of the production logistics distribution resources. The plurality of initial structure codes are subjected to group optimization operation by using a preset group intelligence global optimization algorithm to generate a plurality of first structure codes, including: From the plurality of initial structure codes, based on the delivery cost fitness corresponding to each of the initial structure codes, a preset number of first teacher structure codes are selected; Using the hybrid teaching strategy, all the first teacher structure codes are updated to generate a second teacher structure code, and based on the second teacher structure code and the hybrid teaching strategy, population individual teaching is performed on the first student structure codes in the plurality of the initial structure codes to generate a second student structure code, wherein the first student structure code includes all the first structure codes except the first teacher structure code; Using the hybrid learning strategy, the second student structure code is updated with population individual information learning to generate multiple first candidate student structure codes, wherein the first structure code includes the second teacher structure code and the first candidate student structure code, wherein based on the distribution cost fitness of the distribution configuration scheme corresponding to the first structure code and the swarm intelligence global optimization algorithm, the multiple first structure codes are population updated and optimized to generate multiple candidate structure codes, including: Determine all the first structure codes to be subjected to the Nth population update optimization, obtain candidate structure codes, and determine the first difference between the distribution cost fitness corresponding to each candidate structure code and the corresponding historical structure code, wherein N=1, the historical structure code includes the initial structure code, N>1, the historical structure code includes the first structure code subjected to the N-1th population update optimization, and N is a positive integer; Determine whether the first difference is greater than a preset threshold value, and if it is determined that the first difference is not greater than the preset threshold value, select a third teacher structure code from a plurality of the candidate structure codes according to the delivery cost fitness corresponding to the candidate structure code; Using the hybrid teaching strategy, all the third teacher structure codes are updated to generate a fourth teacher structure code, and based on the fourth teacher structure code and the hybrid teaching strategy, the third student structure code in the plurality of candidate structure codes is subjected to population individual teaching to generate a fourth student structure code, wherein the third student structure code includes all the candidate structure codes except the third teacher structure code; Using the hybrid learning strategy, the fourth student structure code is updated with population individual information learning to generate multiple second candidate student structure codes, wherein the candidate structure code corresponding to the Nth population update optimization includes the fourth teacher structure code and the second candidate student structure code, wherein, when it is determined that the first difference is greater than a preset threshold, the method further includes: Obtain the fifth teacher structure code and the third candidate student structure code generated by the N-1 population update optimization; Determine that the candidate structure codes include the fifth teacher structure code and the third candidate student structure code.

2. The method according to claim 1, characterized in that: Updated instruction on structure coding, including: Determine the current candidate structure code to be updated for teaching, wherein the current candidate structure code includes one of the following: the sixth teacher structure code selected based on the distribution cost fitness, the fifth student structure code to be taught to the population individual, each of the current candidate structure codes includes a plurality of first sub-codes, and the gene value of the first sub-code is used to represent the distribution configuration information for distributing a distribution material; Using a preset collaborative algorithm and a first fitness corresponding to the first sub-code, all the first sub-codes corresponding to each of the current candidate structural body codes are processed, and a preset number of first random sub-codes are randomly inserted into the processed first sub-codes to generate corresponding second sub-codes, wherein the gene value of the first random sub-code is used to characterize the distribution configuration information of a randomly generated distribution material, and the first fitness is used to characterize the distribution efficiency of the distribution configuration information corresponding to the corresponding sub-code; Based on the second sub-code, the first target structure code for completing the current teaching is generated, and according to the first fitness corresponding to the second sub-code, the distribution cost fitness corresponding to the first target structure code is determined, wherein the first target structure code includes one of the following structure codes for completing the teaching update: the seventh teacher structure code corresponding to the sixth teacher structure code, and the sixth student structure code corresponding to the fifth student structure code.

3. The method according to claim 2, characterized in that Using the hybrid learning strategy, the corresponding student structure code is processed, including: Determine a plurality of current student structure codes to be updated for population individual information learning, wherein each of the current student structure codes includes a plurality of third sub-codes, and the gene values ​​of the third sub-codes are used to represent the distribution configuration information for distributing a distribution material; The self-learning strategy and the group learning strategy are used in turn to process all the third sub-codes to generate a fourth sub-code, wherein the self-learning strategy is used to guide the third sub-codes to search and update the best sub-code among all the third sub-codes corresponding to the respective corresponding gene values ​​and each current student structure code, and the group learning strategy is used to guide the candidate sub-codes generated by completing the self-learning to search and update based on all the candidate sub-codes corresponding to the multiple current student structure codes; A preset number of second random sub-codes are randomly inserted into the fourth sub-code to generate a corresponding target sub-code, wherein the corresponding candidate student structure code includes the target sub-code.

4. The method according to claim 1, characterized in that: Detecting a target structure code among the plurality of candidate structure codes includes: Determining the delivery cost fitness corresponding to each of the candidate structure codes; Among the multiple candidate structure codes, the candidate structure code with the highest fitness is selected to obtain the target structure code.

5. The method according to claim 1, characterized in that: Get the available cloud vehicle supply information, including: Acquire a distribution vehicle parameter table, wherein the distribution vehicle parameter table includes the material target information, the distribution vehicle target information, and the correspondence relationship information between the material target information and the distribution vehicle target information; In the distribution vehicle parameter table, query the distribution vehicle target information corresponding to the material target information; In the obtained evaluation parameter table, the evaluation parameters corresponding to the delivery vehicle target information are queried, and the delivery vehicle target information and the evaluation parameters corresponding to the delivery vehicle target information are used as the cloud vehicle supply information, wherein the evaluation parameter table includes the delivery vehicle target information, the evaluation parameters and the correspondence information between the delivery vehicle target information and the evaluation parameters, and the evaluation parameters are obtained by processing the delivery vehicle target information using a fuzzy evaluation method to obtain parameters for evaluating the service quality of the corresponding delivery vehicle.

6. The method according to claim 1, characterized in that In the case where it is determined that the delivery carrier configuration does not include the target configuration, the method includes: Acquire current delivery vehicle configuration information from the planned logistics delivery planning information, wherein the delivery vehicle configuration information includes the configured target delivery vehicle for each material to be delivered; The number of distribution times implemented by the target distribution vehicle corresponding to each material to be distributed is modified to generate target distribution configuration information, wherein the configuration result of the production logistics distribution resources includes the target distribution configuration information.

7. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for configuring production logistics distribution resources described in any one of claims 1 to 6 are implemented.

Images