A production logistics optimization method, device, equipment and medium based on an improved hybrid genetic algorithm

Abstract

The application relates to the technical field of production logistics AGV path planning. A production logistics optimization method, device, equipment and medium based on an improved hybrid genetic algorithm are disclosed. The method comprises the following steps: a vertical production logistics system model is constructed, the production logistics system model comprises a crane, an automatic guided vehicle and a plurality of material storage areas; a production logistics system basic database is constructed, and a mapping relationship between the production logistics system basic database and the vertical production logistics system model is constructed to obtain a production logistics system simulation model; and based on a preset improved hybrid genetic algorithm, a target moving path of the automatic guided vehicle between the crane and the plurality of material storage areas is planned according to the production logistics system simulation model. In the application, the target moving path of the automatic guided vehicle between the crane and the plurality of material storage areas is planned based on the improved hybrid genetic algorithm, the automatic guided vehicle transportation time can be significantly shortened, and the overall operation efficiency of the production process can be improved.

Description

Cross-references to related applications

[0001] This application claims priority to Chinese Patent Application No. 202511600600.6, filed on November 4, 2025, entitled "A Method, Apparatus, Equipment and Medium for AGV Path Planning in Port Logistics", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of AGV path planning technology for production logistics, and more specifically, to a production logistics optimization method, apparatus, equipment, and medium based on an improved hybrid genetic algorithm. Background Technology

[0003] Against the backdrop of accelerating globalization and digitalization, production logistics systems, as the core hubs of the global trade network, are undertaking increasingly heavy logistics tasks. With the continuous prosperity of international trade and the increasing diversity of goods, there is a widespread expectation for improved logistics efficiency. Businesses want to gain an advantage in fierce market competition, and governments want to promote high-quality economic development and efficient resource utilization by optimizing logistics efficiency.

[0004] However, in an increasingly complex and dynamic environment, production logistics faces multiple challenges, such as the daily changes in warehousing not keeping up with the surge in cargo flow and low efficiency in resource allocation. These problems not only limit the efficiency of production operations but also have a significant impact on the stable growth of the global economy. Therefore, optimizing the production logistics system has become an urgent task.

[0005] Therefore, finding a production logistics AGV path planning optimization method based on intelligent optimization algorithms is of great significance in both theory and practice, and shows broad application prospects. Summary of the Invention

[0006] In view of the above situation, this application provides a production logistics optimization method, apparatus, equipment and medium based on an improved hybrid genetic algorithm, which aims to solve the above problems or at least partially solve the above problems.

[0007] In a first aspect, this application provides a production logistics optimization method based on an improved hybrid genetic algorithm, including: A vertical production logistics system model is constructed, which includes cranes, automated guided vehicles, and multiple material storage areas; A basic database for the production logistics system is constructed, along with the mapping relationship between the basic database and the vertical production logistics system model, to obtain a simulation model of the production logistics system. Based on a pre-defined improved hybrid genetic algorithm, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned according to the production logistics system simulation model.

[0008] For example, a basic database for a production logistics system is constructed, along with a mapping relationship between the basic database and the vertical production logistics system model, to obtain a production logistics system simulation model, including: The system acquires the straight-line distance from the crane to each material storage area, the Euclidean distance between each material storage area, the transport speed of the automated guided vehicle, and the response time of the yard crane operation in the material storage area, and constructs the basic database of the production logistics system. Based on the vertical production logistics system model, a production logistics weighted network diagram is obtained. In the production logistics weighted network diagram, the cranes and each of the material storage areas are nodes, and the lines connecting adjacent nodes are the movement paths of the automated guided vehicles. Based on the straight-line distance from the crane to each material storage area and the Euclidean distance between each material storage area, the lines connecting adjacent nodes in the production logistics weighted network diagram are weighted to obtain the simulation model of the production logistics system.

[0009] For example, based on a preset improved hybrid genetic algorithm, and according to the production logistics system simulation model, the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas is planned, including: The problem of planning the target movement path of the automated guided vehicle between the crane and multiple material storage areas is transformed into a constrained traveling salesman problem. Based on the improved hybrid genetic algorithm, the traveling salesman problem is solved according to the production logistics system simulation model to obtain the target movement path of the automated guided vehicle between the crane and multiple material storage areas.

[0010] For example, the problem of planning the target movement path of an automated guided vehicle (AGV) between a crane and multiple material storage areas can be transformed into a constrained traveling salesman problem, including: The constraints include the operational sequence of the automated guided vehicle (AGV) when moving between the crane and the multiple material storage areas, and the constraints also include the AGV starting from the crane node and returning to the crane node after traversing all material storage area nodes; Construct an objective function that minimizes the travel time of the automated guided vehicle between the crane and the multiple material storage areas.

[0011] For example, based on the improved hybrid genetic algorithm and according to the production logistics system simulation model, the traveling salesman problem is solved to obtain the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas, including: Population initialization, generating a path vector set, the path vector set including multiple path vectors, each path vector being a movement path of the automated guided vehicle between the crane and the multiple material storage areas; The fitness of each path vector is calculated based on a pre-set fitness function. Based on a pre-set roulette wheel algorithm, the path vector for genetic inheritance into the next generation is determined according to the fitness of each path vector; Based on a pre-set crossover algorithm, crossover operations are performed on the path vectors of genetic inheritance into the next generation to obtain the crossover path vectors. Based on a pre-set mutation algorithm, the cross-path vector is mutated to obtain the mutated path vector. Based on a pre-set optimization mechanism, a target path vector is determined from the path vector of the genetic material entering the next generation and the path vector after mutation. The target path vector is decoded to obtain the target movement path of the automated guided vehicle between the crane and the multiple material storage areas.

[0012] For example, based on a pre-set optimization mechanism, determining a target path vector from the genetic path vector into the next generation and the mutated path vector includes: Based on a pre-set elite retention mechanism, an initial screening path vector is determined from the genetic path vector into the next generation and the mutated path vector. Based on a pre-set swarm search optimization algorithm, the updated path vector is determined according to the initial screening path vector; The population is updated based on the updated path vector until the iteration terminates, and the target path vector is determined.

[0013] For example, the method further includes: A production logistics simulation model based on Plant Simulation was constructed. Based on the genetic algorithm, the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas was planned to obtain the predicted movement path and the parameters of the first iteration process. Based on the preset improved hybrid genetic algorithm, according to the production logistics system simulation model, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned to obtain the target movement path and the parameters of the second iteration process. The production logistics optimization method based on the improved hybrid genetic algorithm is verified based on the estimated movement path and the first iteration process parameters, and the target movement path and the second iteration process parameters.

[0014] Secondly, this application provides a production logistics optimization device based on an improved hybrid genetic algorithm, comprising: The first construction module is used to construct a vertical production logistics system model, which includes cranes, automated guided vehicles and multiple material storage areas. The second construction module is used to construct the basic database of the production logistics system and the mapping relationship between the basic database of the production logistics system and the vertical production logistics system model, so as to obtain the simulation model of the production logistics system. The solution module is used to plan the target movement path of the automated guided vehicle between the crane and multiple material storage areas based on a preset improved hybrid genetic algorithm and the production logistics system simulation model.

[0015] Thirdly, this application provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of a production logistics optimization method based on an improved hybrid genetic algorithm as described in the first aspect.

[0016] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a production logistics optimization method based on an improved hybrid genetic algorithm as described in the first aspect.

[0017] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: This application constructs a production logistics system simulation model and plans the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas based on an improved hybrid genetic algorithm. In practice, it intelligently optimizes the AGV path planning method, which can significantly reduce the transportation time of the AGV, improve the overall efficiency of production operations, and is of great significance for alleviating production station congestion and shortening the dwell time of goods. It also enhances production efficiency competitiveness and economic benefits. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1This is a schematic diagram of an application environment for a production logistics optimization method based on an improved hybrid genetic algorithm according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a production logistics optimization method based on an improved hybrid genetic algorithm according to an embodiment of the present invention. Figure 3 This is a schematic diagram of a production logistics weighting network in one embodiment of the present invention; Figure 4 This is a schematic diagram of the target movement path of the AGV in one embodiment of the present invention; Figure 5 This is a schematic diagram of a model framework built in Plant Simulation according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the optimization results obtained by using the default genetic algorithm in Plant Simulation in one embodiment of the present invention; Figure 7 In one embodiment of the present invention, the spatial distribution of the path and the AGV's driving direction are visualized after being solved by the default genetic algorithm in Plant Simulation. Figure 8 This is a schematic diagram of the optimization results obtained by the improved hybrid genetic algorithm in one embodiment of the present invention; Figure 9 This is a schematic diagram of a production logistics optimization device based on an improved hybrid genetic algorithm in one embodiment of the present invention; Figure 10 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention; Figure 11 This is another structural schematic diagram of a computer device according to one embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such use can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the term "comprising" and its variations should be interpreted as open-ended terms meaning "including but not limited to."

[0021] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0022] As mentioned earlier, current medical guidance services are provided through manual processes or web pages or apps with fixed symptom and disease options, all of which have limitations. To address this technical issue, this application provides a production logistics optimization method based on an improved hybrid genetic algorithm.

[0023] The production logistics optimization method based on an improved hybrid genetic algorithm provided in this invention can be applied to, for example... Figure 1 In this application environment, the device communicates with the server via a network. The server can construct a vertical production logistics system model using the device. This model includes cranes, automated guided vehicles (AGVs), and multiple material storage areas. It also constructs a basic database for the production logistics system and establishes a mapping relationship between this database and the vertical production logistics system model, resulting in a production logistics system simulation model. Based on a pre-defined improved hybrid genetic algorithm, the server plans the target movement path of the AGVs between the cranes and multiple material storage areas according to the simulation model. This application constructs a production logistics system simulation model and plans the target movement path of the AGVs between the cranes and multiple material storage areas using an improved hybrid genetic algorithm. In practice, this intelligently optimizes the AGV path planning method in production logistics, significantly reducing AGV transportation time, improving overall production efficiency, alleviating production congestion, shortening intermediate and product dwell time, and enhancing production competitiveness and economic benefits. The device can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The invention will be described in detail below through specific embodiments.

[0024] Please see Figure 2 As shown, Figure 2 A flowchart illustrating a production logistics optimization method based on an improved hybrid genetic algorithm provided in this embodiment of the invention includes the following steps: S10: Construct a vertical production logistics system model, which includes cranes, automated guided vehicles and multiple material storage areas.

[0025] In one embodiment, based on the analysis of the distribution type of production intermediates and product yards, a vertical layout is more suitable for AGV operations than a horizontal layout. AGV transportation only requires a 90° turn, which can shorten the travel distance and horizontal transportation distance, and it is highly compatible with automation systems. Therefore, a vertical production logistics system is chosen as the research object. The core of the production logistics system model includes cranes, AGVs, and various material storage areas. It is constructed according to the spatial layout relationship of each functional area (cranes, automated guided vehicles (AGVs), and multiple material storage areas). The crane is located in the upper-central front area of ​​the production logistics system model, and a fixed AGV receiving point is set below its landside extension, serving as the "crane-transport vehicle" handover hub. The unloading operation is only initiated when the AGV arrives. The AGV serves as the core transportation carrier connecting the crane and the material storage area.

[0026] S20: Construct a basic database for the production logistics system, and establish the mapping relationship between the basic database and the vertical production logistics system model to obtain a simulation model of the production logistics system.

[0027] In one embodiment, step S20 includes: S21: obtaining the straight-line distance from the crane to each material storage area, the Euclidean distance between each material storage area, the transportation speed of the automated guided vehicle, and the response time of the yard crane operation in the material storage area, and constructing the basic database of the production logistics system.

[0028] S22: A production logistics weighted network diagram is obtained based on the vertical production logistics system model. The cranes and material storage areas in the production logistics weighted network diagram are nodes, and the lines connecting adjacent nodes are the movement paths of the automated guided vehicles. S23: Based on the straight-line distance from the crane to each material storage area and the Euclidean distance between each material storage area, the connection lines of adjacent nodes in the production logistics weighted network diagram are weighted to obtain the simulation model of the production logistics system.

[0029] In one embodiment, the vertical production logistics system model is abstracted into a weighted network graph, such as... Figure 3 As shown in the diagram. The nodes represent cranes and various material storage areas, and the lines connecting adjacent nodes are edges, representing different functional areas and the material flow between them. The automated guided vehicles (AGVs) move along these edges. Figure 3As shown, specifically, the pentagonal nodes 1 to 4 represent automated cranes, while the circular nodes 5 to 24 represent material storage areas. The lines connecting the nodes represent the operating paths of the Automated Guided Vehicles (AGVs). The numbers on the labels between two lines represent the weight of the path, corresponding to the travel distance. For example, the weight of the line between the horizontal line containing node 1 and the horizontal line containing node 5 is 12; the weight of the line between the horizontal line containing node 5 and the horizontal line containing node 9 is 12; the weight of the line between the horizontal line containing node 9 and the horizontal line containing node 13 is 60; the weight of the line between the horizontal line containing node 13 and the horizontal line containing node 17 is 12; the weight of the line between the horizontal line containing node 17 and the horizontal line containing node 21 is 12; the weight of the line between the vertical line containing node 1 and the vertical line containing node 2 is 16; the weight of the line between the vertical line containing node 2 and the vertical line containing node 3 is 24; and the weight of the line between the vertical line containing node 3 and the vertical line containing node 4 is 16. The weight represents the distance of the path, configured based on the straight-line distance from the crane to each material storage area and the Euclidean distance between the material storage areas. Nodes are connected to other nodes by lines, reflecting the flow of goods. Figure 3 The center lines are horizontal and vertical lines; in other embodiments, diagonal lines may also be included. The weight of the diagonal lines is calculated using the weights of the horizontal and vertical lines at the corresponding nodes. The response time for material storage area crane operations is also configured at each node.

[0030] In one embodiment, this application addresses the path planning of an automated guided vehicle (AGV) between a crane and multiple material storage areas. The planning results can be referenced. Figure 4 As shown, Figure 4 Node 1 is a crane, nodes 2 to 9 are material storage areas, and the lines connecting the two nodes are the paths of automated guided vehicles (AGVs).

[0031] S30: Based on a preset improved hybrid genetic algorithm, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned according to the production logistics system simulation model.

[0032] In one embodiment, step S30, based on a preset improved hybrid genetic algorithm and according to the production logistics system simulation model, plans the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas, including: S31: The problem of planning the target movement path of the automated guided vehicle between the crane and multiple material storage areas is transformed into a constrained traveling salesman problem. S32: Based on the improved hybrid genetic algorithm, the traveling salesman problem is solved according to the production logistics system simulation model to obtain the movement path of the automated guided vehicle between the crane and multiple material storage areas.

[0033] In one embodiment, the optimization problem of AGV transportation between a crane and multiple material storage areas is explored, based on the principle of even distribution among multiple material storage areas. Since the path distance between the crane and each material storage area varies, and the running time of the yard cranes in each material storage area is inconsistent, the processing time of each material storage area differs. Therefore, determining the optimal processing order of the material storage areas to reduce the transportation time between any two material storage areas, and further reduce the total transportation time, becomes the optimization objective of this application.

[0034] In one embodiment, constraints are constructed, including the operational sequence of the automated guided vehicle (AGV) as it moves between the crane and the plurality of material storage areas, and the constraints also include the AGV departing from the crane node and returning to the crane node after traversing all material storage area nodes.

[0035] In one embodiment, the sequence of operations for the automated guided vehicle (AGV) to move between the crane and the plurality of material storage areas includes: a. The crane lifts the designated stockpile of raw materials from the ship according to the loading and unloading sequence list, transfers it to the AGV at the designated loading and unloading node, and the crane continues to perform the next task; b. After the AGV transports the raw materials to the designated loading / unloading storage area node within the target raw material yard, it transfers the container to the yard crane at that storage area node. The AGV then continues to the next loading / unloading area node. c. The yard crane precisely places the raw materials lifted from the AGV into the designated location within the yard. The yard crane then returns to the loading / unloading node, ready for the next loading / unloading operation.

[0036] The work sequence constraints of the automated guided vehicles (AGVs) in steps a to c, when moving between the crane and the multiple material storage areas, are reflected in the work response time.

[0037] The constraint condition for the automated guided vehicle to start from the crane node and return to the crane node after traversing all material storage area nodes is the path constraint condition.

[0038] In one embodiment, an objective function is constructed to minimize the travel time of the automated guided vehicle between the crane and the plurality of material storage areas.

[0039] In one embodiment, under the premise of satisfying the constraints, we seek an optimal path that starts from the crane, traverses all material storage areas, and finally returns to the crane, minimizing the total transportation time T(R). The objective function is as follows:

[0040] In the formula, Represents the nodes in the distance matrix D To node The transportation time; the distance matrix D is obtained based on the weighted production logistics network diagram. n is the number of nodes in the material storage area, and satisfies non-negativity ( ≥0), symmetry ( = ) and diagonal zero value ( =0), where the subscripts i and j represent different material storage area nodes in the distance matrix. i can be understood as representing a node. j represents a node R = [r1, r2, ..., r9, r 10 [1, P, 1] represents the expanded complete path, where P represents the sequence of all intermediate nodes in the path except for the first and last crane nodes. Fixed first and last crane nodes are added to individual P to construct a closed path R. First, path R is initialized starting from node 1. Nodes of individual P are added sequentially to path R. Crane node 1 is added to the end of path R to form a closed path R.

[0041] For example, in one embodiment, it is necessary to plan the path of the AGV at a crane node 1 and eight material storage area nodes 2 to 9. The path must start and end at node 1 (crane), that is, the path form is 1→ →…→ →1, where { , ..., Let} represent all permutations of nodes 2 to 9. Let the path sequence be... ,in , This refers to the number of material storage areas.

[0042] In one embodiment, step S30, based on the improved hybrid genetic algorithm and according to the production logistics system simulation model, solves the traveling salesman problem to obtain the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas, including: S321: Population initialization, generating a path vector set, the path vector set including multiple path vectors, each path vector being a movement path of the automated guided vehicle between the crane and the multiple material storage areas.

[0043] In one embodiment, this application uses coding design to generate a path vector set based on the discrete combination characteristics of automated guided vehicle (AGV) path planning. The specific process is as follows: The coding design takes into account the discrete combination characteristics of the automated guided vehicle path planning and adopts the Integer Permutation Encoding (IPE) method. The encoding is defined as follows: an individual is represented as a permutation vector (path vector) of length n, P = [p1, p2, ..., pn], where p i ∈{2, 3, ..., n} and satisfy , , here i , j The nodes are represented by the vector; the individuals are represented by vectors that do not contain crane nodes. It is understandable that the mapping relationships between the actual crane node codes, the actual material storage area node codes, and the individuals in the population can be pre-defined so that subsequent population generation can effectively utilize the relevant basic data.

[0044] From a physical perspective, each individual corresponds to the sequence of n material storage areas visited by the AGV after it departs from the crane. This encoding method naturally satisfies the path constraint of "no repeated visits". The initial population is generated through uniform random sampling. First, permutation generation is performed by calling P = randperm(n) + 1 for each individual. Here, randperm(n) generates a random permutation from 1 to n, and adding 1 maps it to a material storage area number from 2 to n. Then, validity verification is performed to ensure that all individuals in the population are valid permutations.

[0045] For example, in one embodiment, it is necessary to plan the path of an AGV to a crane node 1 and eight material storage area nodes 2 to 9. The path vector set includes path vectors such as P=[2, 3, 4, 5, 6, 7, 8, 9], P=[2, 4, 3, 5, 6, 7, 8, 9]..., and the path vector set includes all permutations of 2 to 9.

[0046] S322: Calculate the fitness of each path vector based on a pre-set fitness function.

[0047] In one embodiment, the fitness function f(R) maps the total transport time T(R) to a quantification of the optimization objective:

[0048] Fitness score is a quantitative indicator for evaluating the quality of an individual and is usually directly related to the objective function of the problem.

[0049] S323: Based on a pre-set roulette wheel algorithm, determine the path vector for genetic inheritance to enter the next generation according to the fitness of each path vector.

[0050] In one embodiment, the probability allocation rule for roulette wheel selection is as follows:

[0051] In the formula p i f is the probability of individual i being selected; i The fitness value of individual i is calculated based on the total transport time T(R) corresponding to individual i; N is the number of individuals in the population. Each individual represents a path vector.

[0052] In one embodiment, the fitness values ​​of all individuals are normalized using formula (3) to establish the probability distribution p for each individual. i This step ensures that the sum of the probabilities of all individuals being selected is 1, satisfying the basic requirements of probability theory. Based on the normalized probability distribution, a cumulative probability vector c=[ ],in The cumulative probability vector is used for the random selection operation in the roulette wheel selection process. For each selection operation, a random number is generated. Then, select the option that satisfies the condition. Individual k is the path vector for genetic inheritance to enter the next generation.

[0053] Step S33 determines the path vectors for all N genes to enter the next generation.

[0054] In one embodiment, step S33 further includes, based on an elite retention mechanism, sorting the fitness of all path vectors in the initial population from low to high according to the fitness of each path vector; determining, based on the sorting result, a first portion of path vectors to be inherited into the next generation; the first portion of path vectors to be inherited into the next generation are selected according to a preset fitness value or a preset selection ratio, selecting path vectors with low fitness; and, based on a roulette wheel algorithm, determining a second portion of path vectors to be inherited into the next generation according to the remaining path vectors in the population other than the first portion of path vectors to be inherited into the next generation; the first portion of path vectors to be inherited into the next generation and the second portion of path vectors to be inherited into the next generation constitute all N path vectors to be inherited into the next generation.

[0055] S324: Based on a pre-set crossover algorithm, perform crossover operations on the path vectors of genetic inheritance into the next generation to obtain the crossover path vectors; In one embodiment, the crossover operator is Ox, which is a widely used crossover operator in genetic algorithms. It generates new offspring by recombining the chromosome sequences of two parent individuals. It can bring about new genetic combinations by exchanging controlled segments between parent chromosomes while maintaining the relative order of elements. Two parent chromosome sequences of individuals are randomly selected from the path vector of genetic inheritance into the next generation. and These two individuals will serve as the basis for generating new offspring. In the parent individual... Two cut points, s and e, are randomly selected from the gene sequence (e.g., s=3, e=5), and the sequence is preserved. The fragments F = [P1(e), ..., P1(s)]. New offspring are produced through a multipoint crossover mechanism, which involves selecting multiple crossover points, splitting parental gene fragments, and recombinating them to create new genetic combinations.

[0056] from Elements not present in fragment F are selected sequentially and filled into the vacant positions of the offspring in their original order. The offspring structure is C = [non-F element P2, F, non-F element P2]. This construction method ensures that offspring individuals maintain the relative order of the parent pathways while introducing new genetic variations.

[0057] For example, two parent individuals are P1=[2, 4, 3, 5, 9, 7, 6, 8] and P2=[3, 6, 2, 5, 8, 7, 4, 9]. If s=3 and e=5, then F1=[9, 5, 3] and F2=[8, 5, 2]. Substituting F1 into P2, we get offspring C2=[3, 6, 9, 5, 3, 7, 4, 8]. Substituting F2 into P1, we get offspring C1=[2, 4, 8, 5, 2, 7, 6, 9].

[0058] S325: Based on a pre-set mutation algorithm, perform mutation operation on the cross-path vector to obtain the mutated path vector.

[0059] In one embodiment, randomly swapping two elements in an individual's gene sequence increases population diversity, prevents the algorithm from getting stuck in local optima, and introduces new gene combinations to help the algorithm explore a wider solution space, thereby improving global search capabilities: with a preset mutation probability P m (like Two distinct positions sˊ and eˊ are randomly selected (e.g., sˊ=2, eˊ=6). The elements (nodes) at these two positions will be used for subsequent element swapping operations. Swapping the elements at these two positions creates new offspring individuals. This step introduces new genetic variation by altering the individual's gene sequence. For example, consider an original individual [3, 7, 2, 5, 8, 4, 6, 9], after swapping positions 2 and 6, the offspring individuals become [3, 4, 2, 5, 8, 7, 6, 9].

[0060] S326: Based on a pre-set optimization mechanism, determine the target path vector from the genetic path vector into the next generation and the mutated path vector.

[0061] In one embodiment, the optimization mechanism includes updating the population through an optimization algorithm and determining the target path vector after the iteration termination condition (such as satisfying the maximum number of iterations) is met.

[0062] In one embodiment, the path vector of the genetic material entering the next generation (i.e., the parent population P) t , t∈[1,N]) and the mutated path vector (i.e., the offspring population C) t A temporary population O is formed by t∈[1,N]). t The optimization algorithm is used to find the individual with the lowest fitness in the temporary population, and the population in step S31 is updated using the individual with the lowest fitness.

[0063] In one embodiment, S3261: Based on a pre-set elite retention mechanism, a preliminary screening path vector is determined from the genetic path vector into the next generation and the mutated path vector; S3262: Based on a pre-set group search optimization algorithm, determine the updated path vector according to the initial screening path vector; S3263: Update the population according to the updated path vector until the iteration terminates, and determine the target path vector.

[0064] In one embodiment, the optimization algorithm includes an elite-preserving algorithm and a temporary population O. t The individuals will be sorted in ascending order based on their fitness values, and the top N individuals will be retained as the initial screening path vectors to form the initial screening population O. t The mathematical expression for ´ is as follows:

[0065] The `Sort` function sorts the individuals in the temporary population (the set of parent and offspring individuals) in ascending order of their fitness values. The `Top-N` function selects the top N individuals with the lowest fitness from the sorted population to form the initial screening population O. t ´.

[0066] In one embodiment, the optimization algorithm further includes a group search optimization (GSO) algorithm. Based on a pre-set GSO algorithm, an update path vector is determined in the initial screening population (i.e., searching for the individual with the lowest fitness in the initial screening population). Each screening yields one update path vector, and finally N update path vectors are obtained, resulting in a new generation population P. t+1 With the new generation population P t+1 The original population (i.e., the population in step S31) is updated until the iteration terminates, and the target path vector (the individual with the lowest fitness) is determined.

[0067] After each iteration, the improved hybrid genetic algorithm in this application checks whether the termination condition is met, such as reaching the maximum number of iterations or the fitness value no longer significantly improving. If the termination condition is met, the improved hybrid genetic algorithm stops iterating and outputs the current optimal solution; otherwise, the improved hybrid genetic algorithm continues to execute the next generation of iterations. The improved hybrid genetic algorithm terminates iteration when the search reaches a preset number of iterations and outputs the historical optimal path and its corresponding transportation time.

[0068] S327: Decode the target path vector to obtain the target movement path of the automated guided vehicle between the crane and the multiple material storage areas.

[0069] In one embodiment, the target movement path includes the optimal AGV path sequence and the corresponding total transportation distance. The method also includes visually displaying the spatial distribution of the path and the AGV's travel direction to provide a basis for decision-making in actual production logistics scheduling. For example, in one embodiment, it is necessary to plan the path of an AGV from one crane node 1 to eight material storage area nodes 2 to 9. The output optimal path sequence is in the format "1-4-2-3-8-5-9-6-7-1", where r2→r9 is the optimal access order of the eight material storage areas, and the total transportation distance corresponding to this path is 136. The graphical interface displays the optimal path as follows: Figure 4 As shown.

[0070] The key aspects of this application lie in the collaborative design and parameterization of each operational step: Encoding Design: Integer Permutation Encoding (IPE) is used to directly represent the access order of the material storage area, naturally satisfying the "no duplicate access" constraint, which perfectly aligns with the essence of the TSP problem. Selection Strategy: A roulette wheel selection method is used instead of simple random selection. This ensures that superior individuals with lower fitness have a greater chance of being selected for inheritance, thereby accelerating convergence and improving search efficiency. Crossover Operation: The sequential crossover operator (OX) is used. This operator effectively preserves the relative order (i.e., path segments) in the parent gene sequence when generating offspring, which is crucial for generating high-quality, valid path solutions. Mutation Operation: Exchange mutation is used (with a fixed mutation probability P). m =0.3), by randomly swapping the positions of two material storage areas along the path, actively injecting new genes into the population is the most crucial means to maintain population diversity and avoid the algorithm getting trapped in local optima. Employing Group Search Optimization (GSO) or an elite retention strategy to preserve the best individuals of the current generation for direct entry into the next generation during iterations is an important mechanism to ensure algorithm convergence and prevent result degradation.

[0071] In one embodiment, the method further includes: A production logistics simulation model based on Plant Simulation was constructed. Based on the genetic algorithm, the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas was planned to obtain the predicted movement path and the parameters of the first iteration process. Based on the preset improved hybrid genetic algorithm, according to the production logistics system simulation model, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned to obtain the target movement path and the parameters of the second iteration process. The production logistics optimization method based on the improved hybrid genetic algorithm is verified based on the estimated movement path and the first iteration process parameters, and the target movement path and the second iteration process parameters.

[0072] In one embodiment, a production logistics simulation model framework is built in Plant Simulation based on a vertical production logistics system, such as... Figure 5 As shown, the framework layout is explained in detail below: The crane (S10) is located in the upper center of the diagram; Material storage areas (S2-S9) are located at the bottom of the diagram, and there are a total of 8 material storage areas. StoreZones: Processes the above 9 nodes and calculates the distances between them. It handles node information processing and distance calculation. First, it initializes variables and clears related table data. Then, it iterates through the nodes, filters out nodes belonging to the "Place" class, records their information in the table, and constructs a node mapping. Next, it calculates the distances between nodes and stores the results. Finally, it displays a prompt message to show the path length. Data table Distances: Distances between storage areas and between cranes and storage areas; Data Zones: Used to manage and display information for all material storage areas; GAWizard: A genetic algorithm wizard in Plant Simulation used to optimize logistics operations in production. The optimization parameters are sequences in a Zones table, and fitness is calculated using the LengthOfZone method. The `ShowZone` method connects the nodes using a connector. The code first iterates through all nodes, deletes the network arc object with the inner class name "NwArc", and handles index updates caused by node deletions. Then, it creates a connector object, sequentially connects the nodes in the region table, and closes the loop connection.

[0073] LengthOfZone: Used to calculate the length of the material storage zone, serving as the fitness of the genetic code. Implements circular path distance calculation. Initializes the total distance, traverses the region table to calculate and accumulate the distances between adjacent nodes, and finally closes the loop and outputs the total distance.

[0074] Furthermore, the solution results obtained using the default genetic algorithm in Plant Simulation are as follows: In the Plant Simulation environment, the Traveling Salesman Problem (TSP) is formalized as a constrained variant of the TSP. Specifically, the problem involves eight stations, each representing a material storage area. An AGV (Automated Guided Vehicle) must depart from a crane, visit all material storage areas in a specific order, and eventually return to the crane. The core of the problem is to find a path that satisfies the job sequence constraints while minimizing the total transport time.

[0075] In the Plant Simulation environment, the default genetic algorithm parameters are as follows: The population size is 100, the number of iterations is 20, the selection strategy is to randomly select 1 offspring from 4 offspring for reproduction, the fitness is referenced to the minimum value (the lower the fitness value, the better), the crossover operator is sequential crossover, and the crossover probability is 0.8. Optimization trends such as Figure 6 As shown, the optimization results are as follows: Figure 7As shown in the figure, the results demonstrate the performance of the default genetic algorithm in solving the automated guided vehicle (AGV) path planning problem. Figure 6 The horizontal axis represents the number of iterations in the genetic algorithm, and the vertical axis represents the fitness value, where a lower fitness value indicates a better path. Figure 6 It contains three curves, representing the changes in the fitness values ​​of the best, average, and worst individuals with each iteration. Figure 6 The following points can be observed: Fast convergence: In the first few generations (about the first 5 generations), the fitness values ​​of the best solution, the average solution, and the worst solution all decrease rapidly. This shows that the algorithm can quickly find a better solution in the initial stage, and the global search capability of the genetic algorithm enables it to converge quickly in this stage. It can quickly explore the solution space and find a better path by relying on crossover and mutation operations.

[0076] Gradually stabilizing: Around the 5th generation, the downward trend of the three curves begins to slow down and gradually stabilizes. This means that the algorithm is gradually approaching the optimal solution and the improvement is becoming smaller. At this stage, the algorithm mainly relies on local search to fine-tune the current solution, thereby further improving the quality of the solution.

[0077] Stability: From the 10th generation onwards, the three curves generally stabilize and the fitness values ​​do not change significantly. This indicates that the algorithm has basically converged and found a relatively optimal solution. The small difference between the best, average, and worst solutions indicates that the algorithm has good stability and can reliably find a relatively optimal solution.

[0078] Optimal Solution: Ultimately, the fitness value of the best solution stabilized at approximately 173, as did the fitness values ​​of the average and worst solutions. This indicates that the algorithm successfully found a relatively optimal AGV path planning scheme, significantly reducing the total transportation time.

[0079] Furthermore, the solution results of the improved hybrid genetic algorithm in Matlab are as follows: In Matlab, the parameters for the improved hybrid genetic algorithm are as follows: The population size is 100, the number of iterations is 25, the selection strategy is roulette wheel selection, the fitness reference rule is to pursue the minimum value (i.e., the smaller the fitness value, the better), the crossover operator is sequential crossover, the crossover probability is set to 0.7, and the mutation probability is set to 0.3. Optimization trends such as Figure 8 As shown, the optimization results are as follows: Figure 4 As shown, the total transportation time shows a significant decreasing trend with the increase of the number of iterations, indicating that the improved hybrid genetic algorithm is constantly searching for better solutions.

[0080] In the first three generations, the decrease in total transport time is particularly significant starting from the second generation, dropping from approximately 230 units initially to approximately 170 units, demonstrating the algorithm's rapid convergence in the early iterations. This rapid convergence at this stage can likely be attributed to the high diversity of the population and the efficient exploration of the knowledge space by genetic operators such as crossover and mutation.

[0081] After the 7th generation, the total transport time continues to decrease and then stabilizes at approximately 136 units. This indicates that the algorithm is gradually approaching the optimal solution or a local optimum. At this point, population diversity may decrease, and the exploration ability of genetic operators weakens, leading to a smaller improvement margin. During this stage, the algorithm primarily reduces fitness by fine-tuning the current solution.

[0082] also, Figure 8 No significant oscillations or rebounds were observed, indicating that the algorithm has good stability and can avoid getting trapped in local optima and remain in a relatively good solution region. This result is attributed to reasonable parameter settings and the selection of genetic operators, such as appropriate crossover and mutation probabilities, as well as the optimization strategy.

[0083] Furthermore, the improvements of the genetic algorithm are explained in detail, and the convergence speed, final total transport distance, stability, and diversity of the two algorithms are compared. Specifically: The parameter settings are basically the same in both environments, including population size, number of iterations, fitness reference, crossover operator, and crossover probability. This application improves the algorithm by introducing mutation probability and innovatively introducing a roulette wheel selection strategy.

[0084] Further detailed comparison is needed, as follows; Convergence Speed: In the improved hybrid genetic algorithm, due to the introduction of mutation probability, the initial convergence speed is slightly slower than the default genetic algorithm in the Plant Simulation environment. However, this initial increase in diversity helps the algorithm find better solutions in subsequent iterations. The roulette wheel selection strategy exhibits a faster convergence speed because it can more effectively utilize fitness information, prioritizing individuals with lower fitness, thus accelerating the algorithm's convergence. While the "4-to-1" selection strategy in the default PlantSimulation algorithm simplifies the selection process, it may lead to the neglect of individuals with lower fitness, thereby affecting the algorithm's convergence speed.

[0085] Final Fitness Value: The improved hybrid genetic algorithm achieves a significantly lower final fitness value than the default Plant Simulation algorithm. This indicates that the improved genetic algorithm can find better solutions during optimization. The mutation operation allows the algorithm to explore a wider solution space during the search, thus avoiding getting trapped in local optima. The significantly lower final fitness value of the genetic algorithm compared to the default Plant Simulation algorithm demonstrates that the genetic algorithm can find better solutions during optimization. This is because the roulette wheel selection strategy more accurately reflects the fitness of individuals, thereby improving the accuracy of selection.

[0086] Stability and Diversity: The fitness curve in the default genetic algorithm in Plant Simulation is relatively stable in the later stages, showing good stability. However, this stability may come at the cost of solution diversity, making it difficult for the algorithm to escape local optima. Although the fitness curve of the improved hybrid genetic algorithm also tends to stabilize in the later stages, the overall fitness value is lower. The roulette wheel selection strategy helps maintain population diversity, thereby enhancing the algorithm's global search capability.

[0087] Impact of Mutation Rate: The mutation probability introduced in the improved hybrid genetic algorithm plays a crucial role in the optimization process. An appropriate mutation rate not only helps increase population diversity but also prevents premature convergence, thereby improving the algorithm's global search capability.

[0088] Impact of Selection Strategy: The roulette wheel selection strategy in the improved hybrid genetic algorithm plays a crucial role in the optimization process. This strategy determines the selection probability of individuals based on the proportion of their fitness values, making more effective use of fitness information and prioritizing individuals with low fitness. While the "4-to-1" selection strategy in the default genetic algorithm of Plant Simulation helps maintain population diversity, it may affect the algorithm's convergence speed and solution quality.

[0089] In this application, the optimization performance can be significantly improved by combining the elite retention mechanism with the GSO algorithm during the optimization process. The core advantages include accelerating convergence, maintaining solution diversity, enhancing global search capabilities, and optimizing resource allocation efficiency.

[0090] As can be seen, in the above-mentioned scheme, this application constructs a production logistics system simulation model and plans the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas based on an improved hybrid genetic algorithm. In practice, it intelligently optimizes the AGV path planning method, which can significantly reduce the transportation time of the AGV, improve the overall efficiency of production operations, and is of great significance for alleviating production congestion and shortening the time ships spend in port. It also enhances production efficiency competitiveness and economic benefits.

[0091] This application significantly reduces the total AGV transportation time and improves production efficiency: through an improved hybrid genetic algorithm (integrating roulette wheel selection, sequential crossover, exchange mutation, and optimization mechanisms), the algorithm possesses stronger global search capabilities and the ability to avoid premature convergence. This enables the algorithm to find a better AGV path sequence. Through simulation verification and result output, experimental data shows that the total transportation time obtained by the improved hybrid genetic algorithm in this application (136 units) is significantly lower than the result of the default genetic algorithm in Plant Simulation (approximately 173 units), with a transportation efficiency improvement of approximately 21%. This means that the time required for AGVs to complete the same transportation task is greatly shortened, directly reducing the dwell time of intermediate parts and products in production, accelerating the flow of goods, and thus improving the overall throughput and operational efficiency of production.

[0092] This application enhances the convergence and stability of the algorithm, ensuring the reliability and usability of the optimization results. Through a roulette wheel selection strategy, which selects individuals based on their fitness values ​​with probability, superior individuals have a higher probability of being passed on to the next generation, thus accelerating the initial convergence speed and approaching the optimal solution region more quickly. The introduced optimization mechanism ensures that the optimal individuals in each generation are not destroyed by crossover and mutation operations, and are unconditionally retained to the next generation. This effectively prevents algorithm degradation due to random operations, guaranteeing the stability and convergence of the optimization process. Simulation results ( Figure 8 The improved algorithm's curve shows a smooth decline followed by a stable trend, with significantly less fluctuation than the default algorithm. Figure 6 This provides a reliable and consistent optimization solution for production scheduling.

[0093] This application can improve the global optimization capability of the algorithm and avoid getting trapped in local optima: by setting a fixed mutation probability (such as P). m By using operations such as 0.3 and cross-mutation, new gene combinations are actively introduced into the population, significantly increasing population diversity. This allows the algorithm to continuously explore undiscovered regions in the solution space, effectively breaking the limitation of traditional algorithms that are prone to getting trapped in local optima, thus increasing the probability of discovering the globally optimal or better path planning scheme.

[0094] This application provides intuitive and scientific decision support for production scheduling: through visualization, the optimized path sequence (such as "1→4→2→3→8→5→9→6→7→1") and its corresponding total transportation time are graphically displayed in the PlantSimulation interface (nodes, paths, and directions are marked with different colors or icons), transforming abstract optimization results into clear scheduling instructions. This greatly facilitates the understanding and execution by production schedulers, providing them with intuitive and scientific data support and decision-making basis, and reducing the uncertainty and inefficiency caused by relying on experience-based scheduling.

[0095] It should be understood that the sequence number of each step in the above embodiments does not imply 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 invention.

[0096] The production logistics optimization algorithm based on the improved hybrid genetic algorithm proposed in this application can be applied to production logistics scenarios in electronic information manufacturing and high-end equipment production.

[0097] In one embodiment, a production logistics optimization device based on an improved hybrid genetic algorithm is provided, which corresponds one-to-one with the production logistics optimization method based on an improved hybrid genetic algorithm described in the above embodiments. For example... Figure 9 As shown, the production logistics optimization device based on the improved hybrid genetic algorithm includes a first construction module 101, a second construction module 102, and a solution module 103. Detailed descriptions of each functional module are as follows: The first construction module 101 is used to construct a vertical production logistics system model, which includes a crane, an automated guided vehicle, and multiple material storage areas. The second construction module 102 is used to construct a basic database of the production logistics system and the mapping relationship between the basic database of the production logistics system and the vertical production logistics system model, so as to obtain a simulation model of the production logistics system. The solver module 103 is used to plan the target movement path of the automated guided vehicle between the crane and multiple material storage areas based on a preset improved hybrid genetic algorithm and the production logistics system simulation model.

[0098] Specifically, the second construction module 102 is also used to obtain the straight-line distance from the crane to each material storage area, the Euclidean distance between each material storage area, the transportation speed of the automated guided vehicle and the response time of the yard crane operation in the material storage area, and to construct the basic database of the production logistics system. Based on the vertical production logistics system model, a production logistics weighted network diagram is obtained. In the production logistics weighted network diagram, the cranes and each of the material storage areas are nodes, and the lines connecting adjacent nodes are the movement paths of the automated guided vehicles. Based on the straight-line distance from the crane to each material storage area and the Euclidean distance between each material storage area, the lines connecting adjacent nodes in the production logistics weighted network diagram are weighted to obtain the simulation model of the production logistics system.

[0099] Specifically, the solver module 103 is also used to transform the target movement path planning problem of the automated guided vehicle between the crane and multiple material storage areas into a constrained traveling salesman problem. Based on the improved hybrid genetic algorithm, the traveling salesman problem is solved according to the production logistics system simulation model to obtain the movement path of the automated guided vehicle between the crane and multiple material storage areas.

[0100] Specifically, the solver module 103 is also used to construct constraints, which include the operation sequence of the automated guided vehicle when moving between the crane and the multiple material storage areas, and the constraints also include the automated guided vehicle starting from the crane node and returning to the crane node after traversing all material storage area nodes; Construct an objective function that minimizes the travel time of the automated guided vehicle between the crane and the multiple material storage areas.

[0101] Specifically, the solution module 103 is also used for population initialization and generating a path vector set, which includes multiple path vectors, each of which is a movement path of the automated guided vehicle between the crane and the multiple material storage areas; The fitness of each path vector is calculated based on a pre-set fitness function. Based on a pre-set roulette wheel algorithm, the path vector for genetic inheritance into the next generation is determined according to the fitness of each path vector; Based on a pre-set crossover algorithm, crossover operations are performed on the path vectors of genetic inheritance into the next generation to obtain the crossover path vectors. Based on a pre-set mutation algorithm, the cross-path vector is mutated to obtain the mutated path vector. Based on a pre-set optimization mechanism, a target path vector is determined from the path vector of the genetic material entering the next generation and the path vector after mutation. The target path vector is decoded to obtain the target movement path of the automated guided vehicle between the crane and the multiple material storage areas.

[0102] Specifically, the solver module 103 is also used to determine the initial screening path vector from the genetic path vector into the next generation and the mutated path vector based on a pre-set elite retention mechanism. Based on a pre-set swarm search optimization algorithm, the updated path vector is determined according to the initial screening path vector; The population is updated based on the updated path vector until the iteration terminates, and the target path vector is determined.

[0103] Specifically, the device also includes a verification module, which is used to build a production logistics simulation model based on Plant Simulation, and based on a genetic algorithm, to plan the target movement path of the automated guided vehicle between the crane and multiple material storage areas, and to obtain the estimated movement path and the parameters of the first iteration process. Based on the preset improved hybrid genetic algorithm, according to the production logistics system simulation model, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned to obtain the target movement path and the parameters of the second iteration process. The production logistics optimization method based on the improved hybrid genetic algorithm is verified based on the estimated movement path and the first iteration process parameters, and the target movement path and the second iteration process parameters.

[0104] This invention provides a production logistics optimization device based on an improved hybrid genetic algorithm. First, a preliminary ranking result of candidate questions is obtained through semantic matching. Then, a scheme for optimizing the question answering engine based on entity alignment is proposed. The ranking result of candidate questions is re-ranked through entity alignment, so that more matching candidate questions are selected. This can effectively avoid the generalization ability defects of the model, greatly improve the effect of entity matching, and improve the effect of the question answering engine.

[0105] Specific limitations regarding the production logistics optimization device based on the improved hybrid genetic algorithm can be found in the limitations of the production logistics optimization method based on the improved hybrid genetic algorithm mentioned above, and will not be repeated here. Each module in the aforementioned production logistics optimization device based on the improved hybrid genetic algorithm can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0106] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 10As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used for communication with external devices via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a production logistics optimization method based on an improved hybrid genetic algorithm on the server side.

[0107] In one embodiment, a computer device is provided, which may be a device terminal, and its internal structure diagram may be as follows: Figure 11 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a production logistics optimization method based on an improved hybrid genetic algorithm on the device side.

[0108] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: A vertical production logistics system model is constructed, which includes cranes, automated guided vehicles, and multiple material storage areas; A basic database for the production logistics system is constructed, along with the mapping relationship between the basic database and the vertical production logistics system model, to obtain a simulation model of the production logistics system. Based on a pre-defined improved hybrid genetic algorithm, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned according to the production logistics system simulation model.

[0109] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: A vertical production logistics system model is constructed, which includes cranes, automated guided vehicles, and multiple material storage areas; A basic database for the production logistics system is constructed, along with the mapping relationship between the basic database and the vertical production logistics system model, to obtain a simulation model of the production logistics system. Based on a pre-defined improved hybrid genetic algorithm, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned according to the production logistics system simulation model.

[0110] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and device side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0111] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0112] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0113] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A production logistics optimization method based on an improved hybrid genetic algorithm, characterized in that, include: A vertical production logistics system model is constructed, which includes cranes, automated guided vehicles, and multiple material storage areas; A basic database for the production logistics system is constructed, along with the mapping relationship between the basic database and the vertical production logistics system model, to obtain a simulation model of the production logistics system. Based on a pre-defined improved hybrid genetic algorithm, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned according to the production logistics system simulation model.

2. The method according to claim 1, characterized in that, A basic database for the production logistics system is constructed, along with the mapping relationship between this basic database and the vertical production logistics system model, to obtain a simulation model of the production logistics system, including: The system acquires the straight-line distance from the crane to each material storage area, the Euclidean distance between each material storage area, the transport speed of the automated guided vehicle, and the response time of the yard crane operation in the material storage area, and constructs the basic database of the production logistics system. Based on the vertical production logistics system model, a production logistics weighted network diagram is obtained. In the production logistics weighted network diagram, the cranes and each of the material storage areas are nodes, and the lines connecting adjacent nodes are the movement paths of the automated guided vehicles. Based on the straight-line distance from the crane to each material storage area and the Euclidean distance between each material storage area, the lines connecting adjacent nodes in the production logistics weighted network diagram are weighted to obtain the simulation model of the production logistics system.

3. The method according to claim 1, characterized in that, Based on a pre-defined improved hybrid genetic algorithm, and according to the production logistics system simulation model, the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas is planned, including: The problem of planning the target movement path of the automated guided vehicle between the crane and multiple material storage areas is transformed into a constrained traveling salesman problem. Based on the improved hybrid genetic algorithm, the traveling salesman problem is solved according to the production logistics system simulation model to obtain the target movement path of the automated guided vehicle between the crane and multiple material storage areas.

4. The method according to claim 3, characterized in that, The problem of planning the target movement path of an automated guided vehicle (AGV) between a crane and multiple material storage areas is transformed into a constrained traveling salesman problem, including: The constraints include the work sequence of the automated guided vehicle when it moves between the crane and the multiple material storage areas, and the constraints also include the automated guided vehicle starting from the crane node and returning to the crane node after traversing all material storage area nodes; Construct an objective function that minimizes the travel time of the automated guided vehicle between the crane and the multiple material storage areas.

5. The method according to claim 3, characterized in that, Based on the improved hybrid genetic algorithm, and according to the production logistics system simulation model, the traveling salesman problem is solved to obtain the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas, including: Population initialization, generating a path vector set, the path vector set including multiple path vectors, each path vector being a movement path of the automated guided vehicle between the crane and the multiple material storage areas; The fitness of each path vector is calculated based on a pre-set fitness function. Based on a pre-set roulette wheel algorithm, the path vector for genetic inheritance into the next generation is determined according to the fitness of each path vector; Based on a pre-set crossover algorithm, crossover operations are performed on the path vectors of genetic inheritance into the next generation to obtain the crossover path vectors. Based on a pre-set mutation algorithm, the cross-path vector is mutated to obtain the mutated path vector. Based on a pre-set optimization mechanism, a target path vector is determined from the path vector of the genetic material entering the next generation and the path vector after mutation. The target path vector is decoded to obtain the target movement path of the automated guided vehicle between the crane and the multiple material storage areas.

6. The method according to claim 5, characterized in that, Based on a pre-set optimization mechanism, a target path vector is determined from the genetic path vector into the next generation and the mutated path vector, including: Based on a pre-set elite retention mechanism, an initial screening path vector is determined from the genetic path vector into the next generation and the mutated path vector. Based on a pre-set swarm search optimization algorithm, the updated path vector is determined according to the initial screening path vector; The population is updated based on the updated path vector until the iteration terminates, and the target path vector is determined.

7. The method according to claim 1, characterized in that, The method further includes: A production logistics simulation model based on Plant Simulation was constructed. Based on the genetic algorithm, the target movement path of the automated guided vehicle (AGV) between the crane and multiple material storage areas was planned to obtain the predicted movement path and the parameters of the first iteration process. Based on the preset improved hybrid genetic algorithm, according to the production logistics system simulation model, the target movement path of the automated guided vehicle between the crane and multiple material storage areas is planned to obtain the target movement path and the parameters of the second iteration process. The production logistics optimization method based on the improved hybrid genetic algorithm is verified based on the estimated movement path and the first iteration process parameters, and the target movement path and the second iteration process parameters.

8. A production logistics optimization device based on an improved hybrid genetic algorithm, characterized in that, include: The first construction module is used to construct a vertical production logistics system model, which includes cranes, automated guided vehicles and multiple material storage areas. The second construction module is used to construct the basic database of the production logistics system and the mapping relationship between the basic database of the production logistics system and the vertical production logistics system model, so as to obtain the simulation model of the production logistics system. The solution module is used to plan the target movement path of the automated guided vehicle between the crane and multiple material storage areas based on a preset improved hybrid genetic algorithm and the production logistics system simulation model.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the production logistics optimization method based on an improved hybrid genetic algorithm as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the production logistics optimization method based on an improved hybrid genetic algorithm as described in any one of claims 1 to 7.

Images