A method and system for optimizing a garment sewing process pipeline

Abstract

The application discloses a kind of pipeline optimization method and system based on clothing sewing process, it is related to the intelligent optimization technical field of clothing sewing process, including, to all processes in clothing sewing pipeline is given unique identifier, and the production data of each process and station is recorded;According to production data, the multi-intelligent algorithm optimization framework is constructed by integrating genetic algorithm, ant colony algorithm and particle swarm algorithm, and the optimal process scheduling scheme is output;Based on the optimal process scheduling scheme, multi-objective optimization is carried out, and the final optimization scheme is output;The application constructs multi-intelligent algorithm optimization framework by integrating genetic algorithm, ant colony algorithm and particle swarm algorithm, gives full play to the advantage of different algorithms, and the optimal process scheduling scheme is output, which significantly shortens the production cycle and reduces the production cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent optimization of garment sewing process, and in particular to a pipeline optimization method and system based on garment sewing process. BACKGROUND

[0002] The optimization of garment sewing pipeline has always been an important issue in manufacturing industry. In modern garment production, traditional garment sewing pipeline optimization methods mainly rely on experience rules and manual scheduling. Such methods are not powerful enough to deal with complex production environments and are difficult to cope with changing market demands and production conditions.

[0003] In recent years, with the development of computing technology and intelligent algorithms, more and more researches have begun to apply intelligent optimization algorithms to the optimization of garment sewing pipeline. However, existing optimization methods mostly focus on global optimization or local optimization, and it is difficult to find a balance between the two. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a pipeline optimization method based on garment sewing process to solve the problem of combining global and local optimization.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a pipeline optimization method based on garment sewing process, which includes,

[0008] Assigning a unique identifier to all processes in the garment sewing pipeline and recording production data of each process and workstation;

[0009] According to the production data, a multi-intelligent algorithm optimization framework is constructed by integrating genetic algorithm, ant colony algorithm and particle swarm algorithm, and an optimal process scheduling scheme is outputted;

[0010] Based on the optimal process scheduling scheme, multi-objective optimization is performed, and a final optimization scheme is outputted.

[0011] As a preferred scheme of the pipeline optimization method based on garment sewing process, wherein: the unique identifier is assigned to all processes in the garment sewing pipeline, and the production data of each process is recorded, and the specific steps are as follows,

[0012] According to the process configuration of the garment sewing pipeline of the enterprise, a unique identifier is assigned to each process and workstation;

[0013] Obtaining the equipment requirement, personnel requirement, tool requirement, operation time and material requirement of each process and station from the production log of the enterprise, listing the data, and grouping according to the process and station.

[0014] As a preferred scheme of the garment sewing process-based pipeline optimization method, the steps of constructing a multi-intelligent algorithm optimization framework by integrating a genetic algorithm, an ant colony algorithm and a particle swarm algorithm, and initializing parameters of the multi-intelligent algorithm are as follows,

[0015] The genetic algorithm is used to optimize the process arrangement of the garment sewing pipeline, and an optimal process arrangement scheme is output.

[0016] Based on the optimal process arrangement scheme, the ant colony algorithm is used to dynamically adjust the distribution of the processes on each station, and an optimal process distribution scheme is output.

[0017] Based on the optimal process distribution scheme, the particle swarm algorithm is used to adjust the scheduling order of the processes, and an optimal process scheduling scheme is output.

[0018] According to the production target of the enterprise, a target function is set.

[0019] According to the size of the enterprise and the working time of the pipeline, a constraint condition is set.

[0020] As a preferred scheme of the garment sewing process-based pipeline optimization method, the steps of optimizing the process arrangement of the garment sewing pipeline using the genetic algorithm and outputting an optimal process arrangement scheme are as follows,

[0021] The processes are shuffled and arranged in combination to obtain different process arrangement combinations, and the combinations are combined into an initial population.

[0022] The fitness value of each individual in the initial population is obtained by calculating the reciprocal of the target function value of each individual.

[0023] The proportion of the fitness value of each individual in the total fitness value is calculated to form an individual selection probability distribution.

[0024] According to the individual selection probability distribution, half of the individuals are selected as parent individuals for cross operation.

[0025] Two parent individuals are selected according to the individual selection probability distribution, and two crossover points are randomly selected to form a crossover interval.

[0026] In the crossover interval, the processes of the two parent individuals are exchanged to generate two child individuals.

[0027] For the processes outside the crossover interval, the processes are adjusted according to a mapping rule, and the identifier of each process appears only once in the child individuals.

[0028] Through the crossover variation method, for each newly generated offspring individual, two process positions are randomly selected with a certain variation probability, and their sequences are exchanged;

[0029] From the population after crossover and variation, a number of individuals with the highest fitness are selected as excellent individuals and are replaced into the parent population to replace the corresponding number of individuals with the lowest fitness, forming a new generation population, and the crossover and variation are continued;

[0030] According to the number of individuals in the initial population and the change of the fitness value, a fitness value convergence condition is set, and when the fitness value meets the convergence condition, the iteration is terminated, and a number of individuals with the highest fitness value and meeting the constraint condition are output as the optimal process arrangement scheme.

[0031] As a preferred scheme of the garment sewing process-based pipeline optimization method described in the application, wherein: based on the optimal process arrangement scheme, the distribution of processes on each station is dynamically adjusted using an ant colony algorithm, and an optimal process distribution scheme is output, and the specific steps are as follows,

[0032] According to the optimal process arrangement scheme, an information matrix is constructed, and all information concentrations are initially the same;

[0033] The ants start from the starting station, and select subsequent stations according to the station selection probability one by one until the distribution of all processes is completed;

[0034] After each ant completes a path, pheromone is released on the path, and the information concentration on the path is updated;

[0035] According to the number of processes and stations, the maximum number of iterations is set, and when the maximum number of iterations is reached, a number of paths with the highest information concentration are selected as the optimal process distribution scheme.

[0036] As a preferred scheme of the garment sewing process-based pipeline optimization method described in the application, wherein: based on the optimal process distribution scheme, the scheduling order of the processes is adjusted by a particle swarm algorithm, and an optimal process scheduling scheme is output, and the specific steps are as follows,

[0037] According to the optimal process distribution scheme output by the ant colony algorithm, an initial particle swarm is constructed;

[0038] The sequence of the processes in the optimal process distribution scheme is taken as the initial position of the particles, and the initial speed of the particles is set according to the total number of processes;

[0039] According to the objective function, the fitness value of each particle is calculated, and particles meeting the constraint condition are selected;

[0040] The inertia weight of particle motion is dynamically adjusted in a nonlinear manner;

[0041] updating the particle velocity after each movement of the particle;

[0042] synchronously adjusting the position of the particle according to the updated particle velocity;

[0043] in each iteration movement, for each particle, selecting the position of the particle with the highest fitness value as the historical optimal position;

[0044] for all particles, selecting the position of the particle with the highest fitness value as the global optimal position;

[0045] according to the change of the particle swarm size and the fitness value of the particle, setting a global fitness value change threshold, when the highest fitness value of the particle is less than the global fitness value change threshold, determining that the particle swarm algorithm converges;

[0046] selecting a number of process scheduling schemes with the highest fitness value from the global optimal position as the optimal process scheduling scheme output.

[0047] As a preferred scheme of the garment sewing process-based pipeline optimization method, wherein: based on the optimal process scheduling scheme, multi-objective optimization is performed, and a final optimization scheme is output, and the specific steps are as follows,

[0048] based on the optimal process scheduling scheme output by the particle swarm algorithm, calculating the production time, production cost and product unqualified rate in the objective function;

[0049] normalizing the target calculation result, and creating an empty Pareto optimal solution set;

[0050] when each solution D is not dominated by any other solution, D is added to the Pareto optimal solution set;

[0051] when D dominates a solution E in the Pareto optimal solution set, D is removed from the Pareto optimal solution set;

[0052] for each solution in the Pareto optimal solution set, calculating the objective function value, and selecting the solution with the highest value as the final optimization scheme.

[0053] In a second aspect, the present application provides a garment sewing process-based pipeline optimization system, comprising a production data module, an algorithm integration module and a scheduling optimization module,

[0054] the production data module is used to assign a unique identifier to all processes in the garment sewing pipeline, and record the production data of each process and station;

[0055] The algorithm integration module is configured to construct a multi-intelligent algorithm optimization framework by integrating a genetic algorithm, an ant colony algorithm and a particle swarm algorithm according to the production data, and output an optimal process scheduling scheme.

[0056] The scheduling optimization module is configured to perform multi-objective optimization based on the optimal process scheduling scheme, and output a final optimization scheme.

[0057] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, any step of the pipeline optimization method based on a garment sewing process according to the first aspect of the present application is implemented.

[0058] In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, any step of the pipeline optimization method based on a garment sewing process according to the first aspect of the present application is implemented.

[0059] The present application has the following beneficial effects: the present application assigns a unique identifier to each process and station and records detailed production data, ensuring the transparency and traceability of the production process, providing accurate data support for subsequent optimization, integrating a genetic algorithm, an ant colony algorithm and a particle swarm algorithm to construct a multi-intelligent algorithm optimization framework, fully utilizing the advantages of different algorithms, and outputting an optimal process scheduling scheme that significantly shortens the production cycle, reduces production costs, and improves product quality. Through multi-objective optimization, multiple objectives such as production time, cost and unqualified rate are considered, and the weights of each objective are dynamically adjusted according to the actual needs of the enterprise, so that the enterprise can flexibly select the optimal solution at different production stages. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0061] Figure 1 The flowchart of the pipeline optimization method based on a garment sewing process in embodiment 1.

[0062] Figure 2 The module diagram of the pipeline optimization system based on a garment sewing process in embodiment 1. DETAILED DESCRIPTION

[0063] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0064] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details set forth in this description. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the present application.

[0065] Secondly, the "one embodiment" or "embodiment" referred to herein is intended to mean a specific feature, structure, or characteristic under discussion that is included in at least one implementation of the present application. The various appearances of "in one embodiment" or "in an embodiment" in the description are not all referring to the same embodiment, although they can. Descriptions of the embodiments are intended to be illustrative of the present application and not to limit the scope of the present application.

[0066] Embodiment 1, Reference Figure 1 and Figure 2 The first embodiment of the present application provides a pipeline optimization method based on garment sewing process, comprising the following steps:

[0067] S1, a unique identifier is given to all processes in the garment sewing pipeline, and production data of each process and station is recorded.

[0068] According to the process configuration of the garment sewing pipeline of the enterprise, a unique identifier is assigned to each process and station (such as process assignment by numerical sequence and station assignment by alphabetical sequence, etc.) to ensure that each process can be uniquely identified throughout the production process;

[0069] By obtaining the equipment requirements, personnel requirements, tool requirements, operation time and material requirements required by each process and station from the production log of the enterprise, and including them in the data list, and grouping them according to the process and station respectively;

[0070] It should be noted that a process represents an independent work content, and a station refers to a fixed position or work area on the production line. The station is the spatial layout unit of the production line, which determines the specific execution location of the process. There can be multiple stations on the garment sewing pipeline, each responsible for different processes.

[0071] S2, according to the production data, a multi-intelligent algorithm optimization framework is constructed by integrating genetic algorithm, ant colony algorithm and particle swarm algorithm, and the optimal process scheduling scheme is output.

[0072] The genetic algorithm is used to optimize the process arrangement of the garment sewing pipeline, and the optimal process arrangement scheme is output;

[0073] Based on the optimal process arrangement scheme, the ant colony algorithm is used to dynamically adjust the distribution of processes in each station, and the optimal process distribution scheme is output;

[0074] Based on the optimal process allocation scheme, the scheduling sequence of the process is adjusted by the particle swarm algorithm, and the optimal process scheduling scheme is output;

[0075] Specifically, the genetic algorithm is used to optimize the process arrangement through crossover and mutation operations due to its good global search capability;

[0076] The local search capability of the ant colony algorithm is used to optimize the process allocation based on the optimized process arrangement;

[0077] The particle swarm algorithm is used to optimize the process scheduling scheme based on the optimized process allocation;

[0078] According to the production target of the enterprise, the objective function is set, and the expression is as follows:

[0079] F=ω1×f1+ω2×f2+ω3×f3;

[0080] Wherein, F is the value of the objective function, f1 is the production time, f2 is the production cost, f3 is the product unqualified rate, ω1, ω2, ω3 are the weights of f1, f2, f3 respectively, and the sum of the three weights is 1;

[0081] Wherein, ω1, ω2, ω3 are preset according to the order quantity and actual demand of the enterprise, for example, during the peak period of orders, the enterprise may pay more attention to production efficiency, at this time, ω1 can be reduced, while in the off-season, the enterprise may pay more attention to cost control, at this time, ω2 can be reduced;

[0082] According to the size of the enterprise and the working time of the assembly line, the constraint conditions are set:

[0083] The equipment capacity limit, the equipment utilization rate of each station does not exceed its maximum capacity;

[0084] The working hours of the operating personnel of each station do not exceed its working time;

[0085] The material supply of each process is sufficient to avoid downtime due to lack of materials;

[0086] Ensure that the product qualified rate is not lower than the standard set by the enterprise.

[0087] S2.1, use genetic algorithm to optimize the process arrangement of garment sewing assembly line, output the optimal process arrangement scheme.

[0088] The process is disturbed and arranged to obtain different process arrangement combinations, and combined into an initial population;

[0089] For example, if there are 10 processes in the pipeline, each individual can be represented as a sequence of length 10, such as [1, 3, 2, 4, 5, 6, 7, 8, 9, 10], indicating that the first process is ranked first, the third process is ranked second, and so on.

[0090] By calculating the reciprocal of the objective function value of each individual in the initial population, the fitness value of each individual is obtained to determine its advantages and disadvantages in the initial population. The higher the fitness value, the better the individual, indicating that the process arrangement scheme is more reasonable.

[0091] Roulette Wheel Selection is used for selection operation, which retains individuals with higher fitness and maintains a certain diversity to avoid premature convergence of genetic algorithms.

[0092] The proportion of the fitness value of each individual in the total fitness is calculated to form the individual selection probability distribution, expressed as:

[0093]

[0094] where P i represents the probability of the i-th individual being selected, h i is the fitness value of the i-th individual, n is the number of individuals, and i is the index of the number of individuals.

[0095] According to the individual selection probability distribution, half of the individuals are selected as parent individuals for crossover operation.

[0096] PMX (Partial Map Crossover) is used to select two parent individuals according to the individual selection probability distribution and randomly select two crossover points to form a crossover interval.

[0097] Within the crossover interval, the processes of the two parent individuals are exchanged to generate two child individuals.

[0098] For example, the processes in the crossover interval of parent individuals A1 and A2 are [3, 2, 4, 5, 6] and [7, 8, 9, 10, 1] respectively. By directly exchanging these two gene fragments, two child individuals B1 and B2 are generated.

[0099] For processes outside the crossover interval, adjust according to the mapping rule, and each process identifier appears only once in the child individual.

[0100] For example, process 3 in B1 has already appeared in the crossover interval, so it is replaced by the corresponding process 7 in A2. Similarly, process 7 in B2 is replaced by process 3 in A1.

[0101] In order to control the frequency of crossover operation, the crossover probability is set, the initial crossover probability can be set to 0.85, and the crossover probability is gradually reduced as the number of iterations increases, so as to avoid excessive exploration in the later stage, and the expression is:

[0102]

[0103] Wherein, P c (t1) is the t1th crossover probability, T1 is the total number of crossover iterations, t1 is the number of crossover times;

[0104] Through the exchange mutation method, for each newly generated offspring individual, two process positions are randomly selected with a certain mutation probability, and their order is exchanged;

[0105] For convenience of calculation, the mutation probability is the same as the crossover probability;

[0106] From the population after crossover and mutation, select several individuals with the highest fitness as excellent individuals, and replace the corresponding number of individuals with the lowest fitness in the parent population to form a new generation population, and continue crossover and mutation;

[0107] According to the number of individuals in the initial population and the change of fitness value, the convergence condition of fitness value is set, for example, when the fitness value of the individual with the highest fitness in the population changes less than 0.01 in 10 consecutive iterations, when the fitness value meets the convergence condition, the iteration is terminated, and several (the first five) individuals with the highest fitness and meeting the constraint condition are output as the optimal process arrangement scheme.

[0108] S2.2, based on the optimal process arrangement scheme, the ant colony algorithm is used to dynamically adjust the distribution of processes in each station, and the optimal process distribution scheme is output.

[0109] According to the optimal process arrangement scheme, the pheromone matrix is constructed, and the initial pheromone concentration of all is the same, which represents that each process has the same opportunity to be selected in each station;

[0110] Initialize the pheromone evaporation factor, which determines the decay rate of pheromone with time, the initial evaporation factor is set to 0.6, and the evaporation factor is gradually increased as the number of iterations increases, so as to reduce the accumulation of pheromone on the local path and promote the exploration of more new paths by ants, and the expression is:

[0111]

[0112] Wherein, ρ(t2) is the pheromone evaporation factor of the t2th path selection, T2 is the total number of path selection iterations, t2 is the path selection iteration number;

[0113] The pheromone importance factor a and the heuristic information importance factor β reflect the degree of dependence on pheromone concentration and heuristic information when the ants select paths, respectively.

[0114] The ants start from the starting station, select subsequent stations according to the station selection probability, and complete the allocation of all processes until the expression is:

[0115]

[0116] wherein, Pij(k) represents the probability of the lth ant selecting the kth station in the jth process, Pij(k) is the a-th power of the pheromone concentration of the jth process in the kth station, Pij(k) is the β-th power of the heuristic information of the jth process in the kth station, usually refers to the operation time, equipment utilization, etc. of the jth process in the kth station, and N j Nj represents the set of stations available for the jth process;

[0117] After each ant completes a path, it releases pheromone on the path, and updates the pheromone concentration on the path, the expression is as follows:

[0118] τ jk (t2+1) = (1-ρ(t2))×τ jk (t2) + Δτ jk ;

[0119]

[0120] wherein, τ jk (t2+1) is the pheromone concentration of the jth process in the kth station when the lth ant selects the t2+1th time, τ jk (t2) is the pheromone concentration of the jth process in the kth station when the lth ant selects the t2th time, Δτ jk is the pheromone concentration increment, and Q is the total amount of pheromone.

[0121] According to the number of processes and stations, the maximum number of iterations is set, when the maximum number of iterations is reached, a number of paths with the highest pheromone concentration (the first five) are selected as the optimal process allocation scheme output.

[0122] S2.3, based on the optimal process allocation scheme, the scheduling sequence of the process is adjusted through the particle swarm algorithm, and the optimal process scheduling scheme is output.

[0123] According to the optimal process allocation scheme output by the ant colony algorithm, an initial particle swarm is constructed;

[0124] Each particle represents a process scheduling scheme, the position of the particle represents the scheduling order of the process, and the speed of the particle represents the adjustment amplitude of the process;

[0125] The process order in the optimal process allocation scheme is taken as the initial position of the particle, and the initial speed of the particle is set according to 10% of the total number of processes;

[0126] According to the objective function, the fitness value of each particle is calculated, and the particles meeting the constraint condition are screened out;

[0127] The inertia weight of the particle motion is dynamically adjusted in a nonlinear manner, which is represented as:

[0128]

[0129] where w(t3) is the inertia weight of the t3th iteration motion, w min is the minimum value of the inertia weight, usually 0.4, and w max is the maximum value of the inertia weight, usually 0.9, T3 is the total number of iterations of particle motion, and t3 is the number of iteration motion;

[0130] After each motion of the particle, the particle speed is updated, and the expression is as follows:

[0131] v m (t3+1)=w(t3)×v m (t3)+r1(q m -x m (t3))+r2(g m -x m (t3));

[0132] where v m (t3+1) is the velocity vector of the mth particle in the t3+1th iteration motion, v m (t3) is the velocity of the mth particle in the t3th iteration motion, r1 and r2 are random numbers with a value range of [0, 1], which are used to introduce randomness to avoid the particle falling into local optimum, q m is the historical optimal position vector of the mth particle, x i (t3) is the position vector of the mth particle in the t3th iteration motion, and g m is the global optimal position vector of the mth particle, i.e. the position vector of the particle with the highest fitness value among all particles;

[0133] Further, w(t3)×v m (t3) is the inertia term, which represents scaling the speed of the particle to maintain its original motion direction but adjust its size, r1(q m -x m(t3)) is a cognitive item, representing the velocity component of the particle moving towards the historical optimal position, r2(g m -x m (t3)) is a social item, representing the velocity component of the particle moving towards the global optimal position;

[0134] According to the updated particle velocity, the position of the particle is adjusted synchronously, and the expression is as follows:

[0135] x m (t3+1)=x m (t3)+v m (t3+1)

[0136] Wherein, x m (t3+1) is the position vector of the mth particle at the t3+1th iteration movement;

[0137] In each iteration movement, for each particle, the position of the particle with the highest fitness value is selected as the historical optimal position;

[0138] For all particles, the position of the particle with the highest fitness value is selected as the global optimal position;

[0139] According to the change of the particle swarm size and the fitness value of the particle, the global fitness value change threshold is set, when the highest fitness value of the particle is less than the global fitness value change threshold in continuous 10 iteration movements, it is determined that the particle swarm algorithm converges;

[0140] From the global optimal position, a number (the first five) of process scheduling schemes with the highest fitness value are selected as the optimal process scheduling scheme output.

[0141] S3, based on the optimal process scheduling scheme, multi-objective optimization is carried out, and the final optimization scheme is output.

[0142] Based on the optimal process scheduling scheme output by the particle swarm algorithm, the production time, production cost and product unqualified rate in the objective function are calculated;

[0143] In order to facilitate the comparison of the differences between different objectives, the objective calculation results are normalized, and an empty Pareto optimal solution set is created;

[0144] When each solution D is not dominated by any other solution (i.e. there is no other objective calculation result E, which is not worse than D in all objectives), D is added to the Pareto optimal solution set;

[0145] When D dominates a solution E in the Pareto optimal solution set (i.e. D is not worse than E in all objectives and is better than E in at least one objective), D is removed from the Pareto optimal solution set;

[0146] For each solution in the Pareto optimal solution set, the objective function value is calculated, and the solution with the highest value is selected as the final optimization scheme.

[0147] The embodiment also provides a pipeline optimization system based on a garment sewing process, comprising a production data module, an algorithm integration module and a scheduling optimization module, the production data module is used to assign a unique identifier to all processes in the garment sewing pipeline and record the production data of each process and station; the algorithm integration module is used to construct a multi-intelligent algorithm optimization framework by integrating genetic algorithm, ant colony algorithm and particle swarm algorithm according to the production data, and output an optimal process scheduling scheme; the scheduling optimization module is used to perform multi-objective optimization based on the optimal process scheduling scheme and output a final optimization scheme.

[0148] The embodiment also provides a computer device suitable for the pipeline optimization method based on the garment sewing process, comprising a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the pipeline optimization method based on the garment sewing process proposed in the above embodiment.

[0149] The computer device can be a terminal, and the computer device comprises a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device. In addition, the input device can also be an external keyboard, touchpad or mouse, etc.

[0150] The embodiment also provides a storage medium on which a computer program is stored, the program being executed by a processor to implement the method for realizing pipeline optimization based on a garment sewing process as proposed in the above embodiment; the storage medium can be realized by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0151] To sum up, the present application ensures the transparency and traceability of the production process by assigning a unique identifier to each process and station and recording detailed production data, provides accurate data support for subsequent optimization, constructs a multi-intelligent algorithm optimization framework by integrating genetic algorithm, ant colony algorithm and particle swarm algorithm, fully gives play to the advantages of different algorithms, and significantly shortens the production cycle, reduces the production cost, and improves the product quality of the output optimal process scheduling scheme, considers multiple targets such as production time, cost and unqualified rate through multi-objective optimization, dynamically adjusts the weights of the targets according to the actual needs of the enterprise, and enables the enterprise to flexibly select the optimal solution in different production stages.

[0152] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method for optimizing an assembly line based on a garment sewing process, characterized in that: include, Assign unique identifiers to all processes in the garment sewing line and record the production data of each process and workstation; According to the production data, a multi-intelligent algorithm optimization framework is constructed by integrating genetic algorithm, ant colony algorithm and particle swarm algorithm to output the optimal process scheduling plan. The specific steps are as follows: Use genetic algorithms to optimize the process arrangement of garment sewing lines and output the optimal process arrangement plan; Based on the optimal process arrangement plan, the ant colony algorithm is used to dynamically adjust the allocation of processes to each workstation and output the optimal process allocation plan; Based on the optimal process allocation plan, the scheduling order of the processes is adjusted through the particle swarm algorithm, and the optimal process scheduling plan is output. The specific steps are as follows: Construct the initial particle swarm based on the optimal process allocation plan output by the ant colony algorithm; The process sequence in the optimal process allocation scheme is used as the initial position of the particle, and the initial velocity of the particle is set according to the total number of processes; Filter out particles that meet the constraints; The constraints include that the equipment utilization rate of each workstation does not exceed the maximum capacity, the operator hours at each workstation do not exceed the working hours, the material supply of each process is sufficient, and the product qualification rate is not lower than the standard set by the enterprise; Dynamically adjust the inertia weight of particle motion in a nonlinear way; After each particle movement, the particle velocity is updated; According to the updated particle velocity, the particle position is adjusted synchronously; The particles represent the process scheduling plan, the positions of the particles represent the scheduling order of the processes, and the speed of the particles represents the adjustment range of the processes; In each iterative movement, for each particle, the position of the particle with the highest fitness value is selected as the historical optimal position; For all particles, the position of the particle with the highest fitness value is selected as the global optimal position; According to the changes in the particle swarm size and the particle fitness value, a global fitness value change threshold is set. When the highest particle fitness value is less than the global fitness value change threshold, the particle swarm algorithm is determined to have converged. Select several process scheduling solutions with the highest fitness values ​​from the global optimal position as the optimal process scheduling solution output; Based on the optimal process scheduling plan, multi-objective optimization is performed and the final optimization plan is output. The specific steps are as follows: Based on the optimal process scheduling solution output by the particle swarm algorithm, the production time, production cost and product failure rate in the objective function are calculated; Normalize the target calculation results and create an empty Pareto optimal solution set; When each solution D is not dominated by any other solution, D is added to the Pareto optimal solution set; When D dominates a solution E in the Pareto optimal solution set, remove D from the Pareto optimal solution set; For each solution in the Pareto optimal solution set, calculate the objective function value and select the solution with the highest value as the final optimization solution.

2. The assembly line optimization method based on garment sewing technology according to claim 1, characterized in that: The specific steps of assigning unique identifiers to all processes in the garment sewing line and recording the production data of each process are as follows: According to the process configuration of the enterprise's garment sewing line, a unique identifier is assigned to each process and workstation; By obtaining the equipment requirements, personnel requirements, tool requirements, operation time and material requirements for each process and workstation from the company's production log, listing them in a data list and grouping them by process and workstation.

3. The assembly line optimization method based on garment sewing technology according to claim 2, characterized in that: The method uses genetic algorithm to optimize the process arrangement of the garment sewing line and output the optimal process arrangement scheme. The specific steps are as follows: The processes are disrupted and arranged and combined to obtain individuals with different process arrangements and combinations, and then combined into the initial population; By calculating the inverse of the objective function value of each individual in the initial population, the fitness value of each individual is obtained; Calculate the proportion of each individual's fitness value to the total fitness value to form the individual selection probability distribution; According to the probability distribution of individual selection, half of the individuals are selected as parent individuals and crossover operation is performed; Select two parent individuals according to the individual selection probability distribution, and randomly select two crossover points to form a crossover interval; In the crossover interval, the process of exchanging two parent individuals generates two child individuals; For the processes outside the crossover interval, adjustments are made according to the mapping rules, and the identifier of each process appears only once in the offspring individual; Through the exchange mutation method, for each newly generated offspring individual, two process positions are randomly selected with a certain mutation probability and their order is exchanged; From the population after crossover and mutation, several individuals with the highest fitness are selected and retained as excellent individuals and replaced with the corresponding number of individuals with the lowest fitness in the parent population to form a new generation population, and continue crossover and mutation; According to the number of individuals in the initial population and the change of fitness value, the fitness value convergence condition is set. When the fitness value meets the convergence condition, the iteration is terminated and several individuals with the highest fitness value and meeting the constraint conditions are output as the optimal process arrangement plan.

4. The assembly line optimization method based on garment sewing technology according to claim 3, characterized in that: Based on the optimal process arrangement plan, the ant colony algorithm is used to dynamically adjust the allocation of processes on each workstation and output the optimal process allocation plan. The specific steps are as follows: According to the optimal process arrangement scheme, a pheromone matrix is ​​constructed, and the initial concentration of all pheromones is the same; The ant starts from the starting workstation and selects subsequent workstations in sequence according to the workstation selection probability until all processes are assigned; Each time an ant completes a path, it releases pheromones along the path and updates the pheromone concentration along the path; According to the number of processes and workstations, the maximum number of iterations is set. When the maximum number of iterations is reached, several paths with the highest pheromone concentration are selected and output as the optimal process allocation plan.

5. A garment sewing process-based assembly line optimization system, based on the garment sewing process-based assembly line optimization method according to any one of claims 1 to 4, characterized in that: Including production data module, algorithm integration module and scheduling optimization module, The production data module is used to assign unique identifiers to all processes in the garment sewing line and record the production data of each process and workstation; The algorithm integration module is used to construct a multi-intelligent algorithm optimization framework by integrating genetic algorithm, ant colony algorithm and particle swarm algorithm according to production data, and output the optimal process scheduling solution; The scheduling optimization module is used to perform multi-objective optimization based on the optimal process scheduling plan and output the final optimization plan.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the assembly line optimization method based on the garment sewing process according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the assembly line optimization method based on garment sewing process according to any one of claims 1 to 4 are implemented.

Images