A Smart Scheduling Method for Automotive Welding Body Production Line Considering Conveying Patterns

Abstract

This invention belongs to the field of automotive production line scheduling technology, specifically relating to an intelligent scheduling method for automotive welding body production lines that considers conveying patterns. The method includes: mapping the welding body production line scheduling decision scenario to a variable-rhythm synchronous mixed-flow bilateral production line scheduling problem and acquiring relevant data; establishing an optimization model for the variable-rhythm synchronous mixed-flow bilateral production line scheduling problem; and using a genetic algorithm to solve the variable-rhythm synchronous mixed-flow bilateral production line scheduling problem based on the optimization model. This invention achieves intelligent and precise output of scheduling plans by considering the characteristics of the actual production line, which helps reduce human error and time waste caused by manual scheduling, and makes the scheduling plan more consistent with real-world scheduling scenarios, thus promoting more rational resource utilization, cost reduction, and efficiency improvement.

Description

Technical Field

[0001] This invention belongs to the field of automotive production line scheduling technology, specifically relating to an intelligent scheduling method for automotive welding body production lines that takes into account conveying patterns. Background Technology

[0002] Against the backdrop of global competition in the automotive market and diversified customer demands, automobile production faces challenges such as product diversification, cost transparency, accelerated product upgrades, and shortened product lifecycles. The automotive manufacturing system must be able to quickly adjust to market demands to efficiently organize resources. Therefore, rational and efficient production scheduling is crucial for automotive production organization.

[0003] Automobile production is a typical example of mixed-flow, batch production line organization. Its scheduling problem typically refers to how to rationally arrange the production sequence of multiple products to maximize production line efficiency. The scheduling problem of mixed-flow production lines has been studied for nearly sixty years, with numerous scholars conducting scheduling research on various automotive process production lines, and many scheduling systems being applied in the industry. The automotive welding body assembly line is an assembly line characterized by mixed-flow production, two-way assembly operations at each station, and unpaced synchronous conveying between stations. Welding body assembly lines are typically equipped with assembly fixtures. The car body is welded while in a fixed or stationary state at each station. Once all car bodies are completed at their respective stations, the conveyor system is activated, synchronously transporting the welded car bodies from all stations to their next station. In contrast, final assembly and painting production lines typically employ a chain-type, paced synchronous conveyor system, where the vehicle body is conveyed and assembled simultaneously. Component production lines, on the other hand, usually have buffer devices between workstations. Once a workpiece is completed at one workstation, it is immediately conveyed to the next workstation or placed in the buffer device, resulting in unpaced asynchronous conveying. Due to these differences in conveyor systems, the calculation and solution methods for scheduling the welding body production line differ significantly from those for painting, final assembly, and component production lines. However, existing methods primarily design and output scheduling solutions based on the production line patterns of automotive painting and final assembly, neglecting to explore the unique characteristics of the automotive welding body production line. This leads to the welding body production line scheduling problem being solved under excessively simplified conditions, resulting in overly coarse scheduling solutions or the direct adoption of a "uniformly mixed" scheduling scheme, severely impacting the production line's output efficiency. Summary of the Invention

[0004] To address the above problems, this invention provides an intelligent scheduling method for automotive welding body production lines that considers conveying patterns, comprising the following steps:

[0005] S1. Obtain relevant data on the production scheduling decision problem of the welding vehicle body production line, and map the production scheduling decision problem of the welding vehicle body production line into a variable-rhythm synchronous mixed-flow bilateral production line scheduling problem;

[0006] S2. Establish an optimization model for the scheduling problem of a variable-rhythm synchronous mixed-flow bilateral production line;

[0007] S3. Based on the optimization model, a genetic algorithm is used to solve the production scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line.

[0008] Furthermore, the relevant data obtained after mapping in step S1 regarding the scheduling problem of the variable-rhythm synchronous mixed-flow double-sided production line includes: product set M = {1, 2, ..., |M|}, workstation set S = {1, 2, ..., |S|}, and demand set D = {D1, D2, ..., D...} |M|}、The set of production scheduling schemes Q={Q 1 Q 2 ,…Q |TS| The workstation operation time matrix PT; where |M| is the total number of product types, |S| is the total number of workstations, |TS| is the number of production scheduling plans, and D... m Let D be the demand for the m-th product, and Q be the quantity demanded. p It is the p-th production scheduling plan in the set of production scheduling plans. Let |Q| be the product at position q in the p-th production scheduling plan, where |Q| is the quantity of a product included in a production scheduling plan, and |Q| satisfies the following conditions: h is D = {D1, D2, ..., D} |M| The greatest common divisor of the demand for all products in the matrix; the element PT in the workstation operation time matrix. ms =max(PT) m,s,L ,PT m,s,R ) represents the operation time of the m-th product at the s-th workstation, where PT m,s,L Let PT be the processing time of the m-th product to the left of the s-th workstation. m,s,R Let S be the processing time of the m-th product to the right of the s-th workstation. Further, step S2 establishes an optimization model for the scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line, including:

[0009] S21. Given a fixed number of workstations |S|, the goal of maximizing production efficiency is transformed into minimizing the output cycle time of the production line;

[0010] S22. Based on the variable rhythm synchronous conveying effect, the objective of minimizing the output takt time of the production line is converted into the weighted average of the output takt time of all production line states during the cyclic execution of the production sequence;

[0011] S23. Setting constraints for scheduling problems in variable-rhythm synchronous mixed-flow double-sided production lines: Each workstation has two positions, left and right, and a product can only appear in one position of one workstation.

[0012] Furthermore, step S22 transforms the objective of minimizing the production line output cycle time into a weighted sum of the output cycle times of all production line states during the cyclic execution of the production sequence, resulting in the following objective function:

[0013]

[0014]

[0015] q=μ+|S|±Δ|Q|-s,s=1,2,...,|S|

[0016] Where Obj is the objective function value, representing the weighted average of the output takt times of |Q| production line states in a production sequence; C μ It is the output cycle time of the μth production line state, where μ and q have the same value range μ,q=1,2,...,|Q|; Q represents q The operating time of the product at the s-th workstation, x q,s It is a decision variable; if the q-th product in the production sequence Q appears at the s-th workstation, then x... q,s =1, otherwise x q,s =0; q = μ + |S| ± Δ|Q| - s is the relationship expression between workstations, production sequences, and production line states in a variable-rhythm synchronous assembly line, indicating that the s-th workstation in the μ-th production line state produces the q-th product Q in the production sequence Q. q It is a formal representation of the operation mechanism of the variable rhythm synchronous assembly production line; |Q| is the total number of production line states, |S| is the total number of workstations, and Δ is an arbitrary positive integer such that the parameters satisfy μ,q=1,2,...,|Q|, s=1,2,...,|S|.

[0017] Furthermore, step S3, based on the optimization model, uses a genetic algorithm to solve the scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line, including:

[0018] S31. Set the parameters for the feasible solution of the problem in the genetic operation process: population size PSIZE, selection probability P s Crossover probability P c Probability of mutation P m Set the parameters in the iteration termination condition: number of iterations, iteration time, and iteration termination threshold;

[0019] S32. Encode feasible solutions to the problem and initialize the chromosome population based on the number of possible production scheduling schemes;

[0020] S33. Calculate the objective function value for each chromosome in the initial chromosome population, and use the reciprocal of the objective function value as the fitness value of the corresponding chromosome;

[0021] S34. Normalize each fitness value, and use the roulette wheel betting method to select PSIZE×P based on the normalized fitness values. s One chromosome;

[0022] S35. Using the sequential crossover method to calculate PSIZE×P s Crossover operation is performed on each chromosome;

[0023] S36. Use the two-point flipping method to perform mutation operations on chromosomes after the crossover operation;

[0024] S37. Obtain a new population and iterate until the iteration termination condition is met, then output the optimal chromosome.

[0025] Furthermore, step S32 encodes a feasible solution to the problem and initializes the chromosome population, including:

[0026] S321. Based on the product set M = {1, 2, ..., |M|} and the demand set D = {D1, D2, ..., D...}, |M| The encoding yields |TS| chromosomes, where each chromosome represents a production scheduling plan;

[0027] S322. If |TS|≤PSIZE, then use these |TS| chromosomes to form the initial chromosome population; if

[0028] If |TS|>PSIZE, then select PSIZE chromosomes to form the initial chromosome population.

[0029] Furthermore, the formula for calculating the chromosome number |TS| in step S321 is as follows:

[0030]

[0031] Where, d m =D m / h, where h is the greatest common divisor of the demand quantities of all products in the demand set D. This refers to the number of products included in a production scheduling plan.

[0032] The beneficial effects of this invention are:

[0033] This invention considers the conveying pattern of the welding vehicle body production line and combines the characteristics of mixed-flow automotive production and double-sided assembly at workstations to design an intelligent solution method for scheduling problems. This method facilitates a global search of the solution space and improves the accuracy of the scheduling output.

[0034] This invention enables intelligent scheduling by taking into account the characteristics of actual production lines, which can achieve intelligent and accurate output of scheduling plans. This helps to reduce human error and time waste caused by manual scheduling, and makes the scheduling plan more in line with the actual scheduling scenario, which is more conducive to the rational use of resources, cost reduction and efficiency improvement. Attached Figure Description

[0035] Figure 1 A schematic diagram of the intelligent scheduling method for an automotive welding body production line considering conveying patterns according to the present invention;

[0036] Figure 2 This is a schematic diagram of the production state of the welding vehicle body production line under the action of the variable rhythm synchronous conveying mode of the present invention.

[0037] Figure 3 This is a flowchart illustrating how the genetic algorithm is used in this invention to intelligently solve the scheduling problem of automotive welding body production lines. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] This invention provides an intelligent scheduling method for automotive welding body production lines that considers conveying patterns. This method maps the scheduling problem of automotive welding body production lines into a scheduling problem of variable-rhythm synchronous mixed-flow bilateral assembly lines by considering the common characteristics of automotive production lines, such as mixed-flow production and two-sided assembly operations at workstations, as well as the unique characteristics of variable-rhythm synchronous conveying in automotive welding body production lines. Furthermore, it decomposes the output rhythm of the production line under the action of variable-rhythm synchronous conveying into the output rhythm of each production line state to model the scheduling problem, and then uses a genetic algorithm to intelligently solve the problem, thereby obtaining the optimal scheduling scheme for automotive welding body production lines.

[0040] The specific solutions proposed in this invention are as follows: Figure 1 As shown, it includes the following steps:

[0041] S1. Obtain relevant data on the production scheduling decision problem of the welding vehicle body production line, and map the production scheduling decision problem of the welding vehicle body production line into a production scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line.

[0042] Specifically, the production scheduling decision problem for the welding vehicle body production line is mapped to the production scheduling problem for a variable-rhythm synchronous mixed-flow two-sided production line. This includes: processing the product types of the mixed-flow production of the welding vehicle body production line into a product set; processing the workstations into a workstation set; processing the product demand into a demand set and possible scheduling schemes; processing the workstation operation time into a workstation operation time set; mapping the production scheduling decision problem for the welding vehicle body production line into the production scheduling problem for a variable-rhythm synchronous mixed-flow two-sided production line, and dividing the production process into production line states.

[0043] Specifically, step S1 obtains relevant data related to the scheduling problem of the variable-rhythm synchronous mixed-flow bilateral production line based on the mapping, including:

[0044] S11. Convert the |M| types of products produced in the mixed-flow welding vehicle body production line into a product set M = {1,2,...,|M|}, where |M| is the total number of product types;

[0045] S12. Convert the |S| workstations in the welding vehicle body production line that can be allocated on both the left and right sides into a set of workstations arranged in series, S={1,2,...,|S|}, where |S| is the total number of workstations;

[0046] S13. Map the production demand of the welding vehicle body production line into a set of |M| product demands, D = {D1, D2, ..., D...}. |M|}, D m Let ∈D be the demand for the m-th product; define d m =D m / h, where h is D = {D1,D2,...,D} |M| If the greatest common divisor of the demand for various products in a given set is found, then the set of production scheduling schemes can be represented as Q = {Q}. 1 Q 2 ,…Q |TS|}, |TS| represents the number of production plans. This represents the p-th production scheduling plan (also known as the production sequence) in the set of production scheduling plans. Q p It is a production sequence consisting of |M| types of products and |Q| products, where Represents the production sequence Q p The q-th product in the set M, and the type of this product belongs to product set M; and when the production sequence Q... p If the product is produced repeatedly h times, then the planned demand D = {D1, D2, ..., D...} |M| That can be satisfied.

[0047] S14. Add the operating time of the conveyor device of the welding vehicle body production line, the product's entry time at the first station of the production line, the product's exit time at the last station of the production line, the equipment preparation time, and the tool preparation time to the station time, and generate a station operation time matrix PT; the elements PT in the station operation time matrix PT ms =max(PT) m,s,L ,PT m,s,R ) represents the operation time of the m-th product at the s-th workstation, where PT m,s,L Let PT be the processing time of the m-th product to the left of the s-th workstation. m,s,R Let m be the processing time of the m-th product to the right of the s-th workstation;

[0048] S15. The production scheduling problem for welding vehicle body production lines is transformed into a variable-rhythm synchronous mixed-flow two-sided production line scheduling problem, i.e., given the production line product set M, workstation set S, product demand set D, and workstation operation time PT. m,s,L PT m,s,R In this scenario, how do we find a production scheduling scheme Q that maximizes the efficiency of the production line? The calculation of production efficiency needs to consider the characteristics of variable-rhythm synchronous conveying: On the welding assembly line, workpieces are welded while stationary at their workstations. The conveyor can only start after all workpieces at all workstations have been completed, synchronously transporting all workpieces to their next workstation. Due to this characteristic of variable-rhythm synchronous conveying, any product needs to pass through |S| workstations of the production line in a variable-rhythm sequence, and the products at each of the |S| workstations must move synchronously to their respective next workstations, such that a production scheduling scheme Q = {Q1, Q2, ... Q} |Q| The production process of the |Q| products in the welding vehicle body production line can be divided into |Q| production line states, such as... Figure 2 As shown;

[0049] S2. Establish an optimization model for the scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line.

[0050] Specifically, step S2 establishes an optimization model for the scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line, including:

[0051] S21. Set production scheduling targets and establish objective functions. Production scheduling can achieve multiple objectives. Under the condition of a fixed number of workstations |S|, this invention transforms the objective of maximizing production efficiency into minimizing the output cycle time of the production line;

[0052] S22. Based on the characteristics of variable rhythm synchronous conveying, the objective of minimizing the output takt of the production line is transformed into minimizing the weighted average of the output takt of all production line states during the cyclic execution of the production sequence;

[0053] Specifically, step S22 transforms the objective of minimizing the production line output cycle time into minimizing the weighted average of the output cycle times of all production line states during the cyclic execution of the production sequence, resulting in the following objective function:

[0054]

[0055]

[0056] q=μ+|S|±Δ|Q|-s,s=1,2,...,|S|

[0057] Where Obj is the objective function value, representing the weighted average of the output takt times of |Q| production line states in a production sequence; C μ The output cycle time of the μth production line state, μ and q have the same value range μ,q=1,2,...,|Q|; Product Q q The working time at the s-th workstation is x q,s It is a decision variable, if the production sequence Q p If the qth product appears at the sth workstation, then x q,s =1, otherwise x q,s =0; q = μ + |S| ± Δ|Q| - s is the relationship expression between workstations, production sequences, and production line states in a variable-rhythm synchronous assembly line, representing the production sequence Q produced at workstation s in production line state μ. p The qth product Q q It is a formal representation of the operation mechanism of the variable rhythm synchronous assembly production line; |Q| is the total number of production line states, |S| is the total number of workstations, and Δ is an arbitrary positive integer such that the parameters satisfy μ,q=1,2,...,|Q|, s=1,2,...,|S|;

[0058] S23. Set constraints for the scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line: Each workstation has two positions, left and right; the workstation cycle time is the maximum of the cycle times of the left and right sides; any product in the production sequence can only appear in one position of one workstation; decision variables can only take values ​​of 0 and 1; workstation operation time cannot be negative; specifically expressed as:

[0059] PT ms =max(PT) m,s,L ,PT m,s,R )

[0060]

[0061] x q,s ∈{0,1},PT m,s≥0S3. Based on the optimization model, a genetic algorithm is used to solve the production scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line.

[0062] Specifically, step S3, based on the optimization model, uses a genetic algorithm to solve the scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line, such as... Figure 3 As shown, it includes:

[0063] S31. Set the parameters for the feasible solution of the problem in the genetic operation process: population size PSIZE, selection probability P s Crossover probability P c Probability of mutation P m Set the parameters in the iteration termination condition: number of iterations, iteration time, and iteration termination threshold.

[0064] S32. Encode feasible solutions to the problem and initialize the chromosome population based on the number of possible production scheduling schemes.

[0065] Specifically, step S32 encodes feasible solutions to the problem and initializes the chromosome population, including:

[0066] S321. Based on the product set M = {1, 2, ..., |M|} and the demand set D = {D1, D2, ..., D...}, |M| The chromosomes are encoded, where each chromosome represents a birth planning scheme {Q1, Q2, ... Q}. |Q| The population size can be determined based on the number of all possible production scheduling schemes, |TS|, where |TS| is calculated as follows:

[0067]

[0068] Where, d m =D m / h, where h is the greatest common divisor of the demand quantities of all products in the demand set D.

[0069] S322. If |TS|≤PSIZE, that is, if all possible production schemes |TS| are smaller than the set population size PSIZE, then use these |TS| chromosomes to form the initial chromosome population; if |TS|>PSIZE, then select PSIZE chromosomes from them to form the initial chromosome population.

[0070] S33. Evaluate chromosomes based on the objective function Obj of the variable-rhythm synchronous mixed-flow double-sided production line scheduling problem: Calculate the objective function value of each chromosome in the initial chromosome population, and use the reciprocal of the objective function value as the fitness value of the corresponding chromosome. That is, the smaller the output rhythm, the greater the fitness value of the scheduling scheme.

[0071] S34. Perform chromosome selection based on the evaluation results: Normalize the fitness value of each chromosome in the initial chromosome population, and select PSIZE×P chromosomes using a roulette wheel betting method based on the normalized fitness values. s Each chromosome is used for subsequent crossover and genetic operations.

[0072] S35. Using the sequential crossover method to calculate PSIZE×P s Crossover operation is performed on each chromosome: First, from PSIZE × P... s Of the chromosomes, with probability P c Select a pair of chromosomes as parent chromosomes; then randomly select a gene position as the crossover point. The gene before the crossover point is called the head gene, and the gene after the crossover point is called the tail gene; then exchange the head genes of the two parents; since the product types and quantities contained in the head genes of the two parents may be different, the offspring chromosome generated after the head exchange may have redundant and missing product types and quantities. Check the redundant and missing product types and quantities in the new gene; sequentially search for redundant products in the tail gene and replace the redundant products with missing products to form the new chromosome after the crossover.

[0073] S36. Use the two-point flipping method to perform mutation operations on chromosomes after the crossover operation: for each chromosome, according to probability P... m The mutation operation is performed by reversing some genes. Specifically, for the chromosome selected for mutation, two mutation points are first randomly determined, and then the gene between the two mutations is flipped from left to right.

[0074] S37. Information Acquisition and New Population Generation: The individual with the highest fitness value in the first iteration of the population is set as the global best individual. The chromosome fitness values ​​in each iteration of the population are acquired. If a better fitness value is found, the global best chromosome is updated. The chromosomes resulting from crossover inheritance and the global best chromosome are combined to form a new population.

[0075] S38. End and output the optimal production schedule: Based on the iteration termination condition set by the algorithm, stop the search for the optimal solution of the algorithm, and output the globally optimal chromosome individual, which is the optimal production schedule for the welding car body production line.

[0076] In one embodiment, a welding vehicle body production line with three workstations (A, B, and C) is used, which can produce two products, a and b, in mixed-flow mode. The production demand ratio of the products is 5:3 for product a and product b. The operation time (cycle time) of each product at each workstation is shown in Table 1.

[0077] Table 1 shows the processed workstation operation time matrix.

[0078]

[0079] After the intelligent scheduling method for welding vehicle body production lines, which considers the conveying pattern, proposed in this invention was gradually implemented, the optimal scheduling scheme was obtained as [b,a,a,a,b,a,b,a], with an optimal target value of 14.125. The output cycle time of the optimal scheduling scheme is significantly lower than the maximum station operation cycle time of product b, which is 16. This indicates that the production line with variable-rhythm synchronous conveying has higher efficiency, which is consistent with existing research.

[0080] If we disregard the special characteristics of the conveyor pattern in the welding body production line, i.e., according to the conveyor pattern of the automobile assembly or painting production line, the output cycle time is (14×5+16×3) / 8=14.75, which is significantly higher than the optimal target value that can be obtained by considering the production scheduling problem of the conveyor pattern. This shows that considering the production scheduling problem of the conveyor pattern can yield a more accurate production scheduling plan.

[0081] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "setting," "connection," "fixing," "rotation," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal connection of two components or the interaction between two components. Unless otherwise explicitly limited, those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0082] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart scheduling method for an automotive welding body production line considering conveying patterns, characterized in that, Includes the following steps: S1. Obtain relevant data on the production scheduling decision problem of the welding vehicle body production line, and map the production scheduling decision problem of the welding vehicle body production line into a variable-rhythm synchronous mixed-flow bilateral production line scheduling problem; S2. Establish an optimization model for the scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line, including: S21. Given a fixed number of workstations |S|, the goal of maximizing production efficiency is transformed into minimizing the output cycle time of the production line; S22. Based on the variable rhythm synchronous conveying function, the objective of minimizing the output takt time of the production line is transformed into minimizing the weighted average of the output takt time of all production line states during the cyclic execution of the production sequence; S23. Set constraints for scheduling problems in variable-rhythm synchronous mixed-flow double-sided production lines: Each workstation has two positions, left and right, and a product can only appear in one position of one workstation; Step S22 transforms the objective of minimizing the production line output cycle time into minimizing the weighted average of the output cycle times of all production line states during the cyclic execution of the production sequence, resulting in the following objective function: in, It is the objective function value, represented as the weighted average of the output takt times of the |Q| production lines in a production sequence; It is the first The output cycle time of each production line status and Having the same range of values ; Indicates product The working time at the s-th workstation It is a decision variable, if the production sequence If the qth product appears at the sth workstation, then... ,otherwise ; It is an expression relating workstations, production sequences, and production line states in a variable-rhythm synchronous assembly line, denoted as the first... The s-th workstation in the production line status is producing the production sequence. The qth product |Q| represents the total number of production line states, and |S| represents the total number of workstations. It is any positive integer that makes the parameter satisfy... , ; S3. Based on the optimization model, a genetic algorithm is used to solve the production scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line.

2. The intelligent scheduling method for an automotive welding body production line considering conveying patterns according to claim 1, characterized in that, The relevant data obtained after mapping in step S1 for the scheduling problem of the variable-rhythm synchronous mixed-flow double-sided production line includes: product set M={1,2,...,|M|}, workstation set S={1,2,...,|S|}, and demand set D={D1,D2,...,D...}. |M| Production scheduling plan set Workstation operation time matrix Where |M| is the total number of product types, |S| is the total number of workstations, |TS| is the number of production scheduling plans, and D... m Let D represent the demand for the m-th product. It is the first in the production scheduling plan set. Production scheduling plan , It is the first In the production scheduling plan, the first Location-based products It is the number of products included in a production scheduling plan, and satisfy , It is D={D1,D2,...,D |M| The greatest common divisor of the demand for all products in the matrix; elements in the workstation operation time matrix. This represents the processing time of the m-th product at the s-th workstation, where Let m be the processing time of the m-th product to the left of the s-th workstation. Let m be the processing time of the m-th product to the right of the s-th workstation.

3. The intelligent scheduling method for an automotive welding body production line considering conveying patterns according to claim 1, characterized in that, Step S3, based on the optimization model, uses a genetic algorithm to solve the scheduling problem of a variable-rhythm synchronous mixed-flow double-sided production line, including: S31. Set the parameters for the feasible solution of the problem in the genetic operation process: population size PSIZE, selection probability Crossover probability Probability of mutation Set the parameters in the iteration termination condition: number of iterations, iteration time, and iteration termination threshold; S32. Encode feasible solutions to the problem and initialize the chromosome population; S33. Calculate the objective function value for each chromosome in the initial chromosome population, and use the reciprocal of the objective function value as the fitness value of the corresponding chromosome; S34. Normalize each fitness value, and use the roulette wheel betting method to select PSIZE×P based on the normalized fitness values. s One chromosome; S35. Using the sequential crossover method to calculate PSIZE×P s Crossover operation is performed on each chromosome; S36. Use the two-point flipping method to perform mutation operations on chromosomes after the crossover operation; S37. Obtain a new population and iterate until the iteration termination condition is met, then output the optimal chromosome.

4. The intelligent scheduling method for an automotive welding body production line considering conveying patterns according to claim 3, characterized in that, Step S32 encodes feasible solutions to the problem and initializes the chromosome population, including: S321. Given the product set M = {1, 2, ..., |M|} and the demand set D = {D1, D2, ..., D...}, |M| The encoding yields |TS| chromosomes, where each chromosome represents a production scheduling plan; S322. If |TS|≤PSIZE, then use these |TS| chromosomes to form the initial chromosome population; if |TS|>PSIZE, then select PSIZE chromosomes from them to form the initial chromosome population.

5. The intelligent scheduling method for an automotive welding body production line considering conveying patterns according to claim 4, characterized in that, The formula for calculating the chromosome number |TS| in step S321 is: in, , It is the greatest common divisor of the demand for all products in the demand set D. The number of products included in a production scheduling plan.

Images