A method and device for collaborative optimization of production scheduling and distribution under mass customization

Abstract

The application discloses a production scheduling and distribution collaborative optimization method and device under mass customization, and relates to the technical field of production scheduling and distribution collaborative optimization. The method comprises the following steps: obtaining production data to be optimized; inputting the production data to be optimized into a constructed production scheduling and distribution collaborative optimization model; adopting a multi-objective evolutionary algorithm IMOEA / COD based on collaboration and decomposition to solve the production scheduling and distribution collaborative optimization model, and obtaining an optimized production scheduling and distribution scheme. The application can solve how to collaboratively control two stages of production and distribution under mass customization, and formulate a reasonable production scheduling and distribution scheme, so as to realize the reduction of maximum completion time, distribution cost and lead time / tardiness cost. The application considers the mutual influence of the two stages of production and distribution, can fully utilize existing resources, and effectively solves the production scheduling and distribution collaborative optimization problem under mass customization.

Description

Technical Field

[0001] This invention relates to the field of production scheduling and distribution collaborative optimization technology, and in particular to a method and apparatus for production scheduling and distribution collaborative optimization under mass customization. Background Technology

[0002] With rapid economic and technological development, the variety of goods is increasing, and customer demands are becoming more diversified and personalized. Traditional mass production models are struggling to adapt to market development, posing new challenges to the manufacturing industry. Mass customization enables the high-speed, low-cost production of customized products, providing manufacturing enterprises with a low-cost, diversified solution for survival in fierce market competition. Many manufacturing companies have already implemented this production model.

[0003] Mass customization production mainly consists of two stages: component processing and product assembly. The component processing stage involves mixed-line production of standard and customized parts. Standard parts are needed for all products and are produced in large quantities. Customized parts reflect the product's individual characteristics and require production based on orders. Product assembly is driven by customer demand.

[0004] Meanwhile, for manufacturing enterprises, production and distribution are two crucial value-added activities, and their operational efficiency and costs impact the overall effectiveness of the supply chain. In the traditional "production-distribution" operation model, production and distribution are operated independently by two departments, lacking timely information sharing and operational coordination, making it difficult to achieve optimal overall operation. Especially under mass customization models, inefficient decisions in production and distribution directly affect customer satisfaction and enterprise operational efficiency. Therefore, it is necessary to combine production and distribution, developing reasonable and efficient collaborative optimization methods to maximize the overall benefits for both the enterprise and its customers. Summary of the Invention

[0005] This invention addresses the problem of how to rationally formulate a collaborative optimization scheme for the production and distribution stages under mass customization in order to improve the overall efficiency of customers and enterprises.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] On the one hand, this invention provides a method for optimizing production scheduling and delivery coordination under mass customization. This method is implemented by electronic devices and includes:

[0008] S1. Obtain the production data to be optimized; the production data includes order demand information, workshop processing information, and delivery vehicle information.

[0009] S2. Input the production data to be optimized into the constructed production scheduling and delivery collaborative optimization model.

[0010] S3. The IMOEA / COD multi-objective evolutionary algorithm based on collaboration and decomposition is used to solve the collaborative optimization model of production scheduling and delivery, and the optimized production scheduling and delivery scheme is obtained.

[0011] Optionally, the order demand information in S1 includes: the time window, demand quantity, volume, geographical location of the customer to which each order belongs, and the unit time cost of early or delayed delivery.

[0012] Workshop processing information includes: workshop geographical location, order processing information, and standard parts processing information; among which, order processing information includes: order assembly time, customized parts required for assembling the order, and process information for each customized part; standard parts processing information includes: process information for each standard part, batch number, and sub-batch quantity.

[0013] Delivery vehicle information includes: vehicle capacity limits, driving speed, fixed usage costs, and delivery costs per unit distance.

[0014] Optionally, the construction process of the production scheduling and delivery collaborative optimization model in S2 includes:

[0015] S21. Set the parameters for the production scheduling and delivery collaborative optimization model; among which, the parameters include: population size PN, iteration number Iter, and crossover probability P. c Probability of mutation P m And the neighborhood update threshold T.

[0016] S22. Set the objective function and constraints to obtain the production scheduling and delivery collaborative optimization model; where the objective function is to minimize the maximum completion time, minimize the delivery cost, and minimize the cost of early or late delivery.

[0017] The constraints include: any operation can only be assigned to one machine at any given time; the next operation can only begin on the same machine after the previous one is completed; the completion time of any process for a component depends on the start time and processing time; the completion time of order assembly depends on the start time and assembly time; the next process for the same component must be completed before the next process can begin; assembly can only begin after all customized parts required for order assembly are completed; each component starts processing from time 0; the number of orders loaded in each vehicle cannot exceed the capacity limit; the departure time of each vehicle cannot be earlier than the completion time of the orders in the vehicle; the order delivery time is equal to the delivery time of the previous order in the same vehicle or the vehicle's departure time, plus the travel time; each vehicle departs from the workshop, completes its task, and returns to the workshop; each order has one and only one vehicle responsible for delivery; decision variables take values ​​of 0 or 1; constraint variables take non-negative values.

[0018] Optionally, in S3, the multi-objective evolutionary algorithm IMOEA / COD based on cooperation and decomposition is adopted to solve the collaborative optimization model of production scheduling and distribution, and an optimized production scheduling and distribution plan is obtained, including:

[0019] S31. Initialize the population.

[0020] S32. Assign the weight vectors of sub-problems to each individual in the initial population.

[0021] S33. Select two parent individuals in the population to perform the crossover operation to generate an individual x cross .

[0022] S34. Perform the mutation operation on the individual x cross to generate an offspring individual x child .

[0023] S35. Perform the cooperative decoding operation on the offspring individual x child to obtain the production scheduling and distribution plan.

[0024] S36. Update the neighborhood of the individual x child .

[0025] S37. Determine whether the preset iteration number Iter is satisfied; if so, output the external archive, that is, the non-dominated solution set obtained after iteration, to obtain the optimized production scheduling and distribution plan; otherwise, go to execute S33.

[0026] Optionally, the initialization of the population in S31 includes:

[0027] S311. Use the genetic algorithm to solve the distribution stage successively with the distribution cost and the early or tardy cost as the objectives to obtain the distribution plans P1 and P2.

[0028] S312. Generate a random number r between 0 and 1; if r ≤ 0.25, then execute S313; if 0.25 < r ≤ 0.5, then select the P1 as the distribution plan P c of the current individual and execute S314; if 0.5 < r ≤ 0.75, then select the P2 as the distribution plan P c of the current individual and execute S314; if 0.75 < r ≤ 1, then execute S315.

[0029] S313. Randomly generate the individual encoding and store it in the initial population P init , and execute S316; where the individual encoding includes the production part encoding I1 of the individual and the distribution part encoding I2 of the individual.

[0030] S314. Use the encoding of P c as the distribution part encoding I2 of the individual; according to Pc The vehicle departure time and order allocation information are used to prioritize the production of orders in vehicles that depart earlier. Customized parts required for assembly orders are randomly sorted and stored in the individual's production part code I1. Standard parts sub-batches are randomly inserted into I1. The individual codes are then stored in the initial population P. init Execute S316.

[0031] S315. Randomly generate the production part code I1 for each individual; based on the order of customized parts in I1, and under the premise of prioritizing the delivery of orders corresponding to the customized parts produced first and the vehicle capacity limit, generate the delivery part code I2 for each individual; store the individual code in the initial population P. init Execute S316.

[0032] S316. Determine the initial population P init If the population size is less than the preset population size PN, proceed to step S312; otherwise, output the initial population size P. init .

[0033] Optionally, in S33, two parent individuals in the selected population undergo a crossover operation to generate individual x. cross ,include:

[0034] S331. Generate a random number r2 between 0 and 1. If r2 ≤ 0.5, randomly select two parent individuals x1 and x2 from the population; otherwise, randomly select two parent individuals x1 and x2 from the neighborhood of individual x.

[0035] S332. Generate a random number r3 between 0 and 1, only if r3 ≤ the preset crossover probability P. c At that time, perform a crossover operation on the parent individuals x1 and x2, and randomly select one individual from the crossover individuals as individual x. cross .

[0036] Optionally, in S34, the pair of individual x cross Perform a mutation operation to generate a offspring individual x. child ,include:

[0037] Generate a random number r4 between 0 and 1, only if r4 ≤ preset mutation probability P. m At that time, for individual x cross Perform a mutation operation to generate a offspring individual x. child .

[0038] Optionally, in S35, the pair of offspring individuals x child Perform collaborative decoding to obtain production scheduling and delivery plans, including:

[0039] S351, Obtain offspring individual x childThe delivery portion is coded as I2. The nearest insertion method is used to solve the problem with delivery cost as the objective, and the vehicle departure time is adjusted to generate delivery plan D.

[0040] S352. Take the vehicle departure time of delivery plan D as the optimal production completion time for the orders in the vehicle, and obtain the offspring individual x. child The production part is encoded as I1, and i = 0.

[0041] S353. If i is less than the total number of genes in I1, then execute S354; otherwise, generate a production scheduling and distribution plan S. pd Execute S358.

[0042] S354, Obtain the i-th gene g of I1 i If g i If it is a standard part, then execute S355; otherwise, execute S356.

[0043] S355, if g i If scheduling has not started, the earliest start time of the current process is 0; otherwise, the earliest start time of the current process is the completion time of the previous process. Apply the gap squeezing method for forward scheduling; let i = i + 1, and proceed to execute S353.

[0044] S356, if g i If the order to which it belongs has not been assembled, then obtain g. i The optimal production completion time for the order; apply the gap squeezing method for reverse scheduling and execute S357; otherwise, execute S357.

[0045] S357, if g i If scheduling has not yet started, the optimal completion time for the current process is g. i The start time of assembly for the order; otherwise, the optimal completion time of the current process is the start time of the subsequent process; apply the gap squeezing method for reverse scheduling; let i = i + 1, and proceed to execute S353.

[0046] S358, if S pd If the start time of the intermediate production section is not negative, then execute S359; otherwise, adjust S sequentially. pd The production and distribution sections enable S pd It is feasible; execute S359.

[0047] S359, Output Production Scheduling and Distribution Plan S pd .

[0048] On the other hand, the present invention provides a production scheduling and delivery coordination optimization device under mass customization. This device is applied to realize the production scheduling and delivery coordination optimization method under mass customization. The device includes:

[0049] The acquisition module is used to acquire production data to be optimized; the production data includes order demand information, workshop processing information, and delivery vehicle information.

[0050] The input module is used to input the production data to be optimized into the pre-built production scheduling and delivery collaborative optimization model.

[0051] The output module is used to solve the production scheduling and delivery collaborative optimization model using the IMOEA / COD multi-objective evolutionary algorithm based on collaboration and decomposition, and obtain the optimized production scheduling and delivery scheme.

[0052] Optionally, order demand information includes: the time window, demand quantity, volume, geographical location of the customer to which each order belongs, and the unit time cost of early or delayed delivery.

[0053] Workshop processing information includes: workshop geographical location, order processing information, and standard parts processing information; among which, order processing information includes: order assembly time, customized parts required for assembling the order, and process information for each customized part; standard parts processing information includes: process information for each standard part, batch number, and sub-batch quantity.

[0054] Delivery vehicle information includes: vehicle capacity limits, driving speed, fixed usage costs, and delivery costs per unit distance.

[0055] Optionally, the input module is further used for:

[0056] S21. Set the parameters for the production scheduling and delivery collaborative optimization model; among which, the parameters include: population size PN, iteration number Iter, and crossover probability P. c Probability of mutation P m And the neighborhood update threshold T.

[0057] S22. Set the objective function and constraints to obtain the production scheduling and delivery collaborative optimization model; where the objective function is to minimize the maximum completion time, minimize the delivery cost, and minimize the cost of early or late delivery.

[0058] The constraints include: Any operation can only be assigned to one machine at a certain moment; The next operation of the same machine can only start after the previous one is completed; The completion time of any process of a component depends on the start time and processing time; The completion time of order assembly depends on the start assembly time and assembly time; The next process of the same component can only start after the previous process is completed; Assembly of an order can only start after all customized parts required for the order are completed; Each component starts processing from time 0; The number of orders loaded in each vehicle cannot exceed the capacity limit; The departure time of each vehicle cannot be earlier than the production completion time of the orders in the vehicle; The order delivery time is equal to the delivery time of the previous order in the same vehicle or the vehicle departure time, plus the driving time; Each vehicle departs from the workshop, returns to the workshop after completing the task; Each order is delivered by one and only one vehicle; The decision variable takes values of 0 or 1; The constraint variable takes non - negative values.

[0059] Optionally, the output module is further configured to:

[0060] S31. Initialize the population.

[0061] S32. Assign a weight vector of sub - problems to each individual in the initial population.

[0062] S33. Select two parent individuals in the population to perform a crossover operation to generate an individual x cross .

[0063] S34. Perform a mutation operation on the individual x cross to generate an offspring individual x child .

[0064] S35. Perform a collaborative decoding operation on the offspring individual x child to obtain a production scheduling and distribution plan.

[0065] S36. Update the neighborhood of the individual x child .

[0066] S37. Determine whether the preset iteration number Iter is satisfied; if so, output the external archive, that is, the non - dominated solution set obtained after iteration, to obtain an optimized production scheduling and distribution plan; otherwise, go to execute S33.

[0067] Optionally, the output module is further configured to:

[0068] S311. Use the genetic algorithm to solve the distribution stage successively with the distribution cost and the early or tardy cost as the objectives, and obtain distribution plans P1 and P2.

[0069] S312. Generate a random number r between 0 and 1; if r ≤ 0.25, then execute S313; if 0.25 < r ≤ 0.5, then select the P1 as the distribution plan P of the current individual c, execute S314; if 0.5 < r ≤ 0.75, then select the P2 as the delivery plan P of the current individual c , execute S314; if 0.75 < r ≤ 1, then execute S315.

[0070] S313. Randomly generate individual codes and store them in the initial population P init , execute S316; where the individual code includes the production part code I1 of the individual and the delivery part code I2 of the individual.

[0071] S314. Take the code of P c as the delivery part code I2 of the individual; according to the vehicle departure time and order allocation information in P c , under the premise of giving priority to the production of orders in the vehicle that departs early, randomly sort the customized parts required for the assembly order and store them in the production part code I1 of the individual; randomly insert the standard part sub-batch into I1; store the individual code in the initial population P init , execute S316.

[0072] S315. Randomly generate the production part code I1 of the individual; according to the order of the customized parts in I1, under the premise of giving priority to the delivery of the orders corresponding to the customized parts to be produced first and the vehicle capacity limit, generate the delivery part code I2 of the individual; store the individual code in the initial population P init , execute S316.

[0073] S316. Judge whether the population quantity of the initial population P init is less than the preset population quantity PN. If so, go to execute S312; otherwise, output the initial population P init .

[0074] Optionally, the output module is further used for:

[0075] S331. Generate a random number r2 between 0 and 1. If r2 ≤ 0.5, then randomly select two parent individuals x1, x2 from the population; otherwise, randomly select two parent individuals x1, x2 from the neighborhood of the individual x.

[0076] S332. Generate a random number r3 between 0 and 1. Only when r3 ≤ the preset crossover probability P c , perform a crossover operation on the parent individuals x1, x2, and randomly select an individual from the crossed individuals as the individual x cross .

[0077] Optionally, the output module is further used for:

[0078] Generate a random number r4 between 0 and 1. Only when r4 ≤ the preset mutation probability P mAt that time, for individual x cross Perform a mutation operation to generate a offspring individual x. child .

[0079] Optionally, the output module is further used for:

[0080] S351, Obtain offspring individual x child The delivery portion is coded as I2. The nearest insertion method is used to solve the problem with delivery cost as the objective, and the vehicle departure time is adjusted to generate delivery plan D.

[0081] S352. Take the vehicle departure time of delivery plan D as the optimal production completion time for the orders in the vehicle, and obtain the offspring individual x. child The production part is encoded as I1, and i = 0.

[0082] S353. If i is less than the total number of genes in I1, then execute S354; otherwise, generate a production scheduling and distribution plan S. pd Execute S358.

[0083] S354, Obtain the i-th gene g of I1 i If g i If it is a standard part, then execute S355; otherwise, execute S356.

[0084] S355, if g i If scheduling has not started, the earliest start time of the current process is 0; otherwise, the earliest start time of the current process is the completion time of the previous process. Apply the gap squeezing method for forward scheduling; let i = i + 1, and proceed to execute S353.

[0085] S356, if g i If the order to which it belongs has not been assembled, then obtain g. i The optimal production completion time for the order; apply the gap squeezing method for reverse scheduling and execute S357; otherwise, execute S357.

[0086] S357, if g i If scheduling has not yet started, the optimal completion time for the current process is g. i The start time of assembly for the order; otherwise, the optimal completion time of the current process is the start time of the subsequent process; apply the gap squeezing method for reverse scheduling; let i = i + 1, and proceed to execute S353.

[0087] S358, if S pd If the start time of the intermediate production section is not negative, then execute S359; otherwise, adjust S sequentially. pd The production and distribution sections enable S pd It is feasible; execute S359.

[0088] S359, Output Production Scheduling and Distribution Plan S pd .

[0089] On the one hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, which is loaded and executed by the processor to implement the above-mentioned production scheduling and delivery collaborative optimization method under mass customization.

[0090] On the one hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement the above-mentioned production scheduling and delivery collaborative optimization method under mass customization.

[0091] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0092] The above solution establishes a collaborative optimization model for the production scheduling and delivery coordination optimization problem under mass customization. Considering the overall benefits to both the enterprise and customers, the optimization objective is set as minimizing the maximum completion time, delivery cost, and early / late delivery costs. To achieve collaborative optimization of production and delivery, a multi-objective evolutionary algorithm based on collaboration and decomposition is designed. The designed collaborative initialization strategy can generate high-quality initial solutions, and the designed collaborative decoding strategy can balance the paradoxes among multiple objectives. The designed method can effectively solve the production scheduling and delivery coordination optimization problem under mass customization, making full use of existing resources and improving customer satisfaction and enterprise operational efficiency.

[0093] This invention can avoid the inefficiencies and irrationalities caused by separate production and distribution decisions, and helps to reduce maximum completion time, distribution costs, and early / late costs. Attached Figure Description

[0094] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0095] Figure 1 This is a schematic diagram of the production scheduling and delivery collaborative optimization method under large-scale customization provided in the embodiments of the present invention;

[0096] Figure 2 This is a flowchart of the collaborative initialization strategy provided in an embodiment of the present invention;

[0097] Figure 3 This is a flowchart of the collaborative decoding strategy provided in an embodiment of the present invention;

[0098] Figure 4 This is a block diagram of the production scheduling and delivery coordination optimization device under large-scale customization provided in the embodiments of the present invention;

[0099] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0100] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0101] like Figure 1 As shown, this embodiment of the invention provides a method for optimizing production scheduling and delivery coordination under mass customization, which can be implemented by electronic devices. Figure 1 The flowchart shown illustrates a method for optimizing production scheduling and delivery collaboration under mass customization. This method's processing flow may include the following steps:

[0102] S1. Obtain the production data to be optimized.

[0103] Production data can include order demand information, workshop processing information, and delivery vehicle information.

[0104] Optionally, the order demand information in S1 may include: the time window, demand quantity, volume, geographical location of the customer to which each order belongs, and the unit time cost of early or delayed delivery.

[0105] Workshop processing information may include: workshop location, order processing information, and standard parts processing information.

[0106] The order processing information may include: the order assembly time, the custom parts required for assembling the order, and the process information for each custom part.

[0107] The processing information for standard parts includes: process information for each standard part, batch number, and sub-batch quantity.

[0108] Delivery vehicle information may include: vehicle capacity limits, driving speed, fixed usage costs, and delivery costs per unit distance.

[0109] S2. Input the production data to be optimized into the constructed production scheduling and delivery collaborative optimization model.

[0110] Optionally, step S2 above may include the following steps S21-S22:

[0111] S21. Set the parameters for the production scheduling and delivery collaborative optimization model.

[0112] The parameters include, but are not limited to: population size PN, number of iterations Iter, and crossover probability P. c Probability of mutation P m And the neighborhood update threshold T.

[0113] S22. Set the objective function and constraints to obtain the production scheduling and delivery collaborative optimization model.

[0114] The objective function is to minimize the maximum completion time, minimize the delivery cost, and minimize the cost of early or late delivery.

[0115] The constraints include: any operation can only be assigned to one machine at any given time; the next operation can only begin on the same machine after the previous one is completed; the completion time of any process for a component depends on the start time and processing time; the completion time of order assembly depends on the start time and assembly time; the next process for the same component must be completed before the next process can begin; assembly can only begin after all customized parts required for order assembly are completed; each component starts processing from time 0; the number of orders loaded in each vehicle cannot exceed the capacity limit; the departure time of each vehicle cannot be earlier than the completion time of the orders in the vehicle; the order delivery time is equal to the delivery time of the previous order in the same vehicle or the vehicle's departure time, plus the travel time; each vehicle departs from the workshop, completes its task, and returns to the workshop; each order has one and only one vehicle responsible for delivery; decision variables take values ​​of 0 or 1; constraint variables take non-negative values.

[0116] In one feasible implementation, the target value is calculated by combining the model to implement constraints such as processing order constraints, component feature constraints, order assembly constraints, vehicle loading constraints, and vehicle departure time constraints.

[0117] Furthermore, the objective function can be expressed by the following equation (1):

[0118] f = min(C) max D cost ,ET cost (1)

[0119] Among them, C max For the maximum completion time, D cost For delivery costs, ET cost The objective function of this application is to minimize the maximum completion time, delivery cost, and early / late costs, in order to determine the costs of early / late delivery.

[0120] Furthermore, the maximum completion time in the objective function is represented by equation (2), the delivery cost by equation (3), and the early / late delivery cost by equation (4):

[0121] C max =max(PC jklDC i (2)

[0122]

[0123]

[0124] Where i, i′ are node indices, i, i′ = 0 represents a workshop, i, i′ = 1, 2, ..., nc represents an order / customer node; nc represents the total number of customers / orders; j is the component index; k is the component batch index, k =

[0125] 1,2,…,K j ;K j K represents the batch number of component j. If j represents a custom part, then K... j =1; l is the process index, l = 1, 2, ..., ON j ON j PC represents the number of operations for component j; jkl DC represents the completion time of the l-th process in the k-th batch of component j; i The delivery time of order i is represented by U; the total number of vehicles used is represented by FC; the fixed usage cost of the vehicles is represented by VC; the delivery cost per unit distance of the vehicles is represented by u; the vehicle index is represented by u = 1, 2, ..., U; W ii′u Let W represent the decision variable. If the transportation between nodes i and i' is completed by vehicle u, then W... iju =1; otherwise, W iju =0; d ii′ Represents the distance between nodes i and i′; e and t represent the unit time cost of delivery earlier and later than the time window, respectively; [E i ,T i ] represents the delivery time window for order i.

[0126] Furthermore, the constraints can include 1)-10):

[0127] 1) The completion time of any process of a component depends on the start time of that process and the processing time, and is expressed by the following formula (5):

[0128]

[0129] Among them, PS jkl This indicates the start time of the l-th process in the k-th batch of component j; b jk M represents the batch size of component j in the k-th batch; jl This represents the set of available machines for the k-th batch and l-th process of component j; m, g represent machine indices, m, g = 1, 2, ..., mp + ma; mp represents the total number of processing machines, and ma represents the total number of assembly machines; pt ilmX represents the unit processing time of the l-th process for component j on machine m; jklm Let X represent the decision variable. If the k-th batch and l-th process of component j is processed on machine m, then X... jklm =1; otherwise, X jklm =0.

[0130] 2) The order assembly completion time depends on the start time of assembly and the assembly time, as expressed by the following formula (6):

[0131]

[0132] Among them, AC i Indicates the assembly completion time of order i; AS i Q represents the start time of assembly for order i; i Indicates the demand quantity for order i; MA represents the assembly machine set; at im X represents the unit assembly time of order i on machine m; im Let X represent the decision variable. If order i is assembled on machine m, then X... im =1; otherwise, X im =0.

[0133] 3) Each operation can only select one machine, as expressed by the following formulas (7) and (8):

[0134]

[0135]

[0136] Equation (7) represents the parts processing stage, and Equation (8) represents the order assembly stage.

[0137] 4) Each component must complete the previous process before starting the next process, as expressed by the following formula (9):

[0138]

[0139] 5) Assembly can only begin after all the customized parts for the assembly order have been processed, as expressed by the following formula (10).

[0140]

[0141] 6) The next task can only begin after the previous task is completed, as expressed by the following formulas (11) and (12):

[0142]

[0143]

[0144]

[0145] Wherein, equation (11) represents the parts processing stage, equation (12) represents the order assembly stage; MP represents the processing machine set; Z jkl,j′k′l′ Denotes the decision variable, if O jkl It is O j′k′l′ The preceding process, Z jkl,j′k′l′ =1; otherwise, Z jkl,j′k′l′ =0; z i,i′ Let Z represent the decision variable. If order i is the preceding order adjacent to order i', then Z... i,i’ =1; otherwise, Z i,i’ =0.

[0146] 7) Each component can be processed from time 0, as expressed by the following formula (13):

[0147]

[0148] 8) The total volume of orders loaded in each vehicle shall not exceed the maximum capacity, as expressed by the following formula (14):

[0149]

[0150] Among them, V iu V represents the decision variable; if order i is delivered by vehicle u, then V iu =1, otherwise V iu =0; Q represents the vehicle capacity limit.

[0151] 9) The departure time of each vehicle shall not be earlier than the time when the order inside the vehicle is assembled, as expressed by the following formula (15):

[0152]

[0153] Among them, s u This indicates the departure time of vehicle u.

[0154] 10) The order delivery time is equal to the delivery time of the previous order in the same vehicle or the vehicle's departure time, plus the travel time, and is expressed by the following formula (16):

[0155]

[0156] Where v represents the vehicle's speed.

[0157] S3. Use IMOEA / COD (Multi-objective Evolutionary Algorithm Based on Cooperation and Decomposition) to solve the collaborative optimization model of production scheduling and distribution, and obtain an optimized production scheduling and distribution plan.

[0158] Optionally, the above step S3 may include the following steps S31 - S37:

[0159] S31. Initialize the population.

[0160] In a feasible implementation, as Figure 2 shown, a collaborative initialization strategy can be used to generate the initial population.

[0161] Specifically, the above step S31 may include the following steps S311 - S316:

[0162] S311. Use the genetic algorithm to solve the distribution stage sequentially with the distribution cost and the early or tardy cost as the objectives, and obtain distribution plans P1 and P2.

[0163] S312. Generate a random number r between 0 and 1; if r ≤ 0.25, then execute S313; if 0.25 < r ≤ 0.5, then select the P1 as the distribution plan P of the current individual c , and execute S314; if 0.5 < r ≤ 0.75, then select the P2 as the distribution plan P of the current individual c , and execute S314; if 0.75 < r ≤ 1, then execute S315.

[0164] S313. Randomly generate individual codes and store them in the initial population P init , and execute S316; where the individual code includes the production part code I1 of the individual and the distribution part code I2 of the individual.

[0165] S314. Use the code of P c as the distribution part code I2 of the individual; according to the vehicle departure time and order allocation information in P c , under the premise of giving priority to the production of orders in the vehicles that depart early, randomly sort the customized parts required for the assembly order and store them in the production part code I1 of the individual; randomly insert the standard part sub-batches into I1; store the individual code in the initial population P init , and execute S316.

[0166] S315. Randomly generate the production part code I1 for each individual; based on the order of customized parts in I1, and under the premise of prioritizing the delivery of orders corresponding to the customized parts produced first and the vehicle capacity limit, generate the delivery part code I2 for each individual; store the individual code in the initial population P. init Execute S316.

[0167] S316. Determine the initial population P init If the population size is less than the preset population size PN, proceed to step S312; otherwise, output the initial population size P. init .

[0168] S32. Assign a weight vector to each individual in the initial population for the subproblem.

[0169] In one feasible implementation, the weight vectors of subproblems are assigned to individuals using methods in the prior art, which will not be elaborated here.

[0170] S33. Select two parent individuals from the population and perform a crossover operation to generate individual x. cross .

[0171] Optionally, step S33 above may include the following steps S331-S332:

[0172] S331. Generate a random number r2 between 0 and 1. If r2 ≤ 0.5, randomly select two parent individuals x1 and x2 from the population; otherwise, randomly select two parent individuals x1 and x2 from the neighborhood of individual x.

[0173] S332. Generate a random number r3 between 0 and 1, only if r3 ≤ the preset crossover probability P. c At that time, perform a crossover operation on the parent individuals x1 and x2, and randomly select one individual from the crossover individuals as individual x. cross .

[0174] In one feasible implementation, selecting the parent individual of individual x for cross-operation includes performing cross-operation on the production section and the distribution section respectively.

[0175] S34, For individual x cross Perform a mutation operation to generate a offspring individual x. child .

[0176] Optionally, step S34 above may include:

[0177] Generate a random number r4 between 0 and 1, only if r4 ≤ preset mutation probability P. m At that time, for individual x cross Perform a mutation operation to generate a offspring individual x. child .

[0178] In one feasible implementation, for individual x cross Performing mutation operations involves performing mutation operations on the production and distribution parts of the code separately.

[0179] S35, Regarding offspring individuals x child Perform collaborative decoding to obtain production scheduling and delivery plans, which are then archived externally.

[0180] Optionally, such as Figure 3 As shown, step S35 above may include the following steps S351-S359:

[0181] S351, Obtain offspring individual x child The delivery portion is coded as I2. The nearest insertion method is used to solve the problem with delivery cost as the objective, and the vehicle departure time is adjusted to optimize order delivery, thus generating delivery plan D.

[0182] S352. Take the vehicle departure time of delivery plan D as the optimal production completion time for the orders in the vehicle, and obtain the offspring individual x. child The production part is encoded as I1, and i = 0.

[0183] S353. If i is less than the total number of genes in I1, then execute S354; otherwise, generate a production scheduling and distribution plan S. pd Execute S358.

[0184] S354, Obtain the i-th gene g of I1 i If g i If it is a standard part, then execute S355; otherwise, execute S356.

[0185] S355, if g i If scheduling has not started, the earliest start time of the current process is 0; otherwise, the earliest start time of the current process is the completion time of the previous process. Apply the gap squeezing method for forward scheduling; let i = i + 1, and proceed to execute S353.

[0186] S356, if g i If the order to which it belongs has not been assembled, then obtain g. i The optimal production completion time for the order; apply the gap squeezing method for reverse scheduling and execute S357; otherwise, execute S357.

[0187] S357, if g i If scheduling has not yet started, the optimal completion time for the current process is g. i The start time of assembly for the order; otherwise, the optimal completion time of the current process is the start time of the subsequent process; apply the gap squeezing method for reverse scheduling; let i = i + 1, and proceed to execute S353.

[0188] S358, if S pd If the start time of the intermediate production section is not negative, then execute S359; otherwise, adjust S sequentially. pd The production and distribution sections were made feasible, in accordance with S359.

[0189] S359, Output Production Scheduling and Distribution Plan S pd .

[0190] S36, Update individual x child The neighborhood of.

[0191] In one feasible implementation, the neighborhood of individual x is updated using methods in the prior art, which will not be elaborated here.

[0192] S37. Determine whether the preset iteration number Iter is met; if yes, output the external archive, i.e. the non-dominated solution set obtained after iteration, and obtain the optimized production scheduling and delivery scheme; otherwise, proceed to execute S33.

[0193] In one feasible implementation, the above iterations are repeated until a preset number of iterations is reached, and the resulting external archive is taken as the final result. That is, the production scheduling and delivery scheme corresponding to the non-dominated solution set that minimizes the maximum completion time, delivery cost, and early / late cost. These schemes can help optimize the overall benefits for enterprises and customers.

[0194] In this embodiment of the invention, a collaborative optimization model is established for the production scheduling and delivery coordination optimization problem under mass customization. Considering the overall benefits to both the enterprise and the customer, the optimization objective is set as minimizing the maximum completion time, delivery cost, and early / late delivery costs. To achieve collaborative optimization of production and delivery, a multi-objective evolutionary algorithm based on collaboration and decomposition is designed. The designed collaborative initialization strategy can generate high-quality initial solutions, and the designed collaborative decoding strategy can balance the paradoxes among multiple objectives. The designed method can effectively solve the production scheduling and delivery coordination optimization problem under mass customization, making full use of existing resources and improving customer satisfaction and enterprise operational efficiency.

[0195] This invention can avoid the inefficiencies and irrationalities caused by separate production and distribution decisions, and helps to reduce maximum completion time, distribution costs, and early / late costs.

[0196] like Figure 4 As shown, this embodiment of the invention provides a production scheduling and delivery coordination optimization device 400 under mass customization. This device 400 is used to implement a production scheduling and delivery coordination optimization method under mass customization. The device 400 includes:

[0197] The acquisition module 410 is used to acquire the production data to be optimized; the production data includes order demand information, workshop processing information and delivery vehicle information.

[0198] Input module 420 is used to input the production data to be optimized into the constructed production scheduling and delivery collaborative optimization model.

[0199] Output module 430 is used to solve the production scheduling and delivery collaborative optimization model using the IMOEA / COD multi-objective evolutionary algorithm based on collaboration and decomposition, so as to obtain the optimized production scheduling and delivery scheme.

[0200] Optionally, order demand information includes: the time window, demand quantity, volume, geographical location of the customer to which each order belongs, and the unit time cost of early or delayed delivery.

[0201] Workshop processing information includes: workshop geographical location, order processing information, and standard parts processing information; among which, order processing information includes: order assembly time, customized parts required for assembling the order, and process information for each customized part; standard parts processing information includes: process information for each standard part, batch number, and sub-batch quantity.

[0202] Delivery vehicle information includes: vehicle capacity limits, driving speed, fixed usage costs, and delivery costs per unit distance.

[0203] Optionally, the input module 420 is further used for:

[0204] S21. Set the parameters for the production scheduling and delivery collaborative optimization model; among which, the parameters include: population size PN, iteration number Iter, and crossover probability P. c Probability of mutation P m And the neighborhood update threshold T.

[0205] S22. Set the objective function and constraints to obtain the production scheduling and delivery collaborative optimization model; where the objective function is to minimize the maximum completion time, minimize the delivery cost, and minimize the cost of early or late delivery.

[0206] The constraints include: Any operation can only be assigned to one machine at a certain moment; The next operation of the same machine can only start after the previous one is completed; The completion time of any process of a component depends on the start time and processing time; The completion time of order assembly depends on the start assembly time and assembly time; The same component must complete the previous process before starting the next one; Assembly of an order can only start after all customized parts required for the order are completed; Each component starts processing from time 0; The number of orders loaded in each vehicle cannot exceed the capacity limit; The departure time of each vehicle cannot be earlier than the production completion time of the orders in the vehicle; The delivery time of an order is equal to the delivery time of the previous order in the same vehicle or the vehicle departure time, plus the driving time; Each vehicle departs from the workshop, returns to the workshop after completing the task; Each order is delivered by one and only one vehicle; The decision variable takes values of 0 or 1; The constraint variable takes non - negative values.

[0207] Optionally, the output module 430 is further configured to:

[0208] S31. Initialize the population.

[0209] S32. Assign a weight vector of sub - problems to each individual in the initial population.

[0210] S33. Select two parent individuals in the population to perform a crossover operation to generate an individual x cross .

[0211] S34. Perform a mutation operation on the individual x cross to generate an offspring individual x child .

[0212] S35. Perform a collaborative decoding operation on the offspring individual x child to obtain a production scheduling and distribution plan.

[0213] S36. Update the neighborhood of the individual x child .

[0214] S37. Determine whether the preset iteration number Iter is satisfied; if so, output the external archive, that is, the non - dominated solution set obtained after iteration, to obtain an optimized production scheduling and distribution plan; otherwise, go to execute S33.

[0215] Optionally, the output module 430 is further configured to:

[0216] S311. Use the genetic algorithm to solve the distribution stage successively with the distribution cost and the early or tardy cost as the objectives, and obtain distribution plans P1 and P2.

[0217] S312. Generate a random number r between 0 and 1; if r ≤ 0.25, then execute S313; if 0.25 < r ≤ 0.5, then select the P1 as the distribution plan P of the current individualc , execute S314; if 0.5 < r ≤ 0.75, then select the said P2 as the delivery plan P of the current individual c , execute S314; if 0.75 < r ≤ 1, then execute S315.

[0218] S313. Randomly generate individual codes and store them in the initial population P init , execute S316; where the individual code includes the production part code I1 of the individual and the delivery part code I2 of the individual.

[0219] S314. Take the code of P c as the delivery part code I2 of the individual; according to the vehicle departure time and order allocation information in P c , under the premise of giving priority to the production of orders in the vehicle that departs early, randomly sort the customized parts required for the assembly order and store them in the production part code I1 of the individual; randomly insert the standard part sub - batch into I1; store the individual code in the initial population P init , execute S316.

[0220] S315. Randomly generate the production part code I1 of the individual; according to the order of the customized parts in I1, under the premise of giving priority to the delivery of the orders corresponding to the customized parts to be produced first and the vehicle capacity limit, generate the delivery part code I2 of the individual; store the individual code in the initial population P init , execute S316.

[0221] S316. Judge whether the population quantity of the initial population P init is less than the preset population quantity PN. If so, go back to execute S312; otherwise, output the initial population P init .

[0222] Optionally, the output module 430 is further used for:

[0223] S331. Generate a random number r2 between 0 and 1. If r2 ≤ 0.5, then randomly select two parent individuals x1, x2 from the population; otherwise, randomly select two parent individuals x1, x2 from the neighborhood of the individual x.

[0224] S332. Generate a random number r3 between 0 and 1. Only when r3 ≤ the preset crossover probability P c , perform a crossover operation on the parent individuals x1, x2 and randomly select an individual from the crossed individuals as the individual x cross .

[0225] Optionally, the output module 430 is further used for:

[0226] Generate a random number r4 between 0 and 1. Only when r4 ≤ the preset mutation probability Pm At that time, for individual x cross Perform a mutation operation to generate a offspring individual x. child .

[0227] Optionally, the output module 430 is further used for:

[0228] S351, Obtain offspring individual x child The delivery portion is coded as I2. The nearest insertion method is used to solve the problem with delivery cost as the objective, and the vehicle departure time is adjusted to generate delivery plan D.

[0229] S352. Take the vehicle departure time of delivery plan D as the optimal production completion time for the orders in the vehicle, and obtain the offspring individual x. child The production part is encoded as I1, and i = 0.

[0230] S353. If i is less than the total number of genes in I1, then execute S354; otherwise, generate a production scheduling and distribution plan S. pd Execute S358.

[0231] S354, Obtain the i-th gene g of I1 i If g i If it is a standard part, then execute S355; otherwise, execute S356.

[0232] S355, if g i If scheduling has not started, the earliest start time of the current process is 0; otherwise, the earliest start time of the current process is the completion time of the previous process. Apply the gap squeezing method for forward scheduling; let i = i + 1, and proceed to execute S353.

[0233] S356, if g i If the order to which it belongs has not been assembled, then obtain g. i The optimal production completion time for the order; apply the gap squeezing method for reverse scheduling and execute S357; otherwise, execute S357.

[0234] S357, if g i If scheduling has not yet started, the optimal completion time for the current process is g. i The start time of assembly for the order; otherwise, the optimal completion time of the current process is the start time of the subsequent process; apply the gap squeezing method for reverse scheduling; let i = i + 1, and proceed to execute S353.

[0235] S358, if S pd If the start time of the intermediate production section is not negative, then execute S359; otherwise, adjust S sequentially. pd The production and distribution sections enable S pd It is feasible; execute S359.

[0236] S359, Output Production Scheduling and Distribution Plan S pd .

[0237] In this embodiment of the invention, a collaborative optimization model is established for the production scheduling and delivery coordination optimization problem under mass customization. Considering the overall benefits to both the enterprise and the customer, the optimization objective is set as minimizing the maximum completion time, delivery cost, and early / late delivery costs. To achieve collaborative optimization of production and delivery, a multi-objective evolutionary algorithm based on collaboration and decomposition is designed. The designed collaborative initialization strategy can generate high-quality initial solutions, and the designed collaborative decoding strategy can balance the paradoxes among multiple objectives. The designed method can effectively solve the production scheduling and delivery coordination optimization problem under mass customization, making full use of existing resources and improving customer satisfaction and enterprise operational efficiency.

[0238] This invention can avoid the inefficiencies and irrationalities caused by separate production and distribution decisions, and helps to reduce maximum completion time, distribution costs, and early / late costs.

[0239] Figure 5 This is a schematic diagram of the structure of an electronic device 500 provided in an embodiment of the present invention. The electronic device 500 can vary considerably due to differences in configuration or performance. It may include one or more central processing units (CPUs) 501 and one or more memories 502. The memory 502 stores at least one instruction, which is loaded and executed by the processor 501 to implement the following method for collaborative optimization of production scheduling and delivery under large-scale customization:

[0240] S1. Obtain the production data to be optimized; the production data includes order demand information, workshop processing information, and delivery vehicle information.

[0241] S2. Input the production data to be optimized into the constructed production scheduling and delivery collaborative optimization model.

[0242] S3. The IMOEA / COD multi-objective evolutionary algorithm based on collaboration and decomposition is used to solve the collaborative optimization model of production scheduling and delivery, and the optimized production scheduling and delivery scheme is obtained.

[0243] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including instructions that can be executed by a processor in a terminal to complete the aforementioned production scheduling and delivery collaborative optimization method under mass customization. For example, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0244] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0245] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for collaborative optimization of production scheduling and delivery under mass customization, characterized in that, The method includes: S1. Obtain the production data to be optimized; wherein, the production data includes order demand information, workshop processing information, and delivery vehicle information; S2. Input the production data to be optimized into the constructed production scheduling and delivery collaborative optimization model; S3. The production scheduling and delivery collaborative optimization model is solved by using the multi-objective evolutionary algorithm IMOEA / COD based on collaboration and decomposition to obtain the optimized production scheduling and delivery scheme. The construction process of the production scheduling and delivery collaborative optimization model in S2 includes: S21. Set the parameters for the production scheduling and delivery collaborative optimization model; wherein, the parameters include: population size PN, iteration number Iter, and crossover probability P. c Probability of mutation P m And the neighborhood update threshold T; S22. Set the objective function and constraints to obtain the production scheduling and delivery collaborative optimization model; wherein, the objective function is to minimize the maximum completion time, minimize the delivery cost, and minimize the cost of early or late delivery; The constraints include: any operation can only be assigned to one machine at any given time; the next operation can only begin on the same machine after the previous one is completed; the completion time of any process for a component depends on the start time and processing time; the completion time of order assembly depends on the start time and assembly time; the next process for the same component must be completed before the next process can begin; assembly can only begin after all customized parts required for order assembly are completed; each component starts processing from time 0; the number of orders loaded in each vehicle cannot exceed the capacity limit; the departure time of each vehicle cannot be earlier than the completion time of the orders in the vehicle; the order delivery time is equal to the delivery time of the previous order in the same vehicle or the vehicle's departure time, plus the travel time; each vehicle departs from the workshop, completes its task, and returns to the workshop; each order has one and only one vehicle responsible for delivery; decision variables take values ​​of 0 or 1; constraint variables take non-negative values. The S3 section employs the IMOEA / COD multi-objective evolutionary algorithm based on cooperation and decomposition to solve the production scheduling and delivery collaborative optimization model, obtaining the optimized production scheduling and delivery scheme, including: S31. Initialize the population; S32. Assign a weight vector of the subproblem to each individual in the initial population; S33. Select two parent individuals from the population and perform a crossover operation to generate individual x. cross ; S34, for the individual x cross Perform a mutation operation to generate a offspring individual x. child ; S35, regarding the offspring individual x child Perform collaborative decoding to obtain production scheduling and delivery plans; S36, Update individual x child ; S37. Determine whether the preset iteration number Iter is met; if yes, output the external archive, i.e., the non-dominated solution set obtained after iteration, to obtain the optimized production scheduling and delivery scheme; otherwise, proceed to execute S33.

2. The method according to claim 1, characterized in that, The order demand information in S1 includes: the time window, demand quantity, volume, geographical location of the customer to which each order belongs, and the unit time cost of early or delayed delivery; The workshop processing information includes: workshop geographical location, order processing information, and standard parts processing information; wherein, the order processing information includes: order assembly time, customized parts required for assembling the order, and process information for each customized part; the standard parts processing information includes: process information for each standard part, batch number, and sub-batch quantity; The delivery vehicle information includes: the delivery vehicle's capacity limit, driving speed, fixed usage cost, and delivery cost per unit distance.

3. The method according to claim 1, characterized in that, The initial population in S31 includes: S311. Using a genetic algorithm, with delivery cost and early or late delivery cost as objectives, solve the delivery stage sequentially to obtain delivery schemes P1 and P2. S312. Generate a random number r between 0 and 1; if r ≤ 0.25, then execute S313; if 0.25 < r ≤ 0.5, then select the said P1 as the delivery plan P of the current individual c , execute S314; if 0.5 < r ≤ 0.75, then select the said P2 as the delivery plan P of the current individual c , execute S314; if 0.75 < r ≤ 1, then execute S315; S313. Randomly generate individual codes and store them in the initial population P. init Execute S316; wherein the individual code includes the individual's production part code I1 and the individual's distribution part code I2; S314, P c The code is used as the individual's delivery part code I2; according to P c The vehicle departure time and order allocation information are used to prioritize the production of orders in vehicles with earlier departure times. Customized parts required for assembly orders are randomly sorted and stored in the individual's production part code I1. Standard parts sub-batches are randomly inserted into I1. The individual codes are then stored in the initial population P. init Execute S316; S315. Randomly generate the production part code I1 for each individual; based on the order of customized parts in I1, and under the premise of prioritizing the delivery of orders corresponding to the customized parts produced first and the vehicle capacity limit, generate the delivery part code I2 for each individual; store the individual code in the initial population P. init Execute S316; S316. Determine the initial population P init If the population size is less than the preset population size PN, proceed to step S312; otherwise, output the initial population size P. init .

4. The method according to claim 1, characterized in that, In step S33, two parent individuals from the selected population undergo a crossover operation to generate individual x. cross ,include: S331. Generate a random number r2 between 0 and 1. If r2 ≤ 0.5, randomly select two parent individuals x1 and x2 from the population; otherwise, randomly select two parent individuals x1 and x2 from the neighborhood of individual x. S332. Generate a random number r3 between 0 and 1, only if r3 ≤ the preset crossover probability P. c At that time, perform a crossover operation on the parent individuals x1 and x2, and randomly select one individual from the crossover individuals as individual x. cross .

5. The method according to claim 1, characterized in that, The individual x in S34 cross Perform a mutation operation to generate a offspring individual x. child ,include: Generate a random number r4 between 0 and 1, only if r4 ≤ preset mutation probability P. m At that time, for the individual x cross Perform a mutation operation to generate a offspring individual x. child .

6. The method according to claim 1, characterized in that, The S35 refers to the offspring individual x child Perform collaborative decoding to obtain production scheduling and delivery plans, including: S351. Obtain the offspring individual x child The delivery portion is coded as I2. The nearest insertion method is used to solve the problem with delivery cost as the objective, and the vehicle departure time is adjusted to generate delivery plan D. S352. Take the vehicle departure time of the delivery plan D as the optimal production completion time for the in-vehicle orders, and obtain the child individual x. child The production part is encoded as I1, let i=0; S353. If i is less than the total number of genes in I1, then execute S354; otherwise, generate a production scheduling and distribution plan S. pd Execute S358; S354, Obtain the i-th gene g of I1 i If g i If it is a standard part, then execute S355; otherwise, execute S356. S355, if g i If scheduling has not started, the earliest start time of the current process is 0; otherwise, the earliest start time of the current process is the completion time of the preceding process; apply the gap squeezing method for forward scheduling; let i = i + 1, and proceed to execute S353; S356, if g i If the order to which it belongs has not been assembled, then obtain g. i The optimal production completion time for the order; apply the gap squeezing method for reverse scheduling and execute S357; otherwise, execute S357. S357, if g i If scheduling has not yet started, the optimal completion time for the current process is g. i The start time of assembly for the order; otherwise, the optimal completion time of the current process is the start time of the subsequent process; apply the gap squeezing method for reverse scheduling; let i = i + 1, and proceed to execute S353; S358, if S pd If the start time of the intermediate production section is not negative, then execute S359; otherwise, adjust S sequentially. pd The production and distribution sections enable S pd It is feasible; execute S359. S359, Output Production Scheduling and Distribution Plan S pd .

7. A production scheduling and delivery coordination optimization device for mass customization, characterized in that, The device includes: The acquisition module is used to acquire production data to be optimized; wherein, the production data includes order demand information, workshop processing information, and delivery vehicle information; The input module is used to input the production data to be optimized into the constructed production scheduling and delivery collaborative optimization model; The output module is used to solve the production scheduling and delivery collaborative optimization model using the IMOEA / COD multi-objective evolutionary algorithm based on collaboration and decomposition, so as to obtain the optimized production scheduling and delivery scheme. The construction process of the production scheduling and delivery collaborative optimization model includes: S21. Set the parameters for the production scheduling and delivery collaborative optimization model; wherein, the parameters include: population size PN, iteration number Iter, and crossover probability P. c Probability of mutation P m And the neighborhood update threshold T; S22. Set the objective function and constraints to obtain the production scheduling and delivery collaborative optimization model; wherein, the objective function is to minimize the maximum completion time, minimize the delivery cost, and minimize the cost of early or late delivery; The constraints include: any operation can only be assigned to one machine at any given time; the next operation can only begin on the same machine after the previous one is completed; the completion time of any process for a component depends on the start time and processing time; the completion time of order assembly depends on the start time and assembly time; the next process for the same component must be completed before the next process can begin; assembly can only begin after all customized parts required for order assembly are completed; each component starts processing from time 0; the number of orders loaded in each vehicle cannot exceed the capacity limit; the departure time of each vehicle cannot be earlier than the completion time of the orders in the vehicle; the order delivery time is equal to the delivery time of the previous order in the same vehicle or the vehicle's departure time, plus the travel time; each vehicle departs from the workshop, completes its task, and returns to the workshop; each order has one and only one vehicle responsible for delivery; decision variables take values ​​of 0 or 1; constraint variables take non-negative values. The production scheduling and delivery collaborative optimization model is solved using the IMOEA / COD multi-objective evolutionary algorithm based on collaboration and decomposition to obtain the optimized production scheduling and delivery scheme, including: S31. Initialize the population; S32. Assign a weight vector of the subproblem to each individual in the initial population; S33. Select two parent individuals from the population and perform a crossover operation to generate individual x. cross ; S34, for the individual x cross Perform a mutation operation to generate a offspring individual x. child ; S35, regarding the offspring individual x child Perform collaborative decoding to obtain production scheduling and delivery plans; S36, Update individual x child ; S37. Determine whether the preset iteration number Iter is met; if yes, output the external archive, i.e., the non-dominated solution set obtained after iteration, to obtain the optimized production scheduling and delivery scheme; otherwise, proceed to execute S33.

Images