Cross-workshop scheduling and process planning integration method based on flexible production

Abstract

The invention discloses a cross-workshop scheduling and process planning integration method based on flexible production, which belongs to the field of intelligent manufacturing and comprises a factory module, a forging and pressing workshop module, a cross-workshop transportation module and a machining workshop scheduling and process planning module. The factory module firstly generates an initialized scheduling population according to forging, vehicle transportation and machining information of all orders, and forging workshop coding information is input into the forging workshop module. The forging and pressing workshop module carries out scheduling and production scheduling solving in a forging and pressing workshop according to forging and pressing workshop coding information and order forging and pressing information, and the cross-workshop transportation module carries out integration and batching according to the completion sequence and the time sequence of all orders in the forging and pressing stage to obtain the transportation completion time of each batch. And transmitting the workpiece information of each batch and the transportation completion time to a machining workshop scheduling and process planning module to obtain the processing completion time and production energy consumption of all orders, and finally determining an optimal workshop allocation scheme and a cross-workshop transportation scheme.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing, and specifically relates to an integrated method for cross-workshop scheduling and process planning based on flexible production. Background Technology

[0002] With the advent of the information age and the rapid development of the manufacturing industry, user demands for products are becoming increasingly diverse and changeable. This has rendered traditional workshop production models inadequate to meet the needs of modern enterprise development. At the same time, environmental degradation is becoming increasingly severe, and countries worldwide have recognized the resource waste and environmental pollution caused by inefficient energy use. In basic industries, forging and machining, as two main processing methods, are widely used in key industries such as equipment manufacturing and aerospace. Both are also energy-intensive industries, necessitating a transformation of traditional production models towards intelligent and efficient production methods through the adoption of advanced strategies.

[0003] With continuous innovation and breakthroughs in intelligent manufacturing technology, smart factories have emerged and are becoming an important direction for the transformation and upgrading of the manufacturing industry. However, facing the complex production environment consisting of multiple workshops, adhering to the independent production model of a single workshop has become a bottleneck. Cross-workshop collaborative production has become a key technology to solve this problem. Modern flexible manufacturing systems are an indispensable part of smart factories, encompassing all aspects of workshop production and possessing high flexibility and collaboration. Process planning and scheduling are two crucial aspects of modern flexible manufacturing systems, and there is a close connection between them. However, in traditional manufacturing models, process planning and scheduling are usually carried out strictly in sequence according to departmental responsibilities, without fully considering the inherent relationship between the two. This often leads to unreasonable utilization of equipment resources and serious resource waste in actual production. Therefore, effectively integrating these two aspects is of great significance for modern factory management and production.

[0004] Based on the above, this patent takes the distributed workshop of a forging and pressing enterprise as the research object, and conducts research on cross-workshop scheduling and process planning integration for the multi-stage actual production process of die forging and machining. It is committed to solving the problem of linkage and collaborative production of multiple workshops, as well as the key problems such as unreasonable resource use and low production efficiency in process planning and scheduling schemes in actual production. It has important practical significance for improving workshop production efficiency and energy utilization. Summary of the Invention

[0005] This invention addresses the need for close coordination among multiple stages of actual production in forging and machining. It proposes an integrated method for cross-workshop scheduling and process planning based on flexible manufacturing to achieve collaborative production across multiple workshops. While considering both intra-workshop scheduling and cross-workshop transportation scheduling, a multi-objective optimization algorithm and decoding / compilation strategy suitable for solving this model are also proposed. The specific implementation steps are as follows:

[0006] Step 1: Cross-workshop scheduling population initialization and forging task transfer. The factory module algorithm parameters are initially set, and a three-layer real number encoding method is used for factory module encoding. Based on the forging information, vehicle transportation information, and machining information of all orders, the initial cross-workshop scheduling code is generated. The factory module encoding information includes the forging workshop selection information code, transportation vehicle selection information code, and machining workshop selection information code for all orders. The encoding information of the forging workshop allocation in the obtained initial cross-workshop scheduling scheme is input into the forging workshop module.

[0007] Step 1.1: Initialize and set the algorithm parameters for the factory module;

[0008] Step 1.2: Enter the forging information, vehicle transportation information, and machining information for all orders;

[0009] Step 1.3: Randomly generate an initial cross-workshop scheduling population, where the encoded information includes the forging workshop selection information, transport vehicle selection information, and machining workshop selection information for all orders;

[0010] Step 1.4: Input the coding information of each order's forging workshop allocation in the initial cross-workshop scheduling scheme into the forging workshop module;

[0011] Step 2: Generation and solution of scheduling schemes within each forging workshop. The intelligent optimization algorithm parameters of the forging workshop module are initialized. Based on the coding information allocated to the forging workshop and the forging information of the orders, a segmented real-number three-layer initial code is generated. The scheduling scheme is decoded using the designed decoding and compilation strategy, and a multi-objective optimization algorithm is used to solve the problem, obtaining the solved scheduling scheme for each forging workshop. The solution results are then transmitted to the cross-workshop transportation module.

[0012] Step 2.1: Initialize and set parameters for the multi-objective optimization algorithm in the forging workshop module;

[0013] Step 2.2: Input the forging workshop allocation information and die forging process information for all orders to be processed. The process information includes heating information, forging information and optional equipment information for each process.

[0014] Step 2.3: Based on the characteristic that the die forging workshop needs to change dies in batches according to orders, a segmented real number three-layer initial code initialization scheduling population is randomly generated, which includes the order sequence and the machine information selected for each heating and forging process;

[0015] Step 2.4: Decode the scheduling scheme based on the encoded information and the temperature regulation decoding and compilation strategy to obtain the completion time and energy consumption of each forging workshop after decoding the scheduling scheme;

[0016] Step 2.5: After the decoding operation is completed, the non-dominated sorting and congestion calculation are performed using the completion time and energy consumption of each forging workshop's order processing as optimization objectives;

[0017] Step 2.6: Generate a new population of forging workshop scheduling schemes through genetic operations such as selection, crossover, and mutation, and then proceed to the operations in Step 2.7;

[0018] Step 2.7: Determine whether the iteration termination condition has been met. If it has, transmit the optimal scheduling scheme of each forging workshop to the cross-workshop transportation module; otherwise, proceed to step 2.4.

[0019] Step 3: Cross-workshop transportation module decoding and order batch transportation. The cross-workshop transportation module performs cross-workshop transportation based on the completion time sequence of orders from each forging workshop. By establishing a cross-workshop transportation model and performing decoding operations according to the vehicle transportation plan and machining workshop allocation plan encoding information, due to the capacity limitations of transportation vehicles, the order workpieces are transported to each target workshop in batches, obtaining the completion time of each batch to the target machining workshop, and transmitting the transportation completion time to the machining workshop scheduling and process planning module.

[0020] Step 3.1: Sort the orders according to the completion time of each forging workshop to obtain the completion time sequence of all orders.

[0021] Step 3.2: Establish a cross-workshop transportation model and carry out cross-workshop transportation according to the order completion time sequence and the assigned transportation vehicles. Due to the capacity limitations of transportation vehicles, the same order may be transported in different batches. Therefore, the order is transported to the target machining workshop in batches, and the completion time of each batch is obtained.

[0022] Step 3.3: Transmit the transportation information and completion time of each batch to the machining workshop scheduling and process planning module;

[0023] Step 4: Batch Integration and Global Optimization Iteration in Each Machining Workshop. Each machining workshop integrates batches based on their transportation information and completion time. Batches within the same time period are grouped together. Then, combining the flexible process information of the order, scheduling and process planning within the machining workshops are solved. After all machining workshops have completed their solutions, the total completion time and production energy consumption are obtained. Finally, the optimal workshop allocation scheme and cross-workshop transportation scheme are determined through optimization iteration of the factory module.

[0024] Step 4.1: Based on the transportation information and completion time of each batch, batches within the same time period are combined into one batch;

[0025] Step 4.2: Combine the process information of each batch of operations to be processed, and randomly generate the initial scheduling population for each batch;

[0026] Step 4.3: Input the machine information in each batch scheduling scheme into the process planning layer, and the process planning layer randomly generates the initial process planning population for each batch of workpieces;

[0027] Step 4.4: The process planning layer uses a multi-objective optimization algorithm to optimize the process scheme and obtain the optimal process scheme for each workpiece;

[0028] Step 4.5: Based on the obtained process plan, delete redundant codes in the scheduling plan, and perform scheduling layer decoding based on the process plan to obtain the operation sequence of all jobs, the corresponding processing machine and processing time, and then calculate the fitness value;

[0029] Step 4.6: Save the decoding-obtained scheduling scheme and process planning scheme, update the original optimal solution set, and determine whether the iteration termination condition of the scheduling layer has been met. If it has been met, output the optimal scheduling scheme and corresponding process scheme for each batch and go to step 4.7. Otherwise, perform genetic operations to generate a new scheduling population and go to step 4.3.

[0030] Step 4.7: Determine whether the iteration termination condition of the factory module has been met. If it has, output the optimal workshop allocation scheme and cross-workshop transportation scheme. Otherwise, perform the genetic operation of the factory module to generate a new cross-workshop scheduling population and go to step 1.4.

[0031] The beneficial effects of this invention are as follows: This patent takes the distributed workshop of a forging and pressing enterprise as the research object, and conducts research on cross-workshop scheduling and process planning integration for the multi-stage actual production process of die forging and machining. This scheduling method is mainly aimed at complex scenarios involving multi-stage production and processing links and multiple parallel production workshops. It can effectively improve workshop production efficiency and reduce production energy consumption, and provides a solution for solving the problem of linkage production of multiple heterogeneous workshops. Attached Figure Description

[0032] Figure 1 Flowchart of the integrated model for cross-workshop scheduling and process planning.

[0033] Figure 2 It is a sequential three-layer real number encoding rule.

[0034] Figure 3 Coding rules for the scheduling population in the forging workshop.

[0035] Figure 4 Gantt chart for using a delayed charging strategy. a) No delayed charging strategy used; b) Delayed charging strategy used.

[0036] Figure 5 This is a nested integrated model diagram of machining workshop scheduling and process planning.

[0037] Figure 6 This refers to the population coding rules for the scheduling layer of the machining workshop.

[0038] Figure 7 Population coding rules for the process planning layer in the machining workshop.

[0039] Figure 8 This is a crossover pattern between the forging workshop sequence and the machining workshop sequence.

[0040] Figure 9 This refers to the sequence crossing method for transport equipment.

[0041] Figure 10 This refers to the variation patterns of the forging workshop sequence and the machining workshop sequence.

[0042] Figure 11 This refers to the variation pattern of transportation equipment sequences. Detailed Implementation

[0043] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings:

[0044] The cross-workshop scheduling and process planning integrated scheduling problem is described as follows: Factory S has P different types of processing workshops to process a batch of orders with a total quantity of K. Each order consists of workpieces of different sizes, quantities, and process characteristics. The processing of each stage needs to be completed within the P selectable workshops. Transportation equipment is required between stages. Within each workshop, the selection of process schemes for each part's characteristics and the selection of processing equipment for each process need to be addressed. The characteristics of each part are not identical, and there may be flexibility factors such as process flexibility, equipment flexibility, and process sequence flexibility. There is also flexibility in the order of workpieces. The ultimate goal is to determine the workshops allocated to each processing stage of each order, the processing sequence of each order, the process route of each part within the workshop, and the processing sequence between parts, to obtain the optimal production plan. The flowchart of the cross-workshop scheduling and process planning integrated model is as follows: Figure 1As shown.

[0045] Step 1: Initialize the factory module algorithm parameters and encode the factory module using a sequential three-level real number encoding method. The first level of encoding is the forging workshop sequence, where each number represents the number of the forging workshop selected by the order during the die forging stage, and the total length is equal to the total number of orders. The second level of encoding is the transportation equipment sequence, where each number represents the number of the transportation equipment during cross-workshop transportation, and the total length is equal to the total number of orders. The third level of encoding is the machining workshop sequence, where each number represents the number of the machining workshop selected by the order during the machining stage, and the total length is equal to the total number of orders. Input the encoding information of the forging workshop allocation in the obtained initial cross-workshop scheduling scheme into the forging workshop module.

[0046] Step 1.1: Initialize the algorithm parameters of the factory module, including the number of iterations, population size, crossover probability, individual length, selection probability, and mutation probability;

[0047] Step 1.2: Input the forging information, vehicle transportation information, and machining information for all orders. The forging process information includes heating information, operation information, and optional machine information. The machining process information includes feature information, operation information, process constraint information, and optional machine information. The transportation vehicle information includes vehicle serial number and capacity.

[0048] Step 1.3: Randomly generate and initialize the cross-workshop scheduling population, which includes the forging workshop selection information, transport vehicle selection information, and machining workshop selection information for all orders. ,in This indicates the forging workshop number selected for order z. This indicates the vehicle number selected for order z. This represents the machining workshop number selected for order z. An example of a sequential three-level real number encoding rule is shown below. Figure 2 As shown;

[0049] Step 1.4: Initialize the coding information assigned to each order forging workshop in the cross-workshop scheduling scheme. Input to the forging workshop module;

[0050] Step 2: Initialize the parameters of the intelligent optimization algorithm for the scheduling problem of the flexible forging hybrid production line workshop, and set the parameters according to the coding information allocated to the forging workshop. The system generates a segmented three-layer real-number initial code based on the heating and forging information of each order. This code corresponds to the order processing sequence of the workpiece, the selection of the workpiece heating furnace, and the selection of the forging machine. The system decodes the scheduling scheme using the designed heat preservation time control decoding and compilation strategy. A multi-objective optimization algorithm is used to solve the problem with the completion time and energy consumption as objectives. The system obtains the scheduling scheme for each forging workshop and transmits the solution results to the cross-workshop transportation module.

[0051] Step 2.1: Initialize the parameters of the multi-objective optimization algorithm for the forging workshop module, including the number of iterations, population size, crossover probability, individual length, selection probability, and mutation probability;

[0052] Step 2.2: Input the forging workshop allocation information and die forging process information for all orders to be processed. The process information includes heating information, forging information and optional equipment information for each process.

[0053] Step 2.3: Randomly generate an initial forging workshop scheduling population. Each individual consists of three layers of real-number encoding. The first layer is the order sequence, where each number represents the order number, and the total length equals the number of orders Zn. The second layer is the furnace selection sequence, where each number represents the furnace number R during the heating stage, arranged according to the order sequence, and the total length equals the total number of workpieces in all orders Pn. The third layer is the forging equipment sequence, where each number represents the machine number D selected for the current forging process of the workpiece, arranged according to the workpiece sequence of the order, and the total length equals the total number of processes DPn for all workpieces in the forging stage. The specific encoding rules are as follows: Figure 3 As shown;

[0054] Step 2.4: Decode the scheduling scheme based on the encoded information and the decoding / compilation strategy. The first line of the encoded chromosome represents the order processing sequence, and the workpieces are heated in this order. When adjustments are made within a reasonable holding time range, and conflicts still exist, a delayed loading strategy is adopted to postpone the loading time so that the forging process can start immediately when the forging press is idle. The completion time PT and energy consumption PE after decoding the scheduling scheme are obtained. The Gantt curves for the loading process without and with the delayed loading strategy are also analyzed. Figure 4 As shown;

[0055] Step 2.5: After the decoding operation is completed, the non-dominated sorting and congestion calculation are performed using the completion time PT and energy consumption PE of each forging workshop's order processing as optimization objectives;

[0056] Step 2.6: Generate a new population of forging workshop scheduling schemes through genetic operations such as selection, crossover, and mutation, and then proceed to the operations in Step 2.7;

[0057] Step 2.7: Determine whether the iteration termination condition has been met. If it has, transmit the optimal scheduling scheme of each forging workshop to the cross-workshop transportation module; otherwise, proceed to step 2.4.

[0058] Step 3: The cross-workshop transportation module performs cross-workshop transportation based on the completion time sequence of orders from each forging workshop. It establishes a cross-workshop transportation model and performs decoding operations based on the vehicle transportation plan and the machining workshop allocation plan encoding information. From the scheduling results obtained by each forging workshop, the specific time for each order to complete the die forging process can be obtained. Arranging the orders in ascending order of completion time is the order in which the transportation equipment transports the orders. Since the carrying capacity of the transportation equipment is limited, the same order may be transported to the target workshop in batches. After all orders have been transported to the target workshop in batches, the decoding of the transportation equipment sequence also ends, thus obtaining the completion time of each batch transported to the target machining workshop.

[0059] Step 3.1: Sort the orders in each forging workshop in ascending order according to their completion time to obtain the completion time sequence of all orders, and transport them across workshops according to the principle of first-to-first-to-ship.

[0060] Step 3.2: Establish a cross-workshop transportation model and an objective function model. Each order is transported across workshops according to its completion time sequence and the assigned transportation vehicle. Due to the capacity limitations of the transportation vehicles, the orders are transported to the target machining workshop in batches, and the completion time of each batch is obtained.

[0061] The definitions of variables and parameters in the mathematical model are shown in Table 1:

[0062] Table 1 Definitions of variables and parameters

[0063]

[0064] Table 1 Definitions of Variables and Parameters (Continued)

[0065]

[0066] The constraints and calculation formulas for the inter-workshop transportation model are as follows:

[0067] formula A model for the start time of order loading is presented, which is jointly determined by the order forging completion time and the arrival time of the transport equipment.

[0068]

[0069] formula A time model for loading order workpieces onto transportation equipment is presented.

[0070]

[0071] formula A model for the permissible transportation time of a vehicle is presented, where the vehicle is in an unavailable state when loaded, and transportation is not permitted in this state.

[0072]

[0073] formula A time model for the start of transportation across stages is presented, which is determined by the completion time of the last loaded workpiece on the transportation equipment.

[0074]

[0075] formula A time model for the first batch of orders arriving at the target workshop in the transportation equipment is presented.

[0076]

[0077] formula A time model for unloading order workpieces from transportation equipment is presented.

[0078]

[0079] formula A model of the permissible transportation time for vehicles during the machining stage is presented. When a vehicle is unavailable during unloading, it is not allowed to be transported.

[0080]

[0081] formula This is the model for the earliest permitted transportation time for vehicles.

[0082]

[0083] formula A time model for order completion in transportation equipment is presented.

[0084]

[0085] formula A time model for the arrival of the x-th order in the transportation equipment to the target workshop is given.

[0086]

[0087] constraint This ensures that each order is assigned to only one transport vehicle during the transportation phase. (Formula) The total weight of the workpieces in the transport equipment at time q is given. Constraints. Ensure that the total weight of all workpieces in the transport equipment at time τ does not exceed its maximum permissible load capacity. (Formula) This refers to the energy consumption of transportation equipment during the transportation process.

[0088]

[0089]

[0090]

[0091]

[0092] The calculation formula for the cross-workshop objective function model is as follows:

[0093] Since the order processing order is not fixed, the completion time can be determined by the maximum completion time of each batch of orders in the array during the machining stage. In other words, it is equivalent to the maximum completion time of the machining workshop. The corresponding model is as follows:

[0094]

[0095] The calculation of total production energy consumption needs to cover the energy consumption of different types of machines in each workshop during operation and idle periods, as well as the energy consumption consumed by the inter-workshop transportation system. Since the inter-workshop scheduling and process planning integration system mainly consists of three parts: the forging workshop system, the inter-workshop transportation system, and the machining workshop scheduling and process planning integration system, the production energy consumption model is as follows:

[0096]

[0097] Step 3.3: After all orders have been transported to the target workshop in batches, the transportation information and completion time of each batch are transmitted to the machining workshop scheduling and process planning module, and the decoding of the transportation equipment sequence is also completed.

[0098] Step 4: Each machining workshop integrates batches based on their transportation information and completion time, grouping batches within the same time period into a single batch. Then, combining the flexible process information of the orders, the workshop begins scheduling and process planning. Based on the data interaction characteristics between process planning and scheduling, a nested integration model is designed, such as... Figure 5As shown, a multi-objective optimization algorithm is used to solve the scheduling and process schemes of each machining workshop. After all machining workshops have been solved, the total completion time and production energy consumption are obtained, and finally the optimal workshop allocation scheme and cross-workshop transportation scheme are determined.

[0099] Step 4.1: Based on the transportation information and completion time of each batch, batches from the same machining workshop within the same time period are integrated into one batch for use in solving subsequent machining workshop solutions;

[0100] Step 4.2: Combine the process information of each batch of jobs to be processed, including feature information, operation information, process constraint information, and available machine information, and randomly generate the initial scheduling population for each batch, such as... Figure 6 As shown, this includes the operation sequence and the machine information corresponding to the operation;

[0101] Step 4.3: Input the machine information from each batch scheduling plan into the process planning layer. Based on the process information of each part and the machine information in the scheduling plan, the process planning layer randomly generates an initial process planning population for each part, such as... Figure 7 As shown, this includes the feature sequence, process selection, and machine information corresponding to the operation;

[0102] Step 4.4: Decode the process coding information to obtain the feature sequence of each workpiece, the process scheme for each feature selection, the corresponding processing machine and processing time, and use a multi-objective optimization algorithm to optimize the process scheme to obtain the optimal process scheme for each workpiece.

[0103] Step 4.5: Based on the process plan obtained from the process planning layer, delete redundant codes in the scheduling plan, and perform scheduling layer decoding based on the process plan to obtain the operation sequence of all operations, the corresponding processing machine and processing time, and then use the total completion time and energy consumption as fitness values ​​for calculation.

[0104] Step 4.6: Save the decoding-obtained scheduling scheme and process planning scheme, update the original optimal solution set, and determine whether the iteration termination condition of the scheduling layer has been met. If it has been met, output the optimal scheduling scheme and corresponding process scheme for each batch and go to step 4.7. Otherwise, perform genetic operations such as selection, crossover, and mutation to generate a new scheduling population and go to step 4.3.

[0105] Step 4.7: Determine whether the iteration termination condition of the factory module has been met. If it has, output the optimal workshop allocation scheme and cross-workshop transportation scheme. Otherwise, perform the genetic operation of the factory module to generate a new cross-workshop scheduling population and go to step 1.4.

[0106] Step 4.7.1: If the optimal workshop allocation scheme and cross-workshop transportation scheme are achieved, output the optimal scheme; otherwise, proceed to step 4.7.2.

[0107] Step 4.7.2: Perform a selection operation on the factory module population. Based on the fitness value of each individual, use the roulette wheel selection method to select individuals from the parent population to pass on their genes to the next generation.

[0108] Step 4.7.3: Perform crossover operations with a certain crossover probability. Different crossover methods are used for the designed sequential real-number gene encoding scheme. Since the forging workshop sequence and the machining workshop sequence have workshop selection restrictions, and the two sequences have the same encoding scheme designed in this paper, no order transformation is required. Therefore, the following method is designed: Figure 8 The crossover method shown does not produce singular solutions, and the transport device sequence has no special restrictions, only requiring gene crossover at the same position. Therefore, a method such as... Figure 9 The crossing pattern shown;

[0109] The specific procedures for the intersection of the forging workshop sequence and the machining workshop sequence are as follows:

[0110] a) First, a binary character array is randomly generated, with the same length as the chromosome. In this array, the position with a value of 1 is regarded as the intersection point of the parent P1, and the position with a value of 0 is regarded as the intersection point of the parent P2.

[0111] b) Copy the value corresponding to the crossover point on the parent P1 chromosome to the same position on the offspring Q1 chromosome, and similarly copy the value corresponding to the crossover point on the parent P2 chromosome to the same position on the offspring Q1 chromosome.

[0112] c) The method for obtaining the offspring Q2 is the same as that for the offspring Q1, and the method for confirming the parent crossover point is the opposite of that for obtaining the parent crossover point of the offspring Q1.

[0113] The specific steps for cross-series transport equipment are as follows:

[0114] a) Randomly select two intersection points and divide the two parental chromosomes P1 and P2 into three parts: 1) left; 2) middle; and 3) right, which are represented as A, B and C respectively.

[0115] b) The middle part B of parent generation P1 is copied to the same position in child generation Q1, and both ends of parent generation P2 are copied to the same position in child generation Q1. The method for obtaining child generation Q2 is the same as that for child generation Q1, but the copying order is reversed between parent generation P1 and P2.

[0116] Step 4.7.4: Perform mutation operations with a certain mutation probability. Different mutation methods are used for the designed sequential real-number gene encoding scheme. Since the forging workshop sequence and the machining workshop sequence only have substitutable options, the following methods are designed: Figure 10 The mutation method based on binary arrays shown, for transport equipment sequences, employs, as follows: Figure 11 The two-point swap gene mutation shown is a gene mutation in which two genes are randomly swapped.

[0117] Step 4.7.5: After the factory module generates a new cross-workshop scheduling population through genetic operations such as selection, crossover, and mutation, it proceeds to step 1.4.

Claims

1. A cross-workshop scheduling and process planning integration method based on flexible manufacturing, characterized in that, The cross-workshop scheduling and process planning integration method based on flexible production includes: Step 1: Cross-workshop scheduling population initialization and forging task transfer; Based on the forging information, vehicle transportation information and machining information of all orders, the factory module generates an initial cross-workshop scheduling population, and inputs the coding information of the forging workshop in the obtained initial cross-workshop scheduling scheme into the forging workshop module; Step 2: Generation and solution of scheduling schemes within each forging workshop; The forging workshop module generates an initial population and solves the scheduling schemes within the forging workshop based on the coding information assigned to the forging workshop, the heating process information of the order, and the forging process information, and then transmits the solution results of each forging workshop to the cross-workshop transportation module; Step 3: Cross-workshop transportation module decoding and order batch transportation; The cross-workshop transportation module decodes the transportation module based on the completion sequence time of all orders in the forging stage, combined with the vehicle transportation and machining workshop allocation coding information. Due to the limited capacity of transportation vehicles, the same order may need to be transported in multiple batches. Therefore, the order workpieces are transported to each target machining workshop in batches, thereby obtaining the completion time of each batch to the target workshop. Step 4: Batch integration and global optimization iteration of each machining workshop; The workpiece information and transportation completion time of each batch are transmitted to the machining workshop scheduling and process planning module for batch integration. Batches within the same time period are integrated into the same batch. Then, the scheduling and process planning solution in the machining workshop is started. After all workshops have been solved, the total completion time and production energy consumption are obtained. Finally, the optimal workshop allocation scheme and cross-workshop transportation scheme are determined through the optimization iteration of the factory module.

2. The cross-workshop scheduling and process planning integration method based on flexible production according to claim 1, characterized in that, In step 1, the information input to the factory module includes forging process information, machining process information, and transport vehicle information for all orders to be processed. The forging process information includes heating information, operation information, and selectable machine information. The machining process information includes feature information, operation information, process constraint information, and selectable machine information. The transport vehicle information includes vehicle serial number and capacity. An initial cross-workshop scheduling population is randomly generated, which includes forging workshop selection information encoding, transport vehicle selection information encoding, and machining workshop selection information encoding. After the initial population is generated, the forging workshop selection information and forging process information in the cross-workshop scheduling scheme are input into the forging workshop module.

3. The cross-workshop scheduling and process planning integration method based on flexible production according to claim 1, characterized in that, In step 2, the forging workshop module randomly generates an initial scheduling population within each forging workshop based on the forging workshop selection information, heating process information, and forging process information of each order in the cross-workshop scheduling scheme. The population information includes the order processing sequence, workpiece heating furnace selection, and forging process machine selection. A multi-objective optimization algorithm is then used to optimize the scheduling scheme within the forging workshop, ultimately obtaining the scheduling scheme for each forging workshop.

4. The cross-workshop scheduling and process planning integration method based on flexible production according to claim 1, characterized in that, In step 3, the completion sequence and completion time sequence of all orders are obtained according to the scheduling plan of each forging workshop. The transportation plan is decoded by combining the transportation vehicle selection and the order machining workshop selection coding information. Since the capacity limitation of the transportation vehicle may cause an order to be transported to the target machining workshop in multiple batches, the time of each batch being transported to the target machining workshop is finally obtained.

5. The cross-workshop scheduling and process planning integration method based on flexible production according to claim 1, characterized in that, In step 4, after the transportation plan decoding operation is completed in the transportation module, each machining workshop integrates batches based on the batch workpiece information and arrival time. Batches within the same machining workshop and at the same time are integrated into the same batch. An initial scheduling population is generated based on the process information of the batch parts. The scheduling layer encoding includes the process information of all parts in the batch and the machine information of each process. The machine information encoding of the scheduling layer is passed to the process planning layer. Then, the process planning layer solves the process based on the machine information and feeds back the process planning information of each workpiece to the scheduling layer for decoding. The decoded scheduling plan and process plan within the machining workshop are obtained. Finally, the optimal workshop allocation plan and cross-workshop transportation plan are obtained through the optimization iteration of the factory module.

Images