A dynamic production scheduling method, module and system based on TW-GA combination

Abstract

This application proposes a dynamic production scheduling method, module and system based on TW-GA combination, starting from the production scheduling business to achieve path planning and optimal result solution algorithm, with the goal of maximizing the release of the overall production capacity of the workshop. Solve the following problems: solve the problem of inconsistent input samples, use the time wheel algorithm to pre-process the production scheduling tasks, and unify the input samples of population genetics; solve the problem of fine-grained production scheduling, realize fine-grained level splitting based on equipment, worker skills and process combinations, and realize fine-grained unit regularization through time windows; solve the problem of simultaneous production scheduling of multiple categories, use multiple time wheels to superimpose projections to realize simultaneous production scheduling of multiple categories; solve the problem of inability to dynamically calculate or high calculation cost, realize local optimization iterative solution based on internal allocation of time windows by dynamically plugging and unplugging time windows, and achieve second-level calculation response under the scale of tens of millions of data; introduce multi-loop calculation process to solve the problem of low probability of obtaining global optimal solution.

Description

Technical Field

[0001] This application relates to the field of flexible customized intelligent manufacturing technology, and specifically to a dynamic production scheduling method, module and system based on TW-GA combination. Background Art

[0002] Currently, the overall level of informatization in the printing and dyeing industry is low. Most companies still rely on simple, pre-planned production scheduling methods, which are considered static scheduling. This typically involves scheduling simple production waves based on the processing tasks and actual production capacity before production begins. Once a production wave is determined, it cannot be changed. Furthermore, scheduling is limited to workshops or production lines, not specific process steps. On-site production in workshops can only be scheduled based on employee experience.

[0003] Most existing flexible customized production scheduling plans are based on waves, which is pre-planned and cannot be adjusted in real time to respond to changes. Most scheduling algorithms use traditional heuristic algorithms, such as simulated annealing, hill climbing algorithms, genetic algorithms, and Tabu algorithms. Similar technologies mostly use a single algorithm or a combination of multiple algorithms to find the optimal solution. Due to the regularity of input conditions and the complexity of the algorithm, the quality of the calculation results is generally poor. The main problems of similar technologies are as follows:

[0004] Inconsistent input samples lead to uncertain scheduling results: The results of heuristic algorithms are significantly affected by the input sample and perturbation factors. Most similar products directly use the process flow configuration information of the product to be produced as a sample and perturb it by randomly feeding the material sequence. This makes it difficult to guarantee the accuracy of the calculation results. The lack of a unified and standardized input sample is a major reason for the poor scheduling results of similar technologies.

[0005] Scheduling based on coarse-grained processes: It is impossible to schedule production at a fine-grained level based on equipment, worker skills, and process combinations. Currently, most technologies on the market are based on coarse-grained scheduling of process flows, and there are generally a large number of processes that are idle or waiting.

[0006] Production scheduling based on a single product category: Production scheduling can only be performed for the process flow of a single product. Mixed scheduling of multiple products and multiple process flows cannot be achieved, resulting in extremely low utilization of equipment and manpower. The workshop can only produce one category at a time.

[0007] Established production schedules are immutable or the cost of changing them is enormous: Whether based on a single or combined algorithm, production schedules cannot quickly adapt to abnormal situations in the workshop (such as equipment failure, worker leave, missing materials, low finished product qualification rate, etc.). These algorithms need to re-sort the inputs and conditions and recalculate them when changes occur. The production schedule calculation takes a very long time, and various abnormal situations frequently occur on the production site. This situation does not meet the actual needs and is the fundamental reason why similar products cannot achieve real-time production scheduling.

[0008] Single calculation link, low global optimization rate: Most similar products use a single calculation link design and adopt heuristic algorithms in diversified flexible customization scenarios. The probability of obtaining the global optimal solution through only one calculation is low. Therefore, most similar products have local optimal results.

[0009] In view of this, it is of great significance to design a dynamic scheduling method and system that can solve problems such as inconsistent input samples, fine-grained scheduling, simultaneous scheduling of multiple categories, and inability to dynamic calculation. Summary of the Invention

[0010] The embodiments of the present application propose a dynamic production scheduling method, module and system based on the TW-GA combination to solve the technical problems mentioned in the above background technology section.

[0011] In a first aspect, an embodiment of the present application provides a dynamic production scheduling method based on a TW-GA combination, comprising the following steps:

[0012] S1. Build time wheel group and product processing timeline;

[0013] S2. Use the task selector in the time wheel group to select the process task to be processed according to the process node of the product processing timeline, and embed the process task to be processed into the time window of the time wheel group;

[0014] S3. Repeat step S2 until there are no more pending processing tasks, and then output a basic production scheduling sample, which includes information about the time window.

[0015] S4. Sort the processing tasks according to N (N≥1) sorting strategies, and repeat steps S1 to S3 to obtain N basic production scheduling samples;

[0016] S5. Randomly select two basic production schedule samples from the N basic production schedule samples, perform time window crossover on the same type of basic time wheels or overlapping time wheels in the two basic production schedule samples, and evaluate the fitness of the results of the time window crossover. Save the individuals with improved fitness as the next generation of basic production schedule samples;

[0017] S6. Repeat step S5 until all N basic production scheduling samples complete the time window cross evolution; and

[0018] S7. Repeat steps S5 to S6 until the system evolves to a preset generation or converges to the evolution stop standard, and then output the production scheduling sample with the highest fitness.

[0019] This method, based on sample generation using product processing timelines and time-wheel combinations, unifies the input sample criteria for population genetics, ensuring the stability of the optimal solution. Furthermore, the introduction of a multi-loop computational process addresses the low probability of achieving a global optimal solution. Therefore, this method significantly reduces the computational and time costs of production scheduling, enabling real-time adjustments to production schedules based on shop floor dynamics.

[0020] In a specific embodiment, in step S1, the time wheel group includes a basic time wheel for equipment machines, a basic time wheel for manual skills, and overlapping time wheels that overlap between equipment machines and manual skills; the product processing timeline includes the processing time of each process of the product and the arrangement order of each process of the product.

[0021] The system creates a basic time wheel for each piece of equipment in the workshop and a time wheel for each worker based on their skills. Specifically, the system creates a basic time wheel for each time-consuming process. Based on this basic time wheel, the system creates overlapping time wheels for the overlap between labor and machine operations. This method enables fine-grained segmentation based on combinations of equipment, worker skills, and processes, and allows for the regularization of fine-grained units through time windows.

[0022] In a specific embodiment, step S1 further includes the following sub-steps:

[0023] S11, constructing a basic time wheel and an overlapping time wheel;

[0024] S12, removing the basic time wheels that have created overlapping time wheels;

[0025] S13, aligning the overlapping time wheels with the basic time wheel of step S12; and

[0026] S14: Start the overlapping time wheel and the basic time wheel of step S13, and advance one unit time to notify the task selector on the corresponding time wheel to select the task to be processed.

[0027] Through the above method, the simultaneous production scheduling of multiple categories can be achieved by using multiple time wheels with superimposed projections.

[0028] In a specific embodiment, in step S2, the task selector in the time wheel group is used to select the corresponding process task to be processed according to the process node of the product processing timeline, and the following conditions are met:

[0029] a. The current time window of the basic time wheel or overlapping time wheel must be able to meet the time consumption requirements of the task to be processed;

[0030] b. The already used basic time wheel or overlapping time wheel gives priority to selecting tasks;

[0031] c. The usage rate of basic time wheels of the same type shall not differ by more than 30%.

[0032] d. If the time wheel group includes overlapping time wheels, select the basic time wheel with the minimum impact number.

[0033] In a specific embodiment, in step S5, time windows are crossed for basic time wheels of the same type or overlapping time wheels in two basic production scheduling samples, and the following conditions are met:

[0034] a. The time window intersection must meet the constraints of the arrangement order of each process of the product;

[0035] b. There is enough time in the time window after the crossover;

[0036] c. Time windows within the same basic time wheel or overlapping time wheels can be self-intersecting;

[0037] d. After the time windows cross, there is only one task node for the same processing step.

[0038] In a specific embodiment, in step S4, the processing tasks are sorted according to N (N≥1) sorting strategies, including sorting by product brand category, sorting by processing time, sorting by the proportion of equipment and machine occupancy time, sorting by the proportion of manual occupancy time, sorting by the number of processes, and heuristic sorting by the order of work sections.

[0039] The system will generate a corresponding sorting combination for each strategy. Different sorting combinations affect the combination and supply order of process processing tasks, which will lead to completely different processing scheduling plans.

[0040] In a second aspect, an embodiment of the present application provides a dynamic production scheduling module based on a TW-GA combination, which includes:

[0041] Construction unit, used to construct time wheel set and product processing timeline;

[0042] The task selection unit is used for the task selector of the time wheel group to select the process tasks to be processed according to the process nodes of the product processing timeline, and embed the process tasks to be processed into the time window of the time wheel group;

[0043] The acquisition unit is used to repeat the operation on the task selection unit until there are no tasks to be processed, and then output a basic production scheduling sample, which includes information about the time window;

[0044] The sample production unit is used to sort the processing tasks according to N (N≥1) sorting strategies. After repeating the operations on the construction unit, task selection unit, and acquisition unit, N basic production scheduling samples are obtained.

[0045] The evolution unit is used to randomly select two basic production schedule samples from N basic production schedule samples, perform time window crossover on the same type of basic time wheels or overlapping time wheels in the two basic production schedule samples, and evaluate the fitness of the results of the time window crossover. The individuals with improved fitness are saved as the next generation of basic production schedule samples;

[0046] Iteration unit, used to repeat the operation on the evolution unit until all N basic production scheduling samples complete the time window cross evolution; and

[0047] The output unit is used to repeat the operations on the evolution unit and the iteration unit until the evolution reaches the preset generation or the evolution convergence reaches the evolution stop standard, and then output the production scheduling sample with the highest fitness.

[0048] On the third aspect, the present application provides a dynamic scheduling system based on the TW-GA combination, including an order task preprocessing module, a workshop situation awareness module and the dynamic scheduling module as mentioned above. The order task preprocessing module provides processing tasks for the dynamic scheduling module, and the workshop situation awareness module provides basic data support for the order task preprocessing module and the dynamic scheduling module.

[0049] In a specific embodiment, the workshop situation awareness module includes a basic data unit, a full-process data acquisition unit, and a status update notification management unit;

[0050] Basic data unit, used to establish workshop material storage, production equipment and machine database, production process skill requirements and time consumption database, production process flow database for various products, and worker skill database;

[0051] The whole-process data collection unit is used to collect worker information, equipment information, logistics information, product processing information, and product inspection information. The whole-process data collection unit is equipped with a risk assessment model.

[0052] Status update notification management unit, event publishing service for message subscription.

[0053] In a specific embodiment, the order task pre-processing module includes a customer order unit, a workshop order replenishment unit and an order to be processed task management unit;

[0054] The customer order unit and the workshop order replenishment unit push the latest order information to the order processing task management unit in real time;

[0055] The pending processing task management unit conducts capacity assessment to determine whether to generate a new workshop production wave. If a new workshop production wave is generated, the order of the workshop production wave is decomposed into process-level tasks.

[0056] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, any of the above methods is implemented.

[0057] The present invention provides a dynamic production scheduling method, module and system based on TW-GA combination, which realizes production scheduling at a fine-grained level based on equipment, worker skills and process combinations. It has the following beneficial effects: (1) It realizes simultaneous production scheduling of multiple categories; (2) It greatly reduces the idle rate of workshop production factors and improves the utilization rate of workshop production factors. (3) The sample generation method based on the product processing timeline and time wheel combination unifies the input standard of the sample and ensures the stability of the optimal solution. (4) It greatly reduces the calculation and time cost of production scheduling, and effectively makes adjustments to production scheduling based on real-time adjustments to the workshop situation. (5) Through two heuristic evolutions, the hit rate of the global optimal solution is greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0059] Figure 1 is a flow chart of a dynamic production scheduling method based on a TW-GA combination according to the present application;

[0060] Figure 2 This is an overall flow chart of the TW-GA combination method according to an embodiment of the present application;

[0061] Figure 3 This is a schematic diagram of the basic structure of the timing wheel according to an embodiment of the present application;

[0062] Figure 4 This is a structural diagram of the product processing timeline of an embodiment of the present application;

[0063] Figure 5 This is a structural diagram of an embodiment of the present application in which a task to be processed is embedded into a time window of a timing wheel group;

[0064] Figure 6 This is a schematic diagram of the structure of population evolution in one embodiment of the present application;

[0065] Figure 7This is a schematic diagram of the structure of the dynamic production scheduling module based on the TW-GA combination of this application;

[0066] Figure 8 This is a schematic diagram of the structure of the dynamic production scheduling system based on the TW-GA combination applied in this application;

[0067] Figure 9 This is a flowchart of the operation of the workshop dynamic perception module implemented in the present application;

[0068] Figure 10 This is a flowchart of the operation of the order task pre-processing module implemented in this application;

[0069] Figure 11 It is a structural diagram of a computer system suitable for implementing the electronic device of the embodiment of the present application. DETAILED DESCRIPTION

[0070] The present application will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to explain the relevant inventions and are not intended to limit the inventions. It should also be noted that, for ease of description, only portions relevant to the relevant inventions are shown in the accompanying drawings. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application may be combined with one another. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the examples.

[0071] Figure 1 The flowchart of the dynamic production scheduling method based on the TW-GA combination according to the present application is shown. Figure 2 The overall flow chart of the TW-GA combination method according to an embodiment of the present application is shown. Figure 1 and Figure 2 , the method comprising:

[0072] S1. Build the time wheel set and product processing timeline.

[0073] Among them, the time wheel group includes the basic time wheel of equipment machines, the basic time wheel of manual skills, and the overlapping time wheels between equipment machines and manual skills; the product processing timeline includes the processing time of each process of the product and the arrangement order of each process of the product.

[0074] In this embodiment, the construction of the time wheel group in step S1 includes the following sub-steps:

[0075] S11, constructing a basic time wheel and an overlapping time wheel;

[0076] The system creates a basic time wheel for each machine in the workshop, and creates a basic time wheel for each worker according to their skills. The time period of all basic time wheels is the standard working hours (4 hours in the morning, 5 hours in the afternoon, and 4 hours in the evening). The time wheel organization structure is shown in the attached figure. Figure 3 As shown, the minimum time scale of the basic time wheel is seconds, and the minimum time window is fixed at 1 second.

[0077] Constructing a product processing timeline means constructing a timeline for each product according to the product process flow. The product processing timeline only defines the constraints of the processing time and sequence of each process. The structure of the product processing timeline is shown in the attached figure. Figure 4 shown.

[0078] The system creates a basic time wheel for each time-consuming process, and based on the basic time wheel, the system creates overlapping time wheels for manual and machine overlap. The specific creation rules are shown in Table 1:

[0079] Table 1

[0080]

[0081] The basic time wheel is an atomic time wheel and is inseparable. The overlapping time wheel is a special, directly usable time wheel that binds a worker's skill to a machine when the machine requires a worker's skill. Overlapping time wheels are mutually exclusive. When a worker's skill becomes active on a specific machine, the overlapping time wheel takes effect immediately, and any other overlapping time wheels occupying that specific worker become inactive.

[0082] For example, a direct-injection workshop has three direct-injection printing and dyeing machines, two drying machines, and two sewing machines, with six staff members (one with image verification skills, one with direct-injection printing and dyeing machine operation skills, one with dryer operation skills, one with quality inspection skills, and two with both direct-injection printing and dyeing machine operation skills and sewing machine operation skills). In this case, the timeline for creating a wheel is shown in Table 2:

[0083] Table 2

[0084] Serial number Time Wheel Name Time wheel type Create the number of time wheels 1 Direct-injection printing and dyeing machine timing wheel Basic time wheel 3 2 Dryer timer Basic time wheel 2 3 Sewing machine timing wheel Basic time wheel 2 4 Worker picture confirmation skill time wheel Basic time wheel 1 5 Workers' direct-injection printing and dyeing machine operation skills time wheel Basic time wheel 3 6 Worker Dryer Operation Skills Time Wheel Basic time wheel 1 7 Worker Quality Inspection Skill Time Wheel Basic time wheel 1 8 Worker sewing machine skills time wheel Basic time wheel 2 9 Combine time wheel 1 and time wheel 5 Overlapping Time Wheels 9 10 Combine time wheel 2 and time wheel 6 Overlapping Time Wheels 2 11 Combine time wheel 3 and time wheel 8 Overlapping Time Wheels 4

[0085] The above method can solve the problem of fine-grained production scheduling, realize fine-grained level splitting based on equipment, worker skills and process combinations, and realize the regularization of fine-grained units through time windows; solve the problem of simultaneous scheduling of multiple categories, and use the method of superimposed projection of multiple time wheels to realize simultaneous scheduling of multiple categories.

[0086] The construction of the time wheel group in step S1 also includes the following sub-steps:

[0087] S12, removing the basic time wheels that have created overlapping time wheels;

[0088] S13, aligning the overlapping time wheels with the basic time wheel in step S12, specifically: setting the time pointers of all remaining time wheels after removal to 0 o'clock.

[0089] S14: Start the overlapping time wheel and the basic time wheel of step S13, and advance one unit time to notify the task selector on the corresponding time wheel to select the task to be processed.

[0090] Through the above method, the simultaneous production scheduling of multiple categories can be achieved by using multiple time wheels with superimposed projections.

[0091] Continue to refer Figure 1 and Figure 2 ,The dynamic scheduling method based on TW-GA combination also includes the following steps:

[0092] S2. Use the task selector in the time wheel group to select the process task to be processed according to the process node of the product processing timeline, and embed the process task to be processed into the time window of the time wheel group.

[0093] That is, the product processing timeline is grouped by the time wheels on the process nodes. The selected process tasks to be processed and the selected time wheels are combined with the product processing timeline. With the process nodes as atoms, the processes on the product processing timeline are embedded in the time windows suitable for the time wheels. The structure is shown as follows Figure 5 shown.

[0094] All the timing wheel groups perform the operation of step S2 in sequence until all the timing wheels have executed the selection of the process tasks to be processed, or there are no process tasks to be processed.

[0095] In this embodiment, the task selector in the time wheel group is used to select the corresponding process task to be processed according to the process node of the product processing timeline. The following conditions must be met:

[0096] a. The current time window of the basic time wheel or overlapping time wheel must be able to meet the time consumption requirements of the task to be processed;

[0097] b. The already used basic time wheel or overlapping time wheel gives priority to selecting tasks;

[0098] c. The usage rate of basic time wheels of the same type shall not differ by more than 30%.

[0099] d. If a timer group includes overlapping timers, select the base timer with the smallest number of impacts. For example, if overlapping timers consist of three equipment units and two manual skills, only two timers are selected based on the two manual skills.

[0100] S3. Repeat step S2 until there are no tasks to be processed, and then output a basic production scheduling sample, which includes information about the time window.

[0101] The above method can solve the problem of inconsistent input samples. The time wheel algorithm is used to pre-process the production scheduling tasks and unify the input samples of population inheritance.

[0102] S4. Sort the processing tasks according to N (N≥1) sorting strategies, and repeat steps S1 to S3 to obtain N basic production scheduling samples.

[0103] In this embodiment, step S4 also includes a feeding initialization stage, specifically sorting the processing tasks according to N (N≥1) sorting strategies. The sorting strategies are established strategies, and there are multiple ones, including at least sorting by product brand category, sorting by processing time, sorting by the proportion of equipment and machine occupancy time, sorting by the proportion of manual occupancy time, sorting by the number of processes, and heuristic sorting by the order of work sections.

[0104] The system generates a corresponding sorting combination for each strategy. Different sorting combinations affect the combination and supply order of process tasks, resulting in completely different processing scheduling plans. That is, for the N different sorting combinations generated during the initial material feeding phase, repeat steps S1 through S3 to obtain N basic scheduling strategies.

[0105] S5. Randomly select two basic production schedule samples from N basic production schedule samples, and perform population evolution on the same type of basic time wheels or overlapping time wheels in the two basic production schedule samples. The specific implementation method is time window crossing, and the fitness evaluation is performed on the results of time window crossing. The individuals with improved fitness are saved as the next generation of basic production schedule samples.

[0106] In this embodiment, in step S5, time window intersection is performed for basic time wheels of the same type or overlapping time wheels in two basic production scheduling samples, and the following conditions must be met:

[0107] a. The time window intersection must meet the constraints of the arrangement order of each process of the product;

[0108] b. There is enough time in the time window after the crossover;

[0109] c. Time windows within the same basic time wheel or overlapping time wheels can self-cross, i.e., self-evolve;

[0110] d. After the time windows cross, there is only one task node for the same processing step.

[0111] The fitness evaluation results of the time window intersection are used to evaluate the fitness. The fitness evaluation model involves five aspects: total processing time, total number of goods moved, total distance of goods moved, total number of machine switches, and total number of manual switches. The individuals with improved fitness after mutation enter the next generation. The schematic diagram is shown in the attached figure. Figure 6 shown.

[0112] The above method can solve the problems of being unable to perform dynamic calculations or having high calculation costs. By dynamically plugging and unplugging the time wheel grid (time window), a local optimization iterative solution method based on the internal allocation of the time wheel grid is realized, achieving a calculation response of seconds under a data scale of tens of millions.

[0113] S6. Repeat step S5 until all N basic production scheduling samples complete the time window cross evolution; and

[0114] S7. Repeat steps S5 to S6 until the system evolves to a preset generation or converges to the evolution stop standard, and then output the production scheduling sample with the highest fitness.

[0115] The overall process of the above method can be roughly divided into two parts: standardized sample set generation and sample evolution. The time wheel group generator is responsible for generating the standardized sample set, and the population genetic generator handles the evolution of the samples. The TW-GA hybrid algorithm includes three parts: the material feeding initialization stage, the basic production scheduling sample generation stage (TW algorithm), and the production scheduling optimal solution solution stage (GA algorithm). The above method unifies the input sample standards of population genetics based on the sample generation method of the product processing timeline and time wheel combination, ensuring the stability of the optimal solution. In addition, the introduction of a multi-loop calculation process can solve the problem of low probability of obtaining the global optimal solution. Therefore, the use of the above method can greatly reduce the computational and time costs of production scheduling, and can effectively make real-time adjustments to production scheduling based on the workshop situation.

[0116] Figure 7 The following is a schematic diagram of the structure of a dynamic production scheduling module based on a TW-GA combination according to an embodiment of the present application. As shown in the figure, the module 200 includes:

[0117] A construction unit 210 is used to construct a time wheel set and a product processing timeline;

[0118] The task selection unit 220 is used for the task selector of the time wheel group to select the process tasks to be processed according to the process nodes of the product processing timeline, and embed the process tasks to be processed into the time window of the time wheel group;

[0119] The acquisition unit 230 is configured to repeat the operation of the task selection unit 220 until there are no tasks to be processed, and then output a basic production scheduling sample, where the basic production scheduling sample includes information about the time window;

[0120] The sample production unit 240 is used to sort the processing tasks according to N (N≥1) sorting strategies, and repeat the operations of the construction unit 210, the task selection unit 220, and the acquisition unit 230 to obtain N basic production scheduling samples.

[0121] The evolution unit 250 is configured to randomly select two basic production schedule samples from the N basic production schedule samples, perform time window crossover on the same type of basic time wheels or overlapping time wheels in the two basic production schedule samples, evaluate the fitness of the time window crossover results, and save the individuals with improved fitness as the next generation of basic production schedule samples;

[0122] The iteration unit 260 is configured to repeat the operation of the evolution unit 250 until all N basic production scheduling samples have completed the time window cross evolution; and

[0123] The output unit 270 is used to repeat the operations on the evolution unit 250 and the iteration unit 260 until the evolution reaches a preset generation or the evolution converges to the evolution stop standard, and output the production scheduling sample with the highest fitness.

[0124] Figure 8 The schematic diagram of the structure of the dynamic production scheduling system based on TW-GA combination is shown in FIG. Figure 8 As shown, the system includes an order task preprocessing module, a workshop situation awareness module and the dynamic scheduling module as mentioned above. The order task preprocessing module provides processing tasks for the dynamic scheduling module, and the workshop situation awareness module provides basic data support for the order task preprocessing module and the dynamic scheduling module.

[0125] Figure 9 The flowchart of the workshop dynamic perception module operation is shown in the present application. Figure 8 and Figure 9 ,The workshop situation awareness module includes a basic data unit, a ,full-process data acquisition unit, and a status update notification ,management unit.

[0126] Among them, the basic data unit is used to establish a workshop material storage library, a production equipment machine library, a production process skill requirement and time consumption library, a production process flow library for various products, and a worker skill library;

[0127] The specific operations are as follows: Model workshop personnel, production equipment, and production materials (people, machines, and materials) using basic data units. Establish a workshop material inventory library (to monitor the status of materials during processing in real time and assist management in allocating materials), a production equipment library (to obtain key equipment operating parameter indicators), a production process skill requirement and time consumption library (to obtain the skills and time parameters required for a single minimum process), a production process flow library for various products (based on the product, combining the processes required for product processing and defining the sequence constraints for the required processes), and a worker skill library (to obtain workers' work skills).

[0128] The whole-process data collection unit is used to collect worker information, equipment information, logistics information, product processing information, and product inspection information. The whole-process data collection unit is equipped with a risk assessment model.

[0129] The specific operations are as follows: relying on the basic data unit, the workshop's full-process data collection unit integrates external systems to obtain the real-time status of people, machines, materials and processing tasks. In terms of personnel, the on-the-job status of workers is collected by connecting to the ERP system and the workstation punch-in system. In terms of machine equipment, information such as the real-time operation of the equipment, whether it is maintained, and whether it has been repaired is obtained by connecting to the factory TPM (equipment management system). In terms of materials, the actual logistics information of the workshop's front warehouse is obtained by connecting to the WMS (warehouse management system). In terms of the real-time status of processing tasks, the real-time processing status information of each product is obtained by connecting to the factory MES (manufacturing execution system). Finished product inspection information is obtained by connecting to the quality control system, so that production scheduling and optimization can be carried out for products with high defective rates.

[0130] Furthermore, the full-process data collection unit incorporates a built-in risk assessment model, dynamically scoring observation items based on a sub-item scoring strategy. Any data change triggers a risk assessment for the corresponding project, impacting the sub-item scoring.

[0131] Status update notification management unit, event publishing service for message subscription.

[0132] Specifically, the status update notification management unit provides a subscription-based event publishing service, providing subscription services based on observation items, scores, and changes. In this invention, the dynamic production scheduling system automatically subscribes to event information related to equipment failures and personnel entry and exit, as well as risk score messages for processing task delay assessments.

[0133] Figure 10 The flowchart of the operation of the order task pre-processing module of this application is shown. Figure 8 and Figure 10The order task preprocessing module includes a customer order placement unit, a workshop order replenishment unit, and an order pending task management unit. The order preprocessing process mainly decomposes the processing steps based on the customized product content in the order. The original order information records the categories of products to be processed, the quantity of each product to be processed, and the customized items for each product. Because the workshop's processing work is based on a single process and has no direct connection to the order, the original order information cannot be used directly by the workshop. It must be decomposed based on the process granularity to generate a pool of pending process tasks.

[0134] Among them, the customer order unit and the workshop order replenishment unit push the latest order information to the order processing task management unit in real time;

[0135] The pending processing task management unit conducts capacity assessment. Specifically, it can determine whether a new workshop production wave is generated by judging whether there are any remaining products in the workshop. If a new workshop production wave is generated, the order of the workshop production wave is decomposed into process-level tasks.

[0136] The task decomposition in the process dimension is centered on the minimum process node, and combines the product category processing process flow, process configuration, personnel skills and other information in the order to form a final resource combination required for process processing, forming a process processing task pool.

[0137] The single task information items of the process processing task pool are shown in Table 3.

[0138] Table 3

[0139]

[0140] When the status update notification management unit notifies you of a workshop status event of interest, the production scheduling system will dynamically adjust the production schedule based on the production schedule. Different strategies will be supported depending on the size of the notification event.

[0141] Major adjustment events primarily include equipment damage, raw material shortages, worker leave requests, new production waves, and adjustments to production waves. Upon receiving a major adjustment event, the system stops the timer, preventing workers and operating machines from receiving new processing tasks (effectively halting the issuance of production tasks in the workshop). After the timer stops, the timer automatically releases pending tasks and restarts the TW-GA combination method to generate a new production schedule.

[0142] Minor adjustment events primarily include: temporarily shutting down equipment for a certain period of time, temporarily adding small batches of products, and temporarily adjusting the production priority of small batches of products. Upon receiving a minor adjustment event, the system will not stop the timer pointer. Instead, it will reschedule tasks based on the time window after the current pointer, implementing corresponding strategies. These strategies primarily include: replacing the time window for processing task tasks and migrating the overall time window of the process's tight timeline (moving it forward or backward).

[0143] This application is a flexible customized intelligent scheduling method based on the TW-GA combination improved by the whole process data of the printing and dyeing workshop. It is a dynamic scheduling method and an effective practice of the Flexible Job-shop Scheduling Problem (FJSP) in the printing and dyeing industry.

[0144] The flexible job shop scheduling problem is a special case of the classic single-product shop scheduling problem (JSP). It addresses the challenges of flexible customization scenarios, characterized by a wide variety of product categories, non-standard products, complex production processes, diverse worker skills, and a high degree of process flexibility. FJSP is more challenging than traditional JSP because it not only considers cross-category job path optimization but also cross-process flow process combinations. It is a typical fusion of path optimization and combinatorial optimization, and falls into the category of NP-HARD problem solving.

[0145] This application provides a flexible, customized, intelligent production scheduling method based on an improved TW-GA combination of data from the entire production process of a printing and dyeing workshop. This method treats the entire workshop as a whole, models its equipment and personnel, and, based on this modeling, integrates a dynamic workshop status perception system to collect real-time data on equipment, personnel, and production status, thereby forming a full-process database. The improved TW-GA combination method analyzes and processes this full-process database and makes corresponding scheduling adjustments.

[0146] Among them, the population genetic algorithm (GA) is designed and proposed based on the evolutionary laws of organisms in nature. It is a computational model of the biological evolution process that simulates the natural selection and genetic mechanisms of Darwin's theory of biological evolution. It is a method of searching for the optimal solution by simulating the natural evolution process. The optimal individual is found by iterating the continuous selection, crossover, and mutation of individuals in the population. The genetic algorithm has a strong global search capability but a poor local search capability, which places high requirements on the quality of the basic population and the disturbance factor. Therefore, this application uses the time wheel algorithm to generate the basic population, which largely guarantees the quality of the individuals and avoids the situation of premature convergence to a great extent.

[0147] The time wheel algorithm is a scheduling model algorithm based on a time wheel. In the time wheel algorithm system, time wheels are created by process, and each process has its own scheduling time wheel. All time wheels are modeled using production working time as the time wheel length and the processing time of the single piece of the process represented as the time grid. After modeling, the time wheels of each process are superimposed and aligned. The time wheel grids are filled with the fragmented processing requirements of the process according to the established strategy, thus forming an initial production schedule. During the specific processing process, if the sending equipment or manual anomalies occur, the time grids within the time wheel's range of influence are automatically closed, and the tasks in the time grids are removed and rescheduled. This time wheel algorithm not only fully utilizes the time window advantages of the time wheel, but also ensures the advantages of queue-based task scheduling.

[0148] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, any of the above methods is implemented.

[0149] like Figure 11 As shown, the computer system 300 includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage unit 308 into a random access memory (RAM) 303. Various programs and data required for the operation of the system 300 are also stored in the RAM 303. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0150] The following components are connected to the I / O interface 305: an input section 306 including a keyboard, a mouse, and the like; an output section 307 including a liquid crystal display (LCD), a speaker, and the like; a storage section 308 including a hard disk and the like; and a communication section 309 including a network interface card such as a LAN card or a modem. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. Removable media 311, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 310 as needed, so that computer programs read therefrom can be installed into the storage section 308 as needed.

[0151] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 309, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, the above-mentioned functions defined in the method of the present application are executed. It should be noted that the computer-readable medium described in the present application can be a computer-readable signal medium or a computer-readable medium or any combination of the above two. The computer-readable medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this application, a computer-readable medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such a propagated data signal may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including, but not limited to, wireless, wire, optical cable, RF, or any suitable combination thereof.

[0152] Computer program code for performing the operations of the present application can be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0153] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0154] The modules described in the embodiments of this application may be implemented in software or hardware. The modules described may also be provided in a processor. For example, a processor may be described as including an acquisition module, an analysis module, and an output module. The names of these modules do not, in some cases, limit the modules themselves.

[0155] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned inventive concept. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A dynamic production scheduling method based on TW-GA combination, characterized in that: The following steps are involved: S1. Build time wheel group and product processing timeline; S2. Using the task selector in the time wheel group, select the process task to be processed according to the process node of the product processing timeline, and embed the process task to be processed into the time window of the time wheel group; S3, repeating step S2 until there are no more pending processing tasks, outputting a basic production scheduling sample, wherein the basic production scheduling sample includes information about the time window; S4. Sort the processing tasks according to N (N≥1) sorting strategies, and repeat steps S1 to S3 to obtain N basic production scheduling samples; S5. Randomly select two basic production scheduling samples from the N basic production scheduling samples, perform time window intersection on the same type of basic time wheels or overlapping time wheels in the two basic production scheduling samples, and perform fitness evaluation on the results of the time window intersection. Save the individuals with improved fitness as the next generation of basic production scheduling samples; S6. Repeat step S5 until all N basic production scheduling samples complete the time window cross-evolution; and S7. Repeat steps S5 to S6 until the evolution reaches the preset generation or the evolution convergence reaches the evolution stop standard, and output the basic production scheduling sample with the highest fitness.

2. The dynamic production scheduling method based on TW-GA combination according to claim 1 is characterized in that: In step S1, the timing wheel group includes the basic timing wheel of the equipment machine, the basic timing wheel of the manual skill, and the overlapping timing wheels between the equipment machine and the manual skill; the product processing timeline includes the processing time of each process of the product and the arrangement order of each process of the product.

3. The dynamic production scheduling method based on TW-GA combination according to claim 1 is characterized in that: Step S1 also includes the following sub-steps: S11, constructing a basic time wheel and an overlapping time wheel; S12, eliminating the basic time wheel that has created the overlapping time wheel; S13, aligning the overlapping time wheel and the basic time wheel of step S12; And S14, start the overlapping time wheel and the basic time wheel of step S13, and notify the task selector on the corresponding time wheel to select the task to be processed every time a unit time is advanced.

4. The dynamic production scheduling method based on TW-GA combination according to claim 1 is characterized in that: In step S2, the task selector in the time wheel group is used to select the corresponding process task to be processed according to the process node of the product processing timeline, and the following conditions are met: a. The current time window of the basic time wheel or the overlapping time wheel must be able to meet the time consumption requirement of the task to be processed; b. Prioritizing the selection of tasks by the basic time wheel or the overlapping time wheel that has been used; c. The usage rates of the basic time wheels of the same type differ by no more than 30%; d. If the timing wheel group includes overlapping timing wheels, select the basic timing wheel with the minimum impact quantity.

5. The dynamic production scheduling method based on TW-GA combination according to claim 1 is characterized in that: In step S5, time windows are crossed for the same type of basic time wheels or overlapping time wheels in the two basic production scheduling samples, and the following conditions are met: a. The time window intersection must meet the constraints of the arrangement order of each process of the product; b. There is sufficient time in the time windows after the intersection; c. The time windows within the same basic time wheel or overlapping time wheel can be self-intersecting; d. After the time windows intersect, there is only one task node of the same process to be processed.

6. The dynamic production scheduling method based on TW-GA combination according to claim 1 is characterized in that: In step S4, the processing process tasks are sorted according to N (N≥1) sorting strategies, including sorting by product brand category, sorting by processing time, sorting by the proportion of equipment and machine occupancy time, sorting by the proportion of manual labor occupancy time, sorting by the number of processes and heuristic sorting by the order of work sections.

7. A dynamic production scheduling module based on TW-GA combination, characterized in that: The modules include: Construction unit, used to construct time wheel set and product processing timeline; A task selection unit, wherein the task selector of the time wheel group selects the process tasks to be processed according to the process nodes of the product processing timeline, and embeds the process tasks to be processed into the time window of the time wheel group; an acquisition unit, configured to repeat the operation on the task selection unit until the to-be-processed process task no longer exists, and output a basic production scheduling sample, wherein the basic production scheduling sample includes information of the time window; The sample production unit is used to sort the processing process tasks according to N (N≥1) sorting strategies, and obtain N basic production scheduling samples after repeating the operations on the construction unit, task selection unit, and acquisition unit. The evolution unit is used to randomly select two basic production scheduling samples from the N basic production scheduling samples, perform time window intersection on the same type of basic time wheels or overlapping time wheels in the two basic production scheduling samples, and perform fitness evaluation on the results of the time window intersection, and save the individuals with improved fitness as the next generation of basic production scheduling samples; An iteration unit is used to repeat the operations on the evolution unit until all N basic production scheduling samples complete the time window cross-evolution; and an output unit is used to repeat the operations on the evolution unit and the iteration unit until the evolution reaches a preset generation or the evolution converges to the evolution stop standard, and then output the basic production scheduling sample with the highest fitness.

8. A dynamic production scheduling system based on TW-GA combination, characterized in that: It includes an order task preprocessing module, a workshop situation awareness module and the dynamic production scheduling module as described in claim 7, the order task preprocessing module provides processing tasks for the dynamic production scheduling module, and the workshop situation awareness module provides basic data support for the order task preprocessing module and the dynamic production scheduling module.

9. The dynamic production scheduling system based on TW-GA combination according to claim 8 is characterized in that: The workshop situation awareness module includes a basic data unit, a full-process data acquisition unit, and a status update notification management unit; The basic data unit is used to establish a workshop material storage library, a production equipment machine library, a production process skill requirement and time consumption library, a production process flow library for various products, and a worker skill library; The whole-process data collection unit is used to collect worker on-the-job information, equipment information, logistics information, product processing information, and product inspection information, and the whole-process data collection unit is equipped with a risk assessment model; The status update notification management unit is used for event publishing services of message subscription.

10. The dynamic production scheduling system based on TW-GA combination according to claim 8, characterized in that: The order task pre-processing module includes a customer order unit, a workshop order replenishment unit and an order pending processing task management unit; The customer order placement unit and the workshop order replenishment unit push the latest order information to the order processing task management unit in real time; The pending-processing task management unit performs capacity assessment to determine whether to generate a new workshop production wave. If a new workshop production wave is generated, the order of the workshop production wave is decomposed into process-level tasks.

11. A computer-readable storage medium, wherein a computer program is stored in the medium, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images