A production scheduling optimization method and system for intelligent manufacturing

Abstract

This invention relates to the field of intelligent manufacturing technology, and in particular to a production scheduling optimization method and system for intelligent manufacturing. The method includes: obtaining the original production schedule; generating an initial schedule based on the original production schedule using an optimization algorithm; deploying the initial schedule and monitoring disturbances in real time; selectively rearranging the initial schedule when a disturbance occurs to generate a final production schedule; and feeding back the monitored disturbance type and the final production schedule, and dynamically adjusting the optimization algorithm parameters based on the disturbance type and the final production schedule.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to a production scheduling optimization method and system for intelligent manufacturing. Background Technology

[0002] Intelligent manufacturing refers to the general term for advanced manufacturing processes, systems, and models that are based on new-generation information technologies (such as the Internet of Things, big data, artificial intelligence, cloud computing, 5G, etc.), permeate all aspects of manufacturing activities such as design, production, management, and service, and have functions such as deep self-sensing of information, intelligent optimization and self-decision-making, and precise control and self-execution.

[0003] Production scheduling in intelligent manufacturing is no longer a simple "shift schedule" in the traditional sense, but a dynamic, real-time, and adaptive optimization process. However, existing optimization scheduling algorithms or models, in pursuit of efficiency, eliminate waste and buffering to the greatest extent, making the system extremely sensitive to disturbances. Even a small disturbance can easily cause the scheduling plan to collapse, and the system is difficult to recover from. Therefore, we propose a technical solution to address these problems. Summary of the Invention

[0004] This invention improves the robustness of the system by rearranging the initial schedule in response to different disturbances, enabling the system to quickly restore the production schedule plan when disturbances occur.

[0005] The technical solution proposed in this invention is: a production scheduling optimization method for intelligent manufacturing, the method comprising:

[0006] Obtain the original production schedule, and generate an initial schedule based on the original production schedule using an optimization algorithm;

[0007] Deploy the initial schedule and monitor disturbances in real time. When disturbances occur, selectively rearrange the initial schedule to generate the final production schedule.

[0008] The algorithm parameters are dynamically adjusted based on the detected disturbance types and the final production schedule.

[0009] Preferably, the step of generating an initial schedule based on the original production schedule using an optimization algorithm includes:

[0010] Obtain the start and finish times of the original generated schedule;

[0011] By analyzing the original production schedule, key tasks and bottleneck resources can be identified.

[0012] Calculate the buffer time based on the characteristics of tasks in the production schedule;

[0013] Insert a buffer time at the back end of the task to obtain the initial schedule.

[0014] Preferably, the step of deploying the initial schedule and monitoring disturbances in real time, and selectively rearranging the initial schedule when a disturbance occurs to generate the final production schedule, includes:

[0015] Collect data from the production management system and IoT sensor data;

[0016] Continuously monitor abnormal events based on production management system data and IoT sensor data to determine if any disturbances have occurred;

[0017] The abnormal events include resource status abnormal events, task progress abnormal events, material change abnormal events, and order change abnormal events; when an abnormal event is identified, i.e., a disturbance occurs, the tasks affected by the disturbance are identified, forming a set of affected tasks, including:

[0018] Extract the time of occurrence of the disturbance;

[0019] Identify tasks that were in progress when the disturbance occurred, tasks that had not yet started, or tasks whose required resources were affected by the disturbance from the initial schedule;

[0020] The set of affected tasks is defined by the tasks that are in progress when the disturbance occurs, the tasks that have not yet started, and the tasks whose required resources are affected by the disturbance.

[0021] Calculate the plan deviation of all tasks within the affected task set. ;in, Indicates task The actual start time under the influence of the disturbance Indicates tasks in the initial schedule The normal start time, Indicates task The priority of the work order;

[0022] Determine the type and intensity of the disturbance, and generate a selective rearrangement strategy based on the disturbance type, including:

[0023] Count the number of affected tasks from the set of affected tasks. ,if and If so, the disturbance is determined to be a local disturbance; among which, Local perturbation threshold Deviation threshold;

[0024] Count the number of critical tasks affected from the set of affected tasks. ,if, If the critical task affected corresponds to a bottleneck resource, then the disturbance is determined to be a global disturbance; among which, Indicates the global perturbation threshold;

[0025] For local disturbances, if In this case, only the start and end times of the ongoing task will be postponed by a fixed time interval. ; Indicates the threshold for moderate disturbance intensity;

[0026] if Then rearrange the time window Tasks that have not yet started within the specified time window are rescheduled; the rescheduling time window ,in, This indicates the current time point, i.e., the time when the disturbance occurred. Indicates the width of the rearrangement time window;

[0027] In response to a global disturbance, all unfinished tasks are rearranged starting from the current time.

[0028] The identification of key tasks and bottleneck resources includes:

[0029] Calculate the earliest start time for each task in the production schedule. Earliest completion time Latest start time and latest completion time ;

[0030] if and If so, then the task is determined to be a critical task;

[0031] A path consisting of critical tasks is a critical task path;

[0032] Obtain resources for each task in the production schedule. Total load time ;

[0033] Calculate the resources for each task utilization rate ;in, Represents the resources for each task Total available time;

[0034] if and Then determine the resource As a bottleneck resource, among which, Indicates the maximum load time. This represents the utilization threshold.

[0035] Preferably, the calculation of the buffer time based on the characteristics of tasks in the production schedule includes:

[0036] Retrieve the historical processing times of all tasks in the production schedule from the database to form a historical processing time series;

[0037] The characteristics of the task are obtained based on the historical processing time series, namely the standard deviation of the historical processing time. ;

[0038] Calculate buffer time separately for each task ;in, Indicates assignment to task buffer time, ; Indicates the adjustment factor. Indicates task Average processing time Indicates task Standard deviation of historical processing time; This represents the safety factor.

[0039] Preferably, the calculation of buffer time based on the characteristics of tasks in the production schedule further includes:

[0040] Identify the critical path; calculate the end-buffer time of the critical path. ;in, Indicates the critical path On the task Quantity; Indicates the critical path On the task The standard deviation of historical processing time;

[0041] Set the conveyor buffer time ;in, Indicates non-critical paths On the task Quantity, Indicates non-critical paths On the task The standard deviation of historical processing time.

[0042] Preferably, the step of inserting a buffer time at the back end of the task to obtain the initial schedule further includes:

[0043] Insert end-buffer time at the end of the critical path. That is, the total processing time of the critical path is calculated by adding the end-of-path buffer time to the average processing time of the critical task at the end of the critical path; ;

[0044] Insert a transport buffer time at the point where a non-critical path merges into a critical path. This refers to the end-of-path delivery buffer time for tasks merging into the critical path from the non-critical path; and obtaining the total processing time of the non-critical path. ;

[0045] Obtain the start time node of the original schedule, and use the total processing time of the critical path, the total processing time of the non-critical path, the critical task path and the non-critical path to form an initial schedule of the aggregated buffer.

[0046] Preferably, the rearrangement time window The tasks that have not yet started are rearranged, including:

[0047] Genetically generated using a genetic algorithm The new schedule for tasks that have not yet started is as follows:

[0048] Initialize the population, that is, for The tasks that have not yet started are encoded to form multiple task execution sequences, and then any one of the task execution sequences is represented by a chromosome;

[0049] Constructing the fitness function Among them, the efficiency objective function ,in, Indicates order priority, Indicates order Delaying time, , Indicates order Completion time, Indicates order Delivery time; Indicates the order quantity; , , Indicates the target weight;

[0050] Stability objective function ;in, Indicates a task that has not yet started. The start time in the new schedule, Indicates a task that has not yet started. The start time in the initial schedule Indicates task priority, This indicates the number of tasks that have not yet started within the reordering time window;

[0051] Priority objective function ;in, Indicates the indicator function, if the order Delay, then Otherwise, it is 0;

[0052] A new generation of population is obtained through selection, crossover, and mutation, and the fitness of the new generation population is evaluated to obtain the value of the fitness function.

[0053] Repeat the above steps to obtain the optimal population, that is, the population with the largest fitness function. After decoding, the optimal task execution sequence is obtained, that is, the new schedule. The new schedule and the already executed initial schedule constitute the final production schedule and are saved locally.

[0054] Preferably, the feedback-monitored disturbance type and final production schedule are used to dynamically adjust the optimization algorithm parameters based on the final production schedule, including:

[0055] Obtain the final production schedule and calculate the consumption ratio of each buffer time, i.e.: ;in, , , This indicates the actual consumption of back-end insertion buffer time, end-of-line buffer time, and delivery buffer time. , , These represent the proportions of back-end insertion time consumption, end-of-line buffer time consumption, and delivery buffer time consumption, respectively.

[0056] Generate decision-making strategies, including:

[0057] if , , If any one of them is greater than 1, then the parameter is adjusted upwards. or ;

[0058] if , , If any one of them is less than 0.5, then the parameter is adjusted downwards. or ;

[0059] Adjusted parameters ; This indicates the average buffer time efficiency ratio. Indicates the percentage of target buffer time consumed; Indicates the learning rate. This represents the optimization algorithm parameters before adjustment, including parameters. or .

[0060] Preferred options also include:

[0061] Obtain the consumption time of each task in the final production schedule, as well as the back-end insertion buffer time, end buffer time, and delivery buffer time of the corresponding task.

[0062] The histogram displays the time consumed by each task and the corresponding backend insertion buffer time, end buffer time, and delivery buffer time. This time, along with the generated final production schedule, is fed back to the technical staff so that they can understand the final production schedule plan and decide whether to execute it. After waiting for a preset time, the final schedule plan is executed automatically, and an automatic execution warning is triggered.

[0063] A production scheduling optimization system for intelligent manufacturing, the system being used to execute the aforementioned production scheduling optimization method for intelligent manufacturing.

[0064] The beneficial effects of this invention are:

[0065] 1. This invention generates an initial schedule by inserting buffer time at different task positions in the original production schedule, thereby improving the system's ability to cope with interference.

[0066] 2. This invention employs different strategies for rescheduling in response to different disturbances, including: a right-shift strategy, which postpones all tasks that have not yet started by a fixed time interval for mild local disturbances; and a local and global rescheduling strategy, which uses local rescheduling for moderate local disturbances and global rescheduling for global disturbances. This strategy adapts to disturbances of different ranges and intensities, ensuring the efficiency of system recovery scheduling and avoiding the time waste caused by a complete rescheduling.

[0067] 3. This invention optimizes the algorithm by dynamically adjusting its parameters, thereby ensuring the system's adaptability. Attached Figure Description

[0068] Figure 1 This is a flowchart of a production scheduling optimization method for intelligent manufacturing according to the present invention. Detailed Implementation

[0069] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious modifications will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0070] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0071] Example 1:

[0072] refer to Figure 1 The technical solution provided by this invention is: a production scheduling optimization method for intelligent manufacturing, the method comprising:

[0073] Step 1: Obtain the original production schedule. Based on the original production schedule, generate an initial schedule using an optimization algorithm. This includes the following steps:

[0074] Step 1.1: Obtain the start and finish times of the original generated schedule.

[0075] Step 1.2: Identify key tasks and bottleneck resources by analyzing the original production schedule, specifically including the following steps:

[0076] Calculate the earliest start time for each task in the production schedule. Earliest completion time Latest start time and latest completion time ;

[0077] if and If so, then the task is determined to be a critical task;

[0078] A path consisting of critical tasks is a critical task path;

[0079] Obtain resources for each task in the production schedule. Total load time ;

[0080] Calculate the resources for each task utilization rate ;in, Represents the resources for each task Total available time;

[0081] if and Then determine the resource As a bottleneck resource, among which, Indicates the maximum load time. This represents the utilization threshold. Resources with higher utilization and longer load durations are considered bottleneck resources because their failures have a greater impact on production scheduling.

[0082] Step 1.3: Based on the characteristics of tasks in the production schedule, calculate the buffer time, which specifically includes the following steps:

[0083] Retrieve the historical processing times of all tasks in the production schedule from the database to form a historical processing time series;

[0084] The characteristics of the task are obtained based on the historical processing time series, namely the standard deviation of the historical processing time. ;

[0085] Calculate buffer time separately for each task .in, Indicates assignment to task buffer time, ;

[0086] Indicates the adjustment factor. This is a local parameter used for fine-tuning based on the criticality of the task or resource. Set a value greater than 1 for tasks on the critical path and bottleneck resources, such as 1.2 or 1.5; set a value less than 1 for other tasks.

[0087] Indicates task The average processing time. Indicates task The standard deviation of historical processing time reflects the volatility of historical processing time. The larger the value, the more unstable the task is and the greater the buffer required.

[0088] Indicates the safety factor. As a global parameter, it determines the overall conservatism of the system. When it is 1, it covers approximately 68% of the fluctuations (assuming the fluctuations follow a normal distribution); when it is 2, it covers 95% of the fluctuations; and when it is 3, it covers 99.7% of the fluctuations.

[0089] Identify the critical path; calculate the end-buffer time of the critical path. ;in, Indicates the critical path On the task The quantity. Indicates the critical path On the task The standard deviation of historical processing time.

[0090] Set the conveyor buffer time ;in, Indicates non-critical paths On the task The quantity. Indicates non-critical paths On the task The standard deviation of historical processing time.

[0091] Step 1.4: Insert buffer time in the backend of the task to obtain the initial schedule, which includes the following steps:

[0092] Insert buffer time at the back end of each task. That is, the average processing time of each task plus the buffer time; therefore, the initial processing time of each task is... ;

[0093] Obtain the start time node of the original schedule, and use the initial processing time of each task and the tasks contained in the original schedule to form the initial schedule of the distributed buffer.

[0094] Insert end-buffer time at the end of the critical path. That is, the total processing time of the critical path is calculated by adding the end-of-path buffer time to the average processing time of the critical task at the end of the critical path; ;

[0095] Insert a transport buffer time at the point where a non-critical path merges into a critical path. This refers to the end-of-path delivery buffer time for tasks merging into the critical path from the non-critical path; and obtaining the total processing time of the non-critical path. ;

[0096] Obtain the start time node of the original schedule, and use the total processing time of the critical path, the total processing time of the non-critical path, the critical task path and the non-critical path to form an initial schedule of the aggregated buffer.

[0097] Step 2: Deploy the initial schedule and monitor disturbances in real time. When a disturbance occurs, selectively rearrange the initial schedule to generate the final production schedule. This includes the following steps:

[0098] Step 2.1: Collect data from the production management system and IoT sensor data (e.g., IoT sensors on production equipment related to production scheduling). By acquiring these two types of data, production scheduling-related data can be obtained, such as order data, material data, and resource status data (e.g., equipment status data).

[0099] Step 2.2: Continuously monitor abnormal events based on production management system data and IoT sensor data to determine whether any disturbances have occurred;

[0100] The abnormal events include resource status abnormal events, task progress abnormal events, material change abnormal events, and order change abnormal events; the resource status includes equipment working status and staff on-duty status; the task progress abnormal events include production tasks starting early or late, completing early or late, and requiring rework; the material change abnormal events include material arrival delays and incomplete materials; and the order change abnormal events include order insertion, order cancellation, and order priority changes.

[0101] Step 2.3: When an abnormal event is detected, i.e., a disturbance occurs, identify the tasks affected by the disturbance, forming a set of affected tasks, including:

[0102] Extract the time of disturbance occurrence; identify tasks that were in progress, tasks that had not yet started, or tasks whose required resources were affected by the disturbance when the disturbance occurred from the initial schedule;

[0103] The set of affected tasks is defined by the tasks that are in progress when the disturbance occurs, the tasks that have not yet started, and the tasks whose required resources are affected by the disturbance.

[0104] Step 2.4: Calculate the plan deviation of all tasks within the affected task set. ;in, Indicates task The actual start time under the influence of the disturbance Indicates tasks in the initial schedule The normal start time, Indicates task The priority of the work order determines the delay cost; higher priority work orders incur greater costs.

[0105] Step 2.5: Determine the type and intensity of the disturbance, and generate a selective rearrangement strategy based on the disturbance type, including:

[0106] Step 2.51: Count the number of affected tasks from the set of affected tasks. ,if and If so, the disturbance is determined to be a local disturbance; among which, Local perturbation threshold Deviation threshold.

[0107] Step 2.52: Count the number of critical tasks affected from the set of affected tasks. ,if, If the critical task affected corresponds to a bottleneck resource, then the disturbance is determined to be a global disturbance; among which, This represents the global disturbance threshold.

[0108] Step 2.53, corresponding to local disturbances, if In this case, only the start and end times of the ongoing task will be postponed by a fixed time interval. ; Indicates the threshold for moderate disturbance intensity;

[0109] if Then rearrange the time window The tasks that have not yet started are reordered, specifically including the following steps:

[0110] The rearrangement time window ,in, This indicates the current time point, i.e., the time when the disturbance occurred. Indicates the width of the rearrangement time window;

[0111] Genetically generated using a genetic algorithm The new schedule for tasks that have not yet started is as follows:

[0112] Initialize the population, that is, for The tasks that have not yet started are encoded to form multiple task execution sequences, and then any one of the task execution sequences is represented by a chromosome;

[0113] Constructing the fitness function Among them, the efficiency objective function ,in, Indicates order priority, Indicates order Delaying time, , Indicates order Completion time, Indicates order Delivery time; Indicates the order quantity; , , This represents the target weight, which can be set higher in the initial stages of the rearrangement. (Stability) To avoid chaos, it can be increased after recovery. (efficiency).

[0114] Stability objective function ;in, Indicates a task that has not yet started. The start time in the new schedule, Indicates a task that has not yet started. The start time in the initial schedule Indicates task priority, This indicates the number of tasks that have not yet started within the reordering time window;

[0115] Priority objective function ;in, Indicates the indicator function, if the order Delay, then Otherwise, it is 0;

[0116] A new generation of population is obtained through selection, crossover, and mutation, and the fitness of the new generation population is evaluated to obtain the value of the fitness function.

[0117] Repeat the above steps to obtain the optimal population, i.e., the population with the largest fitness function. After decoding, the optimal task execution sequence is obtained, i.e., the new schedule. The new schedule allows production to recover quickly and minimizes the impact of disturbances. Local adjustments maintain the stability of the production schedule, avoiding excessive changes under local disturbances.

[0118] The new schedule and the initial schedule that has already been executed constitute the final production schedule, which is stored locally.

[0119] Step 2.54: For the global perturbation, starting from the current time, rearrange all unfinished tasks. The specific rearrangement steps are similar to the task rearrangement process within the rearrangement time window described above, except that the stability objective function... This indicates the total number of tasks that have not yet been completed.

[0120] Step 3: Feedback the detected disturbance types and final production schedule, and dynamically adjust and optimize the algorithm parameters based on the final production schedule.

[0121] Obtain the final production schedule and calculate the consumption ratio of each buffer time, i.e.: ;in, , , This indicates the actual consumption of back-end insertion buffer time, end-of-line buffer time, and delivery buffer time. , , These represent the proportions of back-end insertion time consumption, end-of-line buffer time consumption, and delivery buffer time consumption, respectively.

[0122] Generate decision-making strategies, including:

[0123] if , , If any value in the range is greater than 1 (indicating insufficient buffering), then the parameter should be increased. or ;

[0124] if , , If any of the values ​​is less than 0.5 (indicating that the buffer is too large and affecting efficiency), then the parameter should be adjusted downwards. or ;

[0125] Adjusted parameters .in, This represents the adjusted optimization algorithm parameters. This indicates the average buffer time efficiency ratio. This indicates the percentage of target buffer time consumed (typically 0.7-0.8). Indicates the learning rate. This represents the optimization algorithm parameters before adjustment, including parameters. or .

[0126] The present invention also provides a production scheduling optimization system for intelligent manufacturing, the system being used to execute the aforementioned production scheduling optimization method for intelligent manufacturing.

[0127] Example 2:

[0128] To ensure the optimization algorithm and final production schedule are transparent to technical personnel, we propose the following technical solution based on Example 1:

[0129] Obtain the consumption time of each task in the final production schedule, as well as the back-end insertion buffer time, end-end buffer time, and delivery buffer time of the corresponding task.

[0130] The histogram displays the time consumed by each task, along with the corresponding backend insertion buffer time, end-of-line buffer time, and delivery buffer time. This data, along with the generated final production schedule, is fed back to technical personnel so they can understand the final production schedule and decide whether to execute it. After a preset waiting time, the final schedule is executed automatically, triggering an automatic execution warning. This avoids the black-box problem of the algorithm, ensuring human decision-making authority.

[0131] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. The embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wire segments, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection 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 propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical fibers, RF, etc., or any suitable combination thereof.

[0132] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation that may be implemented in systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0133] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and explained in the embodiments. Without departing from the principles described, the implementation of the present invention may have any changes or modifications.

Claims

1. A production scheduling optimization method for smart manufacturing, characterized by, The method includes: Obtain the original production schedule, and generate an initial schedule based on the original production schedule using an optimization algorithm; this includes: obtaining the start time and finish time of the generated schedule. By analyzing the original production schedule, key tasks and bottleneck resources are identified; based on the characteristics of the tasks in the production schedule, buffer time is calculated, which specifically includes the following steps: obtaining the historical processing time of all tasks in the production schedule from the database to form a historical processing time series; obtaining the characteristics of the task based on the historical processing time sequence, i.e. the standard deviation of the historical processing time ; calculating the buffer time separately for each task ; wherein, the buffer time allocated to the task , ; denotes the adjustment coefficient, denotes the average processing time of the task , denotes the standard deviation of the historical processing time of the task ; denotes the safety coefficient; inserting the buffer time at the end of the task, obtaining the initial schedule; Deploy the initial schedule and monitor disturbances in real time. When disturbances occur, selectively rearrange the initial schedule to generate the final production schedule; including: Collect data from the production management system and IoT sensor data; Based on production management system data and IoT sensor data, we continuously monitor abnormal events to determine whether disturbances have occurred; the abnormal events include resource status abnormal events, task progress abnormal events, material change abnormal events, and order change abnormal events. When an abnormal event is detected, i.e. a disturbance occurs, the tasks affected by the disturbance are identified, forming a set of affected tasks; the disturbance type and intensity are determined, and a selective rescheduling strategy is generated based on the disturbance type; this includes: extracting the disturbance occurrence time; and identifying tasks that were in progress, had not yet started, or whose required resources were affected by the disturbance from the initial schedule when the disturbance occurred. The set of affected tasks is defined by the tasks that are in progress when the disturbance occurs, the tasks that have not yet started, and the tasks whose required resources are affected by the disturbance. Calculate the plan deviation of all tasks within the affected task set. ;in, Indicates task The actual start time under the influence of the disturbance Indicates tasks in the initial schedule The normal start time, Indicates task The priority of the work order; The determination of the disturbance type and disturbance intensity, and the generation of a selective rearrangement strategy based on the disturbance type, includes: Count the number of affected tasks from the set of affected tasks. ,if and If so, the disturbance is determined to be a local disturbance; among which, Local perturbation threshold Deviation threshold; Count the number of critical tasks affected from the set of affected tasks. ,if, If the critical task affected corresponds to a bottleneck resource, then the disturbance is determined to be a global disturbance; among which, Indicates the global perturbation threshold; For local disturbances, if In this case, only the start and end times of the ongoing task will be postponed by a fixed time interval. ; Indicates the threshold for moderate disturbance intensity; if Then rearrange the time window Reorder tasks that have not yet started; including: Genetically generated using a genetic algorithm The new schedule for tasks that have not yet started is as follows: Initialize the population, that is, for The tasks that have not yet started are encoded to form multiple task execution sequences, and then any one of the task execution sequences is represented by a chromosome; Constructing the fitness function Among them, the efficiency objective function ,in, Indicates order priority, Indicates order Delaying time, , Indicates order Completion time, Indicates order Delivery time; Indicates the order quantity; , , Indicates the target weight; Stability objective function ;in, Indicates a task that has not yet started. The start time in the new schedule, Indicates a task that has not yet started. The start time in the initial schedule Indicates task priority, This indicates the number of tasks that have not yet started within the reordering time window; Priority objective function ;in, Indicates the indicator function, if the order Delay, then Otherwise, it is 0; A new generation of population is obtained through selection, crossover, and mutation, and the fitness of the new generation population is evaluated to obtain the value of the fitness function. Repeat the above steps to obtain the optimal population, that is, the population with the largest fitness function. After decoding, the optimal task execution sequence is obtained, that is, the new schedule. The new schedule and the already executed initial schedule constitute the final production schedule and are saved locally. The rearrangement time window ,in, This indicates the current time point, i.e., the time when the disturbance occurred. Indicates the width of the rearrangement time window; In response to a global disturbance, all unfinished tasks are rearranged starting from the current time. The identification of key tasks and bottleneck resources includes: Calculate the earliest start time for each task in the production schedule. Earliest completion time Latest start time and latest completion time ; if and If so, then the task is determined to be a critical task; A path consisting of critical tasks is a critical task path; Obtain resources for each task in the production schedule. Total load time ; Calculate the resources for each task utilization rate ;in, Represents the resources for each task Total available time; if and Then determine the resource As a bottleneck resource, among which, Indicates the maximum load time. Indicates the utilization threshold; The feedback includes the detected disturbance types and the final production schedule. Based on the final production schedule, the optimization algorithm parameters are dynamically adjusted, including: obtaining the final production schedule and calculating the consumption ratio of each buffer time, i.e.: ;in, , , This indicates the actual consumption of back-end insertion buffer time, end-of-line buffer time, and delivery buffer time. , , These represent the proportions of back-end insertion time consumption, end-of-line buffer time consumption, and delivery buffer time consumption, respectively. Generate decision-making strategies, including: if , , If any one of them is greater than 1, then the parameter is adjusted upwards. or ; if , , If any one of them is less than 0.5, then the parameter is adjusted downwards. or ; The optimization algorithm parameters are adjusted using a machine learning model; that is, the adjusted parameters. ; This indicates the average buffer time efficiency ratio. Indicates the percentage of target buffer time consumed; Indicates the learning rate. This represents the optimization algorithm parameters before adjustment, including parameters. or .

2. The production scheduling optimization method for intelligent manufacturing according to claim 1, characterized in that, The calculation of buffer time based on the characteristics of tasks in the production schedule also includes: Identify the critical path; calculate the end-buffer time of the critical path. ;in, Indicates the critical path On the task Quantity; Indicates the critical path On the task The standard deviation of historical processing time; Set the conveyor buffer time ;in, Indicates non-critical paths On the task Quantity; Indicates non-critical paths On the task The standard deviation of historical processing time.

3. The production scheduling optimization method for intelligent manufacturing according to claim 2, characterized in that, The step of inserting a buffer time at the back end of the task to obtain the initial schedule also includes: Insert end-buffer time at the end of the critical path. That is, the end-of-path buffer time is added after the average processing time of the critical task at the end of the critical path; The total processing time of the critical path ; Insert a transport buffer time at the point where a non-critical path merges into a critical path. This means inserting a delivery buffer time at the end of a non-critical path task that merges into a critical path on a non-critical path. Obtain the total processing time of non-critical paths ; Obtain the start time node of the original schedule, and use the total processing time of the critical path, the total processing time of the non-critical path, the critical task path and the non-critical path to form an initial schedule of the aggregated buffer.

4. The production scheduling optimization method for intelligent manufacturing according to claim 3, characterized in that, Also includes: Obtain the consumption time of each task in the final production schedule, as well as the back-end insertion buffer time, end buffer time, and delivery buffer time of the corresponding task. The consumption time of each task is displayed by a histogram, along with the back-end insertion buffer time, end buffer time, and delivery buffer time of the corresponding task. This, along with the generated final production schedule, is fed back to the technical staff so that they can understand the final production schedule plan and decide whether to execute it. After waiting for the preset time, the final scheduling plan will be executed automatically, and an automatic execution warning will be triggered.

5. A production scheduling optimization system for intelligent manufacturing, characterized in that, The system is used to execute a production scheduling optimization method for intelligent manufacturing as described in any one of claims 1-4.

Images