Intelligent scheduling and efficient production optimization method and system for whole process of garment production

Abstract

The application discloses a full-process intelligent scheduling and efficient production scheduling optimization method and system for garment production, comprising the following steps: collecting order process characteristics, equipment operation states, workshop resource configurations and historical production scheduling data through a garment production intelligent scheduling cloud platform, constructing a multi-dimensional production scheduling basic database, initializing production scheduling constraints and order priority rules by using a garment flexible production scheduling dynamic game model, deducing processing time consumption and delivery nodes by using an order delivery period prediction scheduling algorithm, adjusting equipment task allocation by means of an equipment load balancing optimization network, forming an initial scheme, and then performing multi-objective game optimization, outputting optimized scheduling instructions from the cloud platform and synchronizing to a production execution unit. The application coordinates order delivery and equipment load balancing, improves scheduling adaptability and precision, and solves the problems of single target optimization limitation and dynamic adjustment deficiency in the prior art through multi-model cooperation and full-process dynamic linkage.

Description

Technical Field

[0001] This invention relates to the field of garment production process scheduling technology, and in particular to intelligent scheduling and efficient production optimization methods and systems for the entire garment production process. Background Technology

[0002] The current apparel manufacturing industry faces trends towards smaller order batches, more diverse styles, and tighter delivery cycles. The production process involves complex parameters across multiple dimensions, including order process characteristics, equipment operating status, and workshop resource allocation. These parameters are dynamically correlated and subject to constraints and conflicts. Traditional production scheduling relies on experience-based scheduling models, which struggle to handle scenarios such as parallel processing of multiple orders, fluctuating equipment loads, and complex process connections. However, with the widespread adoption of smart manufacturing technologies, apparel companies have an increasingly urgent need for flexible, precise, and efficient end-to-end scheduling. There is a pressing need to build an intelligent scheduling system that integrates multiple models to achieve a dynamic balance between order delivery and equipment utilization, adapting to the complex demands of modern production.

[0003] Existing technologies suffer from two major drawbacks: First, they lack an integrated optimization mechanism that coordinates multiple objectives. Existing solutions often focus solely on shortening order delivery times or balancing equipment load, without establishing a dynamic game theory and multi-parameter coupled optimization model. This makes it difficult for scheduling solutions to simultaneously meet order delivery requirements and the need for efficient equipment operation, and they cannot effectively coordinate multi-dimensional constraints. Second, they lack intelligent scheduling and dynamic adjustment capabilities. Existing systems often rely on fixed algorithms and static data input, failing to form a closed loop of real-time data acquisition, dynamic simulation, and iterative optimization supported by a cloud platform. This makes it difficult to quickly respond to sudden situations such as order changes and equipment failures, and it does not achieve deep linkage between all stages of the production scheduling process, thus limiting the adaptability and optimization accuracy of the scheduling solutions. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides a method and system for intelligent scheduling and efficient production optimization of the entire garment production process.

[0005] The technical solution adopted in this invention is an intelligent scheduling and efficient production optimization method for the entire garment production process, comprising the following steps: S1, collecting garment order process characteristic parameters, equipment operating status parameters, workshop production resource configuration parameters, and historical production execution data through a garment production intelligent scheduling cloud platform to construct a multi-dimensional production scheduling basic database; S2, based on the basic database, initializing the scheduling constraint boundary using a garment flexible scheduling dynamic game model, and determining the order priority weight allocation rules and equipment processing capacity thresholds; S3, calling an order delivery period prediction scheduling algorithm to dynamically extrapolate the processing time of each order's processing steps, the gap between steps, and the redundancy time of abnormal working conditions; S4, iteratively adjusting the multi-equipment processing task allocation scheme through an equipment load balancing optimization network to form an initial production scheduling scheme; S5, combining the garment flexible scheduling dynamic game model to perform multi-objective game optimization on the initial scheme, coordinating order delivery requirements and equipment load balancing objectives; S6, the garment production intelligent scheduling cloud platform outputs optimized full-process production scheduling instructions, synchronized to each production execution unit, for real-time distribution and dynamic updating of the scheduling scheme.

[0006] Furthermore, the expression for the dynamic game model of flexible production scheduling in the garment industry is as follows: For flexible production scheduling, dynamic Boyer equilibrium solution. Assign the processing task quantity for the i-th order to the j-th equipment. Let j be the maximum processing load threshold of the j-th equipment. Let be the process compatibility coefficient between the i-th order and the j-th piece of equipment. Let i be the priority weight of the i-th order. Let be the load adjustment coefficient of the j-th equipment. Total number of orders This represents the total number of devices.

[0007] Furthermore, the expression for the order delivery date prediction and scheduling algorithm is as follows: ,in, For the predicted delivery period of the k-th order, The standard processing time for the I-th process of the k-th order. Let be the process complexity correction value for the i-th process in the k-th order. Let be the equipment processing efficiency coefficient corresponding to the I-th process of the k-th order. Let be the delivery time sensitivity coefficient for the k-th order. The time impact weight for the I-th process is... Let be the abnormal working condition correction coefficient for the k-th order. This represents the total number of processes in a single order.

[0008] Furthermore, the expression for the device load balancing optimization network is: ,in, It serves as an evaluation index for equipment load balance. Let the weight of the processing task of the j-th machine in the i-th order be . Assign the processing task quantity for the i-th order to the j-th equipment. Let j be the rated processing capacity of the j-th equipment. Let be the load balancing weighting coefficient for the j-th device. Let be the load fluctuation suppression coefficient of the j-th equipment. Total number of orders This represents the total number of devices.

[0009] Furthermore, the scheduling decision optimization expression of the intelligent scheduling cloud platform for garment production is as follows: ,in, To optimize the scheduling scheme, the decision value is determined. For decision weight coefficients, For flexible production scheduling, dynamic Boyer equilibrium solution. The average predicted delivery time for all orders. It serves as an evaluation index for equipment load balance. The maximum value of the equilibrium solution in the game. This represents the maximum value of the average predicted delivery cycle. This represents the maximum value of the equipment load balance.

[0010] Furthermore, the multi-objective coordination expression for intelligent scheduling and efficient production optimization of the entire garment production process is as follows: ,in, To optimize the comprehensive evaluation function for multiple objectives, To coordinate the weights for load balancing and load equilibrium, To optimize weights for delivery period, Let i be the game equilibrium solution for the i-th order. For the load balance of the j-th device, For the predicted delivery period of the k-th order, This is the average delivery time for historical orders. Total number of orders This represents the total number of devices.

[0011] Further, S3 includes the following sub-steps: S31, based on the order process feature parameters collected by the intelligent scheduling cloud platform for garment production, extract the fabric characteristic parameters, cutting accuracy requirements, sewing process complexity parameters, and finishing process parameters required for each process, and establish a process feature vector library; S32, input the parameters in the process feature vector library into the order delivery period prediction scheduling algorithm, and combine it with the process time data of similar historical orders to construct a dynamic prediction model for process time; S33, use this model to make a preliminary prediction of the processing time of a single process for each order, and at the same time introduce equipment operating status parameters to dynamically correct the prediction results, and obtain the corrected single process time data; S34, based on the corrected single process time data, considering the connection logic between processes and the parameters of the workshop logistics transfer path, calculate the total processing time of the entire process of each order and the delivery time of the designated node.

[0012] Further, S4 includes the following sub-steps: S41, collecting real-time operating parameters of each device through the equipment load balancing optimization network, including the current processing workload, cumulative running time, fault downtime records, and maintenance cycle parameters, and establishing an equipment load status database; S42, based on the equipment load status database, calculating the current load rate and remaining processing capacity of each device, and determining the target range for equipment load balancing optimization; S43, formulating an initial task allocation scheme based on the order delivery period prediction results and equipment load status, and allocating the tasks of each order process to the corresponding devices; S44, iteratively calculating the initial task allocation scheme through the equipment load balancing optimization network, adjusting the task allocation amount of each device, so that the load rate of all devices is within the target range.

[0013] Further, step S5 includes the following sub-steps: S51, inputting the order priority parameters, equipment load parameters, and delivery date parameters from the initial production scheduling scheme into the apparel flexible production scheduling dynamic game model to determine the game participants and their interests; S52, setting the game objective function, clarifying the coordination relationship between order delivery efficiency and equipment load balance, and constructing a multi-participant game model; S53, solving the model through an iterative game algorithm to obtain a set of candidate production scheduling schemes under different game scenarios; S54, selecting the optimal scheme that meets the order delivery requirements and balances the equipment load from the candidate scheme as the final production scheduling optimization result.

[0014] This system is an intelligent scheduling and efficient production optimization system for the entire garment production process. It includes: a multi-dimensional production data acquisition and transmission unit, used to collect garment order process parameters, equipment operating status parameters, workshop resource configuration parameters, and historical production scheduling data through a distributed sensor network, and transmits the encrypted data to the garment production intelligent scheduling cloud platform; a flexible production scheduling dynamic game model construction and solution unit, connected to the multi-dimensional production data acquisition and transmission unit, which initializes game constraints based on the collected data, constructs a dynamic game model, and solves for the equilibrium solution; and an order delivery date prediction scheduling algorithm calculation unit, connected to the multi-dimensional production data acquisition and transmission unit, which extracts order process parameters and optimizes the production schedule. The system uses sequence characteristic parameters and predictive scheduling algorithms to deduce the total processing time and delivery nodes of orders; the equipment load balancing optimization network operation unit is connected to the flexible production scheduling dynamic game model construction and solution unit and the order delivery period predictive scheduling algorithm operation unit to iteratively optimize the equipment task allocation scheme; the full-process production scheduling scheme generation and issuance unit is connected to the equipment load balancing optimization network operation unit, integrates the game optimization results and load balancing results, generates production scheduling instructions and synchronizes them to each production execution unit; the scheduling scheme dynamic monitoring and adjustment unit is connected to the full-process production scheduling scheme generation and issuance unit, monitors the production execution status in real time, and triggers the optimization adjustment mechanism to update the production scheduling scheme when parameter deviations occur.

[0015] Beneficial Effects: This invention proposes an intelligent scheduling and efficient production optimization method and system for the entire garment production process. Utilizing the collaborative operation of a flexible production scheduling dynamic game model, an order delivery date prediction scheduling algorithm, and an equipment load balancing optimization network, a multi-objective integrated optimization mechanism is established. Order delivery requirements and equipment load balancing objectives are incorporated into a unified game framework. Through multi-parameter coupling analysis and dynamic optimization, precise coordination between the two is achieved, completely resolving the constraint conflict problem caused by the single-objective optimization of existing solutions. Simultaneously, by leveraging an intelligent scheduling cloud platform to build a complete closed loop of real-time data acquisition, dynamic simulation, iterative optimization, and instruction issuance, the scheduling logic of deep linkage across all stages of the process is integrated. This enables rapid response to unexpected situations such as order changes and equipment failures. By dynamically adjusting the production scheduling plan, the adaptability and optimization accuracy of the scheduling are improved, overcoming the shortcomings of existing systems that rely on static data and fixed algorithms and lack dynamic adjustment capabilities. Ultimately, this achieves flexible scheduling and efficient production scheduling throughout the entire garment production process, fully adapting to the industry's development needs of smaller order batches, more diverse styles, and tighter delivery cycles. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the overall process of the method of the present invention. Figure 2 This is a diagram showing the system unit composition of the present invention. Detailed Implementation

[0017] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1 As shown, the intelligent scheduling and efficient production optimization method for the entire garment production process includes the following steps: S1, collecting garment order process characteristic parameters, equipment operating status parameters, workshop production resource configuration parameters, and historical production execution data through the garment production intelligent scheduling cloud platform to construct a multi-dimensional production scheduling basic database; S2, based on the basic database, initializing the scheduling constraint boundary using the garment flexible scheduling dynamic game model, and determining the order priority weight allocation rules and equipment processing capacity thresholds; S3, calling the order delivery period prediction scheduling algorithm to dynamically extrapolate the processing time of each order's processing steps, the gap between steps, and the redundancy time of abnormal working conditions; S4, iteratively adjusting the multi-equipment processing task allocation scheme through the equipment load balancing optimization network to form an initial production scheduling scheme; S5, combining the garment flexible scheduling dynamic game model to perform multi-objective game optimization on the initial scheme, coordinating order delivery requirements and equipment load balancing objectives; S6, the garment production intelligent scheduling cloud platform outputs the optimized full-process production scheduling instructions, synchronized to each production execution unit, for real-time distribution and dynamic updating of the scheduling scheme.

[0019] Step S1 involves the comprehensive collection and standardized integration of multi-dimensional production data through a smart cloud platform for garment production scheduling, constructing a foundational database covering the entire production process. Specifically, this is achieved by utilizing a distributed sensor network, equipment data interfaces, and order management system interfaces located throughout the workshop's production areas to simultaneously collect four types of core data: order process characteristic parameters, including over 20 sub-parameters such as fabric type, number of cutting layers, stitch density, number of patch pockets, zipper installation location, and finishing process type; equipment operating status parameters, including 15 key indicators such as real-time speed, operating current, cumulative runtime, current processing task progress, last maintenance time, and fault record codes for each sewing, cutting, and ironing machine; workshop production resource allocation parameters, including 12 types of resource information such as the number of production teams, skill level distribution of each team, number of workbenches, fabric storage capacity, location of accessory supply points, and number of logistics and transfer tools; and historical production execution data, which includes historical records of over 1,000 similar orders from the past 12 months, such as actual processing time, process connection time, equipment load fluctuation data, and order delivery deviations. All collected data undergoes format standardization conversion, abnormal data removal, and data correlation verification by the cloud platform's data preprocessing module. It is then categorized and stored according to order number, equipment number, resource type, and time dimension, forming a structured, multi-dimensional production scheduling basic database. This database provides complete, accurate, and traceable data support for subsequent model calculations and algorithm deductions. It supports a data update speed of over 100 records per second, ensuring that data timeliness meets dynamic scheduling requirements.

[0020] Step S2, based on the multi-dimensional production scheduling database constructed in Step S1, initializes the scheduling constraint boundaries using a dynamic game model for flexible garment production scheduling, clarifying the order priority weight allocation rules and equipment processing capacity thresholds. In the specific implementation process, key influencing factors such as order process complexity, delivery deadline urgency, order amount, and customer cooperation level are first extracted from the database. The analytic hierarchy process (AHP) is used to determine the weight percentage of each factor, with process complexity accounting for 35%, delivery deadline urgency for 30%, order amount for 20%, and customer cooperation level for 15%. A weighted average is then used to obtain a comprehensive priority score for each order, which is used to divide the order into five priority levels, each corresponding to a different weight coefficient range. Simultaneously, parameters such as rated processing speed, maximum continuous operating time, process compatibility type, and maintenance cycle requirements of each piece of equipment are extracted. Combined with historical operating data showing fluctuations in actual equipment processing efficiency, processing capacity thresholds for each piece of equipment for different process types are determined. For example, the processing capacity threshold for a flat sewing machine is set at 300 pieces per hour for thin fabrics and 180 pieces per hour for thick fabrics. The cutting equipment's cutting capacity threshold varies from 200 to 500 layers per hour for fabrics with different layers. A dynamic game theory model for flexible garment production scheduling integrates order priority rules and equipment processing capacity thresholds into scheduling constraint boundaries, clarifying the input range and constraints for model calculations. These constraint boundaries include four main categories: order processing sequence constraints, equipment task allocation upper limit constraints, process connection logic constraints, and resource occupation conflict constraints. Each constraint dimension is further subdivided into 8 to 12 specific constraint clauses, ensuring that subsequent production scheduling plans are optimized while meeting actual production conditions, laying the foundation for the feasibility of the production scheduling plan.

[0021] Step S3 uses an order delivery date prediction and scheduling algorithm to dynamically extrapolate the processing time of each order, the gap between processes, and the redundancy time due to abnormal conditions, providing accurate time references for production scheduling. In practice, the process breakdown information for each order is first extracted from the basic database, clarifying the processing content, technological requirements, and dependencies of each process, which is broken down into 10 to 15 core processes such as cutting, sewing, overlocking, buttoning, ironing, and packaging. For each process, a process time prediction model is constructed using an algorithm, combining the corresponding process characteristic parameters, equipment operating status parameters, and historical processing time data of similar processes. The model input includes eight key parameters such as fabric thickness, processing accuracy requirements, current equipment operating efficiency, and operator skill level. Through learning and training on over 500 historical similar process data points, accurate prediction of process time is achieved, with the prediction error controlled within 5%. Subsequently, based on the logical dependencies between processes, the connection gaps between each process are calculated. The calculation of these gaps comprehensively considers factors such as logistics transfer distance, equipment changeover time, and material preparation time. For example, the connection gap between the cutting and sewing processes is set at 15 to 30 minutes, with the specific value dynamically adjusted according to the workshop layout and material transfer efficiency. Simultaneously, to cope with unforeseen circumstances such as equipment failure and material shortages, algorithms analyze the probability and duration of various abnormal operating conditions in historical production. Reasonable redundancy time for abnormal operating conditions is reserved for each order, accounting for 10% to 15% of the total predicted time for the order. This redundancy time is dynamically adjusted according to the order priority level, with higher priority orders having appropriately increased redundancy time to ensure order delivery stability. Through this dynamic extrapolation process, the predicted processing time for each process in each order, the connection gaps between processes, and the total predicted delivery cycle are finally output, providing accurate data support for subsequent equipment task allocation and production scheduling optimization.

[0022] Step S4 utilizes the equipment load balancing optimization network to iteratively adjust the multi-equipment processing task allocation scheme, forming an initial production scheduling scheme. The core objective is to achieve a balanced distribution of load across all equipment, avoiding overload or idleness. In the specific implementation process, firstly, real-time load status data of all production equipment is extracted from the basic database, including parameters such as currently allocated tasks, remaining processing capacity, continuous running time, and maintenance requirements. An equipment load status matrix is ​​constructed, with the matrix dimension being the number of equipment multiplied by the number of load evaluation indicators. These load evaluation indicators include six core indicators such as current load rate, cumulative operating load, and load fluctuation coefficient. Based on the order process prediction duration data output in Step S3, an initial task allocation scheme is formulated according to order priority and equipment process adaptability principles. The process tasks of each order are assigned to the corresponding equipment, ensuring that the initial task allocation for each piece of equipment does not exceed its processing capacity threshold. Subsequently, the initial task allocation scheme is evaluated for load balance using the equipment load balancing optimization network. The standard deviation of the load rate for each piece of equipment is calculated; a standard deviation greater than 10% indicates an unbalanced load, requiring iterative adjustment. The adjustment process employs a greedy algorithm, migrating some non-core tasks from equipment with a load rate exceeding 85% to idle or low-load equipment with a load rate below 40%. During the migration, factors such as equipment process adaptability, task migration costs, and order delivery deadlines are comprehensively considered to avoid order delivery delays or decreased processing quality due to task migration. After each adjustment, the equipment load balance is recalculated until the standard deviation of the load rate for each piece of equipment is less than 10%, at which point the equipment load is considered balanced. After 3 to 5 rounds of iterative adjustments, an initial production scheduling plan is formed. This plan clearly defines the task allocation list for each piece of equipment, the start and end times of each process, material supply time nodes, and other key information. It also marks the priority and constraints of each task, providing a foundation for subsequent multi-objective game optimization.

[0023] Step S5, combining the dynamic game model of flexible production scheduling in the apparel industry, performs multi-objective game optimization on the initial production scheduling scheme formed in Step S4, coordinating order delivery needs with equipment load balancing objectives to improve the overall optimization effect of the production scheduling scheme. In specific implementation, the two core objectives of game optimization are first clarified: first, to meet order delivery time requirements and minimize the deviation between actual and predicted delivery times; second, to achieve equipment load balancing and minimize the fluctuation range of load rates for each piece of equipment. Order delivery time parameters, equipment load parameters, and process connection parameters from the initial production scheduling scheme are used as input variables for the game model. The game participants are identified as the order management module and the equipment management module. The order management module aims for optimal order delivery, while the equipment management module aims for optimal equipment load balancing. Game rules are set, including objective weight allocation, conflict coordination mechanisms, and iteration termination conditions. The weights of the order delivery objective and the equipment load balancing objective are dynamically adjusted according to the company's production strategy. Under normal circumstances, both have a weight ratio of 50%, but when order delivery pressure is high, the weight of the order delivery objective can be increased to 60%. The game theory model employs multiple iterative calculations. In each iteration, two participating entities propose optimization and adjustment suggestions based on the current plan. The order management module can suggest adjusting the processing sequence and increasing processing resources for high-priority orders, while the equipment management module can suggest adjusting task allocation ratios and optimizing equipment operating parameters. After each iteration, the comprehensive optimization evaluation score of the plan is calculated. The evaluation indicators include five core indicators such as order delivery deviation rate, equipment load balance, and resource utilization rate. When the comprehensive score change rate is less than 3% for three consecutive iterations, it is determined that a game equilibrium has been reached, and the iteration stops. The final output is an optimized production scheduling plan obtained through multi-objective game optimization. This plan satisfies both the time requirements for order delivery and ensures the balanced and stable equipment load, achieving synergistic optimization of the two core objectives.

[0024] Step S6 involves the intelligent scheduling cloud platform for garment production outputting optimized full-process production scheduling instructions, which are synchronized to each production execution unit. This enables real-time distribution and dynamic updates of the scheduling plan, ensuring its effective implementation and dynamic adjustment. Specifically, the optimized scheduling plan from step S5 is first converted into standardized scheduling instructions. These instructions include 12 key pieces of information: equipment number, task number, process name, start time, end time, processing technology requirements, quality inspection standards, and material requirements list. The instruction format uses an industry-standard communication protocol to ensure accurate parsing by each production execution unit. Through the cloud platform's communication module, the scheduling instructions are wirelessly transmitted to the control systems of each production equipment in the workshop, the mobile terminals of operators, the management system of the material supply department, and the terminal equipment of the quality inspection department. Transmission delay is controlled to within one second, ensuring synchronized instruction reception by each unit. During production execution, the cloud platform uses a distributed sensor network to collect real-time data on the operating status of each device, process progress, material consumption, and quality inspection, updating the data every 5 seconds to dynamically monitor the execution of the production scheduling plan. When abnormal situations such as equipment failure, material shortage, or order changes are detected, the cloud platform automatically triggers a dynamic adjustment mechanism. The abnormal data is input into the garment flexible scheduling dynamic game model and the equipment load balancing optimization network to re-evaluate and optimize the solution, generating adjusted scheduling instructions and distributing them to relevant execution units in real time. Simultaneously, the cloud platform stores and visualizes scheduling plan execution data, adjustment records, and optimization effects in real time, providing production managers with a comprehensive production scheduling monitoring interface. This allows managers to easily monitor production progress, equipment status, and order delivery status, achieving transparent and intelligent management of the entire garment production process.

[0025] Preferably, the expression for the dynamic game model of flexible production scheduling in the garment industry is: For flexible production scheduling, dynamic Boyer equilibrium solution. Assign the processing task quantity for the i-th order to the j-th equipment. Let j be the maximum processing load threshold of the j-th equipment. Let be the process compatibility coefficient between the i-th order and the j-th piece of equipment. Let i be the priority weight of the i-th order. Let be the load adjustment coefficient of the j-th equipment. Total number of orders This represents the total number of devices.

[0026] Specifically, the dynamic game theory model for flexible production scheduling in the garment industry is constructed by integrating game theory and production scheduling theory, based on the dynamic interaction between order task allocation and equipment load constraints in garment production. First, the game participants are defined as the order set and the equipment set. The matching degree between the order task allocation and the equipment load threshold is the core game point. A logarithmic function is introduced to represent the nonlinear relationship between task allocation and equipment adaptability, and a trigonometric function is used to quantify the constraint relationship between task allocation and equipment load. The weighted sum of these two factors forms the calculation logic for the game equilibrium solution. The core basis of this formula is that order task allocation in garment production must simultaneously consider the upper limit of equipment processing capacity and process adaptability. The logarithmic term is used to weaken the impact of extreme task allocation on the equilibrium solution, and the trigonometric function is used to reflect the periodic constraint characteristics of equipment load. Regarding parameter values, order priority weights are determined based on factors such as process complexity and delivery urgency, ranging from 0.1 to 0.9. Equipment load adjustment coefficients are set based on equipment type and operating status, ranging from 0.3 to 1.2. The total number of orders and equipment is determined based on the actual production scale of the workshop, typically ranging from 50 to 200 orders and 30 to 80 equipment. During implementation, parameters such as order workload, equipment load thresholds, and process adaptability coefficients are extracted from the basic database and substituted into the model to calculate the game equilibrium solution, thereby determining the optimal task allocation scheme. This model can coordinate the conflicting interests among multiple orders and multiple devices, achieving a dynamic balance between task allocation and equipment load, providing support for the feasibility and optimization of production scheduling schemes.

[0027] Preferably, the expression for the order delivery date prediction and scheduling algorithm is: ,in, For the predicted delivery period of the k-th order, The standard processing time for the I-th process of the k-th order. Let be the process complexity correction value for the i-th process in the k-th order. Let be the equipment processing efficiency coefficient corresponding to the I-th process of the k-th order. Let be the delivery time sensitivity coefficient for the k-th order. The time impact weight for the I-th process is... Let be the abnormal working condition correction coefficient for the k-th order. This represents the total number of processes in a single order.

[0028] Specifically, the order delivery period prediction and scheduling algorithm calculates the multi-factor coupling relationship of order process time consumption. Combining statistical analysis and time series forecasting theory, it constructs a composite calculation logic of multi-parameter product and summation. The order delivery cycle is decomposed into three parts: processing time of each process, process complexity correction amount, and redundancy time under abnormal conditions. The product term is used to represent the nonlinear relationship between standard process time and process complexity and equipment efficiency. The cubic root term quantifies the impact of process time fluctuations on the delivery period. The weighted sum of the two yields the final prediction result. The formula is based on the fact that the order delivery period in garment production is comprehensively affected by multiple factors such as standard process time, process complexity, and equipment efficiency, and that each factor has a nonlinear relationship with the delivery period. The product term can highlight the dominant role of key factors, and the cubic root term can smooth out the prediction deviation caused by extreme fluctuations. Regarding parameter values, the delivery period sensitivity coefficient is set based on order priority, ranging from 0.8 to 1.5. The weight of process time impact is determined through historical data regression analysis, ranging from 0.2 to 0.7. The abnormal working condition correction coefficient is set based on historical failure probability, ranging from 0.1 to 0.4. The total number of processes per order is determined based on the garment type, typically 10 to 15. During implementation, parameters such as the standard duration of each process, process complexity correction value, and equipment efficiency coefficient are extracted and substituted into the algorithm to calculate the predicted delivery cycle of the order. The algorithm's prediction error is controlled within 5%, providing accurate time references for production scheduling and ensuring timely order delivery.

[0029] Preferably, the expression for the equipment load balancing optimization network is: ,in, It serves as an evaluation index for equipment load balance. Let the weight of the processing task of the j-th machine in the i-th order be . Assign the processing task quantity for the i-th order to the j-th equipment. Let j be the rated processing capacity of the j-th equipment. Let be the load balancing weighting coefficient for the j-th device. Let be the load fluctuation suppression coefficient of the j-th equipment. Total number of orders This represents the total number of devices.

[0030] Specifically, the equipment load balancing optimization network is based on multi-dimensional evaluation indicators of equipment load, integrating mean square error analysis and trigonometric function constraint theory to construct a calculation model combining quadratic and sine functions. The ratio of actual equipment load to rated capacity is used as the core variable. A quadratic function characterizes the square relationship of equipment load rate, highlighting the negative impact of overload operation. A sine function reflects the periodic fluctuation characteristics of equipment load. The weighted sum of the two yields the load balance evaluation index. The formula is established based on the fact that in garment production, equipment load balancing requires simultaneous control of overload rate exceeding limits and fluctuation amplitude. The quadratic function can amplify the adverse effects of overload, while the sine function can adapt to the periodic patterns of equipment operation, achieving accurate evaluation of load balance. Regarding parameter values, the load balance weight coefficient is set according to the importance of the equipment, ranging from 0.5 to 1.0; the load fluctuation suppression coefficient is determined in conjunction with equipment operating stability, ranging from 0.2 to 0.6; and the rated processing capacity of the equipment is determined according to the equipment model and process requirements, typically ranging from 100 to 500. During implementation, parameters such as the current workload, rated capacity, and task weight of the equipment are collected and substituted into the network to calculate the load balance index. When the index exceeds the set threshold, task migration and adjustment are triggered. This network can effectively reduce equipment load fluctuations, avoid equipment overload or idleness, and improve the efficiency of production resource utilization.

[0031] Preferably, the scheduling decision optimization expression of the intelligent scheduling cloud platform for garment production is: ,in, To optimize the scheduling scheme, the decision value is determined. For decision weight coefficients, For flexible production scheduling, dynamic Boyer equilibrium solution. The average predicted delivery time for all orders. It serves as an evaluation index for equipment load balance. The maximum value of the equilibrium solution in the game. This represents the maximum value of the average predicted delivery cycle. This represents the maximum value of the equipment load balance.

[0032] Specifically, the scheduling decision optimization expression of the intelligent scheduling cloud platform for garment production is based on multi-objective optimization theory, integrating normalization processing and exponential function constraints to construct a comprehensive decision model encompassing game equilibrium solution, average delivery cycle, and load balance. First, the game equilibrium solution, average delivery cycle, and load balance are normalized to eliminate dimensional differences. A weighted summation combined with an exponential function is used to quantify the impact of each objective on the scheduling decision, where the exponential term mitigates the extreme influence of load balance on the decision. The formula is established based on the fact that the cloud platform's scheduling decision needs to comprehensively consider three objectives: game equilibrium, delivery cycle, and load balance. The weighted summation reflects the priority of each objective, and the exponential term coordinates the constraint conflicts between multiple objectives, achieving a comprehensive optimal decision. Regarding parameter values, the decision weight coefficients are set according to the enterprise's production strategy, ranging from 0.4 to 0.6. The maximum values ​​of the game equilibrium solution, average delivery cycle, and load balance are determined through historical data statistics, ranging from 100 to 200, 50 to 150, and 0.8 to 1.2, respectively. During implementation, parameters such as the game equilibrium solution, average delivery cycle, and load balance are extracted and substituted into the expression to calculate the comprehensive optimization decision value of the scheduling scheme. The optimal scheduling scheme is selected based on the decision value. This expression can integrate multi-objective optimization requirements and improve the scientificity and rationality of scheduling decisions.

[0033] Preferably, the multi-objective coordination expression for intelligent scheduling and efficient production optimization of the entire garment production process is as follows: ,in, To optimize the comprehensive evaluation function for multiple objectives, To coordinate the weights for load balancing and load equilibrium, To optimize weights for delivery period, Let i be the game equilibrium solution for the i-th order. For the load balance of the j-th device, For the predicted delivery period of the k-th order, This is the average delivery time for historical orders. Total number of orders This represents the total number of devices.

[0034] Specifically, the multi-objective coordination expression for intelligent scheduling and efficient production optimization throughout the garment production process revolves around three core objectives: game equilibrium, load balancing, and delivery time optimization. It integrates proportional analysis and reciprocal constraint theory to construct a comprehensive evaluation model combining ratios and reciprocals. The ratio of the sum of game equilibrium solutions to the sum of equipment load balancing degrees represents the coordination relationship between orders and equipment. The reciprocal of the ratio of the average order delivery cycle to the historical average delivery cycle quantifies the delivery time optimization effect. The weighted sum of these two ratios yields the multi-objective optimization comprehensive evaluation function. The formula is based on the fact that full-process production optimization requires coordinating three objectives: game equilibrium, load balancing, and delivery time. The ratio term reflects the level of collaboration between orders and equipment, while the reciprocal term highlights the importance of delivery time optimization, achieving a dynamic balance among multiple objectives. Regarding parameter values, the coordination weights for game equilibrium and load balancing range from 0.4 to 0.7, the delivery time optimization weight ranges from 0.3 to 0.6, and the historical average order delivery cycle is determined through statistical analysis of historical data from the past 12 months, typically ranging from 3 to 10 days. During implementation, parameters such as the game equilibrium solution of each order, the equipment load balance degree, and the order prediction delivery cycle are extracted and substituted into the expression to calculate the comprehensive evaluation function value. Based on this, the optimal production scheduling plan is selected. This expression can coordinate the optimization needs of multiple objectives, improve the comprehensive performance of the production scheduling plan, and adapt to the complex needs of garment production.

[0035] Preferably, step S3 includes the following sub-steps: S31, based on the order process feature parameters collected by the intelligent scheduling cloud platform for garment production, extract the fabric characteristic parameters, cutting accuracy requirements, sewing process complexity parameters, and finishing process parameters required for each process, and establish a process feature vector library; S32, input the parameters in the process feature vector library into the order delivery period prediction scheduling algorithm, and combine it with the process time data of similar historical orders to construct a dynamic prediction model for process time; S33, use this model to make a preliminary prediction of the processing time of a single process for each order, and at the same time introduce equipment operating status parameters to dynamically correct the prediction results, and obtain the corrected single process time data; S34, based on the corrected single process time data, considering the connection logic between processes and the parameters of the workshop logistics transfer path, calculate the total processing time of the entire process and the delivery time of the designated node for each order.

[0036] Specifically, step S3 constructs a complete order delivery time prediction system through four sub-steps to ensure the accuracy and adaptability of the prediction results. In stage S31, the distributed data acquisition capabilities of the garment production intelligent scheduling cloud platform are utilized to comprehensively extract core process characteristic parameters for each order's process. These include eight indicators of fabric characteristics such as fiber composition, thickness, and elasticity; five parameters of cutting accuracy requirements such as the allowable range of cut piece size error and the number of cutting layers; six aspects of sewing process complexity such as stitch type and number of splices; and four parameters of finishing processes such as ironing temperature range and folding standards. A structured process feature vector library is constructed through classification and coding to provide basic data support for subsequent predictions. In stage S32, the standardized parameters from the vector library are input into the order delivery time prediction scheduling algorithm. Simultaneously, over 1000 sets of historical process time data from similar orders within the past 12 months are imported. Through the algorithm's correlation analysis and pattern mining of the data, a dynamic prediction model for process time based on multi-parameter coupling is constructed. During model training, gradient descent is used to optimize internal parameters, ensuring the model's sensitivity to changes in process characteristics. In stage S33, the model is used to make preliminary predictions of the processing time of each order's single process. Simultaneously, six dynamic parameters related to equipment operation status, including speed, load rate, and fault warning information, are collected in real time. A weighted correction algorithm is used to dynamically adjust the preliminary prediction results, with the correction weight set between 0.3 and 0.7 based on equipment operational stability. The final result is single-process time data with an error controlled within 5%. In stage S34, based on the corrected single-process time data, and considering constraints such as sequential dependencies and parallel possibilities in the process connection logic, while incorporating three parameters from the workshop logistics transfer path—distance and transfer tool speed—a path planning algorithm is used to calculate the total processing time for each order's entire process. This clarifies the delivery times of key nodes such as cutting, sewing, and packaging, providing a precise time benchmark for overall production scheduling.

[0037] Preferably, step S4 includes the following sub-steps: S41, collecting real-time operating parameters of each device through the equipment load balancing optimization network, including the current processing workload, cumulative running time, fault downtime records, and maintenance cycle parameters, and establishing an equipment load status database; S42, based on the equipment load status database, calculating the current load rate and remaining processing capacity of each device, and determining the target range for equipment load balancing optimization; S43, formulating an initial task allocation scheme based on the order delivery period prediction results and equipment load status, and allocating the tasks of each order process to the corresponding devices; S44, iteratively calculating the initial task allocation scheme through the equipment load balancing optimization network, adjusting the task allocation amount of each device, so that the load rate of all devices is within the target range.

[0038] Specifically, step S4 achieves precise load balancing optimization of equipment through four sub-steps, avoiding equipment overload or idleness. In stage S41, the sensor terminals of the equipment load balancing optimization network comprehensively collect real-time operating parameters of each production device, including eight core indicators such as current processing workload, cumulative runtime, number and duration of downtime in fault downtime records, and remaining maintenance intervals in maintenance cycles. Simultaneously, it integrates inherent parameters such as rated power and processing accuracy to construct an equipment load status database encompassing both static attributes and dynamic states. This database is updated every 3 seconds to ensure real-time data accuracy. In stage S42, based on this database, the current load rate of each device is calculated using a load rate calculation algorithm: Load rate = Current workload / Rated processing capacity × 100%. Combining this with the load fluctuation range in historical operating data, the target range for equipment load balancing optimization is determined to be 40% to 85%. This range can fluctuate by 5% depending on the equipment type; for example, the target range for high-precision sewing equipment is 50% to 80%. In stage S43, based on the time consumption data of each process in the order delivery forecast results and the equipment process adaptability parameters in the equipment load status database, an initial task allocation plan is formulated according to the order priority order and the principle of maximizing equipment efficiency. The cutting, sewing, and other process tasks of each order are assigned to the corresponding equipment one by one. During the allocation process, it is ensured that the initial task volume of a single piece of equipment does not exceed 90% of its rated processing capacity. In stage S44, the initial task allocation plan is iteratively calculated through the equipment load balancing optimization network. In each iteration, the standard deviation of the load rate of each piece of equipment is calculated. When the standard deviation is greater than 10%, the task migration mechanism is triggered, and non-core process tasks on equipment with a load rate of more than 85% are migrated to equipment with a load rate of less than 40%. During the migration process, factors such as task migration cost and equipment process adaptability are comprehensively considered. The number of iterations is set to 3 to 5 rounds until the standard deviation of the load rate of each piece of equipment is less than 10%, forming a load-balanced equipment task allocation plan.

[0039] Preferably, step S5 includes the following sub-steps: S51, inputting the order priority parameters, equipment load parameters, and delivery date parameters from the initial production scheduling scheme into the apparel flexible production scheduling dynamic game model to determine the game participants and their interests; S52, setting the game objective function, clarifying the coordination relationship between order delivery efficiency and equipment load balance, and constructing a multi-participant game model; S53, solving the model through an iterative game algorithm to obtain a set of candidate production scheduling schemes under different game scenarios; S54, selecting the optimal scheme that meets the order delivery requirements and balances the equipment load from the candidate scheme as the final production scheduling optimization result.

[0040] Specifically, the multi-objective game optimization process in step S5 achieves coordinated optimization of order delivery and equipment load balancing through four sub-steps. In stage S51, core parameters from the initial production scheduling plan, such as order priority parameters, equipment load rate data, and order delivery date requirements, are all input into the garment flexible scheduling dynamic game model. The game participants are clearly defined as the order management module and the equipment management module. The core objective of the order management module is to meet delivery date requirements, while the core objective of the equipment management module is to achieve load balancing. Eight parameters, including order process complexity and equipment maintenance costs, are extracted as game benefit parameters, providing the basic input for the game model construction. In stage S52, the game objective function is set, clarifying the coordination relationship between order delivery efficiency and equipment load balancing. The initial weights of both are set at 50%, which can be dynamically adjusted according to the company's production strategy. For example, when order delivery pressure is high, the weight of order delivery efficiency is increased to 60%. Simultaneously, constraints for the multi-agent game are constructed, including six clauses such as equipment processing capacity constraints and process connection logic constraints, forming a complete multi-agent game model. In stage S53, an iterative game theory algorithm is used to solve the model. In each round of the game, the order management module and the equipment management module propose task adjustment suggestions based on their own needs. For example, the order management module suggests prioritizing resource allocation to high-priority orders, while the equipment management module suggests adjusting task allocation to reduce overloaded equipment. Through alternating optimization strategies, the game equilibrium point is gradually approached. After each iteration, the comprehensive evaluation score of the solution is calculated, and the evaluation indicators include five items such as delivery time fulfillment rate and load balance. In stage S54, when the change rate of the comprehensive evaluation score is less than 3% for three consecutive iterations, the iteration stops and the optimal solution is selected from the candidate set. The selection criteria are a delivery time fulfillment rate of over 98% and a standard deviation of equipment load rate of less than 8%. This optimal solution satisfies the delivery time requirements of each order and ensures that the load of all equipment is balanced, achieving the synergistic optimization of the two core objectives.

[0041] like Figure 2As shown in the figure, the intelligent scheduling and efficient production scheduling optimization system for the entire process of clothing production includes: a multi-dimensional production data collection and transmission unit, which is used to collect clothing order process parameters, equipment operation status parameters, workshop resource configuration parameters, and historical production scheduling data through a distributed sensor network, and transmit them to the intelligent scheduling cloud platform for clothing production through encrypted transmission; a flexible production scheduling dynamic game model construction and solution unit, which is connected to the multi-dimensional production data collection and transmission unit, initializes game constraint conditions based on the collected data, constructs a dynamic game model, and solves the equilibrium solution; an order delivery period prediction scheduling algorithm operation unit, which is connected to the multi-dimensional production data collection and transmission unit, extracts order process feature parameters, and uses the prediction scheduling algorithm to deduce the total processing duration and delivery nodes of the order; an equipment load balance optimization network operation unit, which is respectively connected to the flexible production scheduling dynamic game model construction and solution unit and the order delivery period prediction scheduling algorithm operation unit, and iteratively optimizes the equipment task allocation plan; a full-process production scheduling plan generation and distribution unit, which is connected to the equipment load balance optimization network operation unit, integrates the game optimization results and the load balance results, generates a production scheduling instruction, and synchronizes it to each production execution unit; a scheduling plan dynamic monitoring and adjustment unit, which is connected to the full-process production scheduling plan generation and distribution unit, monitors the production execution status in real time, triggers an optimization adjustment mechanism when parameter deviations occur, and updates the production scheduling plan.

[0042] The intelligent scheduling and efficient production scheduling optimization method and system for the entire process of clothing production, through the organic integration of a flexible production scheduling dynamic game model, an order delivery period prediction scheduling algorithm, and an equipment load balance optimization network, incorporates order delivery-related parameters and core equipment operation parameters into a unified analysis framework, forming a multi-objective collaborative optimization mechanism. Through dynamic coupling analysis and game optimization of key elements such as order priority, process duration, and equipment load, it achieves precise adaptation of order delivery requirements and equipment load balance goals, completely solving the constraint conflict problem caused by existing solutions focusing on only one objective. The scheduling plan not only meets the time requirements of order delivery but also ensures the efficient and stable operation of equipment, improving the overall utilization efficiency of production resources.

[0043] The method and system utilize the intelligent scheduling cloud platform for clothing production to build a complete operation closed-loop of real-time data collection, dynamic deduction, iterative optimization, and instruction distribution, integrating the linkage logic of all links in the entire production process. By continuously collecting dynamic data such as order changes and equipment status fluctuations, it quickly feeds back to the scheduling model and algorithm for real-time deduction, timely adjusts the production scheduling plan, and can efficiently respond to various sudden working conditions. At the same time, the deep linkage design of all links in the entire process breaks the problem of disconnection between links in traditional scheduling, greatly improving the adaptability and optimization accuracy of the scheduling plan, and comprehensively adapting to the current industry development needs of small batch sizes, diverse styles, and short delivery cycles for clothing production orders.

[0044] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0045] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent scheduling and efficient production optimization of the entire garment production process, characterized in that: Includes the following steps: S1. Collect garment order process characteristic parameters, equipment operating status parameters, workshop production resource configuration parameters, and historical production scheduling execution data through the garment production intelligent scheduling cloud platform to construct a multi-dimensional production scheduling basic database. S2. Based on the basic database, initialize the scheduling constraint boundary using a garment flexible scheduling dynamic game model to determine order priority weight allocation rules and equipment processing capacity thresholds. S3. Call the order delivery date prediction scheduling algorithm to dynamically extrapolate the processing time of each order's processing steps, process connection gaps, and abnormal working condition redundancy time. S4. Iteratively adjust the multi-equipment processing task allocation scheme through an equipment load balancing optimization network to form an initial production scheduling scheme. S5. Combine the garment flexible scheduling dynamic game model to perform multi-objective game optimization on the initial scheme, coordinating order delivery requirements with equipment load balancing objectives. S6. The garment production intelligent scheduling cloud platform outputs optimized full-process production scheduling instructions, synchronized to each production execution unit, for real-time distribution and dynamic updating of the scheduling scheme.

2. The intelligent scheduling and efficient production optimization method for the entire garment production process according to claim 1, characterized in that, The expression for the dynamic game model of flexible production scheduling in the garment industry is as follows: For flexible production scheduling, dynamic Boyer equilibrium solution. Assign the processing task quantity for the i-th order to the j-th equipment. Let j be the maximum processing load threshold of the j-th equipment. Let be the process compatibility coefficient between the i-th order and the j-th piece of equipment. Let i be the priority weight of the i-th order. Let be the load adjustment coefficient of the j-th equipment. Total number of orders This represents the total number of devices.

3. The intelligent scheduling and efficient production optimization method for the entire garment production process according to claim 1, characterized in that, The expression for the order delivery date prediction and scheduling algorithm is as follows: ,in, For the predicted delivery period of the k-th order, The standard processing time for the I-th process of the k-th order. Let be the process complexity correction value for the i-th process in the k-th order. Let be the equipment processing efficiency coefficient corresponding to the I-th process of the k-th order. Let be the delivery time sensitivity coefficient for the k-th order. The time impact weight for the I-th process is... Let be the abnormal working condition correction coefficient for the k-th order. This represents the total number of processes in a single order.

4. The intelligent scheduling and efficient production optimization method for the entire garment production process according to claim 1, characterized in that, The expression for the equipment load balancing optimization network is: ,in, It serves as an evaluation index for equipment load balance. Let the weight of the processing task of the j-th machine in the i-th order be . Assign the processing task quantity for the i-th order to the j-th equipment. Let j be the rated processing capacity of the j-th equipment. Let be the load balancing weighting coefficient for the j-th device. Let be the load fluctuation suppression coefficient of the j-th equipment. Total number of orders This represents the total number of devices.

5. The intelligent scheduling and efficient production optimization method for the entire garment production process according to claim 1, characterized in that, The optimization expression for the scheduling decision of the intelligent scheduling cloud platform for garment production is as follows: ,in, To optimize the scheduling scheme, the decision value is determined. For decision weight coefficients, For flexible production scheduling, dynamic Boyer equilibrium solution. The average predicted delivery time for all orders. It serves as an evaluation index for equipment load balance. The maximum value of the equilibrium solution in the game. This represents the maximum value of the average predicted delivery cycle. This represents the maximum value of the equipment load balance.

6. The intelligent scheduling and efficient production optimization method for the entire garment production process according to claim 1, characterized in that, The multi-objective coordination expression for intelligent scheduling and efficient production optimization of the entire garment production process is as follows: ,in, To optimize the comprehensive evaluation function for multiple objectives, To coordinate the weights for load balancing and load equilibrium, To optimize weights for delivery period, Let i be the game equilibrium solution for the i-th order. For the load balance of the j-th device, For the predicted delivery period of the k-th order, This is the average delivery time for historical orders. Total number of orders This represents the total number of devices.

7. The intelligent scheduling and efficient production optimization method for the entire garment production process according to claim 1, characterized in that, S3 includes the following steps: S31, based on the order process feature parameters collected by the intelligent scheduling cloud platform for garment production, extract the fabric characteristic parameters, cutting accuracy requirements, sewing process complexity parameters, and finishing process parameters required for each process, and establish a process feature vector library; S32, input the parameters in the process feature vector library into the order delivery period prediction scheduling algorithm, and combine it with the process time data of similar historical orders to construct a dynamic prediction model for process time; S33, use this model to make a preliminary prediction of the processing time of a single process for each order, and at the same time introduce equipment operating status parameters to dynamically correct the prediction results, and obtain the corrected single process time data; S34, based on the corrected single process time data, considering the connection logic between processes and the workshop logistics transfer path parameters, calculate the total processing time of each order and the delivery time of the designated node.

8. The intelligent scheduling and efficient production optimization method for the entire garment production process according to claim 1, characterized in that, S4 includes the following steps: S41, collecting real-time operating parameters of each device through the device load balancing optimization network, including the current processing workload, cumulative running time, fault downtime records and maintenance cycle parameters, and establishing a device load status database; S42, based on the device load status database, calculating the current load rate and remaining processing capacity of each device, and determining the target range for device load balancing optimization. S43. Based on the order delivery period forecast and equipment load status, formulate an initial task allocation plan and allocate the tasks of each order process to the corresponding equipment. S44. Iteratively calculate the initial task allocation plan through the equipment load balancing optimization network, adjust the task allocation of each equipment, and ensure that the load rate of all equipment is within the target range.

9. The intelligent scheduling and efficient production optimization method for the entire garment production process according to claim 1, characterized in that, S5 includes the following steps: S51, inputting the order priority parameters, equipment load parameters, and delivery date parameters from the initial production scheduling scheme into the apparel flexible production scheduling dynamic game model to determine the game participants and their interests; S52, setting the game objective function, clarifying the coordination relationship between order delivery efficiency and equipment load balance, and constructing a multi-participant game model; S53, solving the model through an iterative game algorithm to obtain a set of candidate production scheduling schemes under different game scenarios; S54, selecting the optimal scheme that meets the order delivery requirements and balances the equipment load from the candidate scheme as the final production scheduling optimization result.

10. A smart scheduling and efficient production optimization system for the entire garment production process, characterized in that: This system, applied to the intelligent scheduling and efficient production optimization method for the entire garment production process as described in claim 1, includes: a multi-dimensional production data acquisition and transmission unit, used to collect garment order process parameters, equipment operating status parameters, workshop resource configuration parameters, and historical production scheduling data through a distributed sensor network, and transmit them to the garment production intelligent scheduling cloud platform after encryption; a flexible production scheduling dynamic game model construction and solution unit, connected to the multi-dimensional production data acquisition and transmission unit, which initializes game constraints based on the collected data, constructs a dynamic game model, and solves for the equilibrium solution; and an order delivery date prediction scheduling algorithm calculation unit, connected to the multi-dimensional production data acquisition and transmission unit, which extracts order process feature parameters and calculates the algorithm. The predictive scheduling algorithm is used to deduce the total processing time and delivery node of orders; the equipment load balancing optimization network operation unit is connected to the flexible production scheduling dynamic game model construction and solution unit and the order delivery period predictive scheduling algorithm operation unit to iteratively optimize the equipment task allocation scheme; the full-process production scheduling scheme generation and issuance unit is connected to the equipment load balancing optimization network operation unit, integrates the game optimization results and load balancing results, generates production scheduling instructions and synchronizes them to each production execution unit; the scheduling scheme dynamic monitoring and adjustment unit is connected to the full-process production scheduling scheme generation and issuance unit to monitor the production execution status in real time, and triggers the optimization adjustment mechanism when parameter deviation occurs to update the production scheduling scheme.

Images