Textile order allocation method and system

Abstract

The invention provides a textile order allocation method and system, and relates to the technical field of data processing, and the method comprises the steps: obtaining textile order information; determining the emergency degree and the operation complexity of each order according to the textile order information; obtaining the production capacity value of each textile mill; matching the operation complexity of the orders with the production capacity values of the textile mills, and determining the production efficiency when each textile mill carries out production on each order; determining the completion time of each order and the carbon emission of each textile mill according to the production efficiency of each textile mill when each order is produced; according to the urgency degree and the completion time of each order, determining the order completion timeliness; according to the carbon emission of each textile mill, the carbon emission standard exceeding degree of the textile mill is determined; the method comprises the following steps of: constructing a textile order allocation agent based on deep reinforcement learning by taking improvement of order completion timeliness and reduction of carbon emission standard exceeding degree of a textile mill as targets; and textile order distribution is carried out through the intelligent agent.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a textile order allocation method and system. Background Art

[0002] The textile industry is one of the main sources of global carbon emissions. Especially in the production processes such as dyeing, spinning and weaving, large amounts of energy consumption (such as electricity and heat energy) lead to high carbon emissions.

[0003] As people's living standards improve, the demand for diversified textiles is also increasing. Faced with the increasing number of textile orders, manual allocation methods that rely on experience are gradually becoming difficult to adapt. More and more companies are adopting more advanced ERP systems to allocate orders based on pre-set allocation rules, such as giving priority to orders with the most urgent delivery dates, and allocating orders to factories or production lines with the lightest current production loads.

[0004] Although the ERP system can automatically allocate orders according to pre-set allocation rules, this allocation method is difficult to handle multi-objective optimization problems and cannot take into account the important factor of carbon emissions. At the same time, fixed allocation rules lack flexible response mechanisms when faced with fluctuations in order volume, production equipment failures or other unpredictable events, and cannot achieve optimal production efficiency. Summary of the invention

[0005] In order to solve the technical problems that current technology is difficult to handle multi-objective optimization problems and cannot take into account the important factor of carbon emissions, and that fixed allocation rules lack flexible response mechanisms when facing fluctuations in order volume, production equipment failures or other unpredictable events, and cannot achieve optimal production efficiency, the present invention provides a textile order allocation method and system.

[0006] The technical solution provided by the embodiment of the present invention is as follows: First aspect: An embodiment of the present invention provides a textile order allocation method, comprising: S1: Get textile order information; S2: Determine the urgency and operation complexity of each order according to the textile order information; S3: Get the production capacity value of each textile factory; S4: Match the operation complexity of the order with the production capacity value of the textile factory to determine the production efficiency of each textile factory when producing each order; S5: Determine the completion time of each order and the carbon emissions of each textile mill based on the production efficiency of each textile mill when producing each order; S6: Determine the timeliness of order completion based on the urgency and completion time of each order; S7: Determine the carbon emissions exceeding the scale of the textile mills based on the carbon emissions of each textile mill; S8: With the goal of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories, a textile order allocation agent based on deep reinforcement learning is constructed; S9: Allocate textile orders through the textile order allocation agent.

[0007] Second aspect: An embodiment of the present invention provides a textile order distribution system, comprising: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the textile order allocation method as described in the first aspect is implemented.

[0008] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: In the present invention, with the goal of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories, a textile order allocation agent based on deep reinforcement learning is constructed, and then textile orders are allocated by the textile order allocation agent. The two optimization goals of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories can be taken into account at the same time. The dynamic environmental state can be learned during the order allocation process, and the fluctuation of order volume, production equipment failure or other unpredictable events can be better responded to, so that the production efficiency reaches the optimal state. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of a flow chart of a textile order allocation method provided by an embodiment of the present invention; Figure 2 A structural schematic diagram of a textile order distribution system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0011] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0012] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0013] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0014] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0015] Reference Manual Attached Figure 1 , showing a flow chart of a textile order allocation method provided in an embodiment of the present invention.

[0016] The embodiment of the present invention provides a textile order allocation method, which can be implemented by a textile order allocation device, and the textile order allocation device can be a terminal or a server. The processing flow of the textile order allocation method can include the following steps: S1: Get textile order information.

[0017] Optionally, the textile order information includes: product details, product quantity, customer level, order channel, total amount and agreed delivery period, etc.

[0018] It should be noted that different textile orders can be completed in different textile mills. At the same time, different textile mills have different equipment, technical levels and production capacities. Some textile mills may have higher production efficiency and technical advantages in certain specific processes. For example, weaving factories specialize in spinning, weaving and other processes, and dyeing factories specialize in dyeing and other processes. Different processes in the same textile order can also be completed in different textile mills.

[0019] Furthermore, different textile mills take different time and carbon emissions to process the same order. The same textile mill also takes different time and carbon emissions to process different processes. Therefore, textile orders need to be allocated intelligently.

[0020] S2: Determine the urgency and operational complexity of each order based on the textile order information.

[0021] In a possible implementation, S2 specifically includes sub-steps S201 to S203: S201: Determine the urgency of each order according to the customer level, order channel, total amount and agreed delivery period in the textile order information.

[0022] Optionally, the urgency is specifically:

[0023] in, a i Indicates i The urgency of the order, l i Indicates i Customer level for each order, q i Indicates i The ordering channel for each order, m i Indicates i The total amount of the order, D i Indicates i The agreed delivery period for each order, The weight parameter representing the customer level, Indicates the weight parameter of the order channel. The weight parameter representing the total amount, The weight parameter representing the agreed delivery period.

[0024] Among them, those skilled in the art can set the weight parameters of the customer level according to the actual situation. l 1. Weight parameters of order channels l 2. Weight parameter of total amount l 3 and weight parameters of agreed delivery period l The size of 4 is not limited in the present invention.

[0025] It should be noted that higher-level customers usually have higher priorities. To ensure customer satisfaction, orders from high-level customers should be processed first. Some channels (such as online channels or dedicated channels for large customers) may be more important, and the processing priority of orders will be different. Orders with higher amounts usually mean higher profits, and companies tend to give priority to completing high-value orders. The closer the delivery date is and the more urgent the order is, the system will automatically increase the priority of the order based on the delivery date.

[0026] In the present invention, multiple factors such as customer level, order channel, order amount and delivery deadline are comprehensively considered to determine the urgency of the order, and resources can be allocated preferentially to those orders with high urgency, thereby improving the accuracy and efficiency of order allocation.

[0027] S202: Determine the required processes and the total number of processes of the order according to the product details in the textile order information.

[0028] It should be noted that for each product, the system stores specific processes and the total number of processes.

[0029] S203: Determine the operation complexity of each order based on the required processes and the total number of processes of the order.

[0030] Optionally, the operation complexity is specifically:

[0031] in, b i Indicates i The operational complexity of an order, J i Indicates i The total number of operations for an order, w j Indicates j The difficulty score of each process, The weight parameter representing the total number of processes, A weight parameter representing the difficulty of the process.

[0032] Among them, those skilled in the art can set the weight parameter of the total number of processes according to actual conditions. Weight parameters related to process difficulty The present invention does not limit the size.

[0033] In the present invention, not only the total number of processes required for an order is taken into account, but also the difficulty score of each process, that is, the complexity of each process in actual operation, which can comprehensively evaluate the complexity of the order and more accurately reflect the operational complexity of the order.

[0034] S3: Get the production capacity value of each textile factory.

[0035] It should be noted that the production capacity value is related to many factors such as the factory's equipment, technical level, human resources, management process, etc. The production capacity value of each textile factory can be stored in the system after evaluation and wait for call.

[0036] In the present invention, the production capacity value is an important indicator for evaluating the factory production efficiency and resource utilization in the textile industry, which affects order allocation, production scheduling and overall operational efficiency. By accurately calculating and analyzing the production capacity value, enterprises can reasonably arrange orders, optimize production processes, and improve resource utilization, thereby improving overall production efficiency and market competitiveness.

[0037] S4: Match the operational complexity of the order with the production capacity value of the textile factory to determine the production efficiency of each textile factory when producing each order.

[0038] In a possible implementation, S4 specifically includes: S401: Divide the operation complexity into multiple levels, and determine the operation complexity level of each order.

[0039] S402: Divide the production capacity value into multiple levels and determine the production capacity level of each textile factory.

[0040] S403: Determine the production efficiency of each textile factory when producing each order according to the operation complexity level of each order and the production capacity level of each textile factory:

[0041] in, E ki Indicates k The textile mill completed i The production efficiency of each order, l d Indicates the operational complexity level of the order, l f Indicates the production capacity level of a textile factory. Indicates the production efficiency decay rate.

[0042] It should be noted that when the factory's production capacity level is higher than or equal to the order's operational complexity level, the factory can complete production with 100% efficiency, which means that resources can be fully utilized without inefficiency due to mismatch. When the factory's capacity is insufficient, production efficiency will decline, reflecting the mismatch between capacity and order complexity.

[0043] In the present invention, the order complexity level is matched with the production capacity level of the textile factory, which can ensure the rationality of order allocation. If the order complexity is high and the factory's production capacity is insufficient, it may cause production delays or quality problems; conversely, if a simple order is allocated to a high-capacity factory, resources will be wasted. In this way, order requirements and factory capacity can be effectively matched, ensuring the rational use of resources and improving overall production efficiency.

[0044] S5: Determine the completion time of each order and the carbon emissions of each textile mill based on the production efficiency of each textile mill when producing each order.

[0045] In a possible implementation, the order completion time is specifically:

[0046] in, F i Indicates i The completion time of an order, C ij Indicates i Order No. j The completion time of each process, X ijk Indicates i Order No. j The process is assigned to k The distribution of textile factories is carried out i Order No. j The process is assigned to k When the textile factory executes , when i Order No. j Processes not assigned to k When the textile factory executes , Tr k Indicates k The time it takes for a textile factory to transport products to the warehouse, K Represents the total number of textile mills.

[0047]

[0048] in, C ij Indicates i Order No. j The completion time of each process, B ij Indicates i Order No. j The start time of each process, T ij Indicates i Order No. j The processing time of each process.

[0049]

[0050] in, W ij Indicates i Order No. j The workload of each process, PC kj Indicates k The textile mill completed j The daily productivity of each process, E ki Indicates k The textile mill completed iThe production efficiency of each order.

[0051] In the present invention, the process completion time of each order is calculated step by step. C ij and the time it takes to transport the product to the warehouse Tr k Ability to obtain more accurate order completion time.

[0052] In a possible implementation, the carbon emissions of a textile factory are specifically:

[0053] in, CT k Table Shidi k The carbon emissions of a textile factory, CE k Indicates k The carbon emissions of a textile factory Indicates the carbon emissions per unit transport distance, Td k Indicates k The total distance that a textile mill transports orders to other locations.

[0054]

[0055] in, W ij Indicates i Order No. j The workload of each process, PP kj Indicates k The textile mill completed j Carbon emissions per process, X ijk Indicates i Order No. j The process is assigned to k Distribution of textile mills, PC kj Indicates k The textile mill completed j The daily productivity of each process, E ki Indicates k The textile mill completed i The production efficiency of each order, I Indicates the total number of orders, J Represents the total number of processes, E ki Indicates k The textile mill completed iThe production efficiency of each order.

[0056]

[0057] in, FW k Indicates k The distance from the textile factory to the warehouse, DTS ij Indicates i Order No. j The first process and the j +1 transportation distance between operations.

[0058] It should be noted that after each textile factory completes the final processing of the finished product, it often needs to transport the finished product to a unified central warehouse for easy management and delivery.

[0059] In the present invention, not only the carbon emissions of the textile factory during the production process are considered, but also the carbon emissions generated during the transportation process are considered to accurately determine the carbon emissions of the textile factory during the entire order processing process.

[0060] S6: Determine the timeliness of order completion based on the urgency and completion time of each order.

[0061] In a possible implementation, the timeliness of order completion is specifically:

[0062] in, G Indicates the timeliness of order completion. D i Indicates i The agreed delivery period for each order, F i Indicates i The completion time of an order, a i Indicates i The urgency of the order, I Indicates the total number of orders.

[0063] It should be noted that by introducing urgency, the priority of different orders can be better reflected, ensuring that urgent orders can receive sufficient attention and resource allocation. If orders with higher urgency can be completed on time or ahead of schedule, the timeliness index will be significantly improved, encouraging companies to prioritize the timely completion of urgent orders when arranging production, optimize resource utilization, and ensure that customer needs are met.

[0064] In the present invention, the formula combines the urgency of each order, the actual completion time of the order and the agreed delivery deadline, which can comprehensively evaluate the timeliness of order completion and clearly quantify the timeliness of order completion.

[0065] S7: Based on the carbon emissions of each textile mill, determine the extent to which the carbon emissions of the textile mill exceed the standard.

[0066] In one possible implementation, the carbon emissions of the textile factory exceed the standard by:

[0067] in, H Indicates that carbon emissions exceed the standard. CT k Indicates k The carbon emissions of a textile factory, EN k Indicates k The production carbon emission limit of a textile factory, CL represents the total carbon emission limit of all textile factories. K Represents the total number of textile mills.

[0068] In the present invention, on the one hand, it is evaluated whether the carbon emissions of each textile mill exceed their respective emission limits, and on the other hand, it is evaluated whether the total carbon emissions of all factories exceed the total carbon emission limit of the enterprise as a whole. This can help the enterprise comprehensively monitor and manage the carbon emissions of each textile mill, ensure that carbon emissions comply with regulations and corporate goals, optimize production scheduling, and promote the enterprise's transformation to low-carbon production.

[0069] S8: Build a textile order allocation agent based on deep reinforcement learning with the goal of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories.

[0070] Among them, Deep Reinforcement Learning (DRL) is a technology that combines reinforcement learning (RL) and deep learning (DL), and uses deep neural networks to enhance the ability of intelligent agents to learn strategies in complex environments. It is suitable for processing high-dimensional input data and complex decision-making problems.

[0071] In a possible implementation, S8 specifically includes: S801: Constructing the environment space of textile order allocation agent:

[0072] in, s Indicates the environment, s 1 represents the first environmental parameter, D Represents textile order data, s 2 represents the second environmental parameter, G Indicates the timeliness of order completion.s 3 represents the third environmental parameter, H Indicates that carbon emissions exceed the scale.

[0073] S802: Constructing the action space of the textile order allocation agent:

[0074] in, a Indicates action, X ijk Indicates i Order No. j The process is assigned to k distribution implemented by each textile mill.

[0075] S803: With the goal of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories, a reward function for the textile order allocation agent is constructed:

[0076] in, r represents the reward function, r 1 means the reward coefficient for completing the order in time, r 2 represents the penalty coefficient for exceeding carbon emissions standards.

[0077] Among them, technicians in this field can set the reward coefficient for completing orders in a timely manner according to actual conditions. r 1 and penalty coefficient for exceeding carbon emission standards r The size of 2 is not limited in the present invention.

[0078] In the present invention, the reward function takes into account two optimization objectives at the same time: the timeliness of order completion and the carbon emission excess. By combining the timeliness of order completion and the carbon emission excess in the same reward function, the agent can improve the ability to complete orders on time and reduce the carbon emission excess during the order allocation process.

[0079] S804: Setting constraints for the textile order allocation agent, the constraints include: process start constraint, process sequence constraint and processing times uniqueness constraint.

[0080] Optionally, the process start constraint is specifically:

[0081] in, A ij Indicates i Order No. j The time it takes for the textiles to arrive at the textile mill. B ij Indicates iOrder No. j The start time of a process.

[0082]

[0083] in, Indicates i Order No. j -Completion time of one process, Indicates i Order No. j -1 process and j The transportation time between the processes.

[0084] It should be noted that the process start constraint ensures that each process can only start after its predecessor process is completed and the related products arrive at the textile factory. It ensures the rationality of the production process and prevents unreasonable simultaneous processing or early processing.

[0085] Optionally, the process sequence constraint is specifically:

[0086] in, B ij Indicates i Order No. j The start time of each process, Indicates i Order No. j +1 process start time.

[0087] It should be noted that the process sequence constraint ensures that the processes within an order are completed in sequence, avoiding logical confusion in the production process and ensuring a smooth production process.

[0088] Optionally, the uniqueness constraint of the number of processing times is specifically:

[0089] in, X ijk Indicates i Order No. j The process is in k The implementation status of each textile factory, K Represents the total number of textile mills.

[0090] It should be noted that the uniqueness constraint of the number of processing times ensures that each process will only be assigned to one textile factory for processing, avoiding the situation where the same process is executed repeatedly or processed across multiple factories, thereby preventing unreasonable duplication of operations and waste of resources.

[0091] S9: Allocate textile orders through textile order allocation agent.

[0092] In a possible implementation, S9 specifically includes sub-steps S901 to S907: S901: In the online network, use Strategy in the environment s t Next select an action a t , s t express t The environment of the moment, a t express t The action selected at the moment.

[0093] in, Strategy is a basic and commonly used strategy selection method in reinforcement learning. The probability of choosing the currently known optimal action (i.e., utilizing) is The probability of randomly choosing an action (i.e. exploration) is .

[0094] S902: Execute action a t , get reward r t , r t express t Rewards for every moment.

[0095] S903: When this As t The parameters at each moment are stored in the experience pool. s t+1 express t +1 moment for the environment.

[0096] S904: Calculate various allocation strategies value:

[0097] in, express value, The value represents the expected long-term reward of all actions, represents the allocation strategy, Indicates the allocation strategy of value, E represents the expectation operation, r t express t Rewards for every moment, r t+1 expresst +1 moment reward, r t+2 express t +2 momentary rewards, s t express t The environment of the moment, a t express t The action selected at the moment, Represents the discount factor.

[0098] S905: Determine the optimal allocation strategy value.

[0099] S906: Copy the online network parameters to the target network and use the deep neural network to optimize the optimal network parameters at each iteration. The value is fitted to obtain the optimal Value function.

[0100] S907: Maximizing the Optimum The value function is used as the target to allocate textile orders:

[0101] in, a t express t The action selected at the moment, Indicates the best Value function.

[0102] In the present invention, the agent can The optimal strategy learned by the value function automatically allocates textile orders, and makes reasonable allocations based on various factors (such as production capacity, order urgency, carbon emission restrictions, etc.) to improve production efficiency and resource utilization. Moreover, as the agent learns more experience, the accuracy of allocation will continue to improve. The dynamic environment state can be learned during the order allocation process to better cope with fluctuations in order volume, production equipment failures or other unpredictable events, so that production efficiency can reach the optimal state.

[0103] In a possible implementation, the training method of the textile order allocation agent is specifically as follows: Randomly draw small batches of samples from the experience pool.

[0104] It should be noted that by randomly sampling small batches of samples from the experience pool for training, the agent can avoid data correlation between consecutive time steps. This random sampling method breaks the time dependency of samples, helps reduce deviations and oscillations in training, and improves the stability of learning.

[0105] Construct the mean square error loss function:

[0106] in, L represents the mean square error loss function, express t The target network parameters at each moment, express t Always-on network parameters, D represents the experience pool, M Represents the number of samples drawn from the experience pool.

[0107] It should be noted that the mean square error loss function helps to measure the current prediction value of the agent ( By optimizing this loss function, the agent can gradually narrow the gap between the predicted value and the actual value and improve the accuracy of the strategy.

[0108] Update the target network parameters using gradient descent:

[0109] in, express t +1 moment target network parameters, express t The learning rate at each moment, Represents a gradient operation.

[0110] It should be noted that by updating the target network parameters through the gradient descent method, the agent can gradually approach the optimal solution in an efficient way. The gradient descent algorithm guides the update direction of the model parameters by calculating the derivative (gradient) of the loss function with respect to the model parameters, thereby reducing the prediction error in each iteration.

[0111] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: In the present invention, with the goal of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories, a textile order allocation agent based on deep reinforcement learning is constructed, and then textile orders are allocated through the textile order allocation agent. The two optimization goals of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories can be taken into account at the same time. The dynamic environmental state can be learned during the order allocation process to better cope with fluctuations in order volume, production equipment failures or other unpredictable events, so that production efficiency reaches the optimal state.

[0112] Reference Manual Attached Figure 2 , showing a structural schematic diagram of a textile order distribution system provided by the present invention.

[0113] The present invention also provides a textile order allocation system 20, comprising: Processor 201; The memory 202 stores computer-readable instructions, and when the computer-readable instructions are executed by the processor 201, the textile order allocation method described in the method embodiment is implemented.

[0114] The textile order allocation system 20 provided by the present invention can execute the above-mentioned textile order allocation method and achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0115] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: In the present invention, with the goal of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories, a textile order allocation agent based on deep reinforcement learning is constructed, and then textile orders are allocated by the textile order allocation agent. The two optimization goals of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories can be taken into account at the same time. The dynamic environmental state can be learned during the order allocation process, and the fluctuation of order volume, production equipment failure or other unpredictable events can be better responded to, so that the production efficiency reaches the optimal state.

[0116] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0117] There are a few points to note: (1) The drawings of the embodiments of the present invention only involve structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0118] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of the film or region is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a film, a region or a substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0119] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0120] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A textile order allocation method, characterized in that: include: S1: Get textile order information; S2: Determine the urgency and operation complexity of each order according to the textile order information; S3: Get the production capacity value of each textile factory; S4: Match the operation complexity of the order with the production capacity value of the textile factory to determine the production efficiency of each textile factory when producing each order; S5: Determine the completion time of each order and the carbon emissions of each textile mill based on the production efficiency of each textile mill when producing each order; S6: Determine the timeliness of order completion based on the urgency and completion time of each order; S7: Determine the carbon emissions exceeding the scale of the textile mills based on the carbon emissions of each textile mill; S8: With the goal of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories, a textile order allocation agent based on deep reinforcement learning is constructed; S9: Allocate textile orders through the textile order allocation agent.

2. The textile order allocation method according to claim 1, characterized in that: The S2 specifically includes: S201: Determine the urgency of each order according to the customer level, order channel, total amount and agreed delivery period in the textile order information; S202: Determine the required processes and the total number of processes of the order according to the product details in the textile order information; S203: Determine the operation complexity of each order based on the required processes and the total number of processes of the order.

3. The textile order allocation method according to claim 1, characterized in that: The S4 specifically includes: S401: Divide the operation complexity into multiple levels, and determine the operation complexity level of each order; S402: Divide the production capacity value into multiple levels, and determine the production capacity level of each textile factory; S403: Determine the production efficiency of each textile factory when producing each order according to the operation complexity level of each order and the production capacity level of each textile factory: in, E ki Indicates k The textile mill completed i The production efficiency of each order, l d Indicates the operational complexity level of the order, l f Indicates the production capacity level of a textile factory. Indicates the production efficiency decay rate.

4. The textile order allocation method according to claim 1, characterized in that: The order completion time is: in, F i Indicates i The completion time of an order, C ij Indicates i Order No. j The completion time of each process, X ijk Indicates i Order No. j The process is assigned to k The distribution of textile factories is carried out i Order No. j The process is assigned to k When the textile factory executes , when i Order No. j Processes not assigned to k When the textile factory executes , Tr k Indicates k The time it takes for a textile factory to transport products to the warehouse, K represents the total number of textile mills; in, C ij Indicates i Order No. j The completion time of each process, B ij Indicates i Order No. j The start time of each process, T ij Indicates i Order No. j The processing time of each process; in, W ij Indicates i Order No. j The workload of each process, PC kj Indicates k The textile mill completed j The daily productivity of each process, E ki Indicates k The textile mill completed i The production efficiency of each order.

5. The textile order allocation method according to claim 1, characterized in that: The carbon emissions of the textile mill are as follows: in, CT k surface Shidi k The carbon emissions of a textile factory, CE k Indicates k The carbon emissions of a textile factory Indicates the carbon emissions per unit transport distance, Td k Indicates k The total distance that a textile mill transports orders to other locations; in, W ij Indicates i Order No. j The workload of each process, PP kj Indicates k The textile mill completed j Carbon emissions per process, X ijk Indicates i Order No. j The process is assigned to k Distribution of textile mills, PC kj Indicates k The textile mill completed j The daily productivity of each process, E ki Indicates k The textile mill completed i The production efficiency of each order, I Indicates the total number of orders, J Represents the total number of processes, E ki Indicates k The textile mill completed i Production efficiency of each order; in, FW k Indicates k The distance from the textile factory to the warehouse, DTS ij Indicates i Order No. j The first process and the j +1 transportation distance between operations.

6. The textile order allocation method according to claim 1, characterized in that: The timeliness of order completion is specifically as follows: in, G Indicates the timeliness of order completion. D i Indicates i The agreed delivery period for each order, F i Indicates i The completion time of an order, a i Indicates i The urgency of the order, I Indicates the total number of orders.

7. The textile order allocation method according to claim 1, characterized in that: The carbon emissions of the textile factory exceeded the standard as follows: in, H Indicates that carbon emissions exceed the standard. CT k Indicates k The carbon emissions of a textile factory, EL k Indicates k The production carbon emission limit of a textile factory, CL represents the total carbon emission limit of all textile factories. K Represents the total number of textile mills.

8. The textile order allocation method according to claim 1, characterized in that: The S8 specifically includes: S801: Constructing the environment space of the textile order allocation agent: in, s Indicates the environment, s 1 represents the first environmental parameter, D Represents textile order data, s 2 represents the second environmental parameter, G Indicates the timeliness of order completion. s 3 represents the third environmental parameter, H It means that carbon emissions are exceeding the scale; S802: Constructing the action space of the textile order allocation agent: in, a Indicates action, X ijk Indicates i Order No. j The process is assigned to k distribution implemented by each textile mill; S803: With the goal of improving the timeliness of order completion and reducing the excessive carbon emissions of textile factories, a reward function for the textile order allocation agent is constructed: in, r represents the reward function, ρ 1 means the reward coefficient for completing the order in time, ρ 2 represents the penalty coefficient for exceeding the carbon emission limit; S804: Setting constraints for the textile order allocation agent, wherein the constraints include: process start constraints, process sequence constraints, and processing times uniqueness constraints.

9. The textile order allocation method according to claim 8, characterized in that: The S9 specifically includes: S901: In the online network, use Strategy in the environment s t Next select an action a t , s t express t The environment of the moment, a t express t The actions selected at the moment; S902: Execute action a t , get reward r t , r t express t Rewards to be obtained at all times; S903: When this As t The parameters at each moment are stored in the experience pool. s t+1 express t +1 moment of environment; S904: Calculate various allocation strategies value: in, express value, The value represents the expected long-term reward of all actions, represents the allocation strategy, Indicates the allocation strategy of value, E represents the expectation operation, r t express t Rewards for every moment, r t+1 express t +1 moment reward, r t+2 express t +2 momentary rewards, s t express t The environment of the moment, a t express t The action selected at the moment, represents the discount factor; S905: Determine the optimal allocation strategy value; S906: Copy the online network parameters to the target network, and use the deep neural network to optimize the optimal network parameters at each iteration. The value is fitted to obtain the optimal Value function; S907: To maximize the optimal The value function is used as the target to allocate textile orders: in, a t express t The action selected at the moment, Indicates the best Value function.

10. A textile order distribution system, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the textile order allocation method according to any one of claims 1 to 9 is implemented.

Images