Intelligent order delivery method and delivery system

Abstract

The invention belongs to the technical field of production management, and provides an intelligent order delivery method and delivery system. The method comprises the following steps: selecting actual production data of a plurality of historical orders from a production database to construct a training sample; performing iterative training on the machine learning model based on the training sample until a preset number of iterations is reached so as to obtain an order delivery model; obtaining the specified delivery duration of the order to be delivered, the actual equipment state and the inventory of each production node, inputting the order delivery model, and outputting the predicted delivery duration and the predicted production model; generating a prediction production scheme based on the prediction production model, and performing simulation operation in the digital production line model to obtain a simulation delivery duration; when it is judged that the simulated delivery duration is smaller than the predicted delivery duration and the second duration difference value is smaller than a second preset difference value, an actual production scheme is generated based on the predicted production model to be put into a production line for production. According to the invention, the actual product delivery cycle is adaptive to the specified product delivery cycle.

Description

Technical Field

[0001] The present invention relates to the technical field of production management, and in particular to an intelligent order delivery method and delivery system. Background Art

[0002] In the process of production management, how to achieve on-time delivery of products is of great significance to maintaining the healthy operation of the enterprise and improving its competitiveness.

[0003] For hardware production companies, a complete order production and delivery process includes: raw material procurement, parts processing, component assembly, finished product debugging, and product delivery. The entire production and delivery cycle is generally long, but the time specified for general order delivery is relatively short. Therefore, it is necessary to set up corresponding inventory for each production segment and adjust the production nodes based on the specified order delivery cycle. For example, for a batch of products in an order, it is necessary to predetermine how many products to start production from raw material procurement; how many products to call the parts inventory and start production from parts processing; how many products to call the component inventory and start production from component assembly; and how many products to directly transfer from the finished product inventory and ship and deliver only after debugging.

[0004] However, in the actual production process, enterprises mostly seek automation and intelligence on production equipment in an effort to increase production capacity, but production management is still mainly manual. That is, the production quantity of the above-mentioned products corresponding to different production nodes generally needs to be based on the experience of management personnel, so there is still often a situation where the actual product delivery cycle and the prescribed product delivery cycle cannot be fully adapted. At this time, if the actual product delivery cycle is shorter than the prescribed product delivery cycle, there may be overcapacity; in particular, if the actual product delivery cycle is longer than the prescribed product delivery cycle, it may have an adverse impact on the subsequent operation of the enterprise. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent order delivery method and delivery system to improve the technical problem in the prior art that the product production quantity corresponding to different production nodes generally needs to be based on the experience of management personnel, which leads to the incompatibility between the actual product shipment cycle and the prescribed product delivery cycle.

[0006] To achieve the above object, the present invention proposes the following technical solutions:

[0007] In the first aspect, the technical solution provides an intelligent order delivery method, including:

[0008] Selecting actual production data of several historical orders from a production database to construct several training samples;

[0009] The actual delivery time of the selected historical order is less than the specified delivery time, and the first time difference between the two is less than the first preset difference; the actual production data includes first-category data and second-category data, the first-category data includes: specified delivery time, raw material inventory, component inventory, semi-finished product inventory, finished product inventory and equipment status; the second-category data includes actual delivery time and actual production model; the actual production model includes: the number of products produced since the purchase of raw materials, the number of products produced since the transfer of raw materials from the raw materials warehouse, the number of products produced since the transfer of components from the component warehouse, the number of products produced since the transfer of semi-finished product warehouse, and the number of products shipped directly;

[0010] Iteratively training the machine learning model based on the plurality of training samples until a preset number of iterations is reached to obtain an order delivery model;

[0011] The order delivery model is a convolutional neural network model. During the training process, the first type of data is used as input, the second type of data is used as output, and a binary cross entropy loss function and a gradient descent algorithm are used for parameter adjustment.

[0012] Obtain the specified delivery time of the order to be delivered, and input the actual equipment status of the enterprise and the inventory of each production node into the order delivery model to output the predicted delivery time and the predicted production model;

[0013] Generate a predicted production plan based on the predicted production model, and perform simulated production in a digital production line model based on the predicted production plan to obtain a simulated delivery time;

[0014] The digital production line model is a digital twin model built based on the actual production line, which can be simulated and run independently of the actual production line at the production control end;

[0015] When it is determined that the simulated delivery time is less than the predicted delivery time, and a second time difference between the two is less than a second preset difference, an actual production plan is generated based on the predicted production model to put the production line into production for the orders to be delivered.

[0016] Furthermore, the acquisition of the specified delivery time of the order to be delivered, the actual equipment status of the enterprise, and the inventory of each production node are input into the order delivery model to output the predicted delivery time and the predicted production model; including:

[0017] Obtain the proportion of each product in the total number of products in the orders to be delivered in the forecast production model;

[0018] Based on the proportion of each product, the inventory quantity of each type is supplemented and optimized.

[0019] Furthermore, the actual production plan is generated based on the predicted production model to be put into production line to produce the orders to be delivered; including:

[0020] The digital production line model is communicatively connected with the production line to achieve synchronous operation therebetween;

[0021] Remotely monitor the production line based on the operating status of the digital production line model displayed on a remote client;

[0022] When it is determined that the digital production line model has an operational anomaly, the remote client controls the production line to shut down, and sends an early warning message to the operation and maintenance personnel to troubleshoot and resolve the anomaly;

[0023] The warning information includes abnormal time, abnormal section, abnormal equipment number and abnormal log.

[0024] Further, including:

[0025] When it is determined that the corresponding product in the to-be-delivered order needs to be customized, the design key points of the product are extracted to generate a number of prompt words;

[0026] Inputting the plurality of prompt words into a pre-trained predictive design model to obtain a predicted design duration corresponding to the product;

[0027] The difference between the prescribed delivery time and the predicted design time is calculated to obtain the target delivery time; at this time, when obtaining the predicted production model based on the order delivery model, the target delivery time is input.

[0028] Further, including:

[0029] When it is determined that the corresponding product in the to-be-delivered order needs to be customized, the design key points of the product are extracted to generate a number of prompt words;

[0030] Inputting the plurality of prompt words into a diffusion model to obtain an initial design drawing;

[0031] The initial design drawing is adjusted based on the feedback information from the ordering party to obtain the target design drawing.

[0032] In the second aspect, this technical solution provides an intelligent order delivery system. It includes:

[0033] A sample acquisition module is used to select actual production data of several historical orders from a production database to construct several training samples;

[0034] The actual delivery time of the selected historical order is less than the specified delivery time, and the first time difference between the two is less than the first preset difference; the actual production data includes first-category data and second-category data, the first-category data includes: specified delivery time, raw material inventory, component inventory, semi-finished product inventory, finished product inventory and equipment status; the second-category data includes actual delivery time and actual production model; the actual production model includes: the number of products produced since the purchase of raw materials, the number of products produced since the transfer of raw materials from the raw materials warehouse, the number of products produced since the transfer of components from the component warehouse, the number of products produced since the transfer of semi-finished product warehouse, and the number of products shipped directly;

[0035] A model acquisition module, used for iteratively training the machine learning model based on the plurality of training samples until a preset number of iterations is reached to acquire an order delivery model;

[0036] The order delivery model is a convolutional neural network model. During the training process, the first type of data is used as input, the second type of data is used as output, and a binary cross entropy loss function and a gradient descent algorithm are used for parameter adjustment.

[0037] A forecast production module is used to obtain the specified delivery time of the order to be delivered, and the actual equipment status of the enterprise and the inventory of each production node are input into the order delivery model to output the forecast delivery time and the forecast production model;

[0038] A simulation production module, used for generating a prediction production plan based on the prediction production model, and performing simulation production in a digital production line model based on the prediction production plan to obtain a simulation delivery time;

[0039] The digital production line model is a digital twin model built based on the actual production line, which can be simulated and run independently of the actual production line at the production control end;

[0040] The actual production module is used to generate an actual production plan based on the predicted production model to put the production line into production for the orders to be delivered when it is determined that the simulated delivery time is less than the predicted delivery time and the second time difference between the two is less than the second preset difference.

[0041] Further, including:

[0042] A proportion acquisition module is used to obtain the proportion of each product in the total number of products in the order to be delivered in the predicted production model;

[0043] The inventory optimization module is used to supplement and optimize the inventory quantities of various products based on the proportion of each product.

[0044] Further, including:

[0045] A communication connection module, used to connect the digital production line model to the production line for communication so as to achieve synchronous operation between them;

[0046] A remote monitoring module, used for remotely monitoring the production line from the operating status of the digital production line model displayed on a remote client;

[0047] The abnormality troubleshooting module is used to control the production line to shut down from the remote client when it is determined that the digital production line model has an operational abnormality, and send an early warning message to the operation and maintenance personnel to troubleshoot and resolve the abnormality;

[0048] The warning information includes abnormal time, abnormal section, abnormal equipment number and abnormal log.

[0049] In a third aspect, the technical solution provides an electronic device, comprising at least one processor, wherein the processor is coupled to a memory, wherein a computer program is stored in the memory, and wherein the computer program is configured to execute the method described when executed by the processor.

[0050] In a fourth aspect, the present technical solution provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is used to execute the described method.

[0051] Beneficial effects:

[0052] It can be seen from the above technical solutions that the technical solution of the present invention provides an intelligent order delivery method to solve the technical defect that the actual product shipment cycle and the specified product delivery cycle cannot be adapted.

[0053] The method specifically includes: selecting actual production data of several historical orders from a production database to construct several training samples. Based on the several training samples, the machine learning model is iteratively trained until a preset number of iterations is reached to obtain an order delivery model. The specified delivery time of the order to be delivered is obtained, and the actual equipment status of the enterprise and the inventory of each production node are input into the order delivery model to output the predicted delivery time and the predicted production model. A predicted production plan is generated based on the predicted production model, and simulated production is performed in a digital production line model based on the predicted production plan to obtain a simulated delivery time. When it is judged that the simulated delivery time is less than the predicted delivery time, and the second time difference between the two is less than the second preset difference, an actual production plan is generated based on the predicted production model to be put into the production line to produce the order to be delivered. It can be seen that the technical solution first introduces an order delivery model based on a machine learning algorithm to generate a predicted production model. At the same time, in order to ensure that the predicted production model can realize the delivery of the order on schedule, a digital production line model is also used to perform production prediction on the predicted production model. In this way, the intelligence of the entire production management is realized, and the delivery of the order on schedule is ensured.

[0054] It should be appreciated that all combinations of the foregoing concepts, as well as additional concepts described in greater detail below, may be considered to be part of the inventive subject matter of the present disclosure, provided such concepts are not mutually inconsistent.

[0055] The foregoing and other aspects, embodiments and features of the present invention can be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the present invention, such as the features and / or beneficial effects of the exemplary embodiments, will be apparent from the following description or learned from the practice of the specific embodiments according to the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in various figures may be represented by the same reference numeral. For clarity, not every component is labeled in every figure. Embodiments of various aspects of the present invention will now be described by way of example and with reference to the accompanying drawings, in which:

[0057] Figure 1 This is a flow chart of the intelligent order delivery method described in this embodiment;

[0058] Figure 2 A flowchart for optimizing inventory quantity for this embodiment;

[0059] Figure 3 A flowchart for production monitoring in this embodiment;

[0060] Figure 4A flowchart for product design prediction for this embodiment;

[0061] Figure 5 A flowchart for obtaining design drawings for this embodiment;

[0062] Figure 6 This is a structural block diagram of the intelligent order delivery system described in this embodiment;

[0063] Figure 7 is a structural block diagram of the electronic device described in this embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein should be the common meaning understood by people with general skills in the field to which the present invention belongs.

[0065] The words "first", "second" and similar words used in the specification and claims of this application do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, unless the context clearly indicates otherwise, the singular form of "a", "an" or "the" and other similar words do not indicate a quantitative limitation, but rather indicate the presence of at least one. Words such as "include" or "comprise" and the like mean that the elements or objects appearing before "include" or "comprise" cover the features, wholes, steps, operations, elements and / or components listed after "include" or "comprise", and do not exclude the existence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0066] In the process of hardware production and delivery, since the cycle of complete production delivery is generally long, the time specified for order delivery is short. Therefore, it is necessary to set up corresponding inventory for each production segment, and adjust the production nodes based on the specified order delivery cycle. However, in the actual production management process, the production quantity of products corresponding to different production nodes generally needs to be based on the experience of managers, so there is still often a situation where the actual product shipping cycle and the specified product delivery cycle cannot be fully adapted. Based on this, this embodiment aims to provide an intelligent order delivery method to improve this type of technical defects.

[0067] The intelligent order delivery method described in this embodiment is described in detail below in conjunction with the accompanying drawings.

[0068] like Figure 1 As shown, the method comprises the following steps:

[0069] Step S102: Select actual production data of a number of historical orders from the production database to construct a number of training samples.

[0070] Since the machine learning model needs to generate samples based on the historical orders selected in this embodiment and learn from the experience, in order to ensure the prediction accuracy of the trained machine learning model, the selected historical orders need to have the following characteristics: (1) the actual delivery time is less than the specified delivery time; (2) the first time difference between the actual delivery time and the specified delivery time is less than the first preset difference.

[0071] Specifically, the actual production data includes the first type of data and the second type of data. The first type of data includes: the specified delivery time, raw material inventory, component inventory, semi-finished product inventory, finished product inventory and equipment status. The second type of data includes the actual delivery time and the actual production model; wherein the actual production model includes: the number of products produced from the purchase of raw materials, the number of products produced from the transfer of raw materials from the warehouse, the number of products produced from the transfer of components from the warehouse, the number of products produced from the transfer of semi-finished product from the warehouse, and the number of products shipped directly.

[0072] Step S104: iteratively train the machine learning model based on the plurality of training samples until a preset number of iterations is reached to obtain an order delivery model.

[0073] In this embodiment, the order delivery model is a convolutional neural network model. Specifically, during the training process, the first type of data is used as input, and the second type of data is used as output; and a binary cross entropy loss function and a gradient descent algorithm are used for parameter adjustment.

[0074] Step S106: Obtain the specified delivery time of the order to be delivered, and input the actual equipment status of the enterprise and the inventory of each production node into the order delivery model to output the predicted delivery time and the predicted production model.

[0075] In specific implementation, the actual equipment status includes: equipment operation status, equipment normal maintenance cycle, equipment abnormality frequency and common abnormal recovery time. Furthermore, the equipment operation status includes operation, downtime and shutdown; among which, when the equipment operation status is operation, it indicates that the equipment is producing other orders; when the equipment operation status is downtime, it indicates that the equipment is in an abnormal maintenance state; when the equipment operation status is shutdown, it indicates that there is no order production at present or the equipment is in a normal maintenance state. The equipment abnormality frequency and common abnormal recovery time can reflect the impact that uncertain factors may have on the delivery of pending production orders.

[0076] As a preferred implementation method, it is considered that the general order delivery time is definitely less than the complete product production cycle. For example, the order delivery time is 30 days, while the complete product production cycle is 80 days. Therefore, ensuring the inventory of each production node is of great value to ensure the timely delivery of orders. However, how to optimize the inventory of each production node is currently mostly estimated by personnel combined with existing order production data; therefore, there may be large deviations. And combined with Figure 2 As shown, this embodiment can timely optimize each inventory as different new orders are produced in the following manner:

[0077] Step S107.2: Obtain the proportion of each product in the total number of products in the order to be delivered in the predicted production model.

[0078] Step S107.4: Supplement and optimize the inventory quantity of each category based on the proportion of each product.

[0079] At this time, based on steps S107.2 to S107.4, each inventory quantity can be optimized and replenished in a timely manner after each predicted production model is obtained, ensuring that other subsequent orders can be delivered on schedule.

[0080] Step S108: Generate a predicted production plan based on the predicted production model, and perform simulated production in a digital production line model based on the predicted production plan to obtain a simulated delivery time.

[0081] In this embodiment, the digital production line model is a digital twin model built based on the actual production line, and can be simulated and run independently of the actual production line at the production control end.

[0082] Step S110: When it is determined that the simulated delivery time is less than the predicted delivery time, and a second time difference between the two is less than a second preset difference, an actual production plan is generated based on the predicted production model to put the production line into production for the orders to be delivered.

[0083] During the specific implementation process, if the simulated delivery time is greater than the predicted delivery time; or the simulated delivery time is less than the predicted delivery time, but the second time difference between the two is greater than the second preset difference; then it is necessary to check and optimize the order delivery model or the digital production line model.

[0084] As a further embodiment, in combination Figure 3 As shown, production monitoring is carried out in the following ways during the production process:

[0085] Step S110.2: communicate the digital production line model with the production line to achieve synchronous operation between them.

[0086] As a preferred implementation, in order to confirm the operation consistency between the digital production line model and the production line, the digital production line model needs to be debugged and optimized regularly to avoid remote monitoring accidents caused by inconsistency between the two operations.

[0087] Step S110.4: remotely monitor the production line based on the operating status of the digital production line model displayed on the remote client.

[0088] Step S110.6: When it is determined that the digital production line model has an operating abnormality, the remote client controls the production line to shut down, and sends an early warning message to the operation and maintenance personnel to troubleshoot and resolve the abnormality.

[0089] In this step, the warning information includes: abnormal time, abnormal section, abnormal equipment number and abnormal log. Since the production line is shut down on the remote client, abnormal production is avoided and losses are reduced.

[0090] During the specific hardware construction process, the remote client handles the shutdown of the production line based on the Internet of Things.

[0091] As a preferred implementation, considering that some products need to be adjusted or redesigned based on customer needs. Figure 4 As shown, the following steps are also included:

[0092] Step S102: input the plurality of prompt words into a pre-trained predictive design model to obtain a predicted design duration corresponding to the product.

[0093] In this step, the predictive design model is also a convolutional neural network model. In order to improve the prediction accuracy, an attention mechanism is introduced into the model. At this time, during the training process of the predictive design model, more attention can be paid to the key information in the prompt words, the difficulty of the corresponding product design can be learned, and then the corresponding predicted design time can be predicted and output based on the difficulty.

[0094] Step S104: Calculate the difference between the prescribed delivery time and the predicted design time to obtain a target delivery time.

[0095] At this time, when obtaining the forecast production model based on the order delivery model, the target delivery time is input. At this time, the actual situation used in product design is taken into account in the order production and delivery process through steps S102 to S104, thereby avoiding delays.

[0096] Furthermore, generally speaking, for a company, the types of products it produces are relatively uniform, and the product designs also have the same characteristics. Therefore, in order to improve the efficiency when product design needs to be adjusted, Figure 5 As shown, the following steps are also included:

[0097] Step S102': when it is determined that the corresponding product in the to-be-delivered order needs to be customized, the design key points of the product are extracted to generate a number of prompt words.

[0098] Step S104': input the plurality of prompt words into a diffusion model to obtain an initial design drawing.

[0099] Step S106': adjusting the initial design drawing based on the feedback information from the ordering party to obtain a target design drawing.

[0100] At this point, the AI ​​model can be used to carry out preliminary production of the design drawings, and then targeted adjustments can be made based on customer needs; this greatly shortens the time taken up by product design adjustments, leaving enough time for subsequent actual processing and production, and ensuring that the product is delivered on schedule.

[0101] From the above, it can be seen that the order delivery model built based on the machine learning algorithm in this embodiment generates a predictive production model and the digital production line model to realize the intelligence of the entire production management and ensure the timely delivery of orders. At the same time, inventory optimization is also carried out based on the predictive production model to ensure the healthy operation of the enterprise. Considering that some products require targeted design, machine learning algorithms are introduced to predict product design time to avoid affecting the delivery time, and AI algorithms are introduced for drawing design to shorten the overall development cycle and ensure product delivery.

[0102] The above program can be run in the processor, or it can also be stored in the memory (or computer-readable storage medium), which includes permanent and non-permanent, removable and non-removable media. Information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined in this article, computer-readable media does not include temporary computer-readable media, such as modulated data signals and carrier waves.

[0103] These computer programs can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in one or more blocks can be implemented by different modules corresponding to different steps.

[0104] Based on this, this embodiment also provides an intelligent order delivery system. Figure 6 As shown, the system comprises:

[0105] The sample acquisition module is used to select actual production data of several historical orders from the production database to construct several training samples.

[0106] Among them, the actual delivery time of the selected historical orders is less than the prescribed delivery time, and the first time difference between the two is less than the first preset difference; the actual production data includes first-category data and second-category data, the first-category data includes: prescribed delivery time, raw material inventory, component inventory, semi-finished product inventory, finished product inventory and equipment status; the second-category data includes actual delivery time and actual production model; wherein, the actual production model includes: the number of products produced since the purchase of raw materials, the number of products produced since the transfer of raw materials from the raw materials warehouse, the number of products produced since the transfer of components from the component warehouse, the number of products produced since the transfer of semi-finished product warehouse, and the number of products directly shipped.

[0107] The model acquisition module is used to iteratively train the machine learning model based on the plurality of training samples until a preset number of iterations is reached to acquire an order delivery model.

[0108] Among them, the order delivery model is a convolutional neural network model. During the training process, the first type of data is used as input, the second type of data is used as output, and the binary cross entropy loss function and gradient descent algorithm are used for parameter adjustment.

[0109] The forecast production module is used to obtain the specified delivery time of the orders to be delivered. The actual equipment status of the enterprise and the inventory of each production node are input into the order delivery model to output the forecast delivery time and the forecast production model.

[0110] The simulation production module is used to generate a prediction production plan based on the prediction production model, and to perform simulated production in a digital production line model based on the prediction production plan to obtain a simulated delivery time.

[0111] Among them, the digital production line model is a digital twin model built based on the actual production line, which can be simulated and run independently of the actual production line at the production control end.

[0112] The actual production module is used to generate an actual production plan based on the predicted production model to put the production line into production for the orders to be delivered when it is determined that the simulated delivery time is less than the predicted delivery time and the second time difference between the two is less than the second preset difference.

[0113] The system is used to implement the steps of the above method, so what has been explained will not be repeated here.

[0114] For example, the system further comprises:

[0115] The proportion acquisition module is used to obtain the proportion of each product in the total number of products in the order to be delivered in the predicted production model.

[0116] The inventory optimization module is used to supplement and optimize the inventory quantities of various products based on the proportion of each product.

[0117] For another example, the system further includes:

[0118] The communication connection module is used to connect the digital production line model to the production line for communication so as to realize synchronous operation between them.

[0119] A remote monitoring module is used to remotely monitor the production line based on the operating status of the digital production line model displayed on a remote client.

[0120] The abnormality troubleshooting module is used to control the production line to shut down from the remote client when it is determined that the digital production line model has an operating abnormality, and send an early warning message to the operation and maintenance personnel to troubleshoot and resolve the abnormality.

[0121] The warning information includes abnormal time, abnormal section, abnormal equipment number and abnormal log.

[0122] This embodiment also provides an electronic device, such as Figure 7 As shown, at least one processor is included. The processor is coupled to a memory. A computer program is stored in the memory. The computer program is configured to execute the method when executed by the processor.

[0123] This embodiment further provides a computer-readable storage medium on which a computer program is stored. The computer program is used to execute the method described above.

[0124] Since the system, electronic device and storage medium are all used to implement the method, they also have the technical advantage of matching the actual product shipment cycle with the specified product delivery cycle during actual operation, thereby improving the delivery capability and competitiveness of the corresponding enterprise and maintaining the healthy operation of the enterprise.

[0125] Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. A person with ordinary knowledge in the technical field to which the present invention belongs may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be determined by the definition of the claims.

Claims

1. An intelligent order delivery method, characterized in that: include: Selecting actual production data of several historical orders from a production database to construct several training samples; The actual delivery time of the selected historical order is less than the specified delivery time, and the first time difference between the two is less than the first preset difference; the actual production data includes first-category data and second-category data, the first-category data includes: specified delivery time, raw material inventory, component inventory, semi-finished product inventory, finished product inventory and equipment status; the second-category data includes actual delivery time and actual production model; the actual production model includes: the number of products produced since the purchase of raw materials, the number of products produced since the transfer of raw materials from the raw materials warehouse, the number of products produced since the transfer of components from the component warehouse, the number of products produced since the transfer of semi-finished product warehouse, and the number of products shipped directly; Iteratively training the machine learning model based on the plurality of training samples until a preset number of iterations is reached to obtain an order delivery model; The order delivery model is a convolutional neural network model. During the training process, the first type of data is used as input, the second type of data is used as output, and a binary cross entropy loss function and a gradient descent algorithm are used for parameter adjustment. Obtain the specified delivery time of the order to be delivered, and input the actual equipment status of the enterprise and the inventory of each production node into the order delivery model to output the predicted delivery time and the predicted production model; Generate a predicted production plan based on the predicted production model, and perform simulated production in a digital production line model based on the predicted production plan to obtain a simulated delivery time; The digital production line model is a digital twin model built based on the actual production line, which can be simulated and run independently of the actual production line at the production control end; When it is determined that the simulated delivery time is less than the predicted delivery time, and a second time difference between the two is less than a second preset difference, an actual production plan is generated based on the predicted production model to put the production line into production for the orders to be delivered.

2. The intelligent order delivery method according to claim 1, characterized in that: The obtaining of the specified delivery time of the order to be delivered, the actual equipment status of the enterprise, and the inventory of each production node are input into the order delivery model to output the predicted delivery time and the predicted production model; comprising: Obtain the proportion of each product in the total number of products in the orders to be delivered in the forecast production model; Based on the proportion of each product, the inventory quantity of each type is supplemented and optimized.

3. The intelligent order delivery method according to claim 1, characterized in that: The method of generating an actual production plan based on the predicted production model to put the actual production plan into production line to produce the orders to be delivered includes: The digital production line model is communicatively connected with the production line to achieve synchronous operation therebetween; Remotely monitor the production line based on the operating status of the digital production line model displayed on a remote client; When it is determined that the digital production line model has an operational anomaly, the remote client controls the production line to shut down, and sends an early warning message to the operation and maintenance personnel to troubleshoot and resolve the anomaly; The warning information includes abnormal time, abnormal section, abnormal equipment number and abnormal log.

4. The intelligent order delivery method according to claim 1, characterized in that: Also includes: When it is determined that the corresponding product in the to-be-delivered order needs to be customized, the design key points of the product are extracted to generate a number of prompt words; Inputting the plurality of prompt words into a pre-trained predictive design model to obtain a predicted design duration corresponding to the product; Calculating the difference between the specified delivery time and the predicted design time to obtain the target delivery time; At this time, when obtaining the predicted production model based on the order delivery model, the target delivery time is input.

5. The intelligent order delivery method according to claim 1, characterized in that: include: When it is determined that the corresponding product in the to-be-delivered order needs to be customized, the design key points of the product are extracted to generate a number of prompt words; Inputting the plurality of prompt words into a diffusion model to obtain an initial design drawing; The initial design drawing is adjusted based on the feedback information from the ordering party to obtain the target design drawing.

6. An intelligent order delivery system, characterized in that: include: A sample acquisition module is used to select actual production data of several historical orders from a production database to construct several training samples; The actual delivery time of the selected historical order is less than the specified delivery time, and the first time difference between the two is less than the first preset difference; the actual production data includes first-category data and second-category data, the first-category data includes: specified delivery time, raw material inventory, component inventory, semi-finished product inventory, finished product inventory and equipment status; the second-category data includes actual delivery time and actual production model; the actual production model includes: the number of products produced since the purchase of raw materials, the number of products produced since the transfer of raw materials from the raw materials warehouse, the number of products produced since the transfer of components from the component warehouse, the number of products produced since the transfer of semi-finished product warehouse, and the number of products shipped directly; A model acquisition module, used for iteratively training the machine learning model based on the plurality of training samples until a preset number of iterations is reached to acquire an order delivery model; The order delivery model is a convolutional neural network model. During the training process, the first type of data is used as input, the second type of data is used as output, and a binary cross entropy loss function and a gradient descent algorithm are used for parameter adjustment. A forecast production module is used to obtain the specified delivery time of the order to be delivered, and the actual equipment status of the enterprise and the inventory of each production node are input into the order delivery model to output the forecast delivery time and the forecast production model; A simulation production module, used for generating a prediction production plan based on the prediction production model, and performing simulation production in a digital production line model based on the prediction production plan to obtain a simulation delivery time; The digital production line model is a digital twin model built based on the actual production line, which can be simulated and run independently of the actual production line at the production control end; The actual production module is used to generate an actual production plan based on the predicted production model to put the production line into production for the orders to be delivered when it is determined that the simulated delivery time is less than the predicted delivery time and the second time difference between the two is less than the second preset difference.

7. The intelligent order delivery system according to claim 6, characterized in that: include: A proportion acquisition module is used to obtain the proportion of each product in the total number of products in the order to be delivered in the predicted production model; The inventory optimization module is used to supplement and optimize the inventory quantities of various products based on the proportion of each product.

8. The intelligent order delivery system according to claim 7, characterized in that: include: A communication connection module, used to connect the digital production line model to the production line for communication so as to achieve synchronous operation between them; A remote monitoring module, used for remotely monitoring the production line from the operating status of the digital production line model displayed on a remote client; The abnormality troubleshooting module is used to control the production line to shut down from the remote client when it is determined that the digital production line model has an operational abnormality, and send an early warning message to the operation and maintenance personnel to troubleshoot and resolve the abnormality; The warning information includes abnormal time, abnormal section, abnormal equipment number and abnormal log.

9. An electronic device, characterized in that: The method comprises at least one processor, wherein the processor is coupled to a memory, wherein a computer program is stored in the memory, and wherein the computer program is configured to execute the method according to any one of claims 1 to 5 when executed by the processor.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is used to execute the method according to any one of claims 1 to 5.

Images