Electronic component intelligent production control system and control method thereof

Abstract

The embodiment of the invention discloses an electronic component intelligent production control system and a control method thereof, and relates to the technical field of intelligent manufacturing. The method comprises the steps that historical statistical data corresponding to various electronic components are acquired, and the historical statistical data are acquired based on product sales volume data of component models contained in the various electronic components at a plurality of historical time nodes within a specified historical time period; performing data segmentation on the historical statistical data to obtain a plurality of statistical sequence fragments; performing product demand quantity prediction based on the plurality of statistical sequence fragments to obtain predicted demand quantity of component models included in each electronic component type at a plurality of target time nodes; and determining a target production plan according to the predicted demand quantity of the plurality of target time nodes and the order data corresponding to the plurality of target time nodes, and performing production control based on the target production plan. Through the method, the accuracy of the production plan can be improved, and the risk of excessive production or insufficient production is reduced.

Description

Technical Field

[0001] The present application relates to the field of intelligent manufacturing technology, and more specifically, to an intelligent production control system for electronic components and a control method thereof. Background Art

[0002] Due to fierce market competition, in order to meet customer needs and increase market share, manufacturers of electronic components (such as resistors, capacitors, inductors, diodes, triodes, transistors, etc.) often do not limit their production arrangements to customer orders, but will selectively produce some electronic components that may be used in the market and maintain a certain amount of inventory to avoid market risks. The current inventory management method is generally to predict the market demand quantity through manual statistics (for example, through analysis of product sales data trends), and then prepare inventory based on the demand quantity forecast results. However, the current forecasting method is not only time-consuming and labor-intensive, but the accuracy of the forecast results is usually relatively low, and it is easy to produce large errors, which may eventually lead to inventory backlogs or out-of-stock.

[0003] Based on this, it is necessary to study an intelligent production control system for electronic components and its control method to intelligently manage the production plan and inventory of various electronic components. Summary of the invention

[0004] To solve the above problems, one aspect of an embodiment of this specification provides an electronic component intelligent production control method, the method comprising:

[0005] Obtaining historical statistical data corresponding to various types of electronic components, the historical statistical data being obtained based on product sales data of one or more component models included in each type of electronic component at multiple historical time nodes within a specified historical time period;

[0006] Performing data segmentation on the historical statistical data to obtain a plurality of statistical sequence segments;

[0007] Perform product demand forecasting based on the multiple statistical sequence fragments to obtain forecasted demand for one or more component models included in each electronic component type at multiple target time nodes;

[0008] A target production plan is determined according to the predicted demand quantities at the multiple target time nodes and the order data corresponding to the multiple target time nodes, and production control is performed based on the target production plan.

[0009] In some embodiments, the target time node includes at least a first time node and a second time node, wherein the second time node is located after the first time node; and the data segmentation of the historical statistical data to obtain a plurality of statistical sequence segments includes:

[0010] Determine a first segmentation window size based on a distance between the first time node and the current time node;

[0011] Determine a second split window size based on a distance between the second time node and the current time node;

[0012] The historical statistical data is segmented in reverse order starting from the current time node, and a plurality of statistical sequence segments are obtained based on the first segmentation window size and the second segmentation window size.

[0013] In some embodiments, the product demand forecasting based on the multiple statistical sequence segments to obtain the forecasted demand of one or more component models included in each electronic component type at multiple target time nodes includes:

[0014] Inputting a plurality of first statistical sequence segments obtained by segmentation based on the first segmentation window size into a first demand quantity prediction network to obtain a first predicted demand quantity corresponding to one or more component models included in each electronic component type at the first time node;

[0015] The plurality of second statistical sequence segments obtained by segmentation based on the second segmentation window size are input into a second demand forecasting network to obtain a second predicted demand corresponding to one or more component models included in each electronic component type at the second time node.

[0016] In some embodiments, the first demand forecasting network and the second demand forecasting network include an input layer, a hidden layer and an output layer; wherein,

[0017] The input layer is used to receive corresponding statistical sequence fragments as input data;

[0018] The hidden layer includes a plurality of neurons for extracting and transforming features of the input data to capture data variation trends and patterns in the statistical sequence segments;

[0019] The output layer is used to obtain the predicted demand for one or more component models included in each electronic component type at a corresponding time node based on the output results of the hidden layer.

[0020] In some embodiments, the first demand forecasting network and the second demand forecasting network are trained in the following manner:

[0021] Obtain sample historical statistical data corresponding to each type of electronic components, and a first label and a second label corresponding to each piece of sample historical statistical data, wherein the first label is used to indicate the expected demand corresponding to the sample historical statistical data at a first sample time node, and the second label is used to indicate the expected demand corresponding to the sample historical statistical data at a second sample time node, and the sample historical statistical data is obtained based on product sales data of one or more component models included in each electronic component type at multiple sample historical time nodes within a specified sample historical time period;

[0022] Determine a first sample segmentation window size based on a distance between the first sample time node and a reference time node, and determine a second sample segmentation window size based on a distance between the second sample time node and the reference time node;

[0023] Performing data segmentation on the sample historical statistical data in reverse order starting from the reference time node, and obtaining a plurality of first sample statistical sequence segments based on the first sample segmentation window size, and obtaining a plurality of second sample statistical sequence segments based on the second sample segmentation window size;

[0024] Inputting the plurality of first sample statistical sequence fragments as input data into the first demand forecasting network for processing to obtain a first sample forecasting result, and inputting the plurality of second sample statistical sequence fragments as input data into the second demand forecasting network for processing to obtain a second sample forecasting result;

[0025] The parameters of the first demand forecasting network are iteratively optimized based on the difference between the first sample prediction result and the first label, and the parameters of the second demand forecasting network are iteratively optimized based on the difference between the second sample prediction result and the second label, until the training is completed when the preset conditions are met.

[0026] In some embodiments, determining the target production plan according to the predicted demand quantities at the multiple target time nodes and the order data corresponding to the multiple target time nodes includes:

[0027] For any target component model contained in any electronic component type,

[0028] Calculate the error rate between the first predicted demand quantity and the order quantity corresponding to the target component model at the first time node;

[0029] The second predicted demand corresponding to the second time node is corrected based on the error rate, and the target production volume corresponding to the target component model at the second time node is determined based on the corrected data.

[0030] In some embodiments, calculating the error rate between the first predicted demand quantity and the order quantity corresponding to the target component model at the first time node includes:

[0031] Before reaching the first time node, the error rate is updated according to the change of the order quantity; wherein the error rate is calculated in the following manner:

[0032]

[0033] Wherein, ω represents the error rate between the first predicted demand and order quantity corresponding to the target component model at the first time node; y' 1 represents the first predicted demand volume of the target component model at the first time node; 1 Indicates the order quantity corresponding to the target component model at the first time node.

[0034] In some embodiments, the correcting the second predicted demand corresponding to the second time node based on the error rate includes:

[0035] A correction coefficient is determined according to the error rate, and the product of the second predicted demand corresponding to the second time node and the correction coefficient is used as the corrected data.

[0036] In some embodiments, determining the target production volume corresponding to the target component model at the second time node based on the corrected data includes:

[0037] Obtaining the completed production of the target component model as of the first time node;

[0038] The target production volume planned to be completed during the period from the first time node to the second time node is determined based on the difference between the corrected data and the completed production volume.

[0039] Another aspect of the embodiments of this specification also provides an electronic component intelligent production control system, the system comprising:

[0040] An acquisition module is used to acquire historical statistical data corresponding to various types of electronic components, wherein the historical statistical data is obtained based on product sales data of one or more component models included in each type of electronic component at multiple historical time nodes within a specified historical time period;

[0041] A data segmentation module, used for performing data segmentation on the historical statistical data to obtain multiple statistical sequence segments;

[0042] A demand forecasting module, used to forecast product demand based on the multiple statistical sequence fragments, and obtain forecast demand of one or more component models included in each electronic component type at multiple target time nodes;

[0043] The production plan determination module is used to determine the target production plan according to the predicted demand at the multiple target time nodes and the order data corresponding to the multiple target time nodes, and to perform production control based on the target production plan.

[0044] The electronic component intelligent production control system and control method provided in the embodiments of this specification may bring at least the following beneficial effects:

[0045] (1) By performing data segmentation on historical statistical data to obtain multiple statistical sequence segments, product demand is predicted based on the multiple statistical sequence segments to obtain predicted demand for one or more component models included in each electronic component type at multiple target time nodes, and finally a target production plan is determined based on the predicted demand at the multiple target time nodes and the order data corresponding to the multiple target time nodes, and production control is performed based on the target production plan, which can effectively improve the accuracy of the production plan and reduce the risk of overproduction or underproduction;

[0046] (2) By segmenting the historical statistical data into multiple statistical sequence segments, and then further processing and analyzing these statistical sequence segments, the trend characteristics and periodic laws of electronic component sales can be extracted, thereby providing more accurate data support for subsequent production planning and inventory control. At the same time, the historical statistical data can be flexibly segmented based on forecast demand to adapt to the product demand forecast at different target time nodes;

[0047] (3) By determining the correction coefficient based on the error rate between the first predicted demand and the order quantity corresponding to the target component model at the first time node, the actual market situation and order changes can be better reflected, so that the second predicted demand corresponding to the second time node can be adaptively adjusted based on the actual situation, thereby improving the accuracy of the demand forecast results to a certain extent.

[0048] Additional features will be described in part in the following description. For those skilled in the art, it will become apparent by reviewing the following and accompanying drawings, or it may be understood by the production or operation of the examples. The features of this specification may be implemented and obtained by practicing or using various aspects of the methods, tools, and combinations described in the following detailed examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0050] Figure 1 is an exemplary step flow chart of an electronic component intelligent production control method according to some embodiments of this specification;

[0051] Figure 2 It is a schematic diagram of the data processing flow of intelligent production control of electronic components according to some embodiments of this specification;

[0052] Figure 3 is a flowchart of training steps according to some embodiments of this specification;

[0053] Figure 4 It is a schematic diagram of the data processing flow of intelligent production control of electronic components according to other embodiments of this specification;

[0054] Figure 5 It is an exemplary module diagram of an intelligent production control system for electronic components according to some embodiments of this specification. DETAILED DESCRIPTION

[0056] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0057] It should be understood that the "system", "device", "unit" and / or "module" used in this specification is a method for distinguishing different components, elements, parts, parts or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0058] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0059] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0060] The electronic component intelligent production control system and control method provided in the embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0061] Figure 1 is an exemplary step flow chart of an electronic component intelligent production control method according to some embodiments of this specification. In some embodiments, the electronic component intelligent production control method can be executed by processing logic, which can include hardware (e.g., circuits, dedicated logic, programmable logic, microcode, etc.), software (instructions running on a processing device to perform hardware simulation), etc. or any combination thereof. In some embodiments, Figure 1 One or more operations in the flowchart of the electronic component intelligent production control method shown can be implemented by a processing device and / or a terminal device. For example, the electronic component intelligent production control method can be stored in a storage device in the form of a computer program and / or instruction, and called and / or executed by a processing device and / or a terminal device. Figure 1 The electronic component intelligent production control method provided in the embodiment of the present application may include the following steps:

[0062] Step S110, obtaining historical statistical data corresponding to each type of electronic components, the historical statistical data is obtained based on product sales data of one or more component models included in each electronic component type at multiple historical time nodes within a specified historical time period. In some embodiments, step S110 can be performed by the acquisition module 210 mentioned below.

[0063] In the embodiment of the present application, the specified historical time period may refer to the past month, quarter, year or any other suitable time period, and the multiple historical time nodes within the specified historical time period may be understood as time nodes evenly distributed within the time period or distributed according to a certain specific rule, such as the end of each day, week, or month. By collecting and analyzing the product sales data at these historical time nodes, the sales trends and rules of electronic components in different time periods can be obtained, thereby providing data support for subsequent production planning and inventory control.

[0064] Specifically, in some embodiments of the present application, the historical statistical data can be obtained based on the product sales data of one or more component models contained in each electronic component type (such as resistors, capacitors, inductors, diodes, triodes, transistors, etc.) at multiple historical time nodes within a specified historical time period. For example, for a certain type of electronic component, such as a capacitor, it may include a variety of different models, and these components of different models under the same type may have different specifications or performances. In an embodiment of the present application, the sales data of these different models of capacitors at the end of each month in the past year can be counted to obtain the historical statistical data corresponding to the capacitor category. The method of obtaining the historical statistical data corresponding to other categories of electronic components is similar to this, and will not be repeated in this specification.

[0065] Step S120, performing data segmentation on the historical statistical data to obtain a plurality of statistical sequence segments. In some embodiments, step S120 may be performed by the data segmentation module 220 mentioned below.

[0066] In the embodiment of the present application, in order to predict the product demand at multiple different target time nodes, it is necessary to segment the continuous historical statistical data into multiple statistical sequence segments. Each statistical sequence segment may contain sales data of a certain number of historical time nodes, so as to perform a more detailed analysis of the sales trend of electronic components in different time periods.

[0067] It can be understood that by segmenting the historical statistical data to obtain multiple statistical sequence segments, and then further processing and analyzing these statistical sequence segments, the trend characteristics and periodic laws of electronic component sales can be extracted, thereby providing more accurate data support for subsequent production planning and inventory control.

[0068] In an embodiment of the present application, the specific manner of segmenting the historical statistical data may be determined based on the distance between the target time node and the current time node. For example, in some embodiments, the target time node may include at least a first time node and a second time node, wherein the second time node is located after the first time node. Exemplarily, assuming that the current time node is January 1, 2025, the first time node may be February 1, 2025, and the second time node may be March 1, 2025; or the first time node may be April 1, 2025, and the second time node may be July 1, 2025.

[0069] Furthermore, in an embodiment of the present application, a first segmentation window size can be determined based on the distance between the first time node and the current time node, and a second segmentation window size can be determined based on the distance between the second time node and the current time node. Then, the historical statistical data can be segmented in reverse order starting from the current time node, and multiple statistical sequence fragments can be obtained based on the first segmentation window size and the second segmentation window size, respectively.

[0070] For example, assume that the historical statistical data is the sales data of the past year, that is, from January 1, 2024 to December 31, 2024. If the first time node is February 1, 2025, which is one month away from the current time node January 1, 2025, the first segmentation window size can be set to one month; if the second time node is March 1, 2025, which is two months away from the current time node, the second segmentation window size can be set to two months. When the data is segmented in reverse order, you can first start from the current time node January 1, 2025, take the data of the most recent month as the first statistical sequence segment, and then take the data of each subsequent month in turn as a statistical sequence segment until all the data are taken (the last statistical sequence segment can be directly discarded if it is inconsistent with the corresponding segmentation window size), and obtain multiple first statistical sequence segments segmented based on the first segmentation window size. Similarly, based on the second segmentation window size, you can take data every two months starting from the current time node to obtain multiple second statistical sequence segments based on the second segmentation window size. It should be noted that, in the embodiment of the present application, each of the first statistical sequence segments or the second statistical sequence segments may include one or more data.

[0071] Figure 2 This is a schematic diagram of the data processing flow of the intelligent production control of electronic components according to some embodiments of this specification. Figure 2 In the embodiment of the present application, the data segmentation process can be performed by a first segmentation unit and a second segmentation unit. The first segmentation unit can be used to segment the historical statistical data with the first segmentation window size to obtain a plurality of first statistical sequence segments; the second segmentation unit can be used to segment the historical statistical data with the second segmentation window size to obtain a plurality of second statistical sequence segments.

[0072] It should be noted that, in the embodiment of the present application, the first segmentation unit and the second segmentation unit can be understood as software modules or program units, which can automatically perform data segmentation tasks according to a preset segmentation window size. In practical applications, the first segmentation unit and the second segmentation unit can be integrated into the data processing module of the electronic component intelligent production control system as part of the data processing flow. By using these segmentation units, the system can flexibly segment and process historical statistical data to adapt to product demand forecasts at different target time nodes.

[0073] Step S130, perform product demand forecasting based on the multiple statistical sequence segments to obtain forecasted demand of one or more component models included in each electronic component type at multiple target time nodes. In some embodiments, step S130 may be performed by the demand forecasting module 230 mentioned below.

[0074] Continue to refer to Figure 2 In an embodiment of the present application, a plurality of first statistical sequence segments obtained by segmenting based on the first segmentation window size can be input into a first demand forecasting network to obtain a first predicted demand corresponding to one or more component models included in each electronic component type at the first time node; similarly, a plurality of second statistical sequence segments obtained by segmenting based on the second segmentation window size can be input into a second demand forecasting network to obtain a second predicted demand corresponding to one or more component models included in each electronic component type at the second time node.

[0075] It should be noted that, in the embodiment of the present application, the first statistical sequence fragment and the second statistical sequence fragment input into the first demand forecasting network and the second demand forecasting network are obtained by segmenting the historical statistical data corresponding to a component model under an electronic component type. In other words, in the embodiment of the present application, the historical statistical data corresponding to any component model under any electronic component type can be processed in the above manner to predict its first predicted demand at the first time node and the second predicted demand at the second time node (the second predicted demand contains the part corresponding to the first predicted demand).

[0076] Specifically, in an embodiment of the present application, the first demand forecasting network and the second demand forecasting network can be neural network models such as convolutional neural networks, recurrent neural networks, long short-term memory networks, or gated recurrent unit networks. These neural network models can automatically extract features from data by learning historical statistical data and make predictions based on these features. Exemplarily, in an embodiment of the present application, the first demand forecasting network and the second demand forecasting network can include an input layer, a hidden layer, and an output layer; wherein the number of the hidden layers can be one or more.

[0077] In an embodiment of the present application, the input layer can be used to receive corresponding statistical sequence fragments as input data. For example, the input layer of the first demand forecasting network can be used to use multiple first statistical sequence fragments as input data of the first demand forecasting network (neural network model), and the input layer of the second demand forecasting network can be used to use multiple second statistical sequence fragments as input data of the second demand forecasting network (neural network model).

[0078] In an embodiment of the present application, the first demand forecasting network and the second demand forecasting network may respectively include one or more hidden layers, wherein each of the hidden layers may include multiple neurons (basic components of the network) for extracting and transforming features of the input data to capture data change trends and patterns in the statistical sequence fragments. Exemplarily, in some embodiments, the hidden layer may include different types of layer structures such as convolutional layers and pooling layers to further enhance the network's ability to extract features from input data. Among them, the convolutional layer can extract local features of the input data through convolution operations, while the pooling layer can downsample the features output by the convolutional layer, thereby reducing the dimension and amount of calculation of the data while retaining important feature information. It can be understood that through the combination of these hidden layers, the neural network model can learn more complex and abstract data features, thereby more accurately predicting the demand for electronic components.

[0079] The output layer can be used to obtain the predicted demand of one or more component models included in each electronic component type at the corresponding time node based on the output result of the hidden layer. Specifically, in an embodiment of the present application, the output layer of the first demand prediction network can output the first predicted demand of one or more component models included in each electronic component type at the first time node, and the output layer of the second demand prediction network can output the first predicted demand of one or more component models included in each electronic component type at the first time node.

[0080] Figure 3 is a flowchart of training steps according to some embodiments of this specification. Figure 3 In the embodiment of the present application, the above-mentioned demand prediction network (for example, the above-mentioned first demand prediction network and the second demand prediction network) can be trained by the following steps:

[0081] Step S101, obtaining sample historical statistical data corresponding to various types of electronic components, and a first label and a second label corresponding to each piece of sample historical statistical data.

[0082] In an embodiment of the present application, the first label is used to represent the expected demand corresponding to the sample historical statistical data at the first sample time node, and the second label is used to represent the expected demand corresponding to the sample historical statistical data at the second sample time node. The sample historical statistical data can be obtained based on the product sales data of one or more component models included in each electronic component type at multiple sample historical time nodes within a specified sample historical time period.

[0083] Step S102: determining a first sample segmentation window size based on a distance between a first sample time node and a reference time node, and determining a second sample segmentation window size based on a distance between a second sample time node and the reference time node.

[0084] Step S103, segmenting the sample historical statistical data in reverse order starting from the reference time node, and obtaining a plurality of first sample statistical sequence segments based on the first sample segmentation window size, and obtaining a plurality of second sample statistical sequence segments based on the second sample segmentation window size.

[0085] In the embodiment of the present application, the reference time node can be understood as the last time node in the sample historical statistical data. The first sample segmentation window size and the second sample segmentation window size are determined in a similar manner to the first segmentation window size and the second segmentation window size, and the first sample statistical sequence segment and the second sample statistical sequence segment are determined in a similar manner to the first statistical sequence segment and the second statistical sequence segment, which will not be repeated here.

[0086] Step S104: input the multiple first sample statistical sequence fragments as input data into the first demand forecasting network for processing to obtain a first sample forecasting result, and input the multiple second sample statistical sequence fragments as input data into the second demand forecasting network for processing to obtain a second sample forecasting result.

[0087] Step S105, iteratively optimize the parameters of the first demand forecasting network based on the difference between the first sample prediction result and the first label, and iteratively optimize the parameters of the second demand forecasting network based on the difference between the second sample prediction result and the second label, until the training is completed when the preset conditions are met.

[0088] In an embodiment of the present application, the above-mentioned iterative optimization process can be performed using a neural network optimization algorithm such as a gradient descent method, a stochastic gradient descent method or an Adam optimization algorithm. It can be understood that in an embodiment of the present application, the training goal of the first demand forecasting network and the second demand forecasting network is to make the difference between the prediction result (such as the first sample prediction result and the second sample prediction result) and the actual label (such as the first label and the second label) as small as possible, and the process can be achieved by minimizing the loss function (such as the difference between the prediction result and the actual label). The preset condition may be that the loss function value is less than a preset threshold, or the number of iterations reaches a preset upper limit, etc. It should be noted that in an embodiment of the present application, through the training process, the first demand forecasting network and the second demand forecasting network can have the ability to accurately predict the demand for one or more component models contained in each electronic component type.

[0089] More details about the above-mentioned demand forecasting network training can be regarded as prior art and will not be described in detail in this specification.

[0090] In an embodiment of the present application, after completing the training process of the first demand prediction network and the second demand prediction network, they can be used to process the historical statistical data corresponding to any target component model contained in any electronic component type, so as to obtain the first predicted demand corresponding to the first time node and the second predicted demand corresponding to the second time node.

[0091] Step S140, determining a target production plan according to the predicted demand at the multiple target time nodes and the order data corresponding to the multiple target time nodes, and performing production control based on the target production plan. In some embodiments, step S140 may be performed by the production plan determination module 240 mentioned below.

[0092] In an embodiment of the present application, for any target component model included in any electronic component type, the first predicted demand corresponding to the first time node and the second predicted demand corresponding to the second time node can be obtained through the above steps. Further, the error rate between the first predicted demand corresponding to the target component model at the first time node and the order quantity can be calculated. Exemplarily, in some embodiments, the calculation process of the error rate can be expressed as follows:

[0093]

[0094] Wherein, ω represents the error rate between the first predicted demand and order quantity corresponding to the target component model at the first time node; y' 1 represents the first predicted demand volume of the target component model at the first time node; 1 Indicates the order quantity corresponding to the target component model at the first time node. When ω is greater than 0, it means that the first predicted demand corresponding to the target component model at the first time node is greater than the actual order quantity; when ω is less than 0, it means that the first predicted demand corresponding to the target component model at the first time node is less than the actual order quantity.

[0095] It should be noted that, in this specification, the order quantity corresponding to the target component model at the first time node refers to the number of products corresponding to the target component model in the orders that need to be completed from the current time node to the first time node.

[0096] Further, after obtaining the error rate, the second predicted demand corresponding to the second time node can be corrected based on the error rate, and the target production volume corresponding to the target component model at the second time node can be determined based on the corrected data. Exemplarily, in some embodiments, a correction coefficient can be determined based on the error rate, and the product of the second predicted demand corresponding to the second time node and the correction coefficient is used as the corrected data. The process can be expressed as follows:

[0097] Y′ 2 =(1-ω)*y′ 2

[0098] Among them, Y' 2 represents the corrected data obtained after the second predicted demand is corrected; ω represents the error rate between the first predicted demand and the order quantity corresponding to the target component model at the first time node, (1-ω) represents the correction coefficient; y' 2 Indicates the second predicted demand corresponding to the target component model at the second time node. When ω is greater than 0, the corrected data Y' 2 Less than y' 2 , when ω is less than 0, the corrected data Y' 2 Greater than y' 2 .

[0099] The corrected data Y' obtained after correcting the second predicted demand 2Afterwards, the completed production of the target component model as of the first time node (the completed production is usually greater than or equal to the order quantity as of the first time node) can be obtained, and then the target production volume planned to be completed during the period from the first time node to the second time node can be determined based on the difference between the corrected data and the completed production.

[0100] For example, assuming that the target component model has completed production volume Y3 as of the first time node, the target production volume Y4 planned to be completed during the period from the first time node to the second time node can be expressed as Y4=Y' 2 -Y3.

[0101] It should be pointed out that in the embodiment of the present application, by determining the correction coefficient based on the error rate between the first predicted demand and the order quantity corresponding to the target component model at the first time node, the actual market situation and order changes can be better reflected, so as to adaptively adjust the second predicted demand corresponding to the second time node based on the actual situation, thereby improving the accuracy of the demand forecast result to a certain extent. Furthermore, in the embodiment of the present application, by correcting the second predicted demand corresponding to the second time node according to the correction coefficient, and then determining the target production volume corresponding to the target component model at the second time node based on the corrected data, and performing production control based on this, the accuracy of the production plan can be effectively improved, thereby reducing the risk of overproduction or underproduction.

[0102] In some embodiments of the present application, considering that the order quantity that needs to be completed by the first time node may be updated before the first time node is reached, based on this, in some embodiments, in order to determine the accuracy of the error rate, the error rate can be updated according to the change of the order quantity before the first time node is reached. Accordingly, in the embodiments of the present application, the calculation result of the target production volume can also be updated accordingly according to the updated error rate.

[0103] Figure 4 This is a schematic diagram of the data processing flow of the intelligent production control of electronic components according to other embodiments of this specification. Figure 4In some embodiments of the present application, a third segmentation unit and a third demand forecasting network can be added on the basis of the above-mentioned first segmentation unit, the first demand forecasting network, the second segmentation unit, and the second demand forecasting network. Among them, the third segmentation unit can be used to segment the historical statistical data with a third segmentation window size (the third segmentation window size is determined based on the distance between the third time period and the current time node, and the third time period is located after the second time node) to obtain multiple third statistical sequence fragments, and the third demand forecasting network can be used to predict the third predicted demand corresponding to the target component model at the third time node based on the third statistical sequence fragments (the third predicted demand includes the part corresponding to the first predicted demand and the second predicted demand). The structure and training method of the third demand forecasting network can refer to the first demand forecasting network and the second demand forecasting network, and will not be repeated here.

[0104] Similarly, in some embodiments of the present application, a correction coefficient can be determined based on the above method, and then the third predicted demand can be corrected based on the correction coefficient, and finally the target production volume corresponding to the target component model at the third time node can be determined based on the corrected data.

[0105] It can be understood that by adding a third segmentation unit and a third demand forecasting network to process historical statistical data to determine the third predicted demand corresponding to the target component model at the third time node, and then correcting the third predicted demand in the above manner, and based on this, determining the target production volume corresponding to the target component model at the third time node, it is possible to accurately predict long-term demand, thereby facilitating manufacturers to prepare inventory.

[0106] Figure 5 is a schematic diagram of a module of an electronic component intelligent production control system according to some embodiments of this specification. In some embodiments, Figure 5 The electronic component intelligent production control system 200 shown can be implemented in software and / or hardware. For example, it can be configured in the form of software and / or hardware to a processing device and / or a terminal device to process historical statistical data corresponding to various types of electronic components to determine a target production plan.

[0107] Reference Figure 5 In the embodiment of the present application, the electronic component intelligent production control system 200 may include an acquisition module 210, a data segmentation module 220, a demand forecasting module 230 and a production plan determination module 240. Among them:

[0108] The acquisition module 210 can be used to acquire historical statistical data corresponding to various types of electronic components, and the historical statistical data is obtained based on product sales data of one or more component models included in each electronic component type at multiple historical time nodes within a specified historical time period.

[0109] The data segmentation module 220 may be used to segment the historical statistical data to obtain a plurality of statistical sequence segments.

[0110] The demand quantity prediction module 230 may be used to perform product demand quantity prediction based on the multiple statistical sequence segments, and obtain predicted demand quantities of one or more component models included in each electronic component type at multiple target time nodes.

[0111] The production plan determination module 240 can be used to determine a target production plan according to the predicted demand quantities of the multiple target time nodes and the order data corresponding to the multiple target time nodes, and perform production control based on the target production plan.

[0112] For more details about the above modules, please refer to other places in this manual (for example Figure 1 to Figure 3 part and its related description), which will not be repeated here.

[0113] It should be understood that Figure 5 The electronic component intelligent production control system 200 and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above-mentioned system and its control method can be implemented using computer executable instructions and / or included in a processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of this specification can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software (for example, firmware).

[0114] It should be noted that the above description of the electronic component intelligent production control system 200 is provided for illustrative purposes only and is not intended to limit the scope of this specification. It is understandable that for those skilled in the art, according to the description of this specification, without departing from this principle, the various modules can be arbitrarily combined, or a subsystem can be formed to connect with other modules. For example, Figure 2 The acquisition module 210, data segmentation module 220, demand forecasting module 230 and production plan determination module 240 described above can be different modules in a system, or one module can realize the functions of two or more modules. Such variations are within the protection scope of this specification.

[0115] In summary, the beneficial effects that may be brought about by the embodiments of this specification include but are not limited to: (1) In the electronic component intelligent production control system and control method provided in some embodiments of this specification, by segmenting the historical statistical data to obtain multiple statistical sequence segments, and then based on the multiple statistical sequence segments, the product demand is predicted to obtain the predicted demand of one or more component models included in each electronic component type at multiple target time nodes, and finally the target production plan is determined according to the predicted demand at the multiple target time nodes and the order data corresponding to the multiple target time nodes, and production control is performed based on this, which can effectively improve the accuracy of the production plan and thus reduce the risk of overproduction or underproduction; (2) In the electronic component intelligent production control system and control method provided in some embodiments of this specification, by segmenting the historical statistical data to obtain multiple statistical sequence segments, and then based on the multiple statistical sequence segments, the product demand is predicted to obtain the predicted demand of one or more component models included in each electronic component type at multiple target time nodes, and finally the target production plan is determined according to the predicted demand at the multiple target time nodes and the order data corresponding to the multiple target time nodes, and production control is performed based on this, which can effectively improve the accuracy of the production plan and thus reduce the risk of overproduction or underproduction; (2) By segmenting to obtain multiple statistical sequence segments, and then further processing and analyzing these statistical sequence segments, the trend characteristics and periodic laws of electronic component sales can be extracted, thereby providing more accurate data support for subsequent production planning and inventory control. At the same time, the historical statistical data can be flexibly segmented and processed based on predicted demand to adapt to product demand forecasts at different target time nodes; (3) In the electronic component intelligent production control system and control method provided in some embodiments of the present specification, by determining the correction coefficient based on the error rate between the first predicted demand and the order quantity corresponding to the target component model at the first time node, the actual market situation and order changes can be better reflected, so that the second predicted demand corresponding to the second time node can be adaptively adjusted based on the actual situation, thereby improving the accuracy of the demand forecast results to a certain extent.

[0116] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or may be any other possible beneficial effects.

[0117] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0118] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0119] In addition, it will be understood by those skilled in the art that various aspects of this specification may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of this specification may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of this specification may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0120] A computer storage medium may include a propagated data signal containing computer program code, for example, in baseband or as part of a carrier wave. The propagated signal may be in a variety of forms, including electromagnetic, optical, etc., or a suitable combination. A computer storage medium may be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit the program for use. The program code on the computer storage medium may be transmitted via any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0121] The computer program code required for the operation of each part of this specification can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C language, Visual Basic, Fortran2003, Perl, COBOL2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby and Groovy, or other programming languages, etc. The program code can be run entirely on the user's computer, or run on the user's computer as an independent software package, or run partially on the user's computer and partially on a remote computer, or run entirely on a remote computer or processing device. In the latter case, the remote computer can be connected to the user's computer through any network form, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as software as a service (SaaS).

[0122] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing processing device or mobile device.

[0123] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0124] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0125] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. An intelligent production control method for electronic components, characterized in that: include: Obtaining historical statistical data corresponding to various types of electronic components, the historical statistical data being obtained based on product sales data of one or more component models included in each type of electronic component at multiple historical time nodes within a specified historical time period; Performing data segmentation on the historical statistical data to obtain a plurality of statistical sequence segments; Perform product demand forecasting based on the multiple statistical sequence fragments to obtain forecasted demand for one or more component models included in each electronic component type at multiple target time nodes; A target production plan is determined according to the predicted demand quantities at the multiple target time nodes and the order data corresponding to the multiple target time nodes, and production control is performed based on the target production plan.

2. The electronic component intelligent production control method according to claim 1, characterized in that: The target time node includes at least a first time node and a second time node, wherein the second time node is located after the first time node; the data segmentation of the historical statistical data to obtain a plurality of statistical sequence segments includes: Determine a first segmentation window size based on a distance between the first time node and the current time node; Determine a second split window size based on a distance between the second time node and the current time node; The historical statistical data is segmented in reverse order starting from the current time node, and a plurality of statistical sequence segments are obtained based on the first segmentation window size and the second segmentation window size.

3. The electronic component intelligent production control method according to claim 2, characterized in that: The product demand forecasting based on the multiple statistical sequence segments to obtain the forecasted demand of one or more component models included in each electronic component type at multiple target time nodes includes: Inputting a plurality of first statistical sequence segments obtained by segmentation based on the first segmentation window size into a first demand quantity prediction network to obtain a first predicted demand quantity corresponding to one or more component models included in each electronic component type at the first time node; The plurality of second statistical sequence segments obtained by segmentation based on the second segmentation window size are input into a second demand quantity prediction network to obtain a second predicted demand quantity corresponding to one or more component models included in each electronic component type at the second time node.

4. The electronic component intelligent production control method according to claim 3, characterized in that: The first demand forecasting network and the second demand forecasting network include an input layer, a hidden layer and an output layer; wherein, The input layer is used to receive corresponding statistical sequence fragments as input data; The hidden layer includes a plurality of neurons for extracting and transforming features of the input data to capture data variation trends and patterns in the statistical sequence segments; The output layer is used to obtain the predicted demand for one or more component models included in each electronic component type at a corresponding time node based on the output results of the hidden layer.

5. The electronic component intelligent production control method according to claim 3, characterized in that: The first demand forecasting network and the second demand forecasting network are trained in the following manner: Obtain sample historical statistical data corresponding to each type of electronic components, and a first label and a second label corresponding to each piece of sample historical statistical data, wherein the first label is used to indicate the expected demand corresponding to the sample historical statistical data at a first sample time node, and the second label is used to indicate the expected demand corresponding to the sample historical statistical data at a second sample time node, and the sample historical statistical data is obtained based on product sales data of one or more component models included in each electronic component type at multiple sample historical time nodes within a specified sample historical time period; Determine a first sample segmentation window size based on a distance between the first sample time node and a reference time node, and determine a second sample segmentation window size based on a distance between the second sample time node and the reference time node; Performing data segmentation on the sample historical statistical data in reverse order starting from the reference time node, and obtaining a plurality of first sample statistical sequence segments based on the first sample segmentation window size, and obtaining a plurality of second sample statistical sequence segments based on the second sample segmentation window size; Inputting the plurality of first sample statistical sequence fragments as input data into the first demand forecasting network for processing to obtain a first sample forecasting result, and inputting the plurality of second sample statistical sequence fragments as input data into the second demand forecasting network for processing to obtain a second sample forecasting result; The parameters of the first demand forecasting network are iteratively optimized based on the difference between the first sample prediction result and the first label, and the parameters of the second demand forecasting network are iteratively optimized based on the difference between the second sample prediction result and the second label, until the training is completed when the preset conditions are met.

6. The electronic component intelligent production control method according to claim 3, characterized in that: The determining the target production plan according to the predicted demand quantities at the multiple target time nodes and the order data corresponding to the multiple target time nodes includes: For any target component model contained in any electronic component type, Calculate the error rate between the first predicted demand quantity and the order quantity corresponding to the target component model at the first time node; The second predicted demand corresponding to the second time node is corrected based on the error rate, and the target production volume corresponding to the target component model at the second time node is determined based on the corrected data.

7. The electronic component intelligent production control method according to claim 6, characterized in that: The calculating the error rate between the first predicted demand quantity and the order quantity corresponding to the target component model at the first time node includes: Before reaching the first time node, the error rate is updated according to the change of the order quantity; wherein the error rate is calculated in the following manner: Among them, ω represents the error rate between the first predicted demand and the order quantity corresponding to the target component model at the first time node; y'1 represents the first predicted demand corresponding to the target component model at the first time node; y1 represents the order quantity corresponding to the target component model at the first time node.

8. The electronic component intelligent production control method according to claim 6, characterized in that: The correcting the second predicted demand corresponding to the second time node based on the error rate includes: A correction coefficient is determined according to the error rate, and the product of the second predicted demand corresponding to the second time node and the correction coefficient is used as the corrected data.

9. The electronic component intelligent production control method according to claim 8, characterized in that: The determining, based on the corrected data, the target production volume corresponding to the target component model at the second time node includes: Obtaining the completed production of the target component model as of the first time node; The target production volume planned to be completed during the period from the first time node to the second time node is determined based on the difference between the corrected data and the completed production volume.

10. An intelligent production control system for electronic components, characterized in that: include: An acquisition module is used to acquire historical statistical data corresponding to various types of electronic components, wherein the historical statistical data is obtained based on product sales data of one or more component models included in each type of electronic component at multiple historical time nodes within a specified historical time period; A data segmentation module, used for performing data segmentation on the historical statistical data to obtain multiple statistical sequence segments; A demand forecasting module, used to forecast product demand based on the multiple statistical sequence fragments, and obtain forecast demand of one or more component models included in each electronic component type at multiple target time nodes; The production plan determination module is used to determine the target production plan according to the predicted demand at the multiple target time nodes and the order data corresponding to the multiple target time nodes, and perform production control based on the target production plan.

Images