Demand prediction method and system and related equipment
By integrating and preprocessing the internal and external data of the demand objects, and using the regression model to predict demand data, the problem of difficulty in material consumption monitoring is solved, and accurate prediction of demand data and effective guidance of material procurement is achieved.
Patent Information
- Application Number
- CN202410205976.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-23
- Publication Date
- 2025-03-14
AI Technical Summary
In a highly competitive and globalized manufacturing and supply chain environment, it is difficult to monitor material consumption in real time, resulting in uncertainty in the implementation of production plans and insufficient inventory.
By obtaining the internal and external data of the demand object, integrating it into a unified data set, preprocessing and data correlation analysis are performed, and a regression model is used to predict demand data and resource data based on correlation coefficients.
Accurate prediction of demand data is achieved, and guidance is provided for material procurement, avoid production line shutdown caused by insufficient inventory, and improve the reliability of production plans.
Smart Images
Figure CN119623892A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of computer technology, and specifically to a demand forecasting method, system, and related equipment. Background Art
[0002] In today's highly competitive and globalized manufacturing and supply chain environment, real-time monitoring of material consumption has become a huge challenge. Due to the complexity of the production process and the pursuit of efficiency, it is often difficult to monitor material consumption in a timely and accurate manner. This not only brings certain uncertainties to the execution of production plans, but also easily leads to production line shutdowns caused by insufficient inventory of a certain material, affecting the normal operation of the enterprise.
[0003] Therefore, how to achieve accurate prediction of demand data and provide guidance for material procurement has become a technical problem that technical personnel in this field urgently need to solve. Summary of the invention
[0004] In view of this, the embodiments of the present application provide a demand forecasting method, system and related equipment to achieve accurate forecasting of demand data and provide guidance for the procurement of materials.
[0005] To achieve the above objectives, the embodiments of the present application provide the following technical solutions.
[0006] In a first aspect, an embodiment of the present application provides a demand forecasting method, comprising:
[0007] Acquire internal data and external data of the required object, and integrate the internal data and the external data into the same data set;
[0008] Preprocessing the data set to obtain data to be processed related to the demand object; wherein the data to be processed has multiple data types;
[0009] Performing data correlation analysis on the data to be processed to obtain correlation coefficients between different data types;
[0010] Based on the correlation coefficient, demand data of the demand object and resource data corresponding to the demand object are predicted.
[0011] Optionally, the step of predicting the demand data of the demand object and the resource data corresponding to the demand object based on the correlation coefficient is specifically: using a regression model, taking the data type with a correlation coefficient greater than a preset value as the independent variable of the regression model, and predicting the demand data of the demand object and the resource data corresponding to the demand object.
[0012] Optionally, after the step of adopting a regression model and taking a data type with a correlation coefficient greater than a preset value as an independent variable of the regression model to predict the demand data of the demand object and the resource data corresponding to the demand object, the step further includes:
[0013] Evaluating the regression model by cross-validation to test the performance of the regression model;
[0014] The performance of the regression model is continuously monitored, and the regression model is updated periodically.
[0015] Optionally, in the step of evaluating the regression model by cross-validation and testing the performance of the regression model, the evaluation index for evaluating the regression model includes:
[0016] The mean square error, root mean square error, mean absolute error and R-squared value of the prediction results.
[0017] Optionally, the step of obtaining internal data and external data of the demand object and integrating the internal data and the external data into the same data set includes:
[0018] Acquire the internal data from a database;
[0019] Obtain said external data from the website;
[0020] Performing data conversion on the internal data and the external data, wherein the data conversion includes data type conversion and data format conversion;
[0021] Combine transformed internal and external data into the same dataset.
[0022] Optionally, the database is one or more of an enterprise resource planning system, a warehouse management system, and a manufacturing execution system.
[0023] Optionally, the internal data includes one or more of historical purchase data, historical sales data, cost data, supplier performance data, specification and characteristic data, production data, quality control data, equipment utilization data and inventory data of the demand object;
[0024] The external data includes one or more of currency exchange rate data, weather data, global economic indicator data, logistics cost data and news hotspot data.
[0025] Optionally, the preprocessing of the data set includes:
[0026] Clean the data in the data set, filter out the data to be processed that is related to the required object, and remove other irrelevant data;
[0027] Deduplication of the data in the data set is performed to find out whether there is duplicate data, and if there is duplicate data, the duplicate data is deleted;
[0028] Processing missing values in the data set, calculating the mean or median before and after the missing value, and filling the missing value with the calculated mean or median, or deleting data related to the missing value;
[0029] Processing outliers in the data set, and deleting the outliers.
[0030] In a second aspect, an embodiment of the present application provides a demand forecasting system, including:
[0031] A data collection module, used to obtain internal data and external data of the demand object, and integrate the internal data and the external data into the same data set;
[0032] A data preprocessing module, used to preprocess the data set to obtain the data to be processed related to the demand object; wherein the data to be processed has multiple data types;
[0033] A data analysis module is used to perform data correlation analysis on the data to be processed to obtain correlation coefficients between different data types;
[0034] A prediction module is used to predict the demand data of the demand object and the resource data corresponding to the demand object based on the correlation coefficient.
[0035] In a third aspect, an embodiment of the present application provides an electronic device comprising at least one memory and at least one processor, wherein the memory stores one or more computer executable instructions, and the processor calls the one or more computer executable instructions to execute the demand forecasting method as described in the first aspect above.
[0036] In a fourth aspect, an embodiment of the present application provides a storage medium, which stores one or more computer-executable instructions. When the one or more computer-executable instructions are executed, the demand forecasting method described in the first aspect above is implemented.
[0037] Compared with the prior art, the technical solution of the embodiment of the present application has the following advantages:
[0038] The embodiments of the present application provide a demand forecasting method, system and related equipment, wherein the demand forecasting method includes: obtaining internal data and external data of a demand object, and integrating the internal data and the external data into the same data set; preprocessing the data set to obtain data to be processed related to the demand object; wherein the data to be processed has multiple data types; performing data correlation analysis on the data to be processed to obtain correlation coefficients between different data types; and based on the correlation coefficient, predicting the demand data of the demand object and the resource data corresponding to the demand object.
[0039] It can be seen that the demand forecasting method provided in the embodiment of the present application obtains a data set by acquiring the internal data and external data of the demand object, and preprocesses the data set to obtain the data to be processed related to the demand object, performs data correlation analysis on the data to be processed, obtains the correlation coefficient between different data types, and finally predicts the demand data of the demand object and the resource data corresponding to the demand object based on the correlation coefficient. Through the above-mentioned data processing, accurate prediction of demand data is achieved, providing guidance for the purchase of materials, so that purchasing personnel can purchase materials according to the prediction results, avoiding the production line shutdown caused by insufficient inventory of a certain material. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technical personnel in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0041] Figure 1 is a flow chart of the demand forecasting method provided in an embodiment of the present application;
[0042] Figure 2 is an optional flow chart of step S1 in the demand forecasting method provided in the embodiment of the present application;
[0043] Figure 3 is another flow chart of the demand forecasting method provided in the embodiment of the present application;
[0044] Figure 4 It is a structural diagram of the demand forecasting system provided in the embodiment of the present application. DETAILED DESCRIPTION
[0045] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technical staff in this field without creative work are within the scope of protection of this application.
[0046] As described in the background technology, in today's highly competitive and globalized manufacturing and supply chain environment, real-time monitoring of material consumption has become a huge challenge. Due to the complexity of the production process and the pursuit of efficiency, it is often difficult to monitor material consumption in a timely and accurate manner. This not only brings certain uncertainties to the execution of production plans, but also easily leads to production line shutdowns caused by insufficient inventory of a certain material, affecting the normal operation of the enterprise.
[0047] It can be seen that it is particularly necessary to provide a solution for accurate prediction of demand data.
[0048] In view of this, an embodiment of the present application provides a demand forecasting method, system and related equipment, wherein the demand forecasting method includes: obtaining internal data and external data of a demand object, and integrating the internal data and the external data into the same data set; preprocessing the data set to obtain data to be processed related to the demand object; wherein the data to be processed has multiple data types; performing data correlation analysis on the data to be processed to obtain correlation coefficients between different data types; based on the correlation coefficient, predicting the demand data of the demand object and the resource data corresponding to the demand object.
[0049] It can be seen that the demand forecasting method provided in the embodiment of the present application obtains a data set by acquiring the internal data and external data of the demand object, and preprocesses the data set to obtain the data to be processed related to the demand object, performs data correlation analysis on the data to be processed, obtains the correlation coefficient between different data types, and finally predicts the demand data of the demand object and the resource data corresponding to the demand object based on the correlation coefficient. Through the above-mentioned data processing, accurate prediction of demand data is achieved, providing guidance for the purchase of materials, so that purchasing personnel can purchase materials according to the prediction results, avoiding the production line shutdown caused by insufficient inventory of a certain material.
[0050] The demand forecasting method provided in the embodiments of the present application is described in detail below in conjunction with the embodiments of the present application.
[0051] refer to Figure 1 , Figure 1 The flowchart of the demand forecasting method provided by the embodiment of the present application is shown as an example. Figure 1 As shown, the demand forecasting method comprises:
[0052] Step S1: Acquire internal data and external data of a demand object, and integrate the internal data and the external data into the same data set.
[0053] The internal data and the external data are integrated into the same data set, ensuring that the timestamps (dates) between the data match so as to establish a time series relationship.
[0054] By integrating the internal data and the external data, a unified data set can be formed, which is convenient for subsequent data processing and data analysis. At the same time, the integrated data is more traceable. If abnormal data appears, the source of the problem can be tracked and found in time, providing guarantee for the accuracy and reliability of the data.
[0055] refer to Figure 2 , step S1 may include:
[0056] Step S11: Acquire the internal data from the database.
[0057] The internal data includes one or more of the historical purchase data, historical sales data, cost data, supplier performance data, specification and characteristic data, production data, quality control data, equipment utilization data and inventory data of the demand object.
[0058] Step S12: Acquire the external data from the website.
[0059] In a specific example of obtaining external data from a website, the external data may be obtained by using scrapy crawler technology, wherein the external data includes one or more of currency exchange rate data, weather data, global economic indicator data, logistics cost data, and news hotspot data.
[0060] Step S13: performing data conversion on the internal data and the external data.
[0061] The data conversion includes data type conversion and data format conversion. The data type conversion may be, for example, converting text type data into numeric type data, and the data format conversion may be, for example, converting EXCEL spreadsheet data into CSV (Comma-Separated Values) format data.
[0062] Step S14: Integrate the converted internal data and external data into the same data set.
[0063] Integrate the converted internal and external data into the same data set to form a unified data set for subsequent data processing and analysis. At the same time, the integrated data is more traceable. If abnormal data occurs, the source of the problem can be tracked and found in time, providing guarantee for the accuracy and reliability of the data.
[0064] Step S2: pre-process the data set to obtain the data to be processed related to the demand object.
[0065] The data to be processed has multiple data types.
[0066] The preprocessing of the data set may include: cleaning the data in the data set, screening out the data to be processed related to the required object, and removing other irrelevant data.
[0067] And / or, deduplication is performed on the data in the data set to find out whether there is duplicate data, and if there is duplicate data, the duplicate data is deleted. For example, when the historical purchase data of the demand object is processed, if there are multiple identical price records in the same time period, they are regarded as duplicate data and deleted.
[0068] And / or, processing missing values in the data set, calculating the mean or median before and after the missing values, and filling the missing values with the calculated mean or median, that is, backfilling directly based on the mean and median. For example, if the inventory data of a certain material in a certain time period is missing, the average of the inventory data of the material before and after this period can be used to backfill the data. For some data with missing values, it can also be processed by deleting the data related to the missing values. For example: If the inventory data of a certain material cannot find its corresponding procurement data, then in order to ensure the accuracy of the prediction, the inventory data can be directly deleted.
[0069] And / or, process the outliers in the data set and delete the outliers. Deleting outliers is an important step in processing data that deviates from the normal situation, which can avoid the negative impact of outliers on data analysis. For example, if a certain material has multiple different prices in the same time period, this is an abnormal situation. We need to trace the source of these abnormal data, understand the reasons for the abnormality, and delete the abnormal price data.
[0070] Step S3: Perform data correlation analysis on the data to be processed to obtain correlation coefficients between different data types.
[0071] Perform data correlation analysis on the data to be processed and calculate the correlation coefficients between different data types. For example, when analyzing the correlation coefficient between semiconductor industry news hot spots and wafer raw material prices, the news hot spots data is divided into three levels by month:
[0072] When more than 75% of the hot news data in a month tend to be positive news, it is recorded as "good news" and quantified as 3; when 25%-75% of the hot news data in a month tend to be positive news, it is recorded as "no special news" and quantified as 2; when less than 25% of the news in a month tend to be positive news, it is recorded as "bad news" and quantified as 1.
[0073] The price of wafer raw materials, when the price increases in the month, it is recorded as "increase" and quantified as 1; when the price decreases in the month, it is recorded as "decrease" and quantified as 0.
[0074] Then, the correlation coefficient between the hot news in the semiconductor industry and the price of wafer raw materials is calculated.
[0075] Step S4: Based on the correlation coefficient, predict the demand data of the demand object and the resource data corresponding to the demand object.
[0076] Step S4 specifically includes: adopting a regression model, taking the data type with a correlation coefficient greater than a preset value as the independent variable of the regression model, and predicting the demand data of the demand object and the resource data corresponding to the demand object.
[0077] refer to Figure 3 , Figure 3 Another flow chart of the demand forecasting method provided by the embodiment of the present application is shown as an example. Figure 3 As shown, the step of adopting a regression model, taking the data type with a correlation coefficient greater than a preset value as the independent variable of the regression model, and predicting the demand data of the demand object and the resource data corresponding to the demand object further includes:
[0078] Step S5: Evaluate the regression model by cross-validation to test the performance of the regression model.
[0079] Among them, the evaluation indicators for evaluating the regression model may include: mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE) and R-squared of the prediction results to evaluate the prediction performance of the model.
[0080] The smaller the mean square error, root mean square error, and mean absolute error, the more accurate the prediction results are. The closer the R-squared value is to 1, the more variance of the dependent variable can be explained in the model, and the better the model fit is.
[0081] Step S6: Continuously monitor the performance of the regression model and regularly update the regression model.
[0082] In today's highly competitive and globalized manufacturing and supply chain environment, the market changes rapidly and data is frequently updated. Therefore, the performance of the model needs to be evaluated regularly to ensure that it can adapt to the changing market conditions. If the performance of the model declines, it needs to be adjusted in time to maintain its prediction accuracy for the demand data of the demand object and the resource data corresponding to the demand object.
[0083] Secondly, if other variables are found in actual operations that may affect the prediction results, the model also needs to be updated in a timely manner. For example, if a new production line has recently appeared or a new production technology has been adopted, these factors need to be added to the model and the model needs to be updated. In addition, if the size of the data set has changed significantly, the model also needs to be updated. For example, when the number of customers or sales of a company changes significantly, the model needs to be updated to maintain its accuracy.
[0084] It can be seen that the demand forecasting method provided in the embodiment of the present application obtains a data set by acquiring the internal data and external data of the demand object, and preprocesses the data set to obtain the data to be processed related to the demand object, performs data correlation analysis on the data to be processed, obtains the correlation coefficient between different data types, and finally predicts the demand data of the demand object and the resource data corresponding to the demand object based on the correlation coefficient. Through the above-mentioned data processing, accurate prediction of demand data is achieved, providing guidance for the purchase of materials, so that purchasing personnel can purchase materials according to the prediction results, avoiding the production line shutdown caused by insufficient inventory of a certain material.
[0085] The present application also provides a demand forecasting system. Figure 4 , Figure 4 The schematic diagram of the structure of the demand forecasting system provided by the embodiment of the present application is shown as an example. Figure 4 As shown, the demand forecasting system includes: a data collection module 100, a data preprocessing module 101, a data analysis module 102 and a forecasting module 103.
[0086] The data collection module 100 is used to obtain the internal data and external data of the demand object, and integrate the internal data and the external data into the same data set.
[0087] The data preprocessing module 101 is used to preprocess the data set to obtain the data to be processed related to the demand object; wherein the data to be processed has multiple data types.
[0088] The data analysis module 102 is used to perform data correlation analysis on the data to be processed to obtain correlation coefficients between different data types.
[0089] The prediction module 103 is used to predict the demand data of the demand object and the resource data corresponding to the demand object based on the correlation coefficient.
[0090] An embodiment of the present application provides an electronic device, comprising at least one memory and at least one processor, wherein the memory stores one or more computer executable instructions, and the processor calls the one or more computer executable instructions to execute the demand forecasting method as described above.
[0091] An embodiment of the present application also provides a storage medium, which stores one or more computer-executable instructions. When the one or more computer-executable instructions are executed, the demand forecasting method described above is implemented.
[0092] The above describes multiple implementation schemes provided by the embodiments of the present application. The various optional methods introduced in each implementation scheme can be combined and cross-referenced with each other without conflict, thereby extending a variety of possible implementation schemes, which can all be considered as implementation schemes disclosed and open in the embodiments of the present application.
[0093] Although the embodiments of the present application are disclosed above, the present application is not limited thereto. Any person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be subject to the scope defined by the claims.
Claims
1. A demand forecasting method, characterized in that: include: Acquire internal data and external data of the required object, and integrate the internal data and the external data into the same data set; Preprocessing the data set to obtain data to be processed related to the demand object; wherein the data to be processed has multiple data types; Performing data correlation analysis on the data to be processed to obtain correlation coefficients between different data types; Based on the correlation coefficient, demand data of the demand object and resource data corresponding to the demand object are predicted.
2. The demand forecasting method according to claim 1, characterized in that: The step of predicting the demand data of the demand object and the resource data corresponding to the demand object based on the correlation coefficient is specifically: using a regression model, taking the data type with a correlation coefficient greater than a preset value as the independent variable of the regression model, and predicting the demand data of the demand object and the resource data corresponding to the demand object.
3. The demand forecasting method according to claim 2, characterized in that: After the step of adopting a regression model and taking the data type with a correlation coefficient greater than a preset value as the independent variable of the regression model to predict the demand data of the demand object and the resource data corresponding to the demand object, the method further includes: Evaluating the regression model by cross-validation to test the performance of the regression model; The performance of the regression model is continuously monitored, and the regression model is updated periodically.
4. The demand forecasting method according to claim 3, characterized in that: In the step of evaluating the regression model by cross-validation and testing the performance of the regression model, the evaluation indicators for evaluating the regression model include: The mean square error, root mean square error, mean absolute error and R-squared value of the prediction results.
5. The demand forecasting method according to claim 1, characterized in that: The step of obtaining the internal data and external data of the demand object and integrating the internal data and the external data into the same data set includes: Acquire the internal data from a database; Obtain said external data from the website; Performing data conversion on the internal data and the external data, wherein the data conversion includes data type conversion and data format conversion; Combine transformed internal and external data into the same dataset.
6. The demand forecasting method according to claim 5, characterized in that: The database is one or more of an enterprise resource planning system, a warehouse management system, and a manufacturing execution system.
7. The demand forecasting method according to claim 1, characterized in that: The internal data includes one or more of the historical purchase data, historical sales data, cost data, supplier performance data, specification and characteristic data, production data, quality control data, equipment utilization data and inventory data of the demand object; The external data includes one or more of currency exchange rate data, weather data, global economic indicator data, logistics cost data and news hotspot data.
8. The demand forecasting method according to claim 1, characterized in that: The preprocessing of the data set includes: Clean the data in the data set, filter out the data to be processed that is relevant to the required object, and remove other irrelevant data; and / or, Deduplication of the data in the data set is performed to find out whether there is duplicate data, and if there is duplicate data, the duplicate data is deleted; and / or, Processing missing values in the data set, calculating the mean or median before and after the missing value, and filling the missing value with the calculated mean or median, or deleting data related to the missing value; and / or, Processing outliers in the data set, and deleting the outliers.
9. A demand forecasting system, characterized in that: include: A data collection module, used to obtain internal data and external data of the demand object, and integrate the internal data and the external data into the same data set; A data preprocessing module, used to preprocess the data set to obtain the data to be processed related to the demand object; wherein the data to be processed has multiple data types; A data analysis module is used to perform data correlation analysis on the data to be processed to obtain correlation coefficients between different data types; A prediction module is used to predict the demand data of the demand object and the resource data corresponding to the demand object based on the correlation coefficient.
10. An electronic device, characterized in that: It comprises at least one memory and at least one processor, wherein the memory stores one or more computer executable instructions, and the processor calls the one or more computer executable instructions to execute the demand forecasting method as described in any one of claims 1-8.
11. A storage medium, characterized in that: The storage medium stores one or more computer executable instructions, and when the one or more computer executable instructions are executed, the demand forecasting method according to any one of claims 1 to 8 is implemented.