CN-121998150-A - Productivity prediction method, device, storage medium and equipment

CN121998150ACN 121998150 ACN121998150 ACN 121998150ACN-121998150-A

Abstract

A productivity prediction method, device, storage medium and equipment. The method comprises the steps of obtaining product demand data of a current site, WPH data of each product at the current site and historical WIP data of each product at the current site, carrying out first prediction on future WIP data of the current site based on the product demand data, the WPH data of each product at the current site and the historical WIP data of each product at the current site to obtain a first prediction result of the future WIP data of the current site, and obtaining a second prediction result of the future WIP data of the current site based on the product demand data and the first prediction result by using a preset prediction model to predict the future productivity of the current site based on the second prediction result. By adopting the scheme, the accuracy of productivity prediction can be improved.

Inventors

LIU GUOYANG
ZHOU YIZHONG
SUN JUNLI
BAI XUE

Assignees

中芯国际集成电路制造(北京)有限公司
中芯国际集成电路制造(上海)有限公司

Dates

Publication Date: 20260508
Application Date: 20241106

Claims (10)

1.A capacity prediction method, comprising: Acquiring product demand data of a current site, WPH data of each product at the current site and historical WIP data of each product at the current site; based on the product demand data, WPH data of each product at a current site and historical WIP data of each product at the current site, carrying out first prediction on future WIP data of the current site to obtain a first prediction result of the future WIP data of the current site; And obtaining a second prediction result of the future WIP data of the current site by using a preset prediction model based on the product demand data and the first prediction result so as to predict the future productivity of the current site based on the second prediction result.
2. The method of claim 1, wherein the product demand data includes an identification of each product to be produced and a demand of each product to be produced at the current site.
3. The capacity prediction method as claimed in claim 1, wherein said performing a first prediction on future WIP data of a current site based on the product demand data, WPH data of each product at the current site, and historical WIP data of each product at the current site to obtain a first prediction result of the future WIP data of the current site includes: And reconstructing historical WIP data of each product at the current site by using a sliding window based on the product demand data and the WPH data of each product at the current site to obtain a first prediction result of future WIP data of the current site.
4. The capacity prediction method as claimed in claim 3, wherein the reconstructing historical WIP data of each product at the current site using the sliding window based on the product demand data and WPH data of each product at the current site to obtain the first prediction result of future WIP data of the current site includes: Obtaining a predicted time step of each product at the current site based on the WPH data of each product at the current site, the product demand data and the average WPH data of each product at the current site; and reconstructing the historical WIP data of each product at the current site by using the predicted time step and the preset sliding window size to obtain a first predicted result of the future WIP data of the current site.
5. The capacity prediction method as claimed in claim 3, further comprising, before the reconstructing the historical WIP data of each product at the current site using the sliding window based on the product demand data and the WPH data of each product at the current site: and preprocessing the historical WIP data of each product at the current site.
6. The capacity prediction method as claimed in claim 5, wherein said preprocessing the historical WIP data of each product at the current site includes: supplementing historical WIP data of each product at a current site by adopting a spline interpolation method; And carrying out smoothing treatment on the supplemented data by adopting a Savitzky-Golay method so as to reconstruct a sliding window on the smoothed data.
7. The method of claim 6, wherein said obtaining a second predicted result of future WIP data of said current site using a predetermined prediction model based on said product demand data and said first predicted result comprises: Combining future WIP data of each product in the first prediction result with the product demand data respectively, and inputting each data combination into a preset XGBoost model respectively to obtain a second prediction result of the future WIP data of each product; And obtaining a final productivity parameter value based on the second prediction result of the future WIP data of each product.
8. A productivity prediction apparatus, comprising: The acquisition unit is suitable for acquiring product demand data of the current site, WPH data of each product at the current site and historical WIP data of each product at the current site; The first prediction unit is suitable for carrying out first prediction on future WIP data of the current site based on the product demand data, WPH data of each product at the current site and historical WIP data of each product at the current site to obtain a first prediction result of the future WIP data of the current site; And the second prediction unit is suitable for obtaining a second prediction result of the future WIP data of the current site by utilizing a preset prediction model based on the product demand data and the first prediction result so as to predict the future productivity of the current site based on the second prediction result.
9. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program is executed by a processor to implement the steps of the method of any of claims 1 to 7.
10. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program capable of being run on the processor, characterized in that the processor executes the steps of the method according to any of claims 1 to 7 when the computer program is run on the processor.

Description

Productivity prediction method, device, storage medium and equipment Technical Field The present invention relates to the field of semiconductor technologies, and in particular, to a method and apparatus for predicting productivity, and a storage medium and a device. Background The semiconductor industry has the characteristics of long factory building period and complex process, and in order to enhance the risk resistance, the planning capacity of each platform can be formulated according to the current market prediction during capacity planning. When market demands are vigorous, orders can be selected according to the directions of order quantity, order follow-up continuous degree, customer stability and the like, and the capacity conditions of all product platforms are balanced, so that the maximization of benefits is achieved. When the market is light, products required by the market need to be produced to the maximum extent, and the products are delivered as expected or in advance, so that a little higher customer satisfaction degree is obtained, and the competitive capacity is improved. In both cases, the premise of sales order taking is that a factory has the capacity to produce as expected, and how to evaluate whether the factory has the capacity to meet the order requirement, namely, the prediction of productivity parameters, is very important in the semiconductor industry, even the whole manufacturing industry. However, the current productivity prediction method has poor accuracy in productivity prediction. Disclosure of Invention The invention aims to solve the problem of improving the accuracy of productivity prediction. In order to solve the above problems, an embodiment of the present invention provides a productivity prediction method, including: Acquiring product demand data of a current site, WPH data of each product at the current site and historical WIP data of each product at the current site; based on the product demand data, WPH data of each product at a current site and historical WIP data of each product at the current site, carrying out first prediction on future WIP data of the current site to obtain a first prediction result of the future WIP data of the current site; And obtaining a second prediction result of the future WIP data of the current site by using a preset prediction model based on the product demand data and the first prediction result so as to predict the future productivity of the current site based on the second prediction result. In one possible embodiment, the product demand data comprises the identification of each product to be produced at the current site and the demand of each product to be produced. In one possible embodiment, the performing a first prediction on the future WIP data of the current site based on the product demand data, WPH data of each product at the current site, and historical WIP data of each product at the current site to obtain a first prediction result of the future WIP data of the current site includes: And reconstructing historical WIP data of each product at the current site by using a sliding window based on the product demand data and the WPH data of each product at the current site to obtain a first prediction result of future WIP data of the current site. In a possible embodiment, the reconstructing, based on the product demand data and WPH data of each product at the current site, historical WIP data of each product at the current site using a sliding window, to obtain a first prediction result of future WIP data of the current site includes: Obtaining a predicted time step of each product at the current site based on the WPH data of each product at the current site, the product demand data and the average WPH data of each product at the current site; And reconstructing the historical WIP data of each product at the current site by using the predicted time step and the preset sliding window size to obtain a first predicted result of the future WIP data of the current site. In one possible embodiment, before the reconstructing the historical WIP data of each product at the current site using the sliding window based on the product demand data and the WPH data of each product at the current site, the method further includes: and preprocessing the historical WIP data of each product at the current site. In one possible embodiment, the preprocessing the historical WIP data of each product at the current site includes: supplementing historical WIP data of each product at a current site by adopting a spline interpolation method; And carrying out smoothing treatment on the supplemented data by adopting a Savitzky-Golay method so as to reconstruct a sliding window on the smoothed data. In a possible embodiment, the obtaining, based on the product demand data and the first prediction result, a second prediction result of the future WIP data of the current site by using a preset prediction model includes: Combining future WIP data of eac