CN-121998148-A - Method and device for estimating available time of semiconductor device, storage medium and electronic device

Abstract

A method, a device, a storage medium and an electronic device for estimating the available time of a semiconductor device. The method comprises the steps of obtaining product demand data of semiconductor equipment, determining an available time prediction model corresponding to the semiconductor equipment based on the obtained product demand data, inputting the obtained product demand data into the available time prediction model corresponding to the semiconductor equipment, and predicting the available time of the semiconductor equipment, wherein the available time prediction model is obtained by training an initial available time model by utilizing historical maintenance related data and historical downtime related data of the semiconductor equipment under the condition that the product demand data and the product demand data have the same product combination. By adopting the scheme, the available time of the semiconductor equipment can be accurately estimated.

Inventors

• BAI XUE
• ZHOU YIZHONG
• SUN JUNLI

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 (12)

1. 1. A method for estimating the usable time of a semiconductor device, comprising: Acquiring product demand data of semiconductor equipment; Determining an available time estimation model corresponding to the semiconductor equipment based on the acquired product demand data; Inputting the acquired product demand data into an available time estimation model corresponding to the semiconductor equipment, and estimating the available time of the semiconductor equipment; The available time prediction model is obtained by training an initial available time model by utilizing historical maintenance related data and historical downtime related data of the semiconductor equipment under the condition that the product demand data and the product demand data have the same product combination.
2. 2. The method for estimating a time to availability of a semiconductor device according to claim 1, wherein the maintenance related data includes a number of preventive maintenance, a length of time required for preventive maintenance, and a maintenance time interval.
3. 3. The method for estimating time of availability of semiconductor equipment according to claim 1, wherein said downtime-related data includes the number of times of downtime and the length of time required for downtime.
4. 4. The method for estimating time of availability of semiconductor equipment as claimed in claim 1, wherein the product demand data includes product category identification and key index value of each category of products.
5. 5. The method for estimating a time of availability of semiconductor equipment according to claim 4, wherein said product demand data further comprises a yield number of each kind of product.
6. 6. The method of estimating time of availability of semiconductor equipment according to claim 1, wherein the time of availability estimation model is a neural network model.
7. 7. The method for estimating time of availability of semiconductor equipment as claimed in claim 6, wherein the model for estimating time of availability is trained by: Acquiring a plurality of training samples related to the semiconductor equipment, wherein the training samples comprise a total time interval, and product related data, maintenance related data and downtime related data in the total time interval; Inputting the acquired training sample into the initial available time model to obtain an output error of the initial available time model; When the output error does not meet the precision requirement, calculating the average error of each layer of the initial available time model; And when the average error does not meet the precision requirement, updating the weight of each layer of the initial available time model until the updated weight of each layer meets the precision requirement, and obtaining the available time estimated model.
8. 8. The method for estimating time of availability of a semiconductor device according to claim 7, wherein before training the initial availability time model using the acquired training samples, further comprising: and normalizing the acquired training samples.
9. 9. The method of estimating time of availability of semiconductor equipment according to claim 7, wherein the number of nodes of the neural network model is the same as the number of arguments in the training sample.
10. 10. A semiconductor device availability time estimating apparatus, comprising: an acquisition unit adapted to acquire product demand data of the semiconductor device; The determining unit is suitable for determining an available time estimated model corresponding to the semiconductor equipment based on the acquired product demand data; The estimating unit is suitable for inputting the acquired product demand data into an available time estimating model corresponding to the semiconductor equipment and estimating the available time of the semiconductor equipment; The available time prediction model is obtained by training an initial available time model by utilizing historical maintenance related data and historical downtime related data of the semiconductor equipment under the condition that the product demand data and the product demand data have the same product combination.
11. 11. 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 9.
12. 12. 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 9 when the computer program is run on the processor.

Description

Method and device for estimating available time of semiconductor device, storage medium and electronic device Technical Field The present invention relates to the field of semiconductor technologies, and in particular, to a method and apparatus for estimating an available time of a semiconductor device, a storage medium, and an electronic device. Background Semiconductor devices are highly automated devices, and the time of availability of the semiconductor device affects the yield of the semiconductor device. In capacity calculation, evaluation of semiconductor device yield affects future semiconductor device investment and whether customer's demands can be satisfied. In the calculation of the available time of the current semiconductor device, the past performance of the semiconductor device is mainly relied on as a reference for the future, but in actual production, the available time of the semiconductor device is influenced by a plurality of factors, so that the available time of the semiconductor device needs to be estimated more accurately when the yield of the semiconductor device is calculated more accurately according to the future product demand Disclosure of Invention The invention aims to accurately estimate the available time of semiconductor equipment. In order to solve the above problems, an embodiment of the present invention provides a method for estimating a time of availability of a semiconductor device, including: Acquiring product demand data of semiconductor equipment; Determining an available time estimation model corresponding to the semiconductor equipment based on the acquired product demand data; Inputting the acquired product demand data into an available time estimation model corresponding to the semiconductor equipment, and estimating the available time of the semiconductor equipment; The available time prediction model is obtained by training an initial available time model by utilizing historical maintenance related data and historical downtime related data of the semiconductor equipment under the condition that the product demand data and the product demand data have the same product combination. In one possible embodiment, the service related data includes a number of preventative repairs, a length of time required for preventative repair, and a repair time interval. In one possible embodiment, the downtime-related data includes the number of times downtime occurs, and the length of time required for downtime. In one possible embodiment, the product demand data includes product category identification, and key index values for various categories of products. In one possible embodiment, the product demand data further includes the quantity of output of each type of product. In one possible embodiment, the available time prediction model is a neural network model. In one possible embodiment, the available time estimation model is trained by the following method: Acquiring a plurality of training samples related to the semiconductor equipment, wherein the training samples comprise a total time interval, and product related data, maintenance related data and downtime related data in the total time interval; Inputting the acquired training sample into the initial available time model to obtain an output error of the initial available time model; When the output error does not meet the precision requirement, calculating the average error of each layer of the initial available time model; And when the average error does not meet the precision requirement, updating the weight of each layer of the initial available time model until the updated weight of each layer meets the precision requirement, and obtaining the available time estimated model. In a possible embodiment, before the training of the initial available time model using the acquired training samples, the method further includes: and normalizing the acquired training samples. In one possible embodiment, the number of nodes of the neural network model is the same as the number of arguments in the training sample. The embodiment of the invention also provides a device for estimating the available time of the semiconductor equipment, which comprises: an acquisition unit adapted to acquire product demand data of the semiconductor device; The determining unit is suitable for determining an available time estimated model corresponding to the semiconductor equipment based on the acquired product demand data; The estimating unit is suitable for inputting the acquired product demand data into an available time estimating model corresponding to the semiconductor equipment and estimating the available time of the semiconductor equipment; The available time prediction model is obtained by training an initial available time model by utilizing historical maintenance related data and historical downtime related data of the semiconductor equipment under the condition that the product demand data and the product demand data have the same