In the semiconductor industry, the rising significance of data-driven quality control technologies is evident as substantial data volumes continue to be generated. At Infineon, production testing holds significant weight in automotive microcontroller manufacturing. Thousands of data points are amassed during this process. This thesis aims to analyze and leverage machine learning to enhance testing methodologies by delving into this data. In the production process, testing comprises numerous sequential steps, primarily divided into two pivotal phases: front-end testing, conducted before packaging, and back-end testing, which occurs afterward. This thesis investigates the feasibility of predicting the eventual state of packaged chips based on pre-packaging test results. If successful, this concept could revolutionize the approach to chip manufacturing. By accurately predicting the final state of packaged chips based on pre-packaging tests, It could be potentially possible to avoid the need for extra protective layers, leading to more efficient and cost-effective production processes.
In the semiconductor industry, the rising significance of data-driven quality control technologies is evident as substantial data volumes continue to be generated. At Infineon, production testing holds significant weight in automotive microcontroller manufacturing. Thousands of data points are amassed during this process. This thesis aims to analyze and leverage machine learning to enhance testing methodologies by delving into this data. In the production process, testing comprises numerous sequential steps, primarily divided into two pivotal phases: front-end testing, conducted before packaging, and back-end testing, which occurs afterward. This thesis investigates the feasibility of predicting the eventual state of packaged chips based on pre-packaging test results. If successful, this concept could revolutionize the approach to chip manufacturing. By accurately predicting the final state of packaged chips based on pre-packaging tests, It could be potentially possible to avoid the need for extra protective layers, leading to more efficient and cost-effective production processes.
Labeling techniques definition for defective RRAM hidden fail signatures identification
COLETTO, SIMONE
2023/2024
Abstract
In the semiconductor industry, the rising significance of data-driven quality control technologies is evident as substantial data volumes continue to be generated. At Infineon, production testing holds significant weight in automotive microcontroller manufacturing. Thousands of data points are amassed during this process. This thesis aims to analyze and leverage machine learning to enhance testing methodologies by delving into this data. In the production process, testing comprises numerous sequential steps, primarily divided into two pivotal phases: front-end testing, conducted before packaging, and back-end testing, which occurs afterward. This thesis investigates the feasibility of predicting the eventual state of packaged chips based on pre-packaging test results. If successful, this concept could revolutionize the approach to chip manufacturing. By accurately predicting the final state of packaged chips based on pre-packaging tests, It could be potentially possible to avoid the need for extra protective layers, leading to more efficient and cost-effective production processes.File | Dimensione | Formato | |
---|---|---|---|
Coletto_Simone.pdf
accesso riservato
Dimensione
3.42 MB
Formato
Adobe PDF
|
3.42 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/66755