Deep learning techniques for integrated circuit die performance prediction

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F22%3A00365085" target="_blank" >RIV/68407700:21240/22:00365085 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1557/s43580-022-00308-0" target="_blank" >https://doi.org/10.1557/s43580-022-00308-0</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1557/s43580-022-00308-0" target="_blank" >10.1557/s43580-022-00308-0</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Deep learning techniques for integrated circuit die performance prediction

Popis výsledku v původním jazyce

Predicting integrated circuit (IC) functionality based on process control monitoring (PCM) parameters without individual die testing is a major challenge for manufacturers due to the high cost of electrical die measurement on the wafer. Complex dependencies between individual PCM parameters can be used to explain certain patterns of dice failure using Deep learning (DL) algorithms. However, random failure patterns due to process defects cannot be detected by this method. Combining PCM and in-process defect inspection data can be an ultimate prediction technique. In some cases, however, the availability of defect inspection data is much lower than the availability of PCM data, so direct ensemble training is rather ambiguous. This paper shows how to efficiently utilize both defect and PCM data to train a model to predict IC functionality. Such a hybrid model outperforms PCM-only solutions, and in contrast to a defect-only model predicts also failure areas across the wafer.

Název v anglickém jazyce

Deep learning techniques for integrated circuit die performance prediction

Popis výsledku anglicky

Predicting integrated circuit (IC) functionality based on process control monitoring (PCM) parameters without individual die testing is a major challenge for manufacturers due to the high cost of electrical die measurement on the wafer. Complex dependencies between individual PCM parameters can be used to explain certain patterns of dice failure using Deep learning (DL) algorithms. However, random failure patterns due to process defects cannot be detected by this method. Combining PCM and in-process defect inspection data can be an ultimate prediction technique. In some cases, however, the availability of defect inspection data is much lower than the availability of PCM data, so direct ensemble training is rather ambiguous. This paper shows how to efficiently utilize both defect and PCM data to train a model to predict IC functionality. Such a hybrid model outperforms PCM-only solutions, and in contrast to a defect-only model predicts also failure areas across the wafer.

Druh

J_SC - Článek v periodiku v databázi SCOPUS

CEP obor

—

OECD FORD obor

20201 - Electrical and electronic engineering

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

MRS Advances

ISSN

2059-8521

e-ISSN

2059-8521

Svazek periodika

7

Číslo periodika v rámci svazku

30

Stát vydavatele periodika

GB - Spojené království Velké Británie a Severního Irska

Počet stran výsledku

6

Strana od-do

683-688

Kód UT WoS článku

000838469100001

EID výsledku v databázi Scopus

2-s2.0-85135779047