Scholarlink

<p>Semiconductor manufacturing is increasingly complex, requiring high precision across interconnected processes where minor deviations can significantly impact yield and product reliability. Traditional monitoring approaches often struggle to manage high dimensional and dynamic data environments, creating a need for more advanced analytical solutions. Artificial intelligence (AI) has emerged as a promising tool to enhance fault detection and root cause analysis through data driven insights. This review aims to synthesize empirical evidence on the application of AI in semiconductor manufacturing, with a focus on fault detection, predictive maintenance, and root cause analysis. A narrative review approach was adopted to capture diverse methodologies and technological advancements. Relevant studies published between 2018 and 2026 were identified through databases including IEEE Xplore, Scopus, ScienceDirect, and Google Scholar. Inclusion criteria focused on empirical studies reporting measurable outcomes in real manufacturing settings. Data were extracted systematically and analyzed using thematic synthesis. The findings indicate that AI significantly improves fault detection accuracy through image based and sensor driven analytics. Predictive maintenance models effectively reduce equipment downtime by forecasting failures using time series data. AI driven root cause analysis enhances the identification of underlying process issues through pattern recognition and causal modeling. Integration of AI into manufacturing workflows enables real time decision making and operational optimization, though challenges such as data quality and model interpretability persist. AI plays a transformative role in semiconductor manufacturing by improving efficiency, accuracy, and decision making. However, addressing implementation challenges is essential for achieving sustainable and scalable adoption.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Advancing semiconductor manufacturing through ai-based fault detection and root cause analysis

  • 期刊论文

  • 发表日期：

    2026年7月22日

查看全文
  • Arifa Ahmed,
  • Nur Mohammad,
  • Md Abdur Rob,
  • Jakir Hossain Ridoy,
  • MD Samiul Hasan Aric,
  • S M Mainul Islam,
  • Sadia Sharmin,
  • Nur Vanu,
  • Mst. Rina Parvin

摘要

Semiconductor manufacturing is increasingly complex, requiring high precision across interconnected processes where minor deviations can significantly impact yield and product reliability. Traditional monitoring approaches often struggle to manage high dimensional and dynamic data environments, creating a need for more advanced analytical solutions. Artificial intelligence (AI) has emerged as a promising tool to enhance fault detection and root cause analysis through data driven insights. This review aims to synthesize empirical evidence on the application of AI in semiconductor manufacturing, with a focus on fault detection, predictive maintenance, and root cause analysis. A narrative review approach was adopted to capture diverse methodologies and technological advancements. Relevant studies published between 2018 and 2026 were identified through databases including IEEE Xplore, Scopus, ScienceDirect, and Google Scholar. Inclusion criteria focused on empirical studies reporting measurable outcomes in real manufacturing settings. Data were extracted systematically and analyzed using thematic synthesis. The findings indicate that AI significantly improves fault detection accuracy through image based and sensor driven analytics. Predictive maintenance models effectively reduce equipment downtime by forecasting failures using time series data. AI driven root cause analysis enhances the identification of underlying process issues through pattern recognition and causal modeling. Integration of AI into manufacturing workflows enables real time decision making and operational optimization, though challenges such as data quality and model interpretability persist. AI plays a transformative role in semiconductor manufacturing by improving efficiency, accuracy, and decision making. However, addressing implementation challenges is essential for achieving sustainable and scalable adoption.