PAPAI | Packaging and assembly processes assisted by artificial intelligence

AI for material science in semiconductor packaging

Semiconductor packaging and assembly processes are becoming increasingly complex due to miniaturisation, higher performance requirements and growing product diversity. PAPAI develops AI-driven approaches to improve efficiency in semiconductor packaging and assembly, with a specific focus on sintering processes for power modules.
 

Target group

The project is relevant for:

• Semiconductor manufacturers
• Companies active in semiconductor packaging and assembly
• Power electronics and power module manufacturers
• Material suppliers for semiconductor applications

Context

The semiconductor industry increasingly relies on advanced packaging and assembly technologies to support more compact, powerful and heterogeneous electronic systems. In parallel, manufacturers face growing pressure to reduce development time while improving process efficiency and product performance.

Material development plays a key role in this evolution. In semiconductor packaging, innovative materials such as sintering pastes directly influence the reliability and performance of power modules. However, developing and optimising these materials remains complex and time-consuming.

Artificial intelligence can accelerate material development and process optimisation while reducing material and energy consumption. Yet most AI approaches, especially deep learning methods, require large datasets that are rarely available in industrial material development environments.

There is therefore a need for AI methods capable of delivering reliable insights and optimisation strategies based on limited experimental data.
 

Objectives & results

PAPAI aims to improve the efficiency of semiconductor packaging and assembly processes through AI-assisted material development.

Main objectives: 

• Develop AI models for material development and process optimisation
• Investigate methods to train AI models using limited experimental data
• Improve understanding of the parameters influencing sintering performance
• Support faster formulation development and reduced time-to-market

Key expected results:

• A proof of concept for AI-assisted material development
   

Funding

• Funding agency: VLAIO
• Project type: Xecs