1.408.943.6900

Applying Artificial Intelligence, Machine Learning, and Deep Learning in Smart Electronics Manufacturing

SEMI Global Headquarters 673 Milpitas Blvd. Milpitas, San Jose, CA 95035 Room: Seminar 2 Thursday, February 27
9:00am to 12:00pm

Applying Artificial Intelligence, Machine Learning, and Deep Learning in Smart Electronics Manufacturing 

The focus of this course is to discuss how to apply artificial intelligence, machine learning, and deep learning approaches in surface mount assembly and smart electronics manufacturing. This course will start with a general introduction of artificial intelligence, machine learning, and deep learning and introduce several real-life applications of computer intelligence. Emphasis will be placed on various applications of artificial intelligence and machine learning in the field of printed circuit board assembly processes.  By successfully developing and integrating artificial intelligence methods, the surface mount assembly processes will be automatically optimized and controlled. Then, meaningful patterns from the massive surface mount assembly data can be recognized, critical assembly process parameters that are significantly related to defects in the surface mount assembly lines can be identified, and the potential printed circuit board fabrication failures can be avoided by early detection and prevention. Eventually, fewer defects can be achieved and the operational cost of surface mount assembly can be reduced significantly.    

Syllabus 

  1. Introduction  

  • What is Artificial Intelligence, Machine Learning, and Deep Learning?  

  • General AL, ML, and DL techniques 

  • Real-life applications  

    • Social network, logistics, healthcare, etc. 
  1. Introduction to smart electronics manufacturing laboratory 

  2. Process control & data mining performance metrics    

  3. Stencil printing process noise analysis and dynamic simulation.  

  • Wavelet-based noise segmentation model and analysis  

  • Dynamic stencil printing process interpolation on the printer settings 

  1. Closed-loop feedback control: Stencil printing process (SPP) optimization  

  • Design of experiments to collect initial datasets 

  • Stencil printing result prediction model  

  • Hybrid online optimization model based on state-of-the-art online learning algorithms 

  • Guided evolutionary search (GES) by considering the requirements specifically for SPP optimization  

  1. Stencil printing abnormality and cleaning analysis  

  • Stencil printing capability status predictive modeling using a multi-dimensional temporal recurrent neural network model 

  • Real-time control of the stencil cleaning profile decision-making process  

  1. Online chip mounter offset control using solder paste and automated optical inspection machines  

  • Dynamic placement parameters setting 

  • Sequential prediction model 

  • Region-based optimization model 

  • Chip mounter diagnostic study  

If you will be staying for lunch, please register for the Tech Course Networking Luncheon in addition to your course – it is complimentary. 

Share page with AddThis