• Subscribe

Enhancing decision-making in real-time scheduling: leveraging data and AI technology (Part 2/4)

Understanding and defining AI and ML

Renowned influencer James Moyne joins Samantha Duchscherer in an engaging discussion exploring the importance of integrating additional information and advanced technologies like Artificial intelligence (AI) into the semiconductor industry’s scheduling and dispatching process. The comprehensive series consists of four parts focusing on various subjects such as the importance and advantages of data, AI, and human involvement. It also delves into the obstacles faced and offers insights into the role of digital twin technology.

In this second article of the series, the focus is on defining machine learning (ML) and AI. The article provides examples to illustrate their differences and similarities.

AI vs. ML

The distinction between AI and ML can be somewhat blurred, as there is no exact definition that universally separates the two concepts. However, in general terms, AI refers to the broader field of creating intelligent systems that can perform tasks requiring human-like intelligence. It encompasses various techniques and approaches, including machine learning. On the other hand, ML is a specific subset of AI that focuses on enabling machines to learn from data and make predictions or decisions based on that learning. ML algorithms learn patterns and relationships in data to improve their performance over time, without being explicitly programmed. While AI and ML are closely related, AI extends beyond just ML and includes other techniques such as natural language processing, computer vision, and expert systems. The boundaries between AI and ML can be fluid, with ML often being considered a crucial component of AI.

Different levels of AI

There are various levels of AI, each with its own scope and capabilities:
  • Granular AI operates within defined boundaries or parameters, using machine learning algorithms to make precise predictions or decisions within a known space. It fills in gaps between data points to provide a more granular understanding of the system’s behavior.
  • Exploratory AI goes beyond known boundaries, exploring unknown territories where existing models may not apply. It incorporates external knowledge and data from various sources to gain a broader understanding of complex problems.
  • General AI, or artificial general intelligence (AGI), represents the highest level of AI sophistication. It mimics human-like intelligence (i.e., reasoning, problem solving, and adaptation), possessing a comprehensive understanding of the environment and the ability to adapt to a wide range of tasks.


James and I began our discussion talking about the differences between ML and AI, captured in the video below.
In this short video, Sam and James discuss the intricacies of AI and ML

Chatbots and semiconductor examples

After developing a deeper understanding of AI and ML, our attention shifted toward exploring practical examples:
Sam: Let’s discuss a familiar example for everyone: chatbots. Where do chatbots fit within the AI and ML spectrum?
James: You could argue that chatbots are not AI. For instance, if it’s creating my resume, it’s not creating anything new. It’s just mining data and saying, ‘Based on all this information, I’m doing probabilities and telling you what I think is true.’ Yet, it is AI in the sense that it’s mining an enormous amount of information and connecting the dots. Again, there is no clear-cut definition, but at the end of the day my perspective is that chatbots are mostly just Bayesian inference engines.

Sam: Transitioning to semiconductor examples, I’m assuming building an algorithm on two years of historical data to predict lot cycle time would be ML correct?

James: Correct! What you’re doing in that example is stay within a space that you that you’ve already had some experience with and you’re just adding more information so that the machine learning can develop a model.

Sam: What about development of a reinforcement learning model to find the optimal dispatching parameters needed for an unseen event that would happen in a fab? Where would this example fall on the spectrum?

James: I would say that still, for me, falls into machine learning, although I’m sure there are people out there who would say, ‘No, no, this is artificial intelligence.’

However, in the case of an event that has never occurred before, I might talk to experts and go read books to identify what this type of failure might look like and what the risks and rewards are. When we try to build an algorithm on this new data we bring in, whether real or simulated, this is when we start to get into the intelligence aspect.

Sam: And how do you define intelligence in this context?

James: It is a spectrum from pure artificial to pure human. What will be needed in the long term is a full integration of human and AI; even AI systems need some kind of human checks and balances. The interaction between human and artificial intelligence has to become more asynchronous. For example, a human should be able to provide intelligence to an AI system as soon as that information is obtained and verified, e.g., without being prompted. Conversely, an AI system should know when and how to ask the human for help.

I have a saying for this, ‘No knowledge left behind.’


Machine learning and AI are part of a spectrum of technologies that perform similar functions. For this reason, people are often confused as to which one is at play. Ultimately, the distinction between AI and ML may vary depending on the context and perspective. Both fields continue to evolve, and the boundaries between them may become even more nuanced as new technologies and approaches emerge.

Up next, we’ll dive into the challenges and benefits associated with enabling AI and ML within the scheduling and dispatching framework.

Back to Part 1

About Dr. Moyne

Dr. James Moyne is an Associate Research Scientist at the University of Michigan. He specializes in improving decision-making by bringing more information into the scheduling and dispatching area. Dr. Moyne is experienced in prediction technologies such as predictive maintenance, model-based process control, virtual metrology, and yield prediction. He also focuses on smart manufacturing concepts like digital twin and analytics, working towards the implementation of smart manufacturing in the microelectronics industry.

Dr. Moyne actively supports advanced process control through his co-chairing efforts in industry associations and leadership roles, including the IMA-APC Council, the International Roadmap for Devices and Systems (IRDS) Factory Integration focus group, the SEMI Information and Control Standards committee, and the annual APC-SM Conference in the United States.

With his extensive experience and expertise, Dr. Moyne is highly regarded as a consultant for standards and technology. He has made significant contributions to the field of smart manufacturing, prediction, and big data technologies.

About the Author

Picture of Samantha Duchscherer, Global Product Manager
Samantha Duchscherer, Global Product Manager
Samantha is the Global Product Manager overseeing SmartFactory AI Productivity, Simulation AutoSched® and Simulation AutoMod®. Prior to joining Applied Materials Automation Product Group Samantha was Manager of Industry 4.0 at Bosch, where she also was previously a Data Scientist. She also has experience as a Research Associate for the Geographic Information Science and Technology Group of Oak Ridge National Laboratory. She holds a M.S. in Mathematics from the University of Tennessee, Knoxville, and a B.S. in Mathematics from University of North Georgia, Dahlonega.