Building a Digital Strategy with Intel's Irene Petrick

In the second of our series of A3 interviews with AI leaders, Irene Petrick, the Senior Director of Industrial Innovation in the Internet of Things Group at Intel, discusses how to successfully deploy AI solutions. Petrick, who sits on A3's Artificial Intelligence Technology Strategy Board recommends visualizing the end goal of transformation then identifying high ROI opportunities. assessing the data using edge compute. Check out her advice on how assess data using edge compute to maximize digital transformation.

Irene Petrick
Senior Director of Industrial Innovation
Intel Corporation

How would you advise companies to choose their artificial intelligence projects – and what questions do they need to answer before they begin?

Companies must first ask themselves "What am I trying to achieve with an AI strategy? What can digital transformation do for my company?" Once this vision is in place, companies an begin to assess the low hanging fruit, can begin to identify high ROI opportunities, and then can begin to start on smaller deployments that they can rapid learn from as they head to full scale production. Too often, companies just start with AI projects that are obvious or cool, without having a vision of how those projects will come together if they actually are deployed at scale. This tends to create islands of AI success that aren't linked to the broader production system. (See slides 13 & 14 for some advice to manufacturers on how to maximize value of AI for their business)

slide 13

slide 14

How much talk about AI right now if hype vs. reality? Where is AI having the most impact now in manufacturing and automation? What are some of the effective real-work use-cases for AI that are being deployed today?

My colleague, Dr. Faith McCreary, and I have been studying digital transformation for the past 4 years with over 500 individuals from over 400 companies. When we asked companies where they are on their journey to AI enabled manufacturing, we were surprised at the data. Only 2% said that AI was being extensively used (and yielding value) at their company (See slide 4 for all of the related data on this). We are seeing pockets of successful use of AI in manufacturing operations. Not surprisingly, with COVID impacts on manufacturing, companies have been seeking ways to successfully work remotely at least part of the time. Half of the companies we've spoken to are finding that remote monitoring and maintenance is a good candidate for AI solutions that brings nearly immediate value. Decision support (49%) and workflow automation (44%) are also strong application areas. Overall, COVID has accelerated the adoption of automation and control systems, and these systems tend to pull AI into the manufacturing operations with them. (See slide 5).

slide 4

slide 5

Smart automation systems generate mountains of data. How does a company develop a data strategy that can manage this river of information and leverage it in its operations?


Don't Miss These Industry-Leading Events!

A3 Business Forum

January 20-22, 2025
Orlando, FL

Automate

May 12-15, 2025
Detroit, MI

AISA

November 3-5, 2025
Houston, TX



Yes, companies are drowning in data, but very light on insights from this data. Data strategies must go hand in hand with compute (edge to cloud), storage, and with communications (bandwidth) strategies as these are really intertwined. If too much data is collected at the edge (machine level, for example), then the bandwidth may not be sufficient to handle it, or the storage would be overwhelmed. Companies should first decide WHY they are implementing a particular AI solution. Then they must determine what the infrastructure needs will be to support that solution. Finally, they need to determine WHAT data is essential and WHERE that data needs to be processed. Only then, can the AI solution be implemented which balances the data availability with the data that is actually processed and used. Each deployment and operating environment may make different choices. Interestingly, 93% of our participants suggested that AI solutions will actually be hampered by data challenges.....e.g., do we have the RIGHT data? Is the data COMPLETE? How much CERTAINTY is associated with the data?

Is AI only for the big players? How do small and medium-size companies take advantage of these technologies? How do we democratizing the deployment of AI and smart automation?

Our AI solutions today are often too complex (and complicated) for easy deployment. Some require a lot of data preparation before the solution can even be piloted, yet alone deployed at scale. Most companies don't have the data scientists needed to put data into robust forms, and do not have the digital skills to trouble shoot and maintain a solution once it goes from pilot to scale. Regardless of size, companies should seek vendors that have experience in their particular industrial sector. AI challenges in discrete high volume manufacturing are often different than AI challenges in continuous process manufacturing. Seeking a vendor with domain experience is a key ingredient to success.

While we all know downtime can help relieve stress, several science-backed studies confirm you actually improve productivity when you take time to enjoy life outside of the office. What are some of your favorite things to do in your free time?

Reading, cooking, walking

What else should the industry know about deploying AI solutions?

One thing that comes up again and again in terms of successful deployment of AI solutions is TRUST. We were actually quite surprised at the number of our study participants who highlighted this. (See slides 14-16).

Slide 13

Slide 14

Slide 15

Slide 16


Dr. Irene J. Petrick joined Intel in 2015 and is Senior Director of Industrial Innovation in the Internet of Things Group. Irene focuses on emerging technology, social, and global trends and their combined impact on the industrial space. Her work highlights the transition to intelligent manufacturing, the technology solutions driving computing across the manufacturing enterprise, and the needs of the future workforce. She also explores new manufacturing methods such as 3D printing and the new business models that are enabled by intelligent manufacturing. Prior to joining Intel, Irene was a professor at Penn State and has been actively engaged with companies in their innovation and technology strategies for over 25 years, including work with twelve Fortune 100 companies, the U.S. military, and a wide variety of small to medium sized enterprises. Irene is author or co-author on more than 225 publications and presentations.

BACK TO ARTIFICIAL INTELLIGENCE BLOG