Close this search box.

Bridging the Gap: The future of AI implementation

If you’re not already using artificial intelligence (AI) to support your industrial automation and cybersecurity environments, you likely will be in the near future. Newly released research by CFE Media and Technology, covered in the Bridging the Gap Video/Podcast Series, took a closer look at how AI is impacting industrial automation. In the second episode, we covered how AI is being implemented in industrial automation and cybersecurity spaces. While there are challenges to AI implementation, it is happening – and quickly.

Measuring success in AI implementation

As companies are rushing to implement AI, there are still questions about how it’s being used and how success is being measured. According to the CFE research, 56% of respondents are measuring success through improvements in general efficiency. A high percentage of respondents also cited cost reduction (66%) and an increase in predictive accuracy (55%). But too many still are not measuring at all or haven’t figured out just what to measure yet.

“Even with the ramping-up of AI adoption and experimentation right now, the KPIs for AI still are an exception, rather than the rule,” said Bridging the Gap guest Jeff Winter, an Industry 4.0 and digital transformation thought leader and influencer. “If you actually look at a report that came out by the Manufacturing Leadership Council on industrial AI earlier this year, they found that 61% of AI projects had no metrics to evaluate its success. That’s astonishing. And that tells you that companies are investing in the technology with not entirely knowing how they’re going to grade it.”

The goal should be to view your AI implementation through a business lens.

“You do need to switch to the business outcomes that you’re trying to achieve,” Winter said. “Things like productivity enhancement, cost reduction, labor savings. You need to figure out what is the AI supposed to do, not just reporting on AI usage and the accuracy of the AI.”

Timing of AI implementation

It is steadily becoming common practice for AI to be used in industrial automation, partially to deal with the explosion of data being collected. The No. 1 industry for collecting data is manufacturing, according to Morgan Stanley, at almost twice as much as any other sector. Now, it’s more a question of how it can best be used, what is it good for and what is it bad for?

“People are applying AI just as fast as they can, without really knowing where to apply that technology, and the ramifications, and implications of the processes, and the people that come along with it,” Winter said. “Because the technology is arguably easier than other technologies to implement, they’re putting the cart before the horse.”

According to the research, 24% of respondents said they are already using AI, and it is fully implemented. Almost a quarter of respondents said they would be implementing AI in the next 90 days, and 50% in the next 180 days.

Bridging the Gap series

Other episodes in the Bridging the Gap series will focus on the challenges of AI integration, the strengths and weaknesses of AI and the business impact of AI. As companies progress on their digital transformation journeys, AI will likely be implemented into more and more processes, from predictive maintenance to cybersecurity.

One big concern many people have about AI implementation should come as a surprise to no one. It’s become a common refrain: AI is going to take our jobs. According to Winter, job loss is a concern with many new technologies, but that is seldom the result.

“Of all the technologies that they had, almost every one created more jobs than it displaced, in terms of the technology,” he said. “There’s only one technology that actually displaced more jobs than it replaced, and that happened to be robotics. But AI actually created more jobs than it displaced.”

To view Episode 1 of Bridging the Gap, about the challenges of AI integration, click here.




Keep your finger on the pulse of top industry news