How practical AI solutions could change feed mills of the future

Feed Mill of the Future Conference speakers predict artificial intelligence (AI) integration will alter how the industry approaches mill design and operations.

Elise Schafer headshot Headshot
Ai Technology Artificial Intelligence
LookerStudio | BigStock.com

Feed industry leaders gathered at the Feed Mill of the Future Conference to discuss how artificial intelligence (AI) could revolutionize facility design and operations but cautioned against implementing the technology without clear business objectives or strong protection against cybersecurity threats.

The conference, presented by Feed Strategy and Feed & Grain in partnership with the American Feed Industry Association (AFIA), took place during the International Production & Processing Expo (IPPE) in Atlanta. It concluded with a question-and-answer panel discussion featuring the program’s six speakers.

The panelists agreed AI will significantly impact feed mill design, potentially reshaping how new facilities are conceived and built, stressing that AI’s ability to analyze vast amounts of data could lead to more efficient designs.

“Using all the data available on the web, AI has the ability to essentially compare and contrast what has and hasn’t worked at more than hundreds of thousands of sites and pick the best of all the designs to create a site based on real world solutions," said Matthew Boland, sales territory manager, CPM Automation.

A targeted approach to implementing AI

While the potential is evident, Liam Cassidy, senior executive director of automation for Knobelsdorff, advised rather than implementing AI broadly, begin by solving a specific operational challenge.

“Start with a genuine problem that you have today, and implement an AI-driven, small, scalable solution,” Cassidy said. “Look for some tangible results. See if you’re getting a return on investment.”

The critical role of vendor selection

Many facilities already have the foundation for AI implementation through existing automation systems, but success depends heavily on vendor selection. Ben Allen, CEO of BinSentry, said to ask vendors key questions about the technology they’re investing in now and what they plan to invest in in the future.

From left: Ian Mealey, product marketing director for Format Solutions, a Datacor company; Brandon Bohle, endpoint security team lead at Interstates; Brady Gaalswyk, co-president of Easy Automation; Liam Cassidy, senior executive director of automation for Knobelsdorff; Ben Allen, CEO of BinSentry; and Matthew Boland, sales territory manager, CPM Automation.From left: Ian Mealey, product marketing director for Format Solutions, a Datacor company; Brandon Bohle, endpoint security team lead at Interstates; Brady Gaalswyk, co-president of Easy Automation; Liam Cassidy, senior executive director of automation for Knobelsdorff; Ben Allen, CEO of BinSentry; and Matthew Boland, sales territory manager, CPM Automation.“Ask to see the company’s roadmap, and look for vendors that are keeping up or ahead of technology trends,” Allen said.

Brady Gaalswyk, co-president, Easy Automation Inc., added: “When you purchase automation, you're not just buying a widget, you’re making a big commitment to your future with the automation or software provider. It’s a matter of where they are now and where they’re going to take you in the future, both with the system and the change management and training.”

Data quality is key 

Successful implementation of AI solutions hinges on the quality of the company’s data, as well, said Ian Mealey, product marketing director for Format Solutions, a Datacor company.

“AI-optimization depends on the quality of the data and the tools you have available to be able to respond,” Mealey said.

For instance, when updating feed formulations to remain competitive, AI-driven solutions require accurate and timely facility data inputs to function effectively.

“It comes back to the themes we’ve spoken about — vendor selection, data quality and using the right tools in the right places,” Mealey said.

Practical AI applications for feed mills

The panelists highlighted several practical AI applications, including production scheduling optimization and virtual sensing technology. Boland shared a success story of how virtual sensing, an advanced AI-driven anomaly and pattern detection technology, reduced bag filling overages in a mill by 50%.

Boland said virtual sensing enables feed mills to optimize processes by analyzing multiple data points to estimate values that cannot be directly measured with sensors. By leveraging virtual sensing, which integrates data from up to 20 different sources, the system could accurately predict the final weight of a bag before it reached the scale. This AI-driven solution enhanced efficiency, reduced waste and improved cost management.

Cybersecurity: A growing concern 

Cybersecurity emerged as a critical concern with AI's reliance on large amounts of data. The experts warned that professional hacking organizations are increasingly targeting feed facilities. The rise of AI tools has made it easier for attackers to create harmful programs, yet many facilities still treat cybersecurity as an afterthought.

“These hackers are actually professional companies," Cassidy said. “A client in the milling industry had been attacked and you could look up online that the company responsible for the attack was based in Asia and had a C-suite. Our mindset is these are carried out by lone hackers in a basement, but these are professional, elite companies whose main business model is to attack vulnerable companies.”

The speakers recommended some key approaches to address cybersecurity threats in feed mills. Gaalswyk cautioned attendees to avoid common vulnerabilities, like sharing one app login among all users or selecting easy-to-guess passwords.

Brandon Boehle, endpoint security team lead for Interstates, advised controlling data access.

“Make sure as you’re interacting with the cloud, you’re only connecting and allowing information in and out that you want out in that cloud,” Boehle said. “You don't want to blanket allow everything. Understand what you’re adding to the cloud.”

A strategic approach to AI integration

Successful AI integration requires a holistic approach, considering not just physical layout and equipment, but also cybersecurity, data management and integration capabilities, according to the Feed Mill of the Future speakers. They recommend requesting roadmaps from technology vendors to understand their future development plans and to ensure alignment with a feed mill’s objectives to grow with AI.

AI’s role in feed mill operations will only increase, and those who take a strategic, well-planned approach to AI adoption will be best positioned for future efficiency and innovation.

Page 1 of 3
Next Page