Rockwell Automation finds data gaps are holding back food manufacturers’ AI plans

Food and beverage manufacturers are collecting far more data than they effectively use, creating an obstacle to expanding artificial intelligence across their operations, according to new analysis from Rockwell Automation.

• Nearly 90% of food and beverage manufacturers were using AI and machine learning or planned to introduce the technologies, according to Rockwell Automation’s 2026 manufacturing research.
• Food and beverage manufacturers reported effectively using only about 45% of the data they collected, limiting the information available to support decisions and expand AI applications.
• Industry respondents expected AI-supported manufacturing processes to increase from approximately 37% today to more than 60% by 2030, requiring better connections between operational and business systems.

Released on October 29, the analysis drew on findings from Rockwell’s 2026 State of Smart Manufacturing Report to examine why successful AI pilots can prove difficult to reproduce across production lines, facilities and business functions.

Nearly 90% of food and beverage manufacturers were already using AI and machine learning or planned to do so in the near term. Yet respondents reported effectively using only about 45% of the data they collected, leaving much of their operational information unavailable to decision-makers and AI applications.

“Many manufacturers have moved beyond asking whether AI can deliver value,” said Greg Schmidt, Vice President, CPG Global Industry, Rockwell Automation. “The focus now is on how to scale successful pilots across the enterprise. That requires more than new technology. It requires connected data, integrated operations and a clear link between AI initiatives and business outcomes.”

The accompanying FAQ, From AI Pilots to Enterprise Impact, identified data accessibility, disconnected systems and organizational readiness as obstacles to wider adoption. More than four in ten food and beverage leaders cited capturing, understanding and effectively using data as a significant obstacle to growth.

Other barriers included budget constraints, difficulties executing capital projects, competing priorities and managing organizational change. These could prevent a manufacturer from extending an application that had already demonstrated value in an individual production setting.

Rockwell argued that companies needed to consider repeatability before deploying a pilot. Alongside assessing whether an application delivered a measurable benefit, manufacturers should establish whether it could operate consistently across additional facilities, functions and use cases.

That would require common data models, standardized processes and integrated workflows, supported by clear responsibility for how information was managed and used.

The analysis identified integration between information technology and operational technology as an important requirement. Connecting business systems with production equipment and operational information could give AI applications the context needed to identify patterns and support decisions across different parts of a company.

Industry respondents expected the proportion of AI-supported manufacturing processes to increase from approximately 37% today to more than 60% by 2030. That anticipated expansion would place greater demands on manufacturers’ ability to make reliable data accessible across their operations.

Quality inspection was one example explored in the FAQ. Many food and beverage companies had already deployed vision systems to improve inspection, reduce defects and maintain product consistency. However, most of those systems still operated independently at individual production lines or facilities.

Connecting inspection data to cloud-based IT systems could allow manufacturers to compare results across sites, identify recurring problems and apply successful practices elsewhere. Rockwell presented this as an opportunity to extend the usefulness of existing systems through better connections between facilities.

The FAQ also described an illustrative response to increased orders following a retailer promotion or seasonal demand surge. Information from an enterprise resource planning system could flow into a manufacturing execution system to adjust production schedules and coordinate activity on the factory floor.

Autonomous mobile robots could then stage raw materials at production lines. Adding AI to that connected environment could help teams identify optimization opportunities and anticipate operating problems before they affected performance.

The report’s findings suggested that manufacturers were evaluating these technologies against established operating priorities. Improving quality, cited by 46%, reducing costs, at 40%, and minimizing operational risk, at 36%, remained among the industry’s leading objectives.

According to the FAQ, 48% of manufacturers identified AI and machine learning as the smart manufacturing capability delivering the greatest business impact among the technologies evaluated.

Rockwell said connecting AI projects to specific business objectives could help companies demonstrate their value and secure support for additional deployments. Its recommendations emphasized measurable results, including productivity improvements, lower costs and more consistent quality.

Workforce preparation was another requirement. The report found that 93% of manufacturers expected smart manufacturing initiatives to reshape their workforce, while 40% reported reskilling employees during the previous year.

The analysis recommended pairing technology modernization with workforce development, allowing employees to work with new systems and use the information those systems generated. It also emphasized the need to coordinate changes to technology, processes and responsibilities when extending projects across a business.

For manufacturers preparing new pilots, the FAQ recommended setting measurable business objectives and assessing future deployment requirements from the outset. Its proposed foundations included common data models, repeatable processes, integrated workflows and governance to support consistent use across facilities.

If you liked this, check these out...‍

• Can GenAI transform biomanufacturing? It’s complicated

• Kalsec and 2nd Nature explore AI-led discovery of natural food ingredients

• CSIRO joins global effort to turn agrifood waste into competitive protein using AI and fermentation