How Operational AI Is Turning Supply Chain Data Into Real-Time Decisions Across Manufacturing and Planning
For more than a decade, supply chain technology investments have focused heavily on visibility. Organizations have built dashboards, forecasting platforms, analytic environments, and reporting systems designed to provide greater awareness of operational performance. Yet according to Mike Romeri, founder of A2go, many of today’s supply chain challenges arise after the data has already been collected. “Organizations can see what is happening across their operations, but turning that information into timely and high-quality decisions remains a persistent challenge,” he notes.
That distinction is becoming increasingly important as operational complexity continues to grow. While interest in agentic AI continues to expand, organizations are becoming more selective about where and how they invest. Forecast shows that more than 40% of agentic AI projects will be canceled by the end of 2027 due to escalating costs, unclear business value, or inadequate risk controls. In a poll of 3,412 webinar attendees, only 19% reported making significant investments in agentic AI, while 31% said they were still taking a wait-and-see approach. The findings suggest that organizations are placing greater emphasis on AI initiatives that can demonstrate measurable operational value.
From Romeri’s perspective, the central issue is operational judgment at scale. Teams already have access to enterprise resource forecasting tools and performance metrics. What often remains difficult is determining which actions should be prioritized when demand shifts, supplier conditions change, or production schedules need adjustment. He explains that supply chains generate thousands of insights every day, yet many of those decisions still depend on manual review and individual expertise.
This challenge helps explain why many enterprise AI initiatives have struggled to produce meaningful operational results. “Organizations frequently deploy analytics tools that generate insights but remain disconnected from execution,” Romeri says. “In other cases, AI capabilities are introduced through broad transformation programs that require significant organizational change before value can be realized.” Successful AI adoption depends on embedding intelligence directly into operational workflows rather than treating AI as a standalone capability.
Romeri believes a different category is emerging. Rather than focusing primarily on analytics, operational AI is increasingly being used to support decision-making inside existing business processes. According to the company, A2go develops AI-powered supply chain solutions for manufacturers and distributors and other complex enterprises, and deploys more than 35 specialized AI agents designed to address specific operational challenges such as forecasting, inventory management, scheduling, supplier coordination, and order fulfillment. These agents integrate with existing systems and apply intelligence directly to operational decisions.
“The real opportunity is helping organizations improve the quality and consistency of decisions they are already making every day,” Romeri says. “When operational intelligence becomes part of the workflow, teams can focus their attention where it creates the greatest value.”
In practice, Romeri notes that operational AI functions by continuously monitoring signals across the supply chain. “Demand fluctuations, supplier delays, inventory changes, production constraints, and customer requirements can all be evaluated in real time,” he explains. “AI agents identify deviations, prioritize issues based on business impact, and recommend actions that align operations with current conditions rather than relying solely on static planning cycles.”
One of the most significant developments, he notes, is the ability to capture and reuse operational expertise. From Romeri’s perspective, many organizations rely heavily on experienced planners and operations leaders whose decision-making knowledge exists largely in individual experience. “Operational AI creates an opportunity to codify that expertise and apply it consistently across teams and workflows,” he says. “As a result, operational judgment becomes institutional knowledge rather than remaining dependent on a limited number of individuals.”
Deployment speed is another factor influencing adoption. According to Romeri, organizations increasingly favor modular implementations that allow them to address a specific operational challenge, measure outcomes, and expand from there, with the aim of having all agents communicate and coordinate decisions for optimization. Romeri notes that this approach enables companies to realize value incrementally while building familiarity with AI capabilities.
Evidence of this shift can already be seen in large-scale environments. Romeri points to A2go’s work with JBS Foods, where AI-supported decision systems were incorporated into a complex order fulfillment environment. According to him, the project provided a practical example of how operational AI can be embedded into day-to-day decision processes within a large organization.
The broader market appears to be moving in a similar direction. It is projected that spending on supply chain management software with agentic AI capabilities will increase from less than $2 billion in 2025 to $53 billion by 2030. It is also expected that 60% of enterprises using supply chain management software will adopt agentic AI features by 2030, up from just 5% in 2025. The projections suggest that organizations are increasingly looking to embed AI directly into operational workflows rather than limiting its use to planning and analysis.
According to Romeri, the next phase of supply chain performance will depend on how effectively organizations convert operational information into timely and consistent decisions. “Competitive advantage increasingly comes from decision speed, decision consistency, and operational resilience,” he says. “The organizations that improve those capabilities will be the ones best positioned to navigate complexity as supply chains continue to evolve.”
