Global supply chains are under pressure to move faster and hold up better, but a new industry report suggests many companies still aren’t ready for sudden shocks. The Industrial Agility Assessment 2025 found that only 45% of manufacturers and distributors describe themselves as highly or extremely agile — the lowest confidence level recorded in five years. That figure sits awkwardly next to another one from the same report: 64% of those same businesses say AI adoption is a top priority for their digital transformation plans.
The gap between ambition and readiness is hard to ignore. Companies are investing in artificial intelligence to improve forecasting, inventory control, and route planning, but the tools haven’t yet translated into the resilience executives say they want.
Why Agility Is Flat While AI Investment Rises
One executive quoted in the assessment put it plainly: “You can’t just grow at all costs anymore, you need solid foundations and balance.” That sentiment reflects a broader shift in how supply chain leaders view their operations. Growth alone no longer counts as success if the underlying systems can’t handle disruption.
The report’s authors point to a disconnect between what companies plan to do and what they’ve actually implemented. AI is being positioned as the fix for everything from demand forecasting to equipment maintenance, but the data suggests that most organizations are still in the early stages of figuring out how to use it well.
Related: Retail Stores Turn Safety Policies into Action
There’s a plausible path forward, though it’s not guaranteed. If the current wave of AI pilots matures into fully deployed systems, the agility numbers could climb sharply in the next assessment cycle. But that depends on whether manufacturers can move past isolated experiments and build AI into their core planning processes. The tools are proven; the organizational discipline around them is not.
Forecasting and Maintenance: Where AI Is Changing Daily Work
Demand forecasting is one area where AI is already making a visible difference. Traditional methods lean on historical data and human judgment, which often miss the kind of real-time shifts that define today’s markets. AI systems can scan much larger datasets, spot patterns, and adjust predictions as new information arrives.
The assessment highlights manufacturers using AI simulations to stress-test their own supply chains — running scenarios to see how operations would hold up under different kinds of pressure. As one food manufacturing executive said, “We’re looking back at performance, monitoring in real time, and simulating scenarios to test our ability to adapt. It’s building confidence and resilience.”
That approach moves companies from reacting to problems after they appear to anticipating them before they hit. Inventory levels stay closer to optimal, stockouts drop, and the response to market shifts gets faster.
Related: German Doner Kebab appoints Wayne Shurvinton as CEO
Predictive maintenance is the other major application showing up in the report. Even a short equipment failure can ripple through an entire supply chain, so manufacturers are using AI to monitor sensor data and flag anomalies before they become breakdowns. The goal is to extend asset life and keep production lines moving without constant human oversight.
In the automotive sector, AI is being used on the shop floor to guide workers through visual signals and electronic Kanban systems — tools that show exactly what’s needed and when. One automotive executive described it this way: “We’re using AI to simplify complexity. It’s not just about automation; it’s about guiding people with the right information at the right time.”
That combination of human judgment and machine assistance is helping manufacturers cut errors in low-volume, high-precision environments where mistakes are costly.
The report stops short of declaring a turning point. The numbers show interest, not yet results. But the direction is consistent: AI is moving from boardroom discussion to daily operations, and the companies that figure out how to integrate it properly will likely be the ones reporting higher agility next time around.
