PepsiCo's AI twins spot 90% of supply chain issues early

What happened

Major food and drink manufacturers are deploying AI across their operations—from forecasting consumer demand and simulating supply chain scenarios to automating compliance checks and carbon tracking. PepsiCo is piloting digital twins (virtual models) and AI agents with Siemens and NVIDIA to test production and warehouse upgrades before physical construction; Danone is using predictive modelling to develop new products in areas like gut health; Unilever is applying similar techniques in its beauty and wellness division.

Why it matters

Food manufacturing involves constant trade-offs between quality, cost, service and sustainability. AI helps make those trade-offs more transparent and allows companies to respond faster to consumer needs without compromising safety. The technology also reduces guesswork by analyzing data quickly, even when data is limited or inconsistent—a common problem in food supply chains where supplier information varies.

What to watch

PepsiCo's collaboration can identify up to 90 per cent of potential issues before they physically occur and has enabled teams to validate new configurations within weeks while potentially reducing capital expenditure by up to 15 per cent. The pilot is currently underway in the US before being rolled out globally.