Agentic Control: How a Leading Dairy Manufacturer Increased Profits by €1.5M per Year
Executive Summary
A leading European dairy producer operated one of its large-scale spray drying lines below its true potential.
Residual moisture variability, conservative operator setpoints, and delayed lab feedback forced the plant to overdry product, sacrifice throughput, and consume excess energy. Performance varied significantly between shifts, and control relied heavily on individual operator experience.
Juna AI deployed its Agentic Process Control system — combining a physics-informed digital twin with reinforcement learning agents — to optimize spray dryer control in real time.
The agentic control recommendation system stabilized moisture, increased output, and reduced energy consumption, while operating safely within quality and regulatory constraints.
Impact
What Changed
Juna built a hybrid digital twin of the spray dryer, integrating:
- Heat and mass transfer physics
- Historical production data
- Real-time sensor signals
- Operator knowledge and constraints
On top of this twin, reinforcement learning agents were trained in simulation to learn optimal control strategies before deployment. Once live, the AI Co-Pilot:
- Predicts process behavior ahead of time
- Recommends optimal setpoints (Exhaust Air Temp, Main Air Temp, Fluid Bed Temp)
- Balances moisture precision, throughput, and energy simultaneously
- Continuously adapts to product variation, ambient conditions, and equipment drift
The result is a control system that learns from every batch, reduces operator variability, and runs the spray dryer closer to its true physical optimum — safely, transparently, and autonomously.
