Skip to content

BSF-1 — AI “Light-Off Operation” with zero sterility loss in UHT process.

Nestlé Vietnam · Posted 2 days ago

Consumer goodsComputer Vision
Budget
To be determined
Timeline
Based on solution provider proposal
Apply by
22 Nov 2026, 23:59 ICT

Problem statement

A self-running, self-correcting and self-optimising RTD operation: AI continuously coordinates UHT, detects deviations early, autonomously adjusts approved parameters and stabilises production without routine human intervention – people supervise remotely and handle exceptions only. Today: - Equipment is automated at machine or step level, but deviations, coordination between systems, recovery and optimisation still depend heavily on people; there is no autonomous end-to-end operating system yet.

Expected outcomes and success metrics

- More autonomous operating hours and auto-recovered deviations. - Fewer manual interventions. - Less process variability and quality loss, towards zero sterility loss.

Target users and impacted teams

- Manufacturing - Engineering - Quality - SHE

Data availability and readiness

- PLC tags, process parameters, UHT conditions, alarms - Batch records, quality results, vision-camera data - Equipment condition, maintenance history, past deviation/recovery events Note: Synthetic data could be potentially provided during the process, upon approval of the Nestlé team and in compliance with data law.

Integration, deployment and infrastructure

To be determined Note: Nestlé's current environment includes Microsoft Azure and Microsoft Copilot. Solutions that can work with this environment are welcome, and final integration and deployment requirements will be confirmed with the Nestlé team.

Relevant AI technologies

Computer Vision

Also: Machine Learning, AI agents. Based on Nestlé's description of autonomous detection, correction and recovery using process and vision-camera data.

Sign in to keep reading

You can read the first part of this use case. Sign in or create a free account to read the rest.