Why it matters
Without AI, inconsistent particle size and paste temperature can lead to suboptimal malaxation, higher free acidity, and defects that affect extra virgin status and PDO compliance.
From design to safety
Design
Initial AI design for mill settings
AI defines sensor placements, data interfaces, and control logic for hammer mill, disc crusher, and screens during reception and crushing. It models how particle size and temperature evolve through different screen openings and disc gaps, building a baseline that aligns with IOC trade standards and PDO requirements.
- hammer mill
- screen
- disc crusher
Build
Hardware integration and calibration
Install sensors and actuators, connect SCADA, and calibrate paste size and temperature thresholds for real-time feedback. Define calibration records, field mapping, and validation tests that certify readiness for campaigns.
- hammer mill
- screen
- paste particle size sensor
Operation
Real-time control of crushing and paste formation
AI continuously reads particle size and temperature, clamps motor loads, selects screens, and logs adjustments per batch. It correlates incoming harvest data with target profiles to steer batch-level decisions.
- hammer mill
- paste particle size sensor
- malaxer mixer
- paste temperature sensor
Maintenance
Diagnostics and wear monitoring
Track wear on hammers, screens, and disc crushers, schedule preventive maintenance, and update models after campaigns. Capture maintenance history, parts replacement, and regression tests to sustain accuracy.
- screen
- disc crusher
- paste particle size sensor
Safety
Quality and safety governance
Record calibration, ensure HACCP, track deviations, and require certified personnel sign-offs for critical settings. Maintain audit trails for compliance with IOC and PDO sanitation standards.
- paste temperature sensor
- malaxer mixer
- screen
Part by part
- 1
hammer mill
- Watch for
- Wear on hammers, imbalance, and sudden vibration spikes that threaten particle consistency.
- AI action
- Read hammer mill vibration and rpm, adjust feed rate and rotor speed.
- 2
screen
- Watch for
- Screen blinding, wear, or loosening affecting particle size distribution.
- AI action
- Monitor screen openings, adjust feed distribution, trigger screen change when needed.
- 3
disc crusher
- Watch for
- Wear on discs, uneven gap, and surface chatter impacting paste consistency.
- AI action
- Check disc gap, adjust feed, notify maintenance when wear exceeds threshold.
- 4
paste particle size sensor
- Watch for
- Deviations in particle size distribution beyond target range.
- AI action
- Flag out-of-spec batches and trigger alternative screen or hammer setting.
- 5
paste temperature sensor
- Watch for
- Paste temperature drift during malaxation or transfer.
- AI action
- Balance malaxer jacket temperature, adjust flow, notify operator.
- 6
malaxer mixer
- Watch for
- Inadequate mixing homogeneity or overheating risk.
- AI action
- Modulate paddle speed and duration, compare batch to reference curve.
Worked example
Catching a suboptimal paste profile before malaxation
Illustrative scenario with sample data.
A mid-harvest batch shows rising paste particle size and slight temperature spike during crushing. The system flags a potential mismatch between feed rate, screen size, and disc gap.
Signal in
fields: harvestDate=2026-11-07, batchId=B20261107-03, hammerSpeed=1800rpm, screenOpen=12mm, discGap=0.8mm, pasteTemp=26.5C, particleSize=2.8mm, flowRate=520L/min, ambientTemp=22C
System out
alert: paste particle size high; action: reduce hammer speed to 1700rpm, switch to 10mm screen, adjust disc gap to 0.75mm; log batch B20261107-03 adjustments, notify mill master; QA to sample for polyphenol retention.
Human sign-off: Mill master reviews AI log, approves setting changes, and records final parameters for the campaign; agronomist signs the QA cert after review of batch integrity.
How to roll it out
Phase 1
Weeks 1 to 4
Baseline data collection, sensor calibration, and model validation against initial harvest runs.
Phase 2
Weeks 5 to 8
Live testing with feed-forward controls and operator training on screen and disc adjustments.
Phase 3
Weeks 9 to 12
Operational deployment with ongoing monitoring, dashboards, and integration into QA records.
Where people stay in charge
- All critical settings remain under the authority of the mill master and QA lead to ensure product safety.
- Decisions affecting paste quality must be signed off by the agronomist and certified food safety manager.
- The AI can propose adjustments but cannot override regulatory limits or critical process boundaries.
Questions owners and crews ask
What data does the AI monitor at the hammer mill stage?
It monitors hammer speed, motor load, vibration, feed rate, and distance to screen openings to guide adjustments that stabilize paste particle size.
How does paste particle size influence oil quality?
Particle size affects extraction efficiency, paste temperature dynamics, and malaxation behavior, impacting polyphenol retention and overall oil quality signatures.
Who approves changes to mill settings?
The mill master, QA lead, and certifying agronomist review and sign off on any new settings before production continues.
What records are kept for traceability?
Batch records, sensor logs, parameter changes, operator notes, and QA certificates are archived for PDO and IOC traceability.