Cross-Module Insight — The Wet Fuel Scenario
“AI that connects the dots humans can't — across systems, in real-time.”
📖 The Scene
At 08:15, it started raining at Block C. By 10:05, NayaCore had correlated the rainfall with wet fuel in the warehouse and a boiler efficiency drop — data from three independent modules that no human could connect in real-time. After the 5th occurrence, the pattern graduated to a deterministic rule that now fires in <100ms, saving $2.96 in AI tokens every time it rains.
Siti
Control Room Operator (T2.5)
📱 See It In Action
Real WhatsApp conversations between operators, supervisors, and NayaCore — in Bahasa Indonesia, exactly as they appear on the phone.
10:06 WIB · Confidence: 87%
🔗 Pola Bahan Bakar Basah
NayaCore mendeteksi korelasi dari 3 modul sekaligus:
🌧️ CAKRAWALA:
Hujan deras 72mm (90 menit)
di Block C sejak 08:15
📱 PLANTATION:
Budi: "Fuel basah di gudang,
susah nyala boiler"
🏭 MILLOS:
Boiler efisiensi: 71% (normal: 85%)
Press #3 energi: +18% di atas normal
Sterilizer tekanan: 2.1 bar (min: 2.5)
🔍 ANALISIS AI:
Hujan → fuel basah → boiler lemah →
steam kurang → sterilizer lambat →
press butuh energi lebih
💡 REKOMENDASI:
Naikkan waktu sterilizer dari
90 → 105 menit untuk kompensasi
kualitas steam yang rendah.
🔐 Setujui rekomendasi →10:06
Sterilizer cycle time diperbarui:
90 → 105 menit
Disetujui oleh: Siti Aminah
⏰ 10:15 WIB · WebAuthn verified10:16
Mill 1 · 10:06 WIB
Hujan deras di Block C menyebabkan fuel basah → boiler turun ke 71%.
💡 Sterilizer waktu dinaikkan 90→105 min.
📊 Dampak OEE: -2.1% (sementara)
🏭 Status: menunggu approval Siti.
Cross-Module Insight: The Wet Fuel Incident
How NayaCore correlates weather, WhatsApp chatter, and machine sensors to solve a hidden efficiency loss.
Unstructured Field Chatter
Workers discuss heavy rain and wet fuel in their WhatsApp group. NayaCore's NLP engine passively listens, extracting the condition 'Fuel Basah' (Wet Fuel) and the location 'Block C'.
SCADA & Weather Correlation
The Cakrawala module confirms 72mm of heavy rainfall in Block C. Simultaneously, MillOS V2 detects the boiler efficiency dropping to 71% and the Press requiring 18% more energy to compensate for poor steam quality.
AI Inference & Recommendation Push
NayaCore's AI connects the dots: Rain caused wet fuel, which lowered boiler efficiency and steam pressure, forcing the press to work harder. It proposes an immediate fix (increasing sterilizer time) and pushes the recommendation to the supervisor via WhatsApp.
Execution & Executive Briefing
The supervisor approves the change directly from WhatsApp. The system automatically adjusts the SCADA setpoints. An exception report, complete with an impact analysis PDF, is routed to the Regional Director.
📊 Before vs. After
Traditional Operations vs. NayaCore
| Metric | Traditional | NayaCore |
|---|---|---|
| Root cause identification | Hours to days (check boiler, not weather station) | 107 minutes (auto-correlation across 3 modules) |
| Data sources connected | Single-module SCADA only | Weather + WhatsApp + Sensor (3 modules) |
| Human involvement needed | Manual investigation by experienced engineer | One-tap approval of AI recommendation |
| Pattern memory | ❌ Diagnosed from scratch every rainy season | ✅ Graduates to instant rule after 5 occurrences |
| AI cost over time | N/A | $0.08 → $0.00 per detection (graduated to rule) |
| Offline capability | ❌ Requires cloud analysis | ✅ Graduated rules fire locally, no internet |