⌘K
ScopePeriod

Fleet & Logistics Intelligence

Weighbridge anomaly detection · risk indicators for human reviewAll operations
Tickets (7 days)
588
Mill A · Mill B
Net FFB on tickets
4,467t
7 days · scored tickets
Risk indicators ≥ 0.60
6
1.0% of tickets
Requires review
4
awaiting reviewer
Under investigation
1
evidence gathering
Model precision (90 d)
68%
reviewed flags confirmed as issue or gap

Pattern detectors

Tickets matched per detector · last 7 days · click to filter the ticket table

Synthetic data · simulated model output

Featured · Ticket WB-260928-0187Risk indicator: highRequires Review

Weighbridge Anomaly v2.1 · isolation forest + sequence rules · requires human review

Confidence76%
Anomaly score0.91
Time28 Sep 18:19
TruckTR-027
Supplier on ticketFFB Trader T-12
Gross / net14,215 / 8,095 kg
Tare vs truck master6,120 vs 6,800 kg (-10.0%)
Potential variance≈ 680 kg FFB
Why it was flagged
  • Tare weight deviation (−10.0% vs truck master)
  • Repeated ticket timing (3 tickets within 11 min)
  • Unusual supplier/truck combination (company truck on trader ticket)
Evidence · TR-027 tare history vs truck master (kg)
Shaded band = ±2% tolerance. Last recorded tare 6,120 kg (−10.0%).
Evidence · ticket timing
17:1917:4918:1918:49
FFB Trader T-12 tickets (3 in 11 min) Other Mill A tickets
Contribution to anomaly score (points of 100)
Tare deviation vs truck master
+34 pts
Ticket time clustering
+22 pts
Supplier/truck pairing rarity
+19 pts
GPS dwell at declared CP
+8.0 pts
Net weight vs supplier pattern
+5.0 pts
Operator shift (normal)
−4.0 pts
◀ lowers predictionraises ▶
This is a risk indicator that requires review, not a finding. Model output alone is not evidence of wrongdoing by any driver, supplier or employee. Possible benign causes include a tare captured after the spare wheel or tailgate was removed, or load-cell drift.

Weighbridge tickets

37 of 588 tickets · sortable · anomaly score 0–1

TicketTimeMillTruckSupplierNet (kg)Tare vs masterAnomaly scoreRisk indicatorsStatus
WB-260928-018728 Sep 18:19Mill ATR-027FFB Trader T-128,095−10.0%0.91HighTare weight deviation (−10.0% vs truck master) +2 moreRequires Review
WB-260927-016227 Sep 10:50Mill AEXT-353Smallholder Co-op S-074,761−0.6%0.84HighPotential duplicate ticket (same truck, net weight ±0.4%, 6 min apart)Requires Review
WB-260926-017126 Sep 13:38Mill ATR-014Estate E9,278−0.8%0.77ElevatedNet weight 38% above supplier's 90-day pattern +1 moreRequires Review
WB-260925-030525 Sep 07:03Mill BTR-027Estate D8,832+0.2%0.72ElevatedRoute inconsistency — GPS shows no stop at declared collection pointUnder Investigation
WB-260928-032128 Sep 13:21Mill BTR-016Estate C8,496+0.1%0.69ElevatedRepeated truck pattern — 4 deliveries/day exceeds cycle-time feasibilityRequires Review
WB-260924-015824 Sep 08:57Mill AEXT-306Smallholder Co-op S-078,438−0.1%0.66ElevatedAbnormal tare weight (+7.8%) — possible retained loadClosed - No Issue
WB-260922-031622 Sep 11:13Mill BTR-032Estate C8,710+0.3%0.55WatchNet weight above supplier's 30-day pattern (+18%)Clear
WB-260926-030026 Sep 06:29Mill BTR-008Estate D7,622−0.5%0.54WatchArrival outside usual delivery windowClear
WB-260926-017026 Sep 12:43Mill AEXT-337Smallholder Co-op S-036,684−1.1%0.54WatchLoose-fruit share unusually high for declared originClear
WB-260926-031226 Sep 10:49Mill BEXT-347FFB Trader T-195,410+1.0%0.53WatchShort cycle time vs GPS distanceClear
WB-260922-018822 Sep 18:30Mill AEXT-355Smallholder Co-op S-036,275−1.2%0.53WatchLoose-fruit share unusually high for declared originClear
WB-260928-032428 Sep 14:50Mill BEXT-307Plasma Scheme P-025,627+1.3%0.52WatchShort cycle time vs GPS distanceClear
WB-260923-019523 Sep 21:06Mill AEXT-360Smallholder Co-op S-037,490+0.4%0.52WatchTare within tolerance but trending down over 5 ticketsClear
WB-260922-016422 Sep 10:34Mill AEXT-325Smallholder Co-op S-078,796−1.2%0.50WatchShort cycle time vs GPS distanceClear
WB-260926-019426 Sep 20:49Mill AEXT-345Smallholder Co-op S-074,596−1.5%0.49WatchShort cycle time vs GPS distanceClear

Weighbridge analytics produce risk indicators that require review. They never label a supplier, driver or employee as fraudulent, and model output alone is not evidence of wrongdoing. All tickets are synthetic.

V-TEKI — IT & Business ConsultingDeveloped by V-TEKI · IT & Business ConsultingPrime Agri AI Intelligence Platform — prototype. All operational records are synthetic; model outputs are simulated; financial impacts are scenario estimates; recommendations require human validation. No actual company confidential data is used.