Gujarat deploys AI grain analysers at 10 MSP paddy procurement centres

Gujarat State Civil Supplies Corporation has tested 6,191 paddy samples using AI-based grain analysers across 10 centres in four districts during KMS 2025-26, aiming to speed grading, improve consistency and strengthen foodgrain quality controls.

— Source publishedMon, 7 Sept, 2026, 10:50 IST·First seen Mon, 7 Sept, 2026, 10:52 IST·Source BL · Consumer & Economy

What happened

Gujarat State Civil Supplies Corporation · Gujarat deployed AI grain analysers at 10 MSP paddy procurement centres, speeding quality checks and testing 6,191

Key facts

  • 10 procurement centres
  • 4 districts
  • KMS 2025-26
  • 6,191 paddy samples analysed
  • 3,379 Grade A samples (54.6%)
  • 2,812 Common samples (45.4%)
  • 341 samples rejected
  • 13,334 farmer registrations
  • Dholka rejection rate: 24%
  • Two people required under the existing grading system
  • 100% cross-verification for tur dal and gram samples

Why this matters

Grain-analysis technology providers and procurement-platform firms may find partnership opportunities with state civil-supplies agencies seeking scalable, auditable MSP grading systems.

What to watch

  • Announcement of additional analyser installations, new districts or extension to wheat and other MSP crops.
  • Average sample-processing time and queue reduction versus the former two-person manual process.
  • Rejection, Grade A and Common-grade rates by centre, district, farmer type and arrival period.
  • Gap between AI results, human reassessments and accredited laboratory tests.
  • Volume and outcome of farmer appeals or procurement disputes related to analyser decisions.
  • Tender awards for analysers, maintenance, calibration, software integration or data platforms.
  • Central government or other state procurement agencies adopting similar AI-grading specifications.
  • Audit rejection and grade outcomes centre-by-centre, especially Dholka, against manual checks and referee-lab results.
  • Publish analyser accuracy, turnaround-time and dispute-resolution metrics to build farmer and procurement-agent trust.
  • Create a machine-result appeal workflow with sample retention, human review and calibration logs.
  • Use aggregated quality data to target farmer advisories on drying, cleaning, varietal purity and storage before procurement season.
  • Evaluate integration of analyser results with procurement receipts, warehouse acceptance, payments and traceability systems.
  • Assess whether reduced grading time permits centre consolidation, longer procurement hours or redeployment of inspection staff.

Also reported by