Intelligent fruit & vegetable sorting system

A grading and sorting line for a food-processing plant: hyperspectral imaging, multi-sensor fusion, flexible robotics and a production dashboard — software delivered by Jiding.

Industry
Fruit & vegetable processing
System type
Line control + vision integration + dashboard
Key technology
AI grading · hyperspectral imaging · flexible robotics
Jiding scope
Software design, development and commissioning

Before the project, grading on this line was entirely manual: standards varied between shifts, throughput could not keep up in peak season, and every extra pair of hands handling the fruit added bruising and shrink. Jiding designed and built the software that now runs the line — from the control layer through vision-model integration to the dashboard management reads every morning.

The problem, stated precisely

  • Inconsistent judgement. "Good" and "second grade" depended on who was looking. Buyers noticed.
  • Peak-season throughput. Manual sorting could not scale without adding shifts, and adding shifts made consistency worse.
  • Damage during handling. Every manual pick has a cost in bruising; that cost lands directly on margin.
  • No data. Nobody could answer "what percentage went to second grade yesterday, and why".

What we built

The delivered system has three software components, deliberately decoupled from each other:

  • Vision grading integration. Hyperspectral imaging combined with multi-sensor data feeds an AI grading model. The model outputs a grade and a defect class per fruit, with a confidence value the line can act on.
  • Line control software. Scheduling and dispatch for the flexible robots, coordinated with conveyor speed and grading results, with collision avoidance and damage-free pick paths as the first constraint.
  • Production dashboard. Throughput, grade distribution, shrink and downtime by shift and batch, on the shop-floor screen and on management devices.

Two engineering decisions that shaped the outcome

DecisionWhyTrade-off accepted
Decouple grading from line controlThe plant can change cameras or retrain the model without touching the control softwareA defined interface contract to maintain between the two
Prioritise damage-free over maximum speedShrink costs more than a few percent of cycle timeRobot paths are slightly longer; throughput optimised within that constraint

What the client got

  • Consistent grading across shifts, defined by the model rather than by individual judgement.
  • Throughput that scales with demand without adding inspection staff.
  • Less handling, therefore less bruising and lower shrink.
  • Shift and batch data that turns "we think second grade went up" into a number.
Figures for throughput and shrink improvement are shared under NDA. Industry reference values are available during a scoping call, with the assumptions stated.

Evaluating something similar?

Bring three questions to the first call: where is the bottleneck — judgement, speed or missing data? where is the boundary between the equipment vendor and the software team? and who owns model retraining after go-live? The answers decide whether a line software project succeeds.

MORE CASES

More Related Cases

Cloud-factory mobile app for a precision-electronics plant
Smart Factory IoTIntelligent manufacturing, precision electronics

Cloud-factory mobile app for a precision-electronics plant

A smart-factory mobile app (Android / iOS plus a web management backend) for a precision-electronics...

View Details →
Clinical laboratory · reagent supply-chain management platform
Smart Factory IoTIntelligent laboratory, reagent management platform

Clinical laboratory · reagent supply-chain management platform

A reagent supply-chain management system for a hospital clinical laboratory, extended from the deliv...

View Details →
Fruit & vegetable processing · intelligent sorting system
Smart Factory IoTSmart agriculture, smart manufacturing

Fruit & vegetable processing · intelligent sorting system

A grading and sorting line for a food-processing plant: hyperspectral imaging, multi-sensor fusion, ...

View Details →

View All Cases →

Planning a similar production line or enterprise system?

Start with our pricing logic, then let's talk requirements — cost structure, technology comparison and acceptance criteria, all laid out in the open.