Optical fruit and vegetable grading, sorting and packing line

Optical fruit and vegetable grading, sorting and packing line

This profile defines a sourcing requirement for optical produce grading and packing line. Suitable suppliers, origin, availability, and commercial terms are confirmed for the buyer’s project. The sourcing brief is structured around capacity, application, operating environment, required standards, destination, and delivery scope; the exact configuration requires supplier confirmation.

For a fresh-produce packhouse, a grading line is not simply a conveyor with a camera. The engineering problem starts with crop physiology, incoming product variability, defect definitions, required grade categories, target throughput, packaging mix, cold-chain timing, data traceability, and peak-season service risk.

The practical buyer question is therefore: which exact modules are engineered into the quoted line, for which crop and quality specification, and how will the complete facility be tested before peak harvest?

Packhouse process layers to define before procurement

Process layer	Technical solution scope	Project definition and acceptance
Reception and infeed	Bin and case reception, controlled tipping, depalletizing, accumulation conveyors, crop-specific singulation, and metered infeed.	Bin or case type, tipping method, fruit drop heights, incoming temperature, foreign-material risk, and maximum arrival rate.
Washing and pre-conditioning	Washing, drencher treatment, brushing, waxing, drying, hydrocooling, filtration, and sanitation modules selected for the crop.	Water quality, chemical compatibility, contact time, temperature, drainage, residue control, and changeover cleaning.
Sizing and grading	Mechanical sizing, load-cell weighing, electronic grading, color and diameter classification, and external or internal quality analysis.	Grade rules, size bands, weight tolerance, color thresholds, defect taxonomy, reject routing, and calibration samples.
Optical and internal analysis	RGB or multispectral external inspection, AI-assisted classification, UV options, and near-infrared BRIX, acidity, or internal-defect analysis.	Sensor package, crop and variety, defect library, illumination, model tuning, false-reject and missed-defect limits, and seasonal recalibration.
Packing and weighing	Count or weight filling, netting, clipping, box or crate filling, packing conveyors, fruit orientation, and robotic packing options.	Pack type, nominal weight or count, giveaway tolerance, tray or box dimensions, label position, rework route, and packs per hour.
End-of-line logistics	Palletizing, wrapping, strapping, finished-goods conveying, buffering, and cold-store or dispatch interfaces.	Pallet pattern, load stability, forklift or AGV interface, finished-goods buffer, and dispatch sequence.
Traceability and software	Reception-to-dispatch lot records, barcode, QR or RFID identification, ERP/WMS integration, reports, intermediate-stock control, and automated logistics.	Batch genealogy, label standards, API or export format, retention, user roles, audit trail, backups, and data ownership.
Commissioning and lifecycle support	Factory and site acceptance, operator training, preventive maintenance, remote support, critical spares, warranty, and peak-season service planning.	FAT, SAT, training records, spare-parts package, secure remote access, escalation path, response time, and software support.

Crop coverage and grading-line configuration

The line can be configured for:

• Cherry
• Blueberry
• Black fig
• Potato
• Onion
• Peach, apricot, and nectarine
• Citrus
• Kiwi
• Apple
• Melon
• Watermelon
• Tomato

This crop list matters because a line designed around apple handling should not be assumed to fit cherries, blueberries, figs, nectarines, onions, or citrus without a new technical review. Fruit diameter, shape, stem behavior, skin sensitivity, firmness, surface moisture, temperature, defect visibility, and packaging rules can change the required singulation, transport, camera, ejection, weighing, and packing architecture.

Crop-specific capacity and line-performance planning

Apple grading: capacity depends on fruit size, grade program, and line configuration

A representative apple-grading configuration is designed around an average capacity of 5 tonnes per hour with 20-program recipe memory, inverter-controlled conveyor speed, load-cell weighing, alarm diagnostics, sensor-controlled positioning, and stainless-steel product-contact conveyors. Actual sustained capacity is established with the specified apple size distribution, infeed quality, grading rules, packing format, and allowable handling damage.

A 5 t/h apple-grading reference configuration is meaningful only with its stated fruit-size distribution, inlet quality, lane count, grade split, packing mix, staffing, operating hours, and permissible handling damage. Final line capacity is defined as a sustained acceptance-test result for the buyer’s actual product and pack specification.

Cherry grading: direct packing versus pre-sizing changes the line

Cherry sorting can use direct packing or a two-stage pre-sizing and packing process. A full-surface optical inspection module can rotate and image the fruit for cosmetic-defect classification; the acceptance test defines defect classes, surface coverage, false rejects, missed defects, and sustained throughput.

Full-surface camera presentation means the handling and imaging geometry is designed to expose the fruit surface to the inspection system; it does not by itself guarantee perfect defect detection. Factory acceptance therefore uses a defect-by-defect test set covering cracks, bruising, color, softness, stem condition, pitting, doubles, undersize fruit, and the buyer’s export specification, with agreed detection and false-reject thresholds.

Citrus lines: treatment, quality analysis, grading, and packing are separate decisions

Citrus lines can classify by color and diameter, inspect external and internal defects, and measure sugar content and acidity with the selected optical and near-infrared package. Supporting modules can include drenchers, case tippers, selection conveyors, waxing, brushing, drying tunnels, box filling, electronic scales, netting, and clipping.

This means an RFQ should not simply say “citrus sorting line.” It should state whether the facility needs washing, treatment, waxing, drying, external optical grading, internal quality analysis, weight or count filling, netting, clipping, box filling, pallet preparation, traceability, and cold-store integration.

Root vegetables and bulk agricultural products: weighing and pack format drive the architecture

Bulk weighing and filling configurations can automatically dose 5–50 kg packs for potatoes, onions, carrots, radishes, and similar produce into plastic crates, nets, sacks, or bags.

Those specifications belong to that documented weighing/filling machine context. They should not be presented as the range of every optical produce grading and packing line grader or packing line. In a root-vegetable project, buyers should separately define soil and foreign-material handling, abrasion, product diameter distribution, target pack weight, legal metrology requirements, acceptable giveaway, net or bag specification, and upstream/downstream conveyor control.

Cherry hydrocooling: cooling performance needs its own acceptance protocol

The cherry hydrocooling module uses 0–4 °C process water, closed-loop recirculation, filtration, optional disinfection, and a modular tunnel layout. The thermal design is based on incoming fruit temperature, target pulp temperature, residence time, product load, water-to-product ratio, refrigeration duty, sanitation control, and water-quality management.

Hydrocooling performance is specified through measured fruit-pulp temperature, cooling uniformity, residence time, water hygiene, and post-cooling quality rather than a generic shelf-life percentage. The project defines incoming fruit temperature, target pulp temperature, residence time, water-temperature stability, flow distribution, microbiological control, filtration, disinfection method, water replacement, condensate and drainage management, and temperature measurement points.

Electronic quality sorting: define exactly what the camera is expected to classify

Optical grading architecture distinguishes external quality analysis from internal quality analysis.

External multispectral and AI-assisted inspection

External multispectral inspection can include:

• 360° fruit rotation for surface presentation;
• high-resolution imaging;
• multispectral cameras;
• classification by diameter, shape, color, and external defects;
• AI-assisted analysis;
• optional UV and the UV inspection modules for expanded defect detection.

The external-inspection sensor package can be configured for apple, stone fruit, kiwi, citrus, pear, mango, tomato, avocado, onion, lemon, cherry tomato, pomegranate, and other products after crop- and variety-specific validation. Camera geometry, lighting, spectral bands, presentation rollers, defect library, thresholds, and ejection timing are tuned for the actual product.

A buyer should request a crop-and-variety-specific defect matrix. “AI grading” is not an acceptance criterion. The project should name each commercial grade and each defect class, define minimum visible defect size where relevant, provide representative samples, and agree how false rejects and missed defects will be measured.

Internal quality analysis is a different sensing problem

Near-infrared internal-quality analysis can measure BRIX or sugar level and acidity at reference processing rates of up to 22 fruits per second, subject to crop, fruit size, lane count, presentation, and acceptance criteria.

Near-infrared internal-defect detection can be configured for apples, pears, onions, and related produce, with reference capacities above 500 fruits per minute, 2–10 lanes, fruit from 40 mm diameter, and multiple quality classes. Project acceptance uses representative samples and sustained line loading.

These module-level figures must not be represented as complete packhouse throughput. A complete line can be limited by infeed, singulation, camera presentation, ejection, packing, box supply, palletizing, labor, changeovers, or cold-store logistics.

A better way to specify line capacity

Instead of asking only for “tonnes per hour,” AgriTech.tr recommends defining a capacity envelope:

Capacity variable	Example procurement definition
Incoming product	Named crop, variety, harvest condition, size distribution, temperature, and maximum field debris or wet-product condition.
Grade mix	Expected percentage in each commercial grade, reject percentage, and number of simultaneous outlets.
Product handling	Maximum permitted drop height, transfer count, bruising or damage test, and accumulation time.
Throughput	Sustained tonnes/hour or pieces/hour at the agreed grade mix for a defined test duration.
Packing mix	Percentage of product going to each box, tray, punnet, net, sack, bag, count-fill, or weight-fill format.
Changeover	Maximum time to change variety, grade recipe, pack format, label, or customer order.
Availability	Planned operating hours, maintenance window, critical-spares strategy, and target technical availability.
Saleable output	Accepted packed product per hour after rejects, rework, downtime, pack changes, and quality holds.

This prevents a high mechanical headline capacity from being mistaken for the facility’s real export-ready output.

How to test grading accuracy at FAT and SAT

For a new optical or electronic grading project, the acceptance plan should be agreed before manufacturing is complete.

A practical test can include:

1. Build a reference sample set. Use the actual crop, important varieties, size bands, color ranges, and known defect classes from the buyer’s operation.
2. Create agreed ground-truth grades. Buyer and supplier quality teams should jointly classify the reference fruit before the machine test.
3. Run blinded lots. Feed the test fruit without manually steering known defects toward a preferred lane.
4. Measure the classification matrix. Record correct grade, false reject, false accept, and cross-grade errors for every defined defect or grade.
5. Measure handling damage. Inspect fruit before and after the line using an agreed immediate and delayed inspection method suitable for the crop.
6. Measure sustained throughput. Run the line for an agreed duration at the stated input mix, not only as a short demonstration.
7. Test reject and rework routing. Confirm that product can be recovered, re-inspected, or diverted without losing lot identity.
8. Test data records. Verify the lot, grade, weight, count, operator, order, time, and label data that the project is supposed to retain.
9. Test restart behavior. Simulate power interruption, sensor fault, network loss, camera alarm, full downstream buffer, and emergency stop.
10. Sign an exceptions list. Any remaining mechanical, software, safety, quality, or documentation issue should have an owner and closure date.

The exact acceptance method must be adapted to the crop and contract, but the principle is simple: a demonstration video is not a substitute for a written acceptance protocol.

Packing automation: protect the commercial pack specification

Packing automation can combine box filling, packing tables, punnet filling, robotic packing, fruit orientation, final optical quality control, weighing, netting, clipping, conveying, and pallet wrapping in one coordinated line.

For a Türkiye packhouse project, the buyer should document:

• box, tray, punnet, net, sack, bag, or crate dimensions;
• target count or nominal weight;
• allowable giveaway and underweight rule;
• fruit orientation and presentation requirements;
• packaging-material tolerances;
• label content, barcode format, GS1 or customer-specific rules;
• box closing, clipping, netting, strapping, or wrapping requirements;
• rework logic when a pack fails weight, label, or quality inspection;
• customer-order sequencing and pallet pattern;
• cold-store and dispatch interface.

This is particularly important for exporters serving several customers with different grade names and pack formats in the same week.

Traceability, ERP integration, and data ownership

The traceability layer follows product from reception to warehouse dispatch using barcode, QR or RFID identification, real-time integration, reports, intermediate-stock control, and automated logistics.

Traceability software and packing automation are optional project modules. The signed quotation itemizes licenses, terminals, scanners, printers, databases, ERP/WMS interfaces, support services, data ownership, and acceptance tests.

Before buying, ask:

• Can a finished pallet be traced back to grower, field, harvest date, intake lot, and processing order?
• Can one intake lot be split into several grades and customers without losing genealogy?
• What happens when a label is reprinted?
• Does the system support barcode, QR, RFID, or customer-specific identifiers?
• Which ERP or WMS interfaces are standard and which are custom?
• Is there an API, database export, CSV, or other documented export method?
• Who owns the data?
• Can authorized remote-support personnel access production data, and how is that access approved and logged?
• Where is data stored and backed up?
• What happens to historical records if the service agreement ends?
• How are user roles, audit logs, and remote-access credentials managed?

Installation engineering and packhouse layout

A technically capable grader can still underperform in a poor facility layout. Before contract signature, request a dimensioned line drawing and review:

• intake and finished-goods traffic;
• forklift and pedestrian separation;
• emergency exits and machine access;
• product drop heights and transfer points;
• manual inspection ergonomics;
• rejected-product and rework flow;
• empty-box and packaging-material supply;
• full-box accumulation;
• water, drainage, and sanitation zones;
• electrical panels and cable routing;
• compressed air requirements;
• network and control cabinet locations;
• cold-room doors and temperature transitions;
• maintenance pull space for motors, belts, rollers, cameras, and sensors;
• access for cleaning beneath conveyors and platforms;
• future expansion points.

The RFQ should also identify who is responsible for civil works, mezzanines, drains, utilities, electrical distribution, network cabling, compressed air, refrigeration interface, safety guarding, local permits, and final machine integration.

Service, spare parts, and peak-harvest risk

Lifecycle support is structured around commissioning, operator and maintenance training, a preventive-maintenance plan, safe remote assistance, critical spare parts, peak-season escalation, response targets, and software support. The contract identifies the service location, covered labor and travel, stocked items, lead time for non-stock parts, remote-access controls, and responsibility during the buyer’s production season.

That is a relevant procurement signal, but buyers still need project-specific service terms. Ask for:

• critical spare-parts list with part numbers and recommended on-site stock;
• local stock versus imported lead times;
• remote-support hours and cybersecurity method;
• telephone escalation during peak harvest;
• guaranteed or target response times, if offered;
• travel and labor charging rules;
• software and camera-calibration support;
• preventive-maintenance schedule;
• operator and maintenance training;
• warranty start point and exclusions;
• obsolescence and upgrade policy for cameras, PCs, PLCs, drives, and software.

A line that stops for two days during a six-week cherry campaign has a different business risk from a line with a long annual operating season. Service should therefore be evaluated against the crop calendar.

Project definition checklist for fruit and vegetable grading and packing lines

1. Name the crop and varieties. Do not procure a generic “fruit line.”
2. Define commercial grades and defects. Translate customer specifications into measurable machine rules.
3. State the incoming product envelope. Size, temperature, field condition, wetness, debris, and expected defect mix.
4. Define throughput as sustained saleable output. Include grade mix, packing mix, rejects, rework, and changeovers.
5. Request the complete module list. Infeed, washing, treatment, grading, cameras, internal analysis, weighing, packing, pallet handling, software, and utilities.
6. Separate standard from optional equipment. Especially optical modules, internal quality sensors, robotics, traceability, remote access, and cold-chain equipment.
7. Agree FAT and SAT before delivery. Include quality classification, capacity, handling damage, data, safety, restart, and documentation tests.
8. Audit the line layout. Review product flow, labor flow, maintenance access, hygiene, cold chain, and future expansion.
9. Lock down data ownership and integration. Put ERP/WMS interfaces, labels, API/export, backups, and support access in writing.
10. Build a peak-season service plan. Identify critical spares, response path, maintenance schedule, and responsible people.
11. Check references by crop and line architecture. Similar fruit, capacity, technology modules, pack mix, commissioning year, and service model are more useful than a generic customer list.
12. Treat savings claims as project-specific. Labor, waste, giveaway, throughput, and export-rejection improvements need a documented baseline and measurement method.

AgriTech.tr project scope for grading and packing lines

The sourcing brief can be used to define the application, technical interfaces, documentation, and service requirements before supplier research begins.

For guidance on defining capacity, optical grading requirements, internal quality analysis, packing automation, traceability, hydrocooling, pallet logistics, FAT/SAT criteria, or supplier-comparison questions, contact info@agritech.tr.

The cover and in-article photographs provide fruit-handling and packhouse context and are used under the Pexels License. Photographer and source credits are retained in the reference section.

Project capacity, grading accuracy, defect-detection performance, product damage, labor savings, waste reduction, shelf-life effects, line availability, price, warranty, and service response should be confirmed in project-specific documents and acceptance tests.

Procurement and project delivery

AgriTech.tr can structure the technical requirement and compare current supplier responses for the buyer’s project. Installation, commissioning, operator training, warranty, spare-parts, and after-sales scope must be confirmed in each supplier quotation. The final scope should be documented against the approved application, capacity, site conditions, destination, and delivery schedule in the selected supplier quotation and contract.