Coffee Wet-Mill (Pulping Station), Dry-Mill Hulling & Nairobi Coffee Exchange (NCE) Cupping ERP: Cherry Intake, Parchment Drying, AA/AB/PB Grading & DSS Settlement

Coffee cooperative societies, estate pulping stations (wet mills), licensed commercial dry mills, and specialty green-coffee exporters across Nyeri, Kirinyaga, Murang’a, Kiambu, Embu, Meru, Machakos, Kericho, Bungoma, and Trans Nzoia process some of the world’s highest-scoring washed Arabica coffees (SL28, SL34, Ruiru 11, Batian). Yet across the value chain, manual cherry passbooks, unmonitored parchment drying tables, opaque dry-mill hulling outturn reports, and complex Nairobi Coffee Exchange (NCE) Direct Settlement System (DSS) splits create cherry-to-parchment conversion leakages, over-dried or high-moisture parchment that bleaches green beans and depresses cupping scores, disputes over Milling Loss % and grade distribution (AA vs. AB vs. C/T), and non-compliance with European Union Deforestation Regulation (EUDR) farm-level geolocation mandates.

At devManue, we engineer specialized Coffee Wet-Mill, Commercial Dry-Mill & NCE Auction / Specialty Export ERP platforms that provide unbroken digital custody from the smallholder’s coffee trees and wet-mill cherry hopper to dry-mill optical color sorters, SCA cupping tables, NCE auction catalogs, and direct bank/mobile-money grower settlements.

1. Wet-Mill Cherry Intake, Floatation Cherry Sorting & Fermentation Tank Lot Tracking

Premium Kenyan coffee quality is locked in—or destroyed—within the first 24 hours at the wet mill (factory). Our wet-mill module digitizes cherry reception and wet-processing batch control:

• Digital Cherry Hopper Weighing & Cherry Grade Separation (Cherry I vs. Cherry II / Mbuni): As farmers deliver ripe red cherry to the factory intake hopper, digital scales integrated with grower membership cards record exact kilograms of Cherry Grade I (ripe red) separately from under-ripe/over-ripe Cherry Grade II and sun-dried Mbuni, updating the farmer’s SMS/digital passbook in real time.
• Pulper Disc Calibration, Density Floatation & Fermentation Tank Batching: After disc or drum pulping separates skins from mucilage-coated parchment, water density channels float off lighter Lights (P3) from dense prime Firsts (P1) and Seconds (P2). Each day’s cherry intake is assigned a unique Outturn Batch ID tracked through fermentation tanks (monitoring 16–36 hour mucilage breakdown, water recirculation, and washing channel grading).
• Raised Drying Table Moisture Telemetry & Conditioning Bin Custody: Tracks parchment batch placement across skin-drying tables and raised African drying beds, logging daily portable moisture-meter readings until parchment reaches the statutory 10.5% to 11.5% moisture window before bagging into conditioning store bins.

2. Commercial Dry-Mill Hulling Outturn, Screen-Size Grading (AA/AB/PB) & Milling Loss Engine

When conditioned parchment arrives at a licensed commercial dry mill accompanied by a coffee movement permit, growers and cooperative managers demand transparent visibility into parchment hulling loss (typically 18%–22% husk weight) and the exact yield of high-premium grades (AA, AB, PB) versus lower grades (C, TT, T, UG).

Our dry-mill execution engine links every hopper tip on the hulling line—spanning destoners, hullers/polishers, cylindrical/oscillating screen graders (Screen 18 for AA, Screen 15/16 for AB, pea-berry separators for PB), gravity tables (Catadors separating light TT and T), and bichromatic/trichromatic optical color sorters—to generate an auditable Milling Outturn Statement and calculate Direct Settlement System (DSS) net payouts:

from decimal import Decimal, ROUND_HALF_UP

class CoffeeMillingOutturnAndDSSSettlementEngine:
    """
    Computes dry-mill parchment-to-clean-coffee hulling loss, screen grade
    distribution (AA, AB, PB, C, TT, T), and NCE Direct Settlement System (DSS)
    statutory & grower payout allocations.
    """
    @staticmethod
    def calculate_milling_outturn_and_dss_distribution(
        parchment_intake_kg: Decimal,
        intake_moisture_pct: Decimal,
        clean_grade_weights_kg: dict,  # e.g. {"AA": 4200, "AB": 5100, "PB": 850, "C": 620, "TT": 380, "T": 250}
        grade_auction_prices_usd_per_50kg: dict,
        exchange_rate_kes_per_usd: Decimal,
        milling_fee_per_tonne_parchment_usd: Decimal,
        cherry_intake_for_outturn_kg: Decimal,
        cooperative_retention_pct: Decimal,  # Capped by regulation (e.g. <= 20%)
    ) -> dict:
        total_clean_coffee_kg = sum((Decimal(str(v)) for v in clean_grade_weights_kg.values()), Decimal("0.00"))
        milling_loss_kg = parchment_intake_kg - total_clean_coffee_kg
        milling_loss_pct = ((milling_loss_kg / parchment_intake_kg) * Decimal("100.00")).quantize(
            Decimal("0.01"), rounding=ROUND_HALF_UP
        )
        cherry_to_clean_ratio = (cherry_intake_for_outturn_kg / total_clean_coffee_kg).quantize(
            Decimal("0.01"), rounding=ROUND_HALF_UP
        )
        # Compute gross auction USD value across grades (priced per 50kg bag)
        gross_auction_usd = Decimal("0.00")
        for grade, weight_kg in clean_grade_weights_kg.items():
            bags_50kg = Decimal(str(weight_kg)) / Decimal("50.00")
            price_per_50kg = Decimal(str(grade_auction_prices_usd_per_50kg.get(grade, "0.00")))
            gross_auction_usd += bags_50kg * price_per_50kg

        gross_auction_usd = gross_auction_usd.quantize(Decimal("0.01"), rounding=ROUND_HALF_UP)
        milling_charge_usd = ((parchment_intake_kg / Decimal("1000.00")) * milling_fee_per_tonne_parchment_usd).quantize(
            Decimal("0.01"), rounding=ROUND_HALF_UP
        )
        net_distributable_kes = ((gross_auction_usd - milling_charge_usd) * exchange_rate_kes_per_usd).quantize(
            Decimal("0.01"), rounding=ROUND_HALF_UP
        )
        coop_factory_retention_kes = (net_distributable_kes * (cooperative_retention_pct / Decimal("100.00"))).quantize(
            Decimal("0.01"), rounding=ROUND_HALF_UP
        )
        direct_grower_pool_kes = (net_distributable_kes - coop_factory_retention_kes).quantize(
            Decimal("0.01"), rounding=ROUND_HALF_UP
        )
        rate_per_kg_cherry_kes = (direct_grower_pool_kes / cherry_intake_for_outturn_kg).quantize(
            Decimal("0.01"), rounding=ROUND_HALF_UP
        )
        return {
            "total_clean_coffee_kg": total_clean_coffee_kg,
            "milling_loss_pct": milling_loss_pct,
            "cherry_to_clean_coffee_ratio": cherry_to_clean_ratio,
            "gross_nce_auction_proceeds_usd": gross_auction_usd,
            "direct_grower_pool_kes": direct_grower_pool_kes,
            "net_payout_rate_per_kg_cherry_kes": rate_per_kg_cherry_kes,
        }

3. SCA Q-Grader Cupping Lab, Green-Bean Defect Analysis & Bulking/Blending Recipes

Every milled outturn lot is sampled automatically into the dry-mill or exporter cupping laboratory:

• Green Bean Physical Analysis: Logs screen retention percentages, bean density (g/L), water activity (aw < 0.60), moisture content, and Category 1 / Category 2 defect counts (full black, full sour, pod/cherry, insect-damaged, floater, shell, or parchment).
• SCA 100-Point Sensory Cupping Protocol: Q-Graders score roasted samples on tablets across Fragrance/Aroma, Flavor, Aftertaste, Acidity (phosphoric/malic/citric brightness), Body, Balance, Uniformity, Clean Cup, Sweetness, and Overall—generating interactive spider charts and setting auction reserve prices.
• Export Bulking, GrainPro Liner Stuffing & EUDR Geolocation Dossiers: For exporters blending multiple auction lots into 300-bag (19.2 MT) shipping containers in Sisal/Jute bags with hermetic GrainPro / Ecotact liners, the ERP compiles the exact farm GPS polygons of every contributing smallholder into a downloadable GeoJSON dossier compliant with EU Deforestation Regulation (EUDR) due diligence.