Grower weekly review · Week 32
Turn crop signals into
a clear Monday plan.
A human-in-the-loop assistant that finds meaningful variance, prepares the questions to ask in the greenhouse, and turns the week into accountable actions.
Verify on-site before changing the crop plan.
01 · Observe
Weekly crop signals
01B
Plant growth measurements
Use the average from a consistent set of representative plants and the same leaf-selection method each week.
ESTIMATED LEAF AREA INDEX
LAI turns leaf measurements into canopy area.
Estimated leaf area = shape factor × average length × average width. LAI = estimated leaf area × active leaves per plant × plants per m² ÷ 10,000.
Pepper model supporting the 0.60 default ↗LAI is an estimate—not a direct canopy scan. Keep sample plants, leaf position, and measurement technique consistent; recalibrate the shape factor when cultivar-specific data is available.
Transparent rules first. AI synthesis second. Grower judgment always.
02 · Interpret
What deserves attention
Harvest is materially behind plan
Validate fruit count and average fruit weight by compartment before revising the next two-week forecast.
Root-zone stress risk is elevated
Check first and last drain timing, emitter uniformity, slab EC by zone, and whether irrigation frequency matches current radiation.
Plant registration is shifting from target
Compare the change over several weeks and across representative plants before linking it to vegetative–generative balance.
Light availability missed the crop plan
Reconcile temperature and irrigation targets with realized light; avoid steering the crop as if target light was achieved.
Labour demand is above standard
Separate crop-work hours from harvest and packing, then identify the compartment or task driving the variance.
“This week needs a root-zone and forecast check before targets are rolled forward. 10.0% below the weekly target; Drain is 21% and drain EC is 1.5 above feed; Growth +17% · stem -7%. Start with validate fruit count and average fruit weight by compartment before revising the next two-week forecast.”
Take the structured context into your preferred AI tool.
The prompt preserves the raw data, makes uncertainty explicit, and asks for verification before recommendations.
03 · Deliver
Built for the gap between
software and the greenhouse.
This prototype demonstrates the workflow behind forward-deployed grower support: structure messy operational data, surface the signal, verify it on-site, and close the loop with an owner and timing.