📱Web-app

Gratheon web app is intended to help beekeepers manage apiary data, communicate with installed modular hive devices, analyze images and timeseries data to extract valuable statistics, provide this data analytics over time, generate and deliver alerts, forecast, provide AI suggestions to solve issues.

status: TRL 6

Core domain entities

Entity	Description	Key Properties
Apiary	Set of beehives located together at specific location. Size limited by surrounding land that bees can pollinate	Location (lat/lng), name, active status
Hive	Physical beehive structure with vertical sections. Can be split, merged, or collapsed	Name, color, status, boxes, family, split/merge history, collapse tracking
Family (Colony/Queen)	Bee superorganism led by queen that lays eggs	Race (apis mellifera variation), added year, age (auto-calculated), treatments
Box (Hive Section)	Hollow wooden section that houses frames. Types: Deep (brood), Super (honey), Gate (entrance), Ventilation, Queen Excluder, Horizontal Feeder, Bottom Board	Type, position, color, frames
Frame	Wooden frame with wax foundation inside sections. Types: Foundation, Empty Comb, Void, Partition, Feeder	Type, position, left/right sides
Frame Side	One side of a frame where photos can be uploaded for AI analysis	File references for images, detected resources
Inspection	Snapshot of entire beehive state during beekeeper intervention. Stores JSON data of hive composition at specific time	Hive ID, data (JSON), timestamp
Treatment	Anti-varroa chemical interventions tracked per family/box/hive for medical history	Type, timestamp, target (hive/box/family)
File	Uploaded images (frame photos, bottom board varroa photos). Processed through AI detection pipeline	Hash, dimensions, user ID, file type, detection jobs
Detected Resources	AI-detected cell types on frame photos: Capped brood, Eggs, Honey, Larvae, Nectar, Pollen, Other	Class, coordinates (x,y), radius, probability
Detected Bees/Queens	AI-detected bee and queen positions on frames	Bounding boxes, confidence scores
Detected Varroa	AI-detected varroa mites on bottom board photos	Count, positions (coming soon)
Metrics (Telemetry)	Time-series sensor data from IoT devices	Temperature (°C), humidity (%), weight (kg), timestamp
Entrance Movement	Bee traffic analysis from entrance video cameras	Bees in/out, net flow, speed stats, stationary bees, interactions
Alert	Generated warnings based on metric thresholds and rules	Text, metric type/value, hive ID, delivery status, timestamp
Alert Rule	User-defined conditions that trigger alerts	Metric type, condition, threshold, duration, enabled status, scope (hive/apiary)
Alert Channel	Delivery methods for alerts	Type (email/phone/telegram), contact info, time window, enabled status

Main app use cases

Use-cases is a group of features that when combined serve a great value for the customer

Uploading photos to get colony overview stats

• Create a beehive
• Open specific section, add frames
• Open a specific frame, click "upload frame photo" with bees and comb cells on it
• Wait for backend to process
• Receive AI-assisted statistics about bee counts and cell distribution
• Compare different hives to see which ones are more powerful with real data behind it

Track colony development over time

• Follow steps for adding frame photos of the beehive after an inspection
• Click on "Create inspection button" to create a snapshot of the hive in time
• Check that new Inspection is added under Inspections tab
  • Check how over time each inspection has own distribution of resources

Sending weight telemetry from IoT sensors

• Generate API token in Account settings
• Power on IoT sensors device, follow Docs to connect it to WiFi and send data
• Open target beehive, open Metrics tab and see graphs of weight/temperature
• Open Analytics tab to see data more time-bound and correlate it with other data like weather

Streaming entrance video

• Setup entrance video camera to stream video
• Check that video stream is visible

Monitor varroa mites via bottom board

• Add a bottom board section to your hive
• Upload a photo of the white slideable panel showing varroa mites
• Images are versioned with inspections for historical tracking
• AI-powered varroa counting (coming soon) will provide mite counts and treatment recommendations