Modern apiarists rarely diagnose a hive by eye alone — sensors log temperature, humidity, weight and entrance traffic, and the readings get analysed with tools borrowed from other fields entirely. This lab visualises that pipeline: a sample swarm of agent bees on the left reacts to the hive conditions you set, while the panel on the right turns the same numbers into the kind of charts a hive-monitoring dashboard would show.
Real hive-monitoring research pulls directly from these adjacent disciplines: SIR-style compartment models from epidemiology for disease spread, control charts and z-scores from finance and quality engineering for anomaly detection, and rolling averages from climate science for smoothing noisy sensor data into a trend.
A sample colony of agent bees cycles through a Susceptible–Infected–Recovered infection model while a floating data-lab panel turns population, temperature, mite load and forager traffic into a live colony health index with rolling statistical control bands.
Bee colour (gold / red / teal) shows each agent's SIR compartment; the health index blends infection load, thermal stress and forager throughput; the line chart's μ ± 2σ band and z-score flag statistically unusual readings, exactly as hive-monitoring dashboards do.
Raise mite load or push temperature away from 35°C to watch infection spread and health drop. Pick which metric the chart tracks, then hit "Inject anomaly spike" to see the control band catch an out-of-range reading.
Real precision-apiculture research borrows exactly these tools from other fields: SIR models from epidemiology, control charts and z-scores from finance and quality engineering, and rolling averages from climate science.