A method for mining combined data from in-hive sensors, weather and apiary inspections to forecast the health status of honey bee colonies

•An accurate prediction method forecast the honeybee colony health status.•Honeybees health status from homeostasis, weight, inspections, and weather data.•Prediction model based on the Random Forest algorithm with hit rate over 90.•A method to help the beekeeper avoiding winter colonies losses.•A complementary aspect of sensors and inspection data for warnings honeybees’ problems.