Application of fuzzy logic for honey bee colony state detection based on temperature data

Since honey bees are one of the most important actors in the whole world, it is important to follow the life of these insects in order to preserve them from danger, via a range of risk factors such as Colony Collapse Disorder, pesticides, pests etc. Therefore it is important to identify any abnormalities inside the honey bee colony at an early stage, which may be possible using modern technologies e.g. monitoring systems, data processing, and analysis. This research proposes a solution for honey bee colony state identification using temperature data and fuzzy logic. The detection process proposes a Fuzzy Inference System that receives five input parameters and provides an output (defined as “assessment of the colony”) pointing to (for now) three defined possible states – normal, death, and extreme. The rule base for the inference system was defined taking into account the knowledge of field experts, literature research, previous observations and was based only on temperature data and temperature changes inside the hive during different seasons. The proposed system proved to be quite robust, showing an accuracy value of ~98%, 100% precision and specificity, ~97% recall and ~98% F1 score when tested with validation set.