The paper gives an overview of the state of the art methods and technologies in the field of stream data mining with applications in the Internet of Things systems for supporting fruit cold chain logistics. As the number of sensors used in on-line monitoring of the process is large, the amount of time series data is increasing rapidly. It is challenging to process such data in order to discover patterns, trends and outliers as a consequence of fluctuations of certain process parameters. In particular, the paper discusses methods for mining stream data collected in fruit cold chain aiming at real time control of fruit quality. A model of the centralized IoT system and the part responsible for monitoring fluctuations of temperature, humidity, and concentration of gases is proposed.