In the last few years, several approaches and methods have been proposed to improve early warning systems for managing risks due to rapid slope failures where important infrastructures are the main exposed elements. To this aim, a multi-sensor monitoring system has been installed in an abandoned quarry at Acuto (central Italy) to realise a natural-scale test site for detecting rock-falls from a cliff slope. The installed multi-sensor monitoring system consists of: i) two weather stations; ii) optical cam (Smart Camera) connected to an Artificial Intelligence (AI) system; iii) stress- strain geotechnical system; iv) seismic monitoring device and nano-seismic array for detecting microseismic events on the cliff slope. The main objective of the experiment at this test site is to investigate precursors of rock mass failures by coupling remote and local sensors. The integrated monitoring system is devoted to record strain rates of rock mass joints, capturing their variations as an effect of forcing actions, which are the temperature, the rainfalls and the wind velocity and direction. The preliminary tests demonstrate that the data analysis methods allowed the identification of external destabilizing actions responsible for strain effects on rock joints. More in particular, it was observed that the temperature variations play a significant role for detectable strains of rock mass joints. The preliminary results obtained so far encourage further experiments. © 2017 The Authors.