Vibration Analysis for IoT Enabled Predictive Maintenance

Vibration sensor is becoming an essential part of Internet of Things (IoT), fueled by the quickly evolving technology improving the measurement accuracy and lowering the hardware cost. Vibration sensors physically attach to core equipments in control and manufacturing systems, e.g., motors and tubes, providing key insight into the running status of these devices. Massive readings from vibration sensors, however, pose new technical challenges to the analytical system, due to the non-continuous sampling strategy for sensor energy saving, as well as hardness of data interpretation. To maximize the utility and minimize the operational overhead of vibration sensors, we propose a new analytical framework, especially designed for vibration analysis based on its unique characteristics. In particular, our data engine targets to support Remaining Usefulness Lifetime (RUL) estimation, known as one of the most important problems in cyber-physical system maintenance, to optimize the replacement scheduling over the equipments under monitoring. Our empirical evaluations on real manufacturing sites show that scalable and accurate analysis over the vibration data enables to prolong the average lifetime of the tubes by 1.2x and reduce the replacement cost by 20%.