Prognostics is an important part of advanced manufacturing and maintenance. Recently, many sensors have been
adopted for monitoring units before the end of their cycle time. Also, this sensor data is usually noisy and correlated
which is challenging for traditional techniques to capture the information and estimate the remaining useful life (RUL).
In practical cases, there are many operational conditions that the unit goes through which also complicate the task of
estimating RUL. To encounter these challenges, an ensemble learning model could be used to extract the correlated
information and predict the RUL. Ensemble learning models produce many weak learners each learns part from the
data, and in the end, these weak learners vote to get the best prediction possible.

Keywords
Machine learning, ensemble learning, prognostics, Neural Networks and Remaining life prediction