Data Analytics for Reliability and Risk

This course provides a detailed account of data analytics methods for reliability and risk data from three different yet complementary perspectives: frequentist approaches, Bayesian inference, and machine learning methods. It covers both the fundamental and mathematical underpinnings of the data analytics methods, but also focuses on a hands-on approach with end-to-end projects involving field data from various industry fields, encompassing commonly encountered scenarios in reliability and risk such as: (i) when no failure data is available; (ii) failure data with knowledge of explanatory variables such as current, voltage, differential pressure and temperature; (iii) expert opinion data; (iv) massive and multi-sensor data from both process variables and condition monitoring signals.