The mix of cheap sensors, fast networks, and distributed computing—the recipe for the Internet of Things—is gaining increasing attention in the manufacturing industry, where maintenance must be conducted for both individual assets of interest and complex manufacturing processes. In a talk aimed at data scientists, students, researchers, and nontechnical professionals, Danielle Dean introduces the landscape and challenges of predictive maintenance applications in the manufacturing industry.Predictive maintenance, a technique to predict when an in-service machine will fail so that maintenance can be planned in advance, encompasses failure prediction, failure diagnosis, failure type classification, and recommendation of maintenance actions after failure. Danielle reviews predictive maintenance problems from the perspectives of both the traditional, reliability-centered maintenance field and IoT applications, discussing problem coverage, applicable predictive models based on data available, and what data must be collected to perform predictive maintenance tasks. You'll learn how to bridge the data-driven approach and the problem-driven approach by articulating what types of data are needed for different predictive maintenance applications.Topics include:What data must be gathered for effective predictive maintenance applicationsHow to formulate a predictive maintenance problem into three different machine-learning models (regression, binary classification, and multiclass classification)The step-by-step procedure for data input, data preprocessing, data labeling, and feature engineering from the raw data to prepare the training/testing dataHow various types of learning models can be trained and compared using different algorithms