Includes bibliographical references (pages 315-326) and index.
Contents:
Introduction to data visualization and characterization -- Random vectors and the multivariate normal distribution -- Explaining covariance structure : principal components -- Linear model for numerical and categorical response variables -- Linear mixed effects model -- Diagnosis of variation source using PCA -- Diagnosis of variation sources through random effects estimation -- Analysis of system diagnosability -- Prognosis through mixed effects models for longitudinal data -- Prognosis using Gaussian process model -- Prognosis through mixed effects models for time-to-event data.
Summary:
"Today, we are facing a data rich world that is changing faster than ever before. The ubiquitous availability of data provides great opportunities for industrial enterprises to improve their process quality and productivity. Industrial data analytics is the process of collecting, exploring, and analyzing data generated from industrial operations and throughout the product life cycle in order to gain insights and improve decision-making. This book describes industrial data analytics approaches with an emphasis on diagnosis and prognosis of industrial processes and systems. A large number of textbooks/research monographs exist on diagnosis and prognosis in the engineering eld. Most of these engineering books focus on model-based diagnosis and prognosis problems in dynamic systems. The modelbased approaches adopt a dynamic model for the system, often in the form of a state space model, as the basis for diagnosis and prognosis. Dierent from these existing books, this book focuses on the concept of random effects and its applications in system diagnosis and prognosis. The impetus for this book arose from the current digital revolution. In this digital age, the essential feature of a modern engineering system is that a large amount of data from multiple similar units/machines during their operations are collected in real time. This feature poses signicant intellectual opportunities and challenges. As for opportunities, since we have observations from potentially a very large number of similar units, we can compare their operations, share the information, and extract common knowledge to enable accurate and tailored prediction and control at the individual level. As for challenges, because the data are collected in the field and not in a controlled environment, the data contain signicant variation and heterogeneity due to the large variations in working/usage conditions for dierent units. This requires that the analytics approaches should be not only general (so that the common information can be learned and shared), but also flexible (so that the behaviour of an individual unit can be captured and controlled). The random effects modeling approaches can exactly address these opportunities and challenges"-- Provided by publisher.
This resource is supported by the Institute of Museum and Library Services under the provisions of the Library Services and Technology Act as administered by State Library of Iowa.