Risk assessment for linear assets with estimation and model-based uncertainty

点击翻译按钮获取中文摘要

An asset is anything that has potential value to an organization; asset management is the co-ordinated activity of an organization to realize value from its assets. Riva Modeling is a Toronto-based software company that offers holistic physical asset management solutions that integrate data from a variety of sources and provides data-driven asset management decision support tools. The Centre for Maintenance Optimization and Reliability Engineering (C-MORE) at the University of Toronto is the leading industry-academic research group in asset management. Riva seeks to operationalize ongoing research results being made at C-MORE, specifically by making more accurate remaining useful life estimates by accounting for both estimation uncertainty and possible model-misspecification. Our approach is to embed the problem in a Bayesian framework that adds model-misspecification to the equations. C-MORE's methodology is innovative in the way our formulation of the problem can account for model error that is easier and faster to calculate than currently known techniques. Riva Modeling will play a key role in the research by dedicating staff time to the implementation of the theoretical results we will obtain and by connecting us with real industry datasets through its customer base. Riva will also contribute to the computing needs of C-MORE during the development phase of the project.