Overview: Proposed ExxonMobil Short Course on Structural Reliability and the Stanford RMS Program
PROBABILITY-BASED ENGINEERING ANALYSIS AND DESIGN
This course will survey the topics of probability and stochastic processes, and their application to various problems of engineering analysis and design. Emphasis will be placed on methods that are in common use in industry applications of reliability theory. Topics surveyed will include
Probability theory: Discussion of basic models; historical context of engineering applications
Random process theory: Emphasis on models used to describe random vibration of mechanical systems
Statistical inference: Calibration of probability models to the data at hand
Structural reliability analysis: Discussion of methods to propagate the uncertainty in these models, such as Monte-Carlo simulation, FORM and SORM
Probability-based design: Illustration of methods such as LRFD (load- and resistance-factor design)
Case studies will survey a range of applications. Many results will be drawn from the Stanford Reliability of Marine Structures (RMS) Program, an advanced research program funded by a number of oil companies (including Exxon and Mobil) for more than 10 years. With its focus on overall reliability, the RMS Program encompassed the modelling of metocean conditions, random waves, nonlinear hydrodynamics, and resulting responses of marine structures, risers, moorings, etc. Various numerical algorithms were created during these reliability studies; we will survey here some of the resulting RMS software available to ExxonMobil.
Particular attention will be paid to complex structural loading mechanisms -- wind, wave, seismic, etc. -- whose uncertainties have especially motivated the use of probabilistic methods. Limit states arising from both extreme loading and cumulative damage will be considered. The basis behind some current probability-based structural codes will be discussed.
In most of these topics, the talks will seek to provide a general overview, appropriate for those will little or no background in probability or statistics. Additional material and references will be made available to those seeking greater technical detail. A set of homework exercises and solutions will also be provided.
Proposed Time Frame: 12 Lecture Hours, December 13-15, 2011
Instructor: Steve Winterstein
Steve Winterstein holds degrees from the Massachusetts Institute of Technology in Mathematics (B.Sc.) and Civil Engineering (M.Sc. and Ph.D.). He has taught and conducted research in structural reliability for over 25 years; areas of interest include nuclear power plants, offshore oil platforms, and wind turbine for energy production. He co-directed Stanford's Reliability of Marine Structures program (1988-2002), which combined post-Masters graduate study with advanced research in structural reliability applications.