The Probabilistic Mechanics Lab performs research themed around variability and its impact on mechanical systems. The laboratory works on both the innovation of new approaches in probabilistic analysis and the application of these techniques to a variety of fields including biomechanics, materials and nanotechnology, and design. Probabilistic and stochastic analysis represent an important emerging field and by its nature highly collaborative.
As a complement to experiments, finite element modeling and statistical analysis, probabilistic analysis provides a method of characterizing the potential impact of variability in parameters on performance. Specifically, input parameters are represented as distributions in order to predict a distribution of performance. In addition to understanding the probabilities associated with a specific level of performance, the input parameters or combination of inputs that most affect performance are also identified. The most common probabilistic approach is Monte Carlo simulation, which involves repeated sampling of the input parameters according to their distributions. While robust, Monte Carlo simulation is computationally expensive; much of the lab's research has focused on the application of efficient probabilistic methods.
Research projects have investigated:
- Effects of microstructural variability on fatigue of aluminum alloys
- Monte Carlo prediction of polyethylene properties using molecular dynamics
- Understanding how patient, implant and surgical variability influences joint mechanics
- Development of statistical shape models to characterize anatomic variability