Uncertainty quantification and management using cossan software

In collaboration with the Institute for Risk and Uncertainty (UK) and Institut für Risiko und Zuverlässigkeit (Germany), we are offering a 3-day training course on Uncertainty Quantification using COSSAN Software.

Each day focuses on a specific topic. This allows the participants to attend a specific training day.

Aims and learning outcomes

You will learn the main techniques available for dealing with Risk Analysis and Uncertainty Quantification through an easy to use, yet powerful computational software

Main concepts and techniques
• Random Variables and Random Variables Sets
• Monte Carlo simulation and advanced simulation techniques (Subset simulation, Line Sampling, Importance Sampling, Latin Hypercube Sampling)
• Global and Local Sensitivity analysis
• Surrogate Models (Artificial Neural Networks, Responce surface, Kriging)
• Reliability based and robust design


  • Date: 8th - 10th April, 2019
  • Time: Monday - 10:00 - 16:00
  • Time: Tuesday - 09:00 - 16:00
  • Time: Wednesday - 09:00 - 13:00
  • Location: University of Liverpool

Day 1: 8th April 2019

10:00 Introduction Welcome, aim and structure of the course. Getting started: basic useful computer commands and remote connection
10:30 Lecture COSSAN-X and OpenCossan (Main features and capabilities, toolboxes and wizards)
11:15 Practical Session Installation, getting started and familiarisation with the software. Construct a model and run a deterministic analysis.
12:00 Lecture Introduction to Uncertainty Quantification and modelling of the uncertainties (random variables, stochastic processes, etc.)
12:30 Lunch Break  
13:30 Lecture Basic Monte Carlo method simulation for Uncertainty Quantification and Reliability Analysis
13:45 Practical session Basic Tutorial: Uncertainty Quantification of a simple cantilever beam model
14:30 Lecture Advanced Monte Carlo methods (Importance sampling, Line Sampling, Subset Simulation) for UQ and Reliability Analysis
15:00 Practical session Basic Tutorial: Uncertainty Quantification and Reliability Analysis using advanced methods
15:45 Wrap up A summary of the day

Day 2: 9th April 2019

09:15 Introduction Welcome and presentation of the day
09:30 Lecture When desktop is not enough (Cliff Addison)
10:15 Lecture Basic Linux concepts and essential commands
10:30 Tutorial Excercises on High Performance Computing
12:30 Lunch Break  
13:30 Lecture Connect COSSAN-X with 3rd party solvers (ABAQUS, NASTRAN, LS_DYNA, etc.) and high-performance capabilities
14:00 Practical session Connect COSSAN-X with your model
14:45 Lecture Introduction to Local and Global Sensitivity Analysis
15:30 Practical session Tutorial: Sensitivity Analysis of simple model and of external model
15:50 Wrap up A summary of the day

Day 3: 9th April 2019

09:00 Introduction Presentation of the day
09:15 Lecture Optimisation (Gradient based and gradient free approaches, stochastic optimisation)
09:45 Tutorial Basic Tutorial: Optimization of a simple Cantilever Beam
10:30 Lecture Meta-models (Response surface, Artificial Neural Networks, etc), and Robust and Reliability based optimisation
11:00 Tutorial Tutorial: Meta-models, Robust Optimization and Reliability Based Optimization
11:45 Demo Demonstration of latest COSSAN development
12:15 Discussion User-Developer Interaction
13:00 Lunch Close