Monte Carlo simulations using Matlab

From randomness to statistical results!

Monte Carlo allows us to get statistical trends on systems that normally would be far too complex to study via deterministic approaches. Nowadays engineering and scientific fields are counting more and more on this powerful method to get statistical trends on the behaviour of increasingly growing complexity machinery, systems, and scientific models.

What you’ll learn

  • Monte Carlo.
  • Reliability.
  • Availability.
  • Mantainability.
  • Monte Carlo simulations.
  • Probability.
  • Risk management.
  • Monte Carlo integration.
  • Decision making problems.
  • Decision making tools.
  • Monty Hall thought experiment.
  • Risk analisys through Monte Carlo.

Course Content

  • Introduction –> 1 lecture • 5min.
  • Basic concepts –> 7 lectures • 51min.
  • The Monty Hall problem –> 6 lectures • 34min.
  • Uniform continuous random distribuitons in 2D. (Determining the value of pi) –> 7 lectures • 40min.
  • Monte Carlo integration –> 5 lectures • 32min.
  • Monte carlo applied to reliability studies –> 7 lectures • 59min.
  • Congratulations –> 1 lecture • 1min.

Monte Carlo simulations using Matlab

Requirements

  • Basic programming knowledge on any language (matlab experience is a plus).
  • Basic probabilistic and statistics knowledge is desired.
  • A matlab license.
  • An internet connection.

Monte Carlo allows us to get statistical trends on systems that normally would be far too complex to study via deterministic approaches. Nowadays engineering and scientific fields are counting more and more on this powerful method to get statistical trends on the behaviour of increasingly growing complexity machinery, systems, and scientific models.

Imagine having a system with +1000 components that can each fail at any given time. Imagine furthermore that we need to predict how the complete system is going to fail depending on those +1000 components. The statistical equations product of this would be unbearable to calculate via traditional approaches, thus Monte Carlo has been and is still used widely in the reliability engineering field.

Imagine now having to calculate the area of a completely irregular shape (or the definite integral of a curve that has no mathematical function to describe it so that we could use classic integration methods). Monte Carlo could be user to solve this issue relying on the ever increasing power of our computers.

The previous examples were just two of the applications in which we could use a methodology of work that is revolutionizing the way we study systems. Therefore, Monte Carlo is a must in every engineer or scientist toolbelt.