What is a Monte Carlo Analysis with Fred Schenkelberg
In today’s episode, the guest Fred Schenkelberg explains the Monte Carlo simulation in a fair amount of detail. Before you get in the depth of how the tool works, you need to understand what basically this method is. The Monte Carlo simulation has been around since World War 2 and it is a mathematical technique that works based on probability functions, random variables, and the distribution of statistical data. The main concept of it to give the decision maker the most obvious choices while facing any risks to get the best out of every possible outcome. The tool serves the purpose for getting a better insight of the consequences relative to each choice the person making decisions has to make. When you’re looking for the reliability of your assets and checking the integrity of your different maintenance programs, it is a really powerful tool.
The working of this simulation pretty easy to understand if you have the building models that need to be analyzed. But the results generated by the tool are based on substituted values of the probability function. That’s why when there is a huge range of values to be picked from the distribution plots, the calculations are done over and over which leads to complex variability and the standard distribution. The means, modes, and medians are calculated from the statistical data, the uncertainty factor is determined after the simulation is complete so many recalculations and so on.
The idea behind using the probability distributions is to get the closest outcome to reality so that the risk analysis variables describe the uncertainty in the best ways. As the Monte Carlo simulation picks up random samples from the input probability distributions, the output is likely to be thousands and tens of thousands values in the form of probability distribution—these values are more realistic though considering if you don’t have enough data and just rely merely on guesses of your engineers for a critical decision. That’s why so many iterations are done during the simulation to not only give you an idea what might happen but to tell you how likely it could happen.
After these iterative simulations, the outcomes from the samples are recorded. The best things about the tool is that you can decide on the basis of a lot of distribution; Normal distribution when you have only the expected value or mean value along with the standard deviation that might occur about this mean you defined, Lognormal when you have only positive values with potential to go up the line, Triangular when you define min, max, and most likely value. Similarly, you can use Discrete and Pert distributions depending on what kind of likelihood and specifics you have about the expected outcome.
To get started with Monte Carlo, you just have to have a model in your mind, start collecting variability data about that model, perform different types of analysis—graphical, probability based, sensitivity, scenario based, relational—and the results will help you make much better decisions than before.
- Eruditio, LLC
- A Smarter Way of Preventative Maintenance – Free eBook
- Maintenance Planning & Scheduling: Planning for Profitability Video Course
Fred Schenkelberg Links:
Rooted In Reliability podcast is a proud member of Reliability.fm network. We encourage you to please rate and review this podcast on iTunes and Stitcher. It ensures the podcast stays relevant and is easy to find by like-minded professionals. It is only with your ratings and reviews that the Rooted In Reliability podcast can continue to grow. Thank you for providing the small but critical support for the Rooted In Reliability podcast!