Official websites use .gov
A .gov website belongs to an official government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.
Cost examinations often run into the challenge of data uncertainty, where variables in the calculation (e.g., price of gasoline) will fluctuate over the study period. For instance, consider a manufacturer that will invest in 1 of 3 heating, venting, and air conditioning (HVAC) units. To decide which one, they might conduct a cost study to identify which one has the lowest cost over time. One factor in this study will be the price of energy, which fluctuates over time.
NIST’s new “Monte Carlo Tool” allows users to simulate fluctuating variables, such as energy prices, to assess the potential impact of uncertainty. The tool is developed to follow the simulation segment of ASTM E1369 and implements Monte Carlo analysis, a probabilistic sensitivity analysis used to account for uncertainty. The technique involves a method of model sampling. Specification involves 1) defining which variables are to be simulated, 2) defining the distribution of each of these variables, and 3) selecting the number of iterations performed. The software randomly samples from the probabilities for each input variable of interest.
To illustrate the use of the tool, consider a situation where a firm has to purchase 100 ball bearings at $10 each; however, the price can vary plus or minus $2. In order to address this situation, one can use a Monte Carlo analysis where the price is varied using a triangular distribution with $12 being the maximum, $8 being the minimum, and $10 being the most likely. Moreover, the anticipated results should have a low value of approximately $800 (i.e., 100 ball bearings at $8 each) and a high value of approximately $1200 (i.e., 100 ball bearings at $12 each). The triangular distribution would make it so the $8 price and $12 price have lower likelihoods. In a Monte Carlo analysis, each iteration is similar to rolling a pair of dice, albeit, with the probabilities having been altered. In this case, the dice determine the price of the bearings. The number of iterations is the number of times this simulation is calculated (i.e., the number of times the dice is rolled).
More information is available at:
https://www.nist.gov/services-resources/software/monte-carlo-tool