MODEL • SAMPLE • COMPARE
Monte Carlo Studio.
One model. A thousand possible outcomes.
Explore what uncertainty means for your decision.
What changes when demand falls?
Does less variability reduce risk?
DEFINE THE DECISION
Build your model
Inputs
Use a constant or a probability distribution. Inputs are sampled independently, in the order below.
Calculated outputs
Use variable names, + − * / ^ and parentheses. Outputs can reference earlier outputs.
Formula guide & model assumptions
Functions: min, max, abs, sqrt, round, floor, ceil, exp, log (natural logarithm), pow. Example: max(demand - inventory, 0). ^ means exponentiation. Names are case-sensitive.
Every uncertain input consumes one uniform draw per trial. Normal inputs are continuous and unbounded. Negative draws are retained; rounding or clipping requires an explicit output formula. Discrete probabilities must sum to 1. There are no correlated inputs or time steps in this version.
Models and comparisons autosave in this browser when storage is available. Download JSON for a portable backup. No model data is uploaded.
REPEAT THE EXPERIMENT
Run & explore
Trials: 1–100,000. Seed: 1–4,294,967,295 (whole numbers). Same seed + same input order = same random draws. All comparisons use the active trial count, seed, output and thresholds.
See the range, not just the average.
Run your model to reveal the outcome distribution, risk and percentiles.
MAKE THE RANDOMNESS VISIBLE
Follow one trial
Trace each U through its distribution and into the output formula.
The Sanotronics preset contains the workbook's teaching trace. Custom U values do not change the simulation's seeded draws.
CHANGE ONE ASSUMPTION
Compare scenarios
Save the current model as a comparison, then change an input and run again. Each saved scenario keeps a complete copy of its assumptions.
The class set adds baseline, lower mean (12,000) and lower SD (2,250), without changing your active model.
Classroom demonstration files
Lecture 33 · 1,000 trials · Seed 20261106
Workbook adaptation of Sanotronics, An Introduction to Management Science, 15e, §§12.1–12.2. Numerical approximation agrees with the workbook demonstration to within one cent. See Inverse Transform Lab for the sampling foundations.