Math & Numerical Methods

Monte Carlo & Uncertainty Simulators

A focused Math & Numerical Methods hub for monte carlo & uncertainty tools, keeping related formulas, assumptions, and engineering checks together.

20 simulators

Adjacent categories

Simulator list

Bayesian Calibration via MCMC (Metropolis–Hastings) Simulator
Monte Carlo & Uncertainty
Calibrate a CAE model parameter the Bayesian way: feed in a prior, a likelihood and observed data.
Bootstrap Confidence Interval Simulator
Monte Carlo & Uncertainty
Compute 95% bootstrap confidence intervals for the mean and median in real time. Resample with replacement and vary N and B to watch the distribution converge.
Brownian Motion & Random Walk Simulator
Monte Carlo & Uncertainty
Simulate Brownian motion and random walks in real time. Adjust particles, step size, and trails to verify Einstein's diffusion formula and explore parameters.
Buffon's Needle Simulator — Estimating Pi by Probability
Monte Carlo & Uncertainty
The classical experiment: drop needles at random on a floor of equally spaced parallel lines and estimate pi from the crossing probability. Change the count, length, spa…
Control Variates Simulator
Monte Carlo & Uncertainty
Experience the control variates method, which boosts Monte Carlo accuracy almost for free. Using an auxiliary variable strongly correlated with your target, watch how mu…
FFT Spectrum Analyzer — Time-Domain Signal to Frequency Spectrum
Monte Carlo & Uncertainty
Analyze time-series signals with the Fast Fourier Transform. Tweak window functions, sampling frequency and signal components to feel aliasing and leakage.
Gibbs Sampling Simulator
Monte Carlo & Uncertainty
Draw samples from a bivariate normal distribution with Gibbs sampling, the classic MCMC (Markov chain Monte Carlo) method.
Importance Sampling — Efficient Tail Probability Estimation
Monte Carlo & Uncertainty
Estimate the tail probability P(X>t) of the standard normal in parallel with crude Monte Carlo and importance sampling. Tune the proposal mean and standard deviation and…
Latin Hypercube Sampling — LHS vs Plain Monte Carlo
Monte Carlo & Uncertainty
Side-by-side 2D Latin Hypercube and plain Monte Carlo sampling. See how placing exactly one point per row and column changes the integration accuracy at the same sample …
MCMC Metropolis-Hastings Sampler — Acceptance & Autocorrelation
Monte Carlo & Uncertainty
Draw samples from an arbitrary 1D distribution with Metropolis-Hastings MCMC. Tune the proposal variance sigma and watch the trace, histogram, and autocorrelation update…
Monte Carlo Pi Estimator
Monte Carlo & Uncertainty
Estimate π using random numbers! This Monte Carlo simulation drops points into a square to reveal the value of Pi through probability and error analysis.
Monte Carlo Statistics Simulator — π, Integration, CLT
Monte Carlo & Uncertainty
Explore Monte Carlo methods: estimate π, test the Central Limit Theorem, perform integration, and simulate random processes with this interactive statistics simulator.
Morris Elementary Effects Method Simulator
Monte Carlo & Uncertainty
Run the cheapest global sensitivity method — the Morris Elementary Effects (EE) screening — directly in your browser.
Polynomial Chaos Expansion (PCE) Simulator
Monte Carlo & Uncertainty
For a Gaussian input X ~ N(μ, σ²) and the model Y = a₁X + a₂X², this tool builds a Hermite Polynomial Chaos Expansion in real time and reads off the mean, variance and S…
Quasi-Monte Carlo Simulator — Sobol Sequence
Monte Carlo & Uncertainty
Integrate numerically with low-discrepancy sequences such as the Sobol and Halton sequences. Change the sample count, sequence type and integrand to watch how determinis…
2D Random Walk Simulator — MSD & Diffusion Coefficient
Monte Carlo & Uncertainty
Simulate 2D random walks in real time. Visualize diffusion laws with MSD plots and switch between lattice, Gaussian, and Lévy flight modes.
Rejection Sampling Simulator — Monte Carlo Sample Generation
Monte Carlo & Uncertainty
When direct sampling from a target p(x) is hard, cover it with a proposal q(x) and an envelope M*q(x) >= p(x), then accept each candidate with probability p(x)/(M*q(x)).…
Sobol Sensitivity Indices Simulator
Monte Carlo & Uncertainty
A hands-on global sensitivity analysis tool that decomposes how input uncertainty drives output variance into main effects and interactions.
Central Limit Theorem Simulator — Sampling Distribution Visualization
Monte Carlo & Uncertainty
Repeatedly sample from uniform, exponential, bimodal, or skewed distributions. Vary sample size n and observe how sample means converge to a normal distribution regardle…
Stratified Sampling Simulator
Monte Carlo & Uncertainty
A tool for experiencing stratified sampling, a way to boost the accuracy of Monte Carlo integration for free. Split the interval [0,1] into strata and draw samples evenl…

How to Use

  1. Select a distribution type (Normal, Uniform, Lognormal, Triangular) for each input parameter from the cat-filter dropdown
  2. Enter mean/mode values and standard deviations or bounds based on your uncertainty source (measurement tolerance, material variability, process drift)
  3. Set sample count (typically 5,000–50,000 iterations) and click Run to execute the Monte Carlo sampling
  4. Inspect output histograms, percentile tables, and sensitivity indices to identify dominant uncertainty contributors

Worked Example

Concrete beam deflection analysis: L=4m simply-supported steel beam, E=200GPa (±2% variability), load P=25kN (±5% sensor uncertainty), moment of inertia I=150cm⁴ (±3% section tolerance). Assign Normal distributions to each. With 10,000 Monte Carlo samples, mean deflection δ=8.42mm, 95th percentile=9.28mm, 5th percentile=7.61mm. Sensitivity analysis shows load contributes 65% of variance, material properties 28%, geometry 7%.

Practical Notes

  1. Use Lognormal distributions for fatigue strength and crack growth rates where lower tail asymmetry dominates failure risk
  2. Validate input bounds against tolerance stacks: if assembly clearance is ±0.5mm, use Uniform[−0.5, +0.5] rather than Normal to avoid false tail probability
  3. Always compare 95th percentile output against design limits; Monte Carlo reveals tail risks missed by nominal stress analysis
  4. For brittle materials (ceramics, concrete), increase sample count to 50,000+ to resolve low-probability fracture events accurately