Chapter Four · failure evidence
What Monte Carlo Simulation got wrong, from 82 dissertations
Monte Carlo methods across these studies frequently faced severe computational bottlenecks and convergence failures when applied to high-dimensional systems or rare events. Researchers often rejected or replaced Monte Carlo simulations due to excessive runtime, sampling noise in optimization loops, and poor scaling compared to analytical or stratified alternatives. These records come from PhD theses at 24 institutions, 2021 to 2026. Each links to its thesis. They were extracted by language models reading the full text, so treat each as a lead to read, not a verdict.
Excessive computational cost and memory overhead in high dimensional simulations
Standard and Markov chain Monte Carlo methods become computationally intractable when applied to complex physics models, large spatial networks, or high dimensional latent spaces. The vast number of required sample evaluations demands processing time and memory budgets that exceed practical computing limits.
Tried and failed
Monte Carlo simulation with fine mesh resolution applied to wildfire spread modeling. Outcome: did not converge. Reason: fine mesh stochastic simulations required thousands of runs to reach target error, exceeding practical computational budgets
Wildfire simulations to protect rural communities and avoid dire evacuations · Imperial
Tried and failed
nested Monte Carlo calibration with neural SDEs applied to joint asset price and volatility calibration. Outcome: infeasible cost. Reason: GPU memory limits restricted nested Monte Carlo path counts, causing severe sampling errors and poor fits
Deep learning-based methods in the era of rough volatility · Imperial
Tried and failed
Monte Carlo EM for joint MLE applied to bivariate mixed-effects models. Outcome: infeasible cost. Reason: High computational cost yielded negligible MSE improvement over simpler two-step composite estimators
Small area estimation and graphical model for complex surveys · Iowa State
Considered and rejected
Considered and rejected: Rejected Monte Carlo simulation for transient stability index due to prohibitive computational expense in multimachine systems.
Considered and rejected
Considered and rejected: Rejected the Monte Carlo method for angular directional discretisation due to excessive computational cost required to reduce statistical sampling error.
Considered and rejected
Considered and rejected: Rejected standard Monte Carlo simulation for multiscale tissue uncertainty propagation due to prohibitive computational burden.
Data analytics and mathematical modeling to advance cardiac care · Texas Tech
Considered and rejected
Considered and rejected: Continuous probability density sampling of multi-year electricity/regulation prices in the 15-year Monte Carlo model, rejected due to computational intractability of repeated stochastic optimizations.
Integrated Modeling Approaches to Quantify Vehicle-to-Grid Services in an Evolving Power Sector · MIT
Considered and rejected
Considered and rejected: Rejected Monte Carlo NLS (MC-NLS) formulation for practical circuit simulations due to computational inefficiency.
Switching Dynamics in Ferroelectric Hf₀.₅Zr₀.₅O₂ Devices: Experiments and Models · MIT
Considered and rejected
Considered and rejected: Rejected transported PDF methods solved via stochastic Monte-Carlo particles for practical engine simulations due to prohibitive computational expense.
Conditional source-term estimation evaluations for partially-premixed flames · Oxford
Considered and rejected
Considered and rejected: Rejected detailed micro-modeling of brick-mortar interfaces due to prohibitive computational costs in Monte Carlo simulation reliability workflows
STUDY OF SEISMIC BEHAVIOUR OF MASONRY INFILLED RC FRAMES · DalSpace
Considered and rejected
Considered and rejected: Hamiltonian Monte Carlo / BAT.jl was rejected for KaFit due to the prohibitive computational cost of numerical finite-difference gradients in non-analytic models.
An Optimized Bayesian Analysis Framework for the KATRIN Experiment · MIT
Considered and rejected
Considered and rejected: Rejected exact likelihood integration via MCMC/Monte Carlo due to prohibitive computational costs in high-dimensional latent space q.
Considered and rejected
Considered and rejected: Rejected standard greedy hill-climbing over direct Monte-Carlo simulations of diffusion cascades for influence maximization due to Omega(mnk * POLY(eps^{-1})) runtime.
Considered and rejected
Considered and rejected: Rejected high-resolution photon Monte Carlo radiative transport (Prem et al. 2019) because rovibrational explicit modeling was computationally prohibitive and existing opacity model had <1% error.
Simulating Tvashtar's plume observed during the 2007 New Horizons Io flyby · UT Austin
Considered and rejected
Considered and rejected: Rejected full radiative transfer modeling including infrared emission/absorption because multiple integral calculations are computationally too cumbersome for fast Monte Carlo profile generation.
Considered and rejected
Considered and rejected: Rejected full discrete stochastic Monte Carlo microsimulation in favor of analytical integration/multiplication directly on survival functions to save computation time.
Accounting for Heterogeneity in Health Decision Analysis · Harvard
Tried and failed
ensemble Markov Chain Monte Carlo sampling applied to multi-field cosmological inflation models. Outcome: infeasible cost. Reason: running chains to standard autocorrelation convergence thresholds required computationally prohibitive sample counts
Cosmic Echoes of the Early Universe: From Primordial Black Holes to Gravitational Waves · MIT
Considered and rejected
Considered and rejected: Markov-Chain Monte Carlo (MCMC) sampling for Bayesian harmonic inference; rejected because it became computationally intractable when scaling across spatially coherent networks of thousands of nodes.
Spatiotemporal tidal prediction and analysis through physics-informed machine learning · Oxford
Considered and rejected
Considered and rejected: Rejected direct sampling from raw Monte Carlo posterior histograms via MCMC/Metropolis-Hastings as prohibitively cumbersome and inefficient compared to parametric variational inference.
Considered and rejected
Considered and rejected: Rejected standard Markov Chain Monte Carlo (MCMC) sampling in favor of dynamic nested sampling (dynesty), because Python MCMC implementations do not provide a computationally efficient way to calculate the Bayesian evidence Z.
Providing new constraints on Europa's surface composition · Cornell
Slower convergence and inferior sample efficiency compared to analytical or stratified baselines
Pure Monte Carlo sampling often exhibits higher variance and slower convergence than closed-form expressions, Latin Hypercube Sampling, or semi-analytical methods. As a result, practitioners frequently abandoned standard sampling in favor of deterministic or variance-reduced alternatives that reach target precision with fewer evaluations.
Tried and failed
Monte Carlo simulation for statistical power estimation applied to hypothesis testing power analysis. Reason: produced equivalent power estimates to analytical t-tests while restricting output metrics
Detecting Change in Belowground Carbon Stocks: Statistical Feasibility and Future Opportunities from Loblolly Pine Forests · Virginia Tech
Lost to a baseline
Monte Carlo returns (MCR) and Monte Carlo with value baseline (MCVB) achieved lower return, lower success rate, and lower trajectory ensemble entropy than actor-critic (AC) on the tabular 100-step random walk excursion problem.
Trajectory Ensembles and Machine Learning: From reinforcement learning for rare event sampling to training of neural network ensembles · University of Nottingham Repository
Considered and rejected
Considered and rejected: Rejected Monte-Carlo simulation for compressor station availability evaluation due to high computational time, cost of execution, and lower precision compared to analytical binomial methods.
Introduction of social benefits to the tera – gas turbines and pipelines · Cranfield
Lost to a baseline
Seq-RE achieved higher reported speedup factors against Monte Carlo (e.g. 4.00x10^4 vs 3.46x10^2 on S27) due to comparing against 10^12 MC iterations instead of 10^5.
Efficient Evaluation of Probability and Reliability with Digital Integrated Circuits · Scholarship at UWindsor Institutional Repository
Considered and rejected
Considered and rejected: Rejected using naive Monte Carlo simulation to evaluate expectation in state evolution for high dimensions (p in thousands) due to instability and inefficiency, replacing it with quantile discretization and closed-form conditional expectation.
Algorithmic Analysis And Statistical Inference Of Sparse Models In High Dimension · Penn
Considered and rejected
Considered and rejected: Rejected relying solely on Monte Carlo approximations or amortized explainers in isolation due to the slow inference convergence of Monte Carlo methods and the high approximation error of amortized models
Considered and rejected
Considered and rejected: Rejected naive Monte Carlo pointwise integration of heat content (M paths from K starting positions) in favor of empirical survival distribution estimation conditioned on positive measure domains
Tried and failed
Monte Carlo volume integration over implicit boundaries applied to hydrostatic force computation. Outcome: did not converge. Reason: achieving relative accuracy below 10^-3 was challenging for Monte Carlo and Quasi-Monte Carlo integration
Considered and rejected
Considered and rejected: Rejected standard Bernoulli Monte Carlo sampling for mission success probability due to high discretization error and slow convergence compared to direction-based chi-squared integration.
Considered and rejected
Considered and rejected: Rejected pure Monte Carlo integration for TM optimization due to redundant clustered integration points.
Efficient diagnostics of complex mechanical systems · Leibniz Universität Hannover Repository
Considered and rejected
Considered and rejected: Rejected Monte Carlo sampling for base scenario generation due to slow moment convergence compared to Latin Hypercube Sampling.
Machine Learning based Methods to Improve Power System Operation under High Renewable Pennetration · Virginia Tech
Considered and rejected
Considered and rejected: Rejected Monte Carlo time randomization across the cyclotron period due to introduced systematic seeding error and computational overhead, replacing it with the analytical kernel method
Measuring the anomalous precession frequency wa for the Muon g − 2 experiment · ResearchWorks
Model misspecification and distribution distortion in complex physical systems
Approximations relying on uncorrected Monte Carlo proposals, empirical distribution fits, or simplified transport physics produce inaccurate tail predictions and unphysical states. These sampling setups suffer from severe variance overestimation, unphysical particle losses, or decorrelation over long simulation horizons.
Tried and failed
least squares Monte Carlo simulation applied to multi-uncertainty real options valuation. Outcome: unstable. Reason: regression exhibited severe heteroskedasticity across the multi-uncertainty life-cycle state space
Tried and failed
without-replacement resampling in sequential Monte Carlo applied to constrained language model decoding. Outcome: worse than baseline. Reason: hurt downstream execution accuracy compared to standard multinomial resampling
Scaling Bayesian inference for generative models via probabilistic programming · MIT
Tried and failed
Monte Carlo dropout for uncertainty quantification applied to spatiotemporal neural network predictions. Outcome: worse than baseline. Reason: produced overly narrow prediction intervals and degraded point prediction quality
Innovations in Urban Computing: Uncertainty Quantification, Data Fusion, and Generative Urban Design · MIT
Tried and failed
sampling product of experts without importance weighting applied to sequential Monte Carlo decoding with constraints. Outcome: worse than baseline. Reason: omitting importance weight corrections severely degraded approximation of the global target posterior distribution
Scaling Bayesian inference for generative models via probabilistic programming · MIT
Tried and failed
Monte Carlo energy deposition and light transport simulation applied to heterogeneous powder-matrix scintillator composite. Reason: optical transport and microscale particle interaction modeling in heterogeneous powder suspension was not predictive
Gamma-Blind Fast Neutron Detection for Spent Nuclear Fuel Characterization · EPFL
Tried and failed
Monte Carlo event generator cross section modeling applied to forward lepton-nucleus scattering cross sections. Outcome: did not generalise. Reason: Inadequate modeling of nuclear effects and kinematics overpredicted cross sections at small forward scattering angles.
Lepton-Nucleus Constraints for Neutrino Interactions and Oscillations · MIT
Tried and failed
Monte Carlo particle transport simulation applied to randomly dispersed particulate fuel cores. Reason: extreme sensitivity to stochastic spatial particle distribution caused significant eigenvalue over-prediction
Tried and failed
Monte Carlo particle transport in high-fidelity CAD geometry applied to divertor vacuum pump particle balance. Outcome: unstable. Reason: caused numerical instability and unphysical particle loss instead of reaching steady-state balance
Understanding the Mechanisms that Determine the Edge Electron Density Profile in Tokamaks · MIT
Tried and failed
Monte Carlo simulation with fitted empirical distributions applied to multi-stage manufacturing scrap forecasting. Outcome: did not generalise. Reason: fitted distributions produced large forecast errors and severe variance overestimation at upper confidence intervals
Tried and failed
direct simulation modeling of detector shower shapes applied to calorimeter shower shape distribution tail. Outcome: did not generalise. Reason: out-of-the-box Monte Carlo simulations failed to accurately model the distribution tail
Differential measurements of Z and γ bosons produced with jets at the CMS experiment · Imperial
Tried and failed
multilevel Monte Carlo path simulation applied to chaotic dynamical systems. Outcome: did not converge. Reason: Fine and coarse trajectory realizations decorrelate over long horizons, preventing variance reduction across levels.
Multi-Level Monte Carlo Methods for Uncertainty Quantification and Risk-Averse Optimisation · EPFL
Sampling noise and non-differentiability within iterative optimization loops
Evaluating Monte Carlo approximations inside iterative algorithms or dynamic programming updates introduces severe stochastic noise and non-differentiable surfaces. These fluctuations lead to stagnant convergence, vanishing derivatives, and unstable policy or state recursions.
Tried and failed
nested Monte Carlo risk analysis inside optimization applied to techno-economic system evaluation. Outcome: too slow. Reason: Computational overhead of repeated risk simulations was too high for iterative optimization or sensitivity analysis
Techno-economic environmental risk analysis of sustainable power systems. · Cranfield
Considered and rejected
Considered and rejected: Rejected test-set MSE stopping criteria in dynamic programming because evaluating conditional expectations via Monte Carlo at every test state was computationally prohibitive.
Considered and rejected
Considered and rejected: Directly incorporating stochastic Monte Carlo buckling simulations into the early-stage optimization loop was rejected due to prohibitive computational effort.
Integrative Form Finding and AI-Driven Human-in-the-Loop approach for Structural Optimization in Long-Span Architectural Design · IRIS - POLITO - prod
Considered and rejected
Considered and rejected: Rejected solving the KKT system (2.6) directly via Monte Carlo gradient descent due to computational intractability of evaluating expected overage/underage integrals and derivatives.
Constrained Inventory Optimization on Complex Warehouse Networks · MIT
Tried and failed
single-batch Monte Carlo relaxation applied to coupled multiphysics transient simulations. Outcome: did not converge. Reason: extreme statistical noise stagnates convergence despite dampening numerical oscillations
Modeling Feedback Effects of Transient Nuclear Systems Using Monte Carlo · MIT
Considered and rejected
Considered and rejected: Rejected direct differentiation of empirical Monte Carlo sums because derivatives become piecewise constant or zero, destroying convergence rates.
Multi-Level Monte Carlo Methods for Uncertainty Quantification and Risk-Averse Optimisation · EPFL
Considered and rejected
Considered and rejected: Rejected using the direct mode of Monte Carlo simulated estimates for progress projection due to instability when multiple local modes exist.
Deep Learning Classification of Spinal Osteoporotic Compression Fractures on Radiographs · ResearchWorks
Considered and rejected
Considered and rejected: Rejected Monte Carlo risk evaluation within the trajectory optimization loop due to non-smoothness, non-differentiability, and high sample complexity.
Considered and rejected
Considered and rejected: Rejected Monte Carlo simulation for exogenous state transitions to avoid simulation noise and sample instability in the Bellman recursion.
Three Essays on Sustainable Operations: Renewable Energy Procurement, Battery Storage, and Equitable Work Scheduling · ResearchWorks
Failure to capture spatial correlations and multi-variable parameter dependencies
Sampling randomly from marginal distributions ignores critical inter-variable correlations, load patterns, and structural network constraints. Consequently, the simulations produce physically implausible parameter pairings, chaotic divergence, or distorted uncertainty bounds.
Tried and failed
Monte Carlo scenario generation with rolling volatility applied to stochastic asset allocation models. Outcome: unstable. Reason: scenario generation failed uniformity tests under true empirical volatility without artificial parameter dampening
Improving time series and cross-sectional momentum trading strategies using stochastic programming · Iowa State
Tried and failed
unconstrained Monte Carlo parameter sampling applied to parametric learning production modeling. Outcome: did not generalise. Reason: unbounded parameter combinations generated unrealistic negative gain correlations failing to reproduce empirical meta-analytic patterns
Tried and failed
sequential Monte Carlo simulation applied to capacity adequacy with intermittent generation. Outcome: unstable. Reason: numerical instability and high relative error under high penetration of intermittent sources
Adequacy Assessment of a Combined Generating System Containing Wind Energy Conversion Systems · HARVEST
Tried and failed
dynamic network simulation with Monte Carlo sampling applied to complex organizational communication systems. Outcome: unstable. Reason: extreme sensitivity to initial conditions caused chaotic divergence across trials
Measuring the Communicative Constitution of Partial Organizations as Complex Systems · Virginia Tech
Tried and failed
Monte Carlo sampled independent deterministic inversions applied to correlated geophysical dispersion data. Reason: sampling strategy distorted output uncertainty bounds across frequencies
Considered and rejected
Considered and rejected: Rejected standard Monte Carlo random sampling from marginal parameter distributions because it creates dynamically inconsistent and physically implausible parameter pairings by ignoring variable correlations.
Scenario-Based Methodology for Predicting the Safety Benefits of Emerging Advanced Rider Assistance Systems (ARAS) · Virginia Tech
Considered and rejected
Considered and rejected: Rejected Monte Carlo simulation for demand scenario generation because standard sampling ignores inter-substation load correlation and pattern coincidences
Considered and rejected
Considered and rejected: Rejected standard non-sequential Monte Carlo simulation (NSMCS) and other importance splitting variants (fixed-splitting, fixed number of successes) for generating systems short-term reliability due to non-Markovian static structure and population explosion risks.
OPERATIONAL RELIABILITY AND RISK EVALUATION FRAMEWORKS FOR SUSTAINABLE ELECTRIC POWER SYSTEMS · HARVEST
Considered and rejected
Considered and rejected: Standard Wolff cluster Monte Carlo rejected because suitability objective introduces spatial inhomogeneities that violate the algorithm's symmetry assumptions.
Physical Inference for Optimization and Design · Queens University Institutional Repository
Inaccurate estimation and vanishing signal in rare event and thin region sampling
Direct Monte Carlo simulation struggles to estimate extremely small failure probabilities or particle interactions in optically thin regions. Insufficient sampling of these critical tail events results in zero reaction rates, severe underestimation of cascade contingencies, or tallies with extreme relative error.
Tried and failed
Monte Carlo simulation for rare event reliability applied to multi-state degrading subsea systems. Outcome: infeasible cost. Reason: computationally expensive and inaccurate when estimating extremely small failure probabilities in complex systems
Advanced reliability analysis of complex offshore Energy systems subject to condition based maintenance. · Cranfield
Tried and failed
direct Monte Carlo simulation on FEA models applied to estimating rare event failure probabilities. Outcome: infeasible cost. Reason: evaluating very small failure probabilities requires prohibitive sample sizes with expensive simulations
Structural reliability assessment of complex offshore structures based on non-intrusive stochastic methods · Cranfield
Tried and failed
Standard Monte Carlo error estimation applied to device voltage model reliability estimation. Outcome: worse than baseline. Reason: Produced error probability estimates an order of magnitude higher than importance sampling
SLM/CCDD structures for high speed numerical computing and signal processing · Iowa State
Considered and rejected
Considered and rejected: Standard Monte Carlo simulation was rejected for generating cascading failure datasets because it severely underestimates N-3+ contingencies and yields ~0.1% cascade cases.
Real-time Prediction of Cascading Failures in Power Systems · HARVEST
Tried and failed
Monte Carlo particle transport with low history counts applied to dosimetric tally estimation in complex phantoms. Outcome: no signal. Reason: insufficient particle histories led to ~100% relative error tallies lacking statistical significance
Tried and failed
Monte Carlo particle transport simulation applied to optically thin low cross-section regions. Outcome: no signal. Reason: Inadequate sampling and low capture cross-sections in thin regions caused zero reaction rates with high statistical variance.
Tried and failed
forward Monte Carlo ray tracing applied to radiative transfer to small aperture. Outcome: too slow. Reason: Extremely low probability of rays randomly hitting a small target aperture made convergence computationally intractable.
Considered and rejected
Considered and rejected: Rejected simulating training points solely from ergodic distributions via Monte Carlo, as rare crisis regions lack sufficient mass and induce severe future-iteration instabilities
Entrapment in local optima and poor mixing in multimodal state spaces
Markov chain and continuous-time Monte Carlo algorithms frequently fail to converge when sampling rough energy landscapes, severe rate disparities, or multimodal targets. The samplers become trapped in local modes, failing to mix properly or explore alternative trajectory configurations.
Tried and failed
kinetic Monte Carlo simulation applied to hopping transport with extreme rate disparities. Outcome: did not converge. Reason: extreme transition rate disparities between transport directions prevented convergence, requiring a Master Equation solver
Modelling the microstructure-charge transport relationship in organic semiconductors · Imperial
Tried and failed
Monte Carlo sampled correlation functions for covariance estimation applied to Hamiltonian reconstruction from quantum states. Outcome: did not converge. Reason: Monte Carlo sampling failed to converge energy variance and showed slow convergence of reconstructed coupling parameters
Computational Approaches to Frustrated Many-Body Systems · Cornell
Tried and failed
transition path sampling continuous-time Monte Carlo applied to finite-time tilted trajectory ensembles. Outcome: did not converge. Reason: fails to sample dynamical activity at long trajectory times without guided reference dynamics
Non-equilibrium dynamics and large deviations in stochastic lattice models via tensor networks · University of Nottingham Repository
Considered and rejected
Considered and rejected: Rejected parameterizing initial PDF and optimizing via Maximum Likelihood Estimation with Monte Carlo integration because of convergence to local minima and lack of feasibility guarantees.
Koopman Operator Approach to Uncertainty Quantification and Decision-Making · Georgia Tech
Considered and rejected
Considered and rejected: Rejected Monte-Carlo optimization for molecular orientation exploration due to high risk of trapping in local minima.
Structure Formation during Organic Molecular Beam Deposition · Publikationssystem UB Tuebingen
Tried and failed
Markov Chain Monte Carlo data association applied to multimodal trajectory tracking. Outcome: did not converge. Reason: Failed to explore multiple modes, becoming stuck in local optima and yielding overconfident, incorrect posterior distributions.
Uncertainty Quantification and Structure Discovery for Scalable Behavior Science · MIT
Tried and failed
single-chain Metropolis-Hastings Markov chain Monte Carlo applied to compact graph partitioning and redistricting. Outcome: did not converge. Reason: Severe multimodal distribution hindered mixing, showing no convergence after days of execution.
Understanding and Mitigating Bias in Algorithms, Data, and Society · Georgia Tech
Considered and rejected
Considered and rejected: Rejected joint likelihood inference via Monte Carlo EM due to high computational intensity and convergence issues.
Multivariate One-Sided Tests for Nonlinear Mixed-Effects Models with Incomplete Data · YorkSpace
Left open by the authors
Problems the authors named and did not get to.
Left open
Investigate the relationship between Monte Carlo scenario generation reliability and stochastic programming portfolio solution performance across simulated market conditions. Blocker: No concrete experimental design, mathematical formulation of scenario reliability, or target metrics are specified.
Improving time series and cross-sectional momentum trading strategies using stochastic programming · Iowa State
Left open
Characterize the cost models for various supply chain resilience options and incorporate them into the Value at Risk Monte Carlo simulation framework. Blocker: Empirical cost data for enterprise supply chain resilience investment options is proprietary or restricted.
Strategic Approach for Assessing Supply Chain Resilience Investment Options · MIT
Left open
Implement statistical Weibull Monte Carlo simulation capabilities in Peri-CORE to assess TRISO fuel batch failure probabilities. Blocker: Requires access to the proprietary or unreleased Peri-CORE Abaqus Fortran subroutines developed in the thesis
Peridynamics and finite element simulations of TRISO Fuel under extreme operating conditions · Imperial
Left open
Derive convergence rates for Importance Support Points resampling and lookback adaptation in projected quasi-Monte Carlo methods. Blocker: Requires specialized theoretical mathematical analysis rather than software engineering.
Novel Experimental Design Techniques for Data Science · Georgia Tech
Left open
Run Monte Carlo simulations and sensitivity analyses on the proposed RGC and feedback TIA circuit models across operating conditions. Blocker: Requires the exact custom IC netlists and component models developed in the thesis
High-Speed Optoelectronic Detector Front-End for Optical Coherence Tomography Applications · Harvard
Left open
Calibrate the Opisthorchis viverrini spatial renewal model against empirical infection data from the Lower Mekong Basin using sequential Monte Carlo. Blocker: Access to empirical Opisthorchis viverrini surveillance and mobility data from the Lower Mekong Basin in Laos.
Left open
Test practical identifiability of SEIR and GRM epidemic model parameters using a Monte Carlo simulation approach. Blocker: None
Validating Forecasting Strategies of Simple Epidemic Models on the 2015-2016 Zika Epidemic · Virginia Tech
Left open
Compare Monte Carlo approximations of crop insurance premiums from the fitted bivariate jump-diffusion model to empirical USDA premium values. Blocker: None
Statistical applications in actuarial science: From cryptocurrency to meme stocks to crop insurance · Iowa State
Left open
Evaluate pressure function equilibria using stochastic microsimulation with Monte Carlo modeling baseline and stochastic origin-destination demand. Blocker: None
Bridging the gap: unifying transportation planning and operations through enhanced travel demand modeling · UT Austin
Left open
Develop alternative coupling methods for Multi-Level Monte Carlo to restore variance decay in chaotic and turbulent high Reynolds number fluid flow simulations. Blocker: The unfinished work proposes a broad research direction without specifying concrete coupling formulations or validation benchmark cases
Multi-Level Monte Carlo Methods for Uncertainty Quantification and Risk-Averse Optimisation · EPFL
Checking a claim in this area?
We can run the same search on any method or claim. If nothing turns up, we will say so, and that proves nothing on its own.