design sensitivity studies

This study addresses the lack of a systematic framework for identifying critical input variables and conducting sensitivity analysis under uncertainty in complex simulations, particularly in military decision-making contexts. The authors propose a unified sensitivity analysis framework that integrates local and global methods—including variance-based, derivative-based, screening, and uncertainty quantification techniques—and strategically maps these approaches to specific decision objectives such as factor prioritization, fixing, variance reduction, and mapping. Innovatively, the framework introduces a “sensitivity audit” mechanism to enhance traceability of model assumptions and promote responsible model usage. By providing a structured guide for high-dimensional, complex simulation systems, this work significantly improves model interpretability, transparency, and the credibility of decisions derived from such models.

Model-form uncertainty (MFU)—arising from simplifying modeling assumptions and particularly challenging to quantify during extrapolation—remains a critical, yet poorly addressed, source of epistemic uncertainty in physics-based modeling. Existing approaches heavily rely on calibration data and cannot isolate the independent influence of individual assumptions on predictions. Method: We propose a calibration-free MFU quantification framework that parameterizes modeling assumptions and integrates grouped variance-based sensitivity analysis to explicitly characterize how assumption changes propagate into predictive variance. The method accommodates parameter dependencies and enables assumption importance ranking under extrapolative conditions. Contribution/Results: Experiments demonstrate that our approach effectively identifies the assumptions dominating prediction uncertainty. It provides quantitative guidance for model simplification, verification, and refinement, thereby significantly enhancing the credibility and robustness of complex physics-based models.

This study addresses the long-overlooked issue of generalized uncertainty in spatial dynamic microsimulation, specifically examining whether qualitative modeling choices—such as variable definitions and state-transition rules—exert greater influence on simulation outcomes than conventional parametric and coefficient uncertainties. Method: Leveraging variance-based global sensitivity analysis, we systematically decompose direct and indirect risk propagation pathways over time for georeferenced individuals within the MikroSim employment module. Contribution/Results: Qualitative modeling decisions contribute substantially more to output variability than parameter uncertainty; commonly used aggregate metrics severely underestimate total uncertainty. Consequently, the paper advocates a paradigm shift in simulation design and result reporting—explicitly integrating qualitative modeling choices into formal uncertainty quantification frameworks. This reconfiguration strengthens both the robustness and interpretability of microsimulation models, establishing a methodological foundation for more transparent and defensible policy-relevant analyses.

This paper addresses the lack of robust design foundations for sensitivity analysis in finite-population causal inference. Methodologically, it introduces a novel sensitivity analysis framework grounded in the experimental design distribution—first integrating design-based distributions with partial identification theory to construct model-free, non-asymptotic confidence intervals for the average treatment effect (ATE). It further reinterprets the role of randomization in sensitivity analysis and provides a new design-driven rationale for covariate balance checks. Key contributions include: (1) model-free, finite-population inference under heterogeneous treatment effects; (2) robust ATE confidence intervals with clear identification-theoretic interpretation; and (3) empirical validation across three real-world applications, demonstrating reliability and practicality in small-sample and highly heterogeneous settings.

This study addresses non-expert users by systematically evaluating the workflow feasibility and factor ranking consistency of multiple global sensitivity analysis (GSA) methods across simulation models of varying complexity. Method: It integrates Sobol’ first-order and total-effect indices with regression tree analysis, and—novelty—employs Kendall’s W to quantify inter-method ranking similarity; special attention is given to how parameter range specification affects result robustness. Contribution/Results: (1) Major GSA methods exhibit high consistency in factor importance ranking; (2) Sobol’ indices offer both interpretability and information richness, while regression trees effectively detect interaction effects; (3) Parameter range specification is identified as a critical practical determinant of GSA reliability. Collectively, these findings significantly enhance operationality and methodological rationality for non-experts in tasks such as factor screening, freezing, and prioritization.

This study addresses the urgent need to enhance predictive accuracy and uncertainty quantification in structural health monitoring of aging bridges under increasing traffic loads and intensifying extreme weather events. To this end, an integrated assessment framework is developed by synergistically combining high-fidelity physics-driven numerical models, structural health monitoring (SHM) data, and global sensitivity analysis. A digital twin of an actual steel–concrete composite bridge is established within this framework, enabling effective identification of key parameters governing structural response. The proposed approach significantly improves the predictive capability for bridge behavior under diverse loading scenarios, thereby providing a robust scientific basis for informed operation and maintenance decisions.

This work proposes the first general framework to systematically quantify and apportion epistemic uncertainty arising from substituting true subprocesses with approximate or learned submodels in stochastic simulation and digital twin applications. The framework constructs confidence or credible intervals for performance metrics via bootstrapping and Bayesian model averaging, and employs a tree-based decomposition to allocate total output variability to individual submodels, yielding importance scores. It is compatible with both parametric and nonparametric models, supports frequentist and Bayesian paradigms, and accommodates dynamic initialization scenarios. Validation on synthetic data and a call center digital twin demonstrates that the method effectively reveals each submodel’s contribution to overall uncertainty, significantly enhancing the interpretability and reliability of simulation outcomes.

This work proposes a gradient-based active learning method to enhance the accuracy of global sensitivity analysis under limited computational budgets. By leveraging the posterior gradient distribution of a Gaussian process surrogate model, the approach introduces a novel acquisition function that explicitly accounts for correlations among partial derivatives, enabling intelligent selection of the most informative input samples within a constrained simulation budget. Compared to existing derivative-based global sensitivity measures (DGSM)-oriented strategies, the proposed method offers a more comprehensive and robust framework. Experimental results on multiple benchmark test functions and a real-world pesticide transport environmental model demonstrate that the method significantly outperforms state-of-the-art approaches, yielding notably improved estimates of both Sobol’ indices and DGSM sensitivity measures.

This study addresses the significant uncertainties in biomechanics arising from inter-individual variability and noisy experimental data by proposing a unified framework grounded in Bayesian probability theory. The approach systematically integrates forward uncertainty propagation and inverse parameter inference within a single coherent paradigm. It seamlessly combines input uncertainty characterization, data-driven surrogate modeling, model selection criteria, and information-theoretic optimal experimental design, while naturally linking global sensitivity analysis with spatially correlated random field priors. As the first mechanics-oriented uncertainty quantification methodology to adopt Bayesian inference as a unifying principle, this work delivers a theoretically consistent, computationally efficient, and robust toolkit for both computational and experimental mechanics, substantially enhancing the accuracy of parameter calibration and the reliability of predictive outcomes.