design risk stratification

To address the low efficiency and poor robustness of conventional stratified sampling in network reliability assessment, this paper proposes an imbalanced stratified sampling method. It jointly stratifies based on component clustering and system failure count, employs a conditional Bernoulli model to estimate failure signatures per stratum, and—novelly—couples stratification refinement with the system-level critical failure number $i^*$ to enable threshold-driven pruning of ineffective strata. Furthermore, a heuristic optimal sample allocation strategy is designed specifically for connectivity-based performance functions. Experimental results on two canonical network reliability problems demonstrate that the proposed method significantly outperforms traditional stratified sampling and importance sampling in both accuracy and efficiency: variance reduction exceeds 40%, while robustness and scalability are markedly improved.

To address resource misallocation arising from Euclidean-distance-based Voronoi service area partitioning in complex terrain, this paper proposes the first probabilistic misallocation risk assessment framework. The method models the ratio of travel distance to Euclidean distance using a lognormal distribution, integrates local Voronoi geometric structure to derive misallocation probability, and incorporates spatial stratification and sensitivity analysis to capture spatial heterogeneity—requiring only 30–100 samples for rapid calibration. Empirical evaluation in Extremadura, Spain, identifies misallocation in 15.4% of municipalities, with theoretical prediction intervals closely matching observed outcomes. The algorithm exhibits O(n) time complexity and achieves 95% consistency with high-fidelity spatial models. This work establishes the first formal quantification and context-adaptive evaluation of Voronoi-based service allocation risk.

This paper addresses the fair allocation of scarce resources—such as vaccines and educational seats—under multi-category reserved quotas (e.g., for minority groups or high-risk populations). We propose a novel Threshold model that, for the first time, enables independent ranking within each priority category and arbitrary setting of benefit/admission thresholds, thereby decoupling inter-category priority dependencies and mitigating inefficiencies and incomparability arising from ambiguous eligibility boundaries. Furthermore, we design a Smart Pipeline Matching mechanism that jointly optimizes allocation scale and multiple fairness objectives—even under general preference domains. The framework guarantees strong strategyproofness, Pareto efficiency, fairness compliance, and maximum coverage. Evaluations on simulated Indian university admissions and COVID-19 vaccine distribution demonstrate a 12.7% improvement in coverage and a substantial reduction in inter-group disputes.

In simulation-based feasibility screening under stochastic constraints, existing indifference-zone (IZ) and IZ-free methods suffer from low statistical efficiency when system performances are either close to or far from the constraint threshold. Method: This paper proposes a sequential screening method based on variable relaxation tolerances. It dynamically adjusts decision thresholds and employs a dual-subprocedure architecture integrated with performance estimation. Multi-level hypothesis testing and statistical difference-zone analysis are incorporated to balance statistical rigor and computational efficiency. Contribution/Results: Theoretically, the method guarantees correct identification with a user-specified confidence level. Empirically, it significantly reduces average simulation runs compared to state-of-the-art IZ and IZ-free approaches, thereby substantially improving screening efficiency while maintaining statistical validity.

This study addresses the limitations of traditional risk matrices in supporting fine-grained, context-sensitive risk decision-making within complex dynamic systems. The authors propose a traceable, three-stage risk analysis framework: first, employing a multidimensional polar-coordinate heatmap to enable context-aware risk prioritization; second, constructing Bowtie causal barrier models for high-priority risks; and third, automatically transforming these Bowtie models into Bayesian networks to facilitate dynamic inference and “what-if” scenario analysis. A key innovation lies in explicitly modeling barriers as activated nodes, thereby establishing an integrated pathway from macro-level risk screening to micro-level intervention. Validation in a real-time payment gateway setting demonstrates that the proposed approach significantly enhances the transparency, auditability, and operational readiness of risk analysis.

This work investigates the performance trade-offs of non-adaptive strategies in stochastic load balancing. It proposes a two-stage model: in the first stage, each job reserves up to $k$ machines based on the task size distribution; in the second stage, after observing the actual job size, it is assigned to one of the reserved machines to minimize the expected makespan. The paper establishes, for the first time in this setting, a “power of two choices” theory, showing that under identical machines, reserving just two machines per job suffices to achieve a constant-factor approximation to the omniscient optimal solution. For related machines, it provides an $O(\log m / \log \log m)$-approximation and a bicriteria constant-factor approximation, and further proves that with 2-reservation, one can approximate the adaptively optimal solution.

Traditional disaster modeling often overlooks the system-specific responses of physical infrastructure to climate hazards, leading to biased risk assessments. This study proposes an infrastructure-oriented, system-specific hazard modeling framework that explicitly links hazard definitions to the unique characteristics of individual infrastructure systems, thereby capturing their differential hazard perceptions. By integrating climate data with infrastructure-specific attributes, a threshold-based probabilistic hazard model generates time-varying hazard maps under changing climatic conditions. The approach enables customized exposure assessments across diverse infrastructure types and hazard scenarios, demonstrating both practical utility and scalability for high-resolution risk modeling.

This study addresses the problem of reserving time-windowed airspace corridor capacity for low-altitude drone logistics under demand uncertainty to maximize expected net revenue. The authors propose a two-stage stochastic programming model: in the first stage, reserved capacities are determined for each corridor–time slot pair; in the second stage, realized delivery requests are routed through the network, explicitly capturing the coupling characteristics that capacity is non-transferable and jointly consumed along spatiotemporal paths. Innovatively integrating endogenous capacity reservation with route selection, the work establishes the equivalence between arc-flow and path-packing formulations and develops a Benders decomposition algorithm accelerated by column generation to efficiently obtain integer solutions. Computational experiments demonstrate single-digit LP-Benders gaps on medium-scale networks and scalability to large instances; a case study of Shenzhen reveals that reservations concentrate in structurally central corridors and are highly sensitive to demand levels and pricing.