Distributed antenna layouts optimized for nominal demand may deteriorate under changing hotspots and correlated shadowing. We present SA-MOBO-DAS, a scenario-aware multi-objective Bayesian optimization framework for placing four shared LTE/5G remote antenna units in a synthetic 500 × 500 m urban area. Its contribution is a placement and validation protocol combining worst-band coverage and throughput objectives, frozen operating scenarios, a fixed evaluation budget, and independent post-search risk assessment. A reproducible simulator incorporates 3GPP TR 38.901-inspired path loss, obstruction footprints, correlated shadowing, strongest power association, full-frequency reuse, and load-aware throughput at 2.1 GHz LTE and 3.5 GHz NR. Independent Matérn-5/2 Gaussian processes and a finite-pool approximation of q-noisy expected hypervolume improvement guide an 80-evaluation search. Thirty paired runs used 64 development and 160 held-out scenarios. SA-MOBO-DAS achieved held-out lower-tail hypervolume of 0.006619 (SD 0.000822), with no statistically detected difference from ParEGO (+0.7%; adjusted p = 0.968). Hypervolume exceeded NSGA-II and Sobol by 27.2% and 24.4%, respectively (both adjusted p < 0.001). Direct CVaR optimization yielded 7.6% and 0.9% higher point estimates of stress tail coverage and throughput, respectively, but reduced overall hypervolume by 10.6%. Paired 95% confidence intervals for both stress tail differences included zero.
The introduction of high-speed rail transport imposes new requirements for safety and traffic control. The integration of modern satellite technologies, digital communication systems and data transmission, ensured the highest level of high-speed rail transport quality. As many researchers note, it is important not only to equip rolling stock with satellite communications, but also to ensure effective processing of their motion parameters based on automated information processing and decisionmaking technologies.
This computational study examines whether a published optical calibration identifies mechanical parameters and supports a unique design recommendation for a dual fiber Bragg grating (FBG) cantilever–spring displacement sensor. Within the adopted series compliance model, the 39.47 pm/mm sensitivity identifies the product of strain transfer and equivalent stiffness. The same anchor admits K = 71.969–91.228 N/m for η = 0.80–1.00. A counterexample extracted from archived nonlinear results shows that η = 0.9500 and 0.9595, with compatible reference stiffnesses of 75.979 and 75.183 N/m, retain only Selected 2 and Selected 3, respectively, across four declared tip geometries under unchanged constraints. Thus, the calibration does not identify a robust choice among the five retained candidates within this family. Selected 2’s nominal 43.07% nonlinearity reduction costs 23.54% more force. Retrospective published optical cycles give a fixed zero root mean square error (RMSE) of 153.6 pm on 37 recovered markers; coordinate and ruler uncertainty preclude hardware accuracy claims. Existing finite element, uncertainty, thermal, bonding and wavelength allocation analyses provide supporting conditional assessments. The contribution is a sensor-specific demonstration of which mechanical and design conclusions remain underdetermined by the adopted calibration. Independent physical validation remains necessary for predictive hardware claims.
Ranking reforms can leave changes in university positions unresolved when published scores are rounded, missing or difficult to compare across editions. We examine which observations suffice to certify improvement or worsening across all admissible configurations in a conditional Quacquarelli Symonds (QS) reform scenario for a fixed cohort of 1,422 institutions. The analysis retains source-edition legacy-score intervals, applies the announced reform weights and varies disclosure of International Research Network, Employment Outcomes and Sustainability scores. Publication-constrained marginal bounds screen the cohort, while mixed-integer linear programming constructs jointly feasible extrema and counterexamples. Competitor disclosure supports substantially more direction certificates than focal disclosure alone. Exhaustive comparison establishes minimum disclosure within declared field subsets, and selected competitor observations sharply reduce information volume without establishing global cell-level minimality. Marginal and joint optimization agree on the tested classifications; uncertified marginal results remain inconclusive. Sensitivity checks show material dependence on retaining legacy scores, while tested linkage and shared-scale variants preserve the audited directions. The findings support reproducible information-sufficiency audits under explicit assumptions; prospective forecasting utility remains unvalidated.