Rea Press null Rea Press3042-01993042-0199 Rea Press https://doi.org/10.22105/opt.v3i3.119 Research Article Structural sizing, Adaptive operator selection, Constraint handling, Population entropy, Evaluation accounting, Reproducibility Constraint-Aware Entropy-Adaptive Swarm Search for Civil Infrastructure Sizing: Formulation and Reported Benchmark EvidenceConstraint-Aware Entropy-Adaptive Swarm Search for Civil Infrastructure Sizing: Formulation and Reported Benchmark Evidence Khan Hassan M. Department of Civil Engineering, NED University of Engineering & Technology, Karachi, Pakistan. El-Sayed Ahmed M. Department of Structural Engineering, Ain Shams University, Cairo, Egypt. Boutros Rania Department of Civil and Environmental Engineering, University of Auckland, Auckland, New Zealand. Lewandowski Tomasz Department of Civil Engineering, Wroclaw University of Science and Technology, Wroclaw, Poland. 09 2026 19 09 2026 3 3 © 2026 Rea Press 2026

This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

 Constraint-Aware Entropy-Adaptive Swarm Search for Civil Infrastructure Sizing: Formulation and Reported Benchmark Evidence

SI-CASS is a proposed population-search framework for civil sizing that combines an operator portfolio, structural-response-based repair and entropy-triggered reinjection. The formulation distinguishes simplified optimization constraints from the complete checks required by a project design code. Repair is a local heuristic followed by structural reevaluation; neither catalog rounding nor area scaling guarantees feasibility or code certification. A conditional coverage argument describes requirements for an infinite search extension, while the specified finite-budget process has no established global convergence rate. Reported summaries from six problems and 30 runs retain the original numerical observations, including a 25-bar-truss mean of 247.1 versus 252.4 for SHADE and 263.5 for PSO. The underlying models, units, implementation and run-level outputs are unavailable, and inconsistencies among the tables, statistical adjustments and figures prevent verification of significance and causal component effects. The contribution is the stated search architecture and an explicit account of the evidence needed to assess it. The reported designs are not certified engineering deliverables.

null