Quantum-Inspired Strategies: Bridging Classical and Quantum Computing for Enhanced Optimization in Structural Engineering and Feature Selection

This paper explores a quantum-inspired approach to optimization by incorporating quantum random numbers (QRNs) into classical bio-inspired algorithms. Specifically, it applies Artificial Bee Colony (ABC) and Genetic Algorithms (GAs) to two tasks: parameter fitting of the Extended Base Plate-Embedded Anchor Bolts (EBP-EAB) seismic model and feature selection from high-dimensional seismic datasets. Experiments compare QRNs with pseudorandom numbers (PRNs), showing that although the final metrics are similar, QRNs consistently yield faster convergence. In the EBP-EAB model, QRNs reduced the number of iterations needed to reach the optimal error. In feature selection, QRNs reduced the generations required to achieve high-quality subsets without compromising accuracy. These results suggest that QRNs enhance convergence efficiency, positioning hybrid quantum-classical computing as a practical strategy for complex optimization in structural engineering and machine learning.