Adaptive dimensional search: A new metaheuristic algorithm for discrete truss sizing optimization

• A new metaheuristic called adaptive dimensional search (ADS) is proposed.
• The ADS updates search dimensional parameter at every iteration.
• Several alternative stagnation-control strategies are integrated with the ADS.
• Capabilities and potentials of ADS in structural optimization are investigated.
• Computational efficiency of the ADS is verified thorough numerical examples.

In the present study a new metaheuristic algorithm called adaptive dimensional search (ADS) is proposed for discrete truss sizing optimization problems. The robustness of the ADS lies in the idea of updating search dimensionality ratio (SDR) parameter online during the search for a rapid and reliable convergence towards the optimum. In addition, several alternative stagnation-control strategies are integrated with the algorithm to escape from local optima, in which a limited uphill (non-improving) move is permitted when a stagnation state is detected in the course of optimization. Besides a remarkable computational efficiency, the ease of implementation and capability of locating promising solutions for challenging instances of practical design optimization are amongst the remarkable features of the proposed algorithm. The efficiency of the ADS is investigated and verified using two benchmark examples as well as three real-world problems of discrete sizing truss optimization. A comparison of the numerical results obtained using the ADS with those of other metaheuristic techniques indicates that the proposed algorithm is capable of locating improved solutions using much lesser computational effort.