Artificial satellite search: A new metaheuristic algorithm for optimizing truss structure design and project scheduling

Abstract

The Artificial Satellite Search Algorithm (ASSA), a novel physics-based metaheuristic algorithm designed to emulate the dynamic motion of satellites within a search space, is introduced in this study. The ASSA uses satellites as candidate solutions, which dynamically update their positions to navigate toward the optimal solution. The algorithm simulates satellite behavior using medium Earth orbit and low Earth orbit trajectories, facilitating more effective exploration and exploitation of the search space by accounting for the diverse scenario's satellites encounter relative to the Earth over time. In addition, orbit control mechanism and quantum computing technique are incorporated into the ASSA to further enhance the computational efficiency. Two experiments were conducted to assess ASSA performance. First, the performances of ASSA and seven well-known algorithms were benchmarked on thirty benchmark functions and the CEC-2020 test suite. ASSA outperformed all of the comparison algorithms on the Wilcoxon signed-rank test, earned the highest rank (scoring 2.21 and 3.27 on the thirty benchmark and CEC-2020 test suite functions, respectively) on the Friedman test, and solved 27 out of 30 functions with shorter computational times. Second, ASSA was applied to address three engineering problems, achieving the best weight for truss structure optimization and the highest success rates for project scheduling. In these practical engineering applications, ASSA not only exhibited superior performance compared to alternative methods but also required the fewest evaluations of objective functions. The robustness and ease of implementation of the ASSA makes this new algorithm a versatile solution for various numerical optimization challenges.