Enhanced Parallel Sine Cosine Algorithm for Constrained and Unconstrained Optimization

The sine cosine algorithm’s main idea is the and cosine-based vacillation outwards or towards best solution. first contribution of this paper proposes an enhanced version SCA algorithm called as ESCA algorithm. supremacy proposed over a set state-of-the-art algorithms in terms solution accuracy convergence speed will be demonstrated by experimental tests. When these are transferred to business sector, they must meet time requirements dependent on industrial process. If temporal not met, efficient them up designing parallel algorithms. second major work design several for efficiently exploiting current multicore processor architectures. First, one-level synchronous asynchronous designed. They have two favors; retain behavior provide excellent performance combining coarse-grained parallelism with fine-grained parallelism. Moreover, scalability further improved employing two-level strategy. Indeed, results suggest that reduce computing time, average, 87.4% 90.8%, respectively, using 12 physical processing cores. extra reductions 91.4%, 93.1%, 94.5% 16, 20, 24 cores, including logical Comparison analysis carried out 30 unconstrained benchmark functions three challenging engineering problems. outcomes show behaves outstandingly well exploration exploitation behaviors, local optima avoidance, toward optimum. overall statistically validated non-parametric statistical tests, namely Friedman, Friedman aligned, Quade constrained optimization; metaheuristic; heuristic algorithm; OpenMP; algorithms; optimization