In diverse research domains, optimization problems remain a pivotal topic of discussion. Conventional optimization algorithms, constrained by fixed parameters and structural determinism, often encounter limitations in locating global optima, as they tend to converge on local solutions. Metaheuristic algorithms, characterized by their flexibility, have recently garnered significant attention in academic research. Among these, the Arithmetic Optimization Algorithm (AOA), leveraging the four basic arithmetic operations, exhibits remarkable search capabilities. However, the AOA is susceptible to reliance on initial solutions and struggles to maintain population diversity in later stages. To address these shortcomings, this paper presents an improved version of multi-strategy, termed multi-strategy combined with improved algorithm optimization algorithm (MCWIAOA), which integrates a tent map initialization strategy and an adaptive coefficient of a sine control factor to augment both global and local search efficacies. Comparative evaluations against prevalent metaheuristic algorithms reveal MCWIAOA’s notable advantages in optimizing multiple test functions, thereby validating its efficacy and practical applicability.