<p>Global optimization provides efficient and feasible solutions to complex problems in both theory and practice. Cuckoo search (CS) algorithm exhibits remarkable potential in handling global optimization problems (GOPs), as it mimics the habit of cuckoos to find suitable hosts and breed offspring through efficient search strategies. Existing CS algorithms fail to adequately balance exploitation and exploration when addressing complex GOPs, resulting in premature stagnation and local convergence. To tackle this challenge, we propose a multi-strategy CS algorithm with scheduling mechanism (MSCSSM). By appropriately scheduling various search strategies, MSCSSM focuses on global exploration during the early stages and transitions to local exploitation during the later stages. We first design five search strategies to enhance the algorithm’s flexibility in solving GOPs. Additionally, we conceptualize crowding degree to evaluate the search range of each strategy, enabling the measurement of its bias toward exploitation or exploration. Finally, we develop a multi-strategy scheduling mechanism based on the crowding degree, which adaptively selects appropriate strategies for different search phases. To validate the performance of MSCSSM, a series of comparative experiments between MSCSSM and fifteen excellent algorithms are conducted at CEC 2017. The experimental results confirm that MSCSSM provides competitive performance.</p>