This paper investigates the effectiveness of two restart methods, local and global restarts, in enhancing the performance of the quantum-inspired evolutionary algorithm (QiEA) to solve optimization problems. The study compares the result of these modifications by testing their performance against the widely recognized and challenging IEEE CEC 2022 benchmark functions. The objective is to analyze the impact of the restart methods on the convergence and optimization outcomes of the QiEA. Through experimental evaluation, it was sought to determine whether the inclusion of restart methods in the QiEA framework would lead to improved convergence toward the optimal solution. Also, a novel Greedy Local Restart algorithm is proposed. The results demonstrate that the QiEA variants consistently outperform the basic QiEA, achieving superior optimization results in terms of convergence and accuracy. Restart approaches are thus observed as an indispensable asset in handling complex optimization problems, greatly enhancing the effectiveness and applicability of evolutionary algorithms.