One of the primary goals of optimization approaches is to strike a balance between exploitation and exploration strategies, thereby enhancing the efficiency of the search process. To improve this balance, considerable research efforts have been directed towards refining these strategies. This paper introduces a novel exploration approach for the Black Widow Optimization (BWO) algorithm, termed Improved BWO (IBWO), aimed at achieving a robust equilibrium between global and local search strategies. The proposed approach tracks and remembers the effective research areas during the research iteration and uses them to direct the subsequent research process toward the most promising areas of the search space. Consequently, this method facilitates convergence towards optimal global solutions, leading to the generation of higher-quality solutions. To evaluate its performance, IBWO is compared with five optimization techniques, including BWO, GA, PSO, ABC, and BBO, across 39 benchmark functions. Simulation results demonstrate that IBWO consistently maintains precision in performance, achieving superior fitness values in 87.2%, 74.4%, and 69.2% of total trials across three distinct simulation settings. These outcomes underscore the efficacy of IBWO in effectively leveraging prior search space information to enhance the balance between exploitation and exploration capabilities. The proposed IBWO has broad applicability, addressing real-world optimization challenges in pilgrim crowd management and transportation during Hajj, supply chain logistics, and energy distribution optimization.

Teaching–Learning-Based Optimization (TLBO) is recently being used as a new, reliable, accurate and robust optimization technique scheme for global optimization over continuous spaces. This paper presents an improved variant of TLBO algorithm, called Improved Teaching–Learning-Based Optimization (ITLBO). A performance comparison of the proposed method is provided against the original TLBO and some other algorithms. The improved TLBO algorithm shows a marked improvement in performance over the traditional TLBO on several benchmark optimization problems.