<p>The rapid growth of Internet of Things (IoT) applications has rendered cloud-centric data processing increasingly time-consuming, energy-intensive, and costly. Fog computing offers a promising solution by enabling computation closer to the network edge. However, existing service placement strategies often suffer from an imbalance between exploration and exploitation, as well as slow convergence. This paper proposes an improved opposition-based chaotic sine cosine algorithm (I-OppoCSCA) to address these challenges in fog computing environments. The algorithm enhances the standard SCA through two key innovations: (1) a chaotic opposition-based population initialization method that improves the quality of initial solutions and accelerates convergence, and (2) an enhanced solution update mechanism incorporating the optimal neighborhood method, quadratic interpolation, and quasi-opposition learning to balance global exploration with local exploitation and prevent premature convergence. The efficacy of I-OppoCSCA is evaluated using the IEEE CEC2019 standard benchmark functions to assess its search behavior, followed by practical evaluation using the iFogSim simulation toolkit under realistic fog computing scenarios. Experimental results demonstrate that the proposed approach significantly outperforms state-of-the-art algorithms, achieving up to 18% lower energy consumption, 19% reduced service cost, and 51% faster service time. Statistical significance is validated through Wilcoxon rank-sum tests and convergence analysis, confirming the robustness and reliability of the algorithm. These results highlight the effectiveness of I-OppoCSCA as a high-performance optimization strategy for efficient IoT application placement in fog computing.</p>