<p>Efficient application placement is essential for maximizing resource usage and improving system performance in the constantly evolving IoT-Fog-Cloud environment. Trajectory-aware proactive service placement in fog computing provides notable benefits by considering the movement patterns of users and devices. This technique guarantees that services are consistently deployed near mobile consumers, hence reducing latency and enhancing the quality of service (QoS). This work introduces a novel approach that utilizes Long Short-Term Memory (LSTM) models to forecast the movement trajectories of mobile Internet of Things (IoT) users using real-world data. Expanding on this capacity to predict, a heuristic approach has been proposed that strategically positions applications near mobile users and fog nodes during specified time periods. The trajectory-aware placement technique seeks to decrease latency, energy consumption, and cost in fog computing environments. The effectiveness of the proposed strategy is assessed by doing thorough experimental analysis and comparing its performance to established methodologies such as Kalman filter and random fog node allocation. Substantial improvements in latency, energy usage, and cost-efficiency are demonstrated by the proposed solution.</p>