This review systematically integrates recent advancements in path optimization algorithms for the automated layout of mechanical, electrical, and plumbing (MEP) systems within complex building environments. A hybrid optimization framework is introduced, combining the global search capability of Ant Colony Optimization (ACO) with the local refinement efficiency of the A* algorithm, enhanced by dynamic weight adjustment and context-aware mechanisms. Simulation experiments based on a hospital BIM model demonstrate that the proposed approach improves design efficiency by approximately 25–35% and reduces conflict incidence by around 40%. The framework further incorporates Building Information Modeling (BIM) and real-time clash detection enabled by IoT devices to enable scalable, multi-objective optimization in high-density spatial configurations. The potential of generative artificial intelligence—such as Generative Adversarial Networks (GANs) and diffusion models—is also explored for generating initial pipeline layouts and enhancing spatial adaptability. To support low-carbon building initiatives, the framework is adaptable to LEED-compliant sustainable MEP design practices. Despite notable progress, challenges remain in algorithmic scalability, dynamic constraint modeling, and multi-objective trade-offs. This review identifies research gaps in diameter-aware layout optimization and BIM-driven multi-scale generative modeling, and outlines future directions toward intelligent, high-performance MEP system design in future sustainable buildings.