This study investigates the application of Edge-AI in manufacturing plants and robots to speed up decision-making and improve operational performance. Using four machine learning algorithms—Decision Tree, Support Vector Machine (SVM), Artificial Neural Networks (ANN), and Random Forest—at the edge, this study pursues improving real-time processing ability, lowering latency, and facilitating predictive maintenance for manufacturing systems. The outcomes show a phenomenal improvement in KPIs, with the Edge-AI models achieving an average accuracy of 92%, compared to 85% accuracy achieved by conventional cloud-based systems. In addition, the Edge-AI solution minimized system downtime by 30%, improved production planning by 25%, and lowered total resource consumption by 15%. These outcomes show the potential of Edge-AI to transform industrial processes, offering quicker, more efficient, and less costly solutions. The research advocates the merits of local processing of data, which de-emphasizes central-server dependency and facilitates more scalable and dynamic systems. As businesses transition to Industry 4.0, this research identifies the prospective value of Edge-AI in developing intelligent, more efficient factory spaces that can modify to dynamic needs and facilitate sustainable processes.