Digital Twin (DT) technology promises transformational value in manufacturing—especially for plastic moulding MSMEs where downtime, energy consumption, and quality inconsistencies hamper competitiveness. This paper presents a scalable, low-cost DT framework integrating IoT sensors, edge computing, AI-driven analytics, and dashboards to support predictive maintenance and process optimization. We implement and validate the system in a plastic moulding pilot with demonstrable reductions in machine downtime (−45%), energy use (−4%), and defect rate (−2.3%). A regression-based predictive model yields early alerts, enabling maintenance deployment before failures. We discuss workflow, ROI, and alignment with Industry 4.0 and 5.0, and propose an adoption roadmap for low-capex environments. Findings validate the way DTs are able to democratize smart manufacturing for MSMEs.