This paper presents a framework for Cyber-Physical Production Systems (CPPS) that integrates digital twins with scheduling-based optimization modules, significantly improving production planning, control, and responsiveness. As digital transformation and Industry 4.0 technologies advance, modern manufacturing requires intelligent solutions capable of real-time decision-making and dynamic adaptation. Digital twins—virtual representations of physical systems—enable continuous monitoring and simulation of production environments, providing critical data to support decision-making. At the core of the proposed architecture are optimization modules based on advanced scheduling algorithms that play a key role in ensuring efficient resource allocation, minimizing downtime, and synchronizing production flows. These modules interpret real-time data generated by the digital twins and apply optimization techniques to propose and implement the most effective scheduling strategies. The framework supports a variety of scheduling strategies, including robust, partial, and full rescheduling approaches, to ensure optimal system performance under changing conditions. By leveraging predictive capabilities, manufacturers can anticipate disruptions, evaluate alternative production scenarios, and respond quickly to changing conditions. Integrating these scheduling-based optimization modules with digital twins creates a powerful feedback loop that continuously refines production processes. This approach provides a scalable and practical path toward more flexible, efficient, and resilient manufacturing systems in line with Industry 4.0 objectives.