Cyber Physical Production Systems (CPPS) depend significantly on high-quality sensor data to function optimally, make decisions in real-time, and perform predictive maintenance inter alia. Nevertheless, the quality of sensor data in industrial settings is often affected by various factors such as environmental interference, hardware wear and tear, calibration drift, and intricate system interactions. This study introduces innovative methods to improve sensor data quality in CPPS through systematic digitalization strategies. By employing a use case methodology, we explore real-world production scenarios to pinpoint common data quality challenges and devise specific solutions. Our strategy integrates signal processing techniques, algorithms for detecting anomalies to establish robust frameworks for data validation and correction. The proposed methods offer practical, scalable solutions that can be adapted to various production environments, thereby enhancing the reliability and efficiency of cyber physical manufacturing systems. To illustrate the feasibility of our approach, we utilise the case study of a test bed.