Digital twins are key elements in the field of capturing and deploying knowledge and are recognize as a power knowledge management tool. But how can those twins be easily generated and maintained? And moreover, where can digital twins be deployed during the product life cycle? This contribution aims at showing industrial examples of how to deploy digital twins for very efficient experimental design, outperforming classical explorative Design of Experiments approaches. Even more, deploying the twin in real time, they can be used to ensure process robustness and also increase productivity significantly. We will show how the digital twin can directly, and not indirectly via other process variables, optimize productivity. Hence digital twins are an essential tool inside of the digital transformation of biopharmaceutical industry.