Silicon (Si) bulk single crystals are mostly grown by the Czochralski (Cz) technique. To obtain highquality crystals by this technique, the understanding and precise control of transport phenomena occurring in the Si melt during growth are essential. Numerical simulation and artificial intelligent techniques such as Neural Networks (NNs) and Physics Informed Neural Networks (PINNs) provide a very powerful tool for this purpose. Each technique has its own advantages and disadvantages. Numerical simulation requires a long computing time for large scale computations. NNs produce results in a very short time but may not always satisfy the associated governing equations of the system. PINNs on the other hand give results as fast as NNs while satisfying the governing equations. Thus, in this study we developed the needed digital twin of the Cz-Si growth process by applying PINNs.