Bi-directional long short-term memory (Bi-LSTM) is an innovative neural network paradigm that is used to predict future occurrences by learning the bi-directional long-term dependencies of time steps and serial data. This paper presents a model that can dynamically predict the pulling speed of a Czochralski (Cz) single-crystal furnace by modeling the time series and environmental parameters. The proposed model is validated using real data from a silicon monocrystal factory. The results show that the proposed model achieves a mean squared error (MSE) of 10.83965, a mean absolute error (MAE) of 2.44357, and goodness of fit (R2) of 0.90085. Thus, the proposed model achieves a significant improvement over existing models. These results verify the validity of modeling the pulling speed of single-crystal furnace devices with a Bi-LSTM model using the time series and environmental parameters. Therefore, the proposed model can serve as a reference for modeling the parameters of such devices.