Additive manufacturing is driving major innovations in many areas such as biomedical engineering. Recent advances have enabled 3D printing of biocompatible materials and cells into complex 3D functional living tissues and organs using bioink. Inkjet-based bioprinting fabricates the tissue and organ constructs by ejecting droplets onto a substrate. Compared with microextrusionbased and laser-Assisted bioprinting, it is very difficult to predict and control the droplet formation process (e.g., droplet velocity and size). To address this issue, this paper presents a new data-driven approach to predict droplet velocity and size in the inkjet-based bioprinting process. An imaging system was used to monitor the droplet formation process. To investigate the effects of excitation voltage, dwell time, and rise time on droplet velocity and droplet size, a full factorial design of experiments was conducted. Two predictive models were developed to predict droplet velocity and droplet size using random forests. The accuracy of the two predictive models was evaluated using the relative error. Experimental results have shown that the predictive models are capable of predicting droplet velocity and size with sufficient accuracy.