High passenger density, prolonged exposure, and close interpersonal distance create a high infection risk (IR) in minibuses. While improving natural ventilation induced by turbulent airflows is essential for controlling IR in minibuses, comprehensive studies on its effectiveness are lacking. To address this, we conducted computational fluid dynamics simulations studies coupling indoor–outdoor turbulent airflows to examine the impact of window opening locations, window opening sizes, and initial droplet diameters (dp) on the ventilation airflow and dispersion of pathogen-laden droplets. Results show that the surrounding turbulent flow patterns create higher surface pressure at bus rear than bus front, which is a key factor influencing bus ventilation. When all windows are closed, ventilation is primarily provided by skylights at bus rooftops. Ventilation through only two skylights resulted in an air change rate per hour (ACH) of 17.55 h−1, leading to high IR of passengers. In contrast, fully opening front and rear windows increases ACH by 27.28-fold to 478.79 h−1, significantly reducing IR by 1–2 orders of magnitude compared to skylight ventilation. Expanding window opening sizes can effectively enhance ventilation when both front and rear windows open (attributed to the pumping effect), while is ineffective when only front windows open. To reduce IR in minibuses, we recommend opening multiple windows at the bus front and rear. Even if the total opening area of the front and rear windows is only two-thirds of that of the front window, its ACH is 2.8 times more than only opening front windows.