Opening windows in coach buses is a practical approach to improving natural ventilation and mitigating infection risk (IR). Due to human behavior and weather conditions, the intermittent window opening strategy (IWOS) is a more common practice than keeping windows constantly open. Despite its prevalence, there are no studies exploring IWOS specifically in vehicles. We employed indoor-outdoor coupled CFD simulations to assess the effects of various IWOS on pathogen-laden droplet (PLD) dispersion and IR in a coach bus that occurred a COVID-19 outbreak in Hunan, China. Results reveal that after ventilating through two skylights for 600–1800 s, opening front and rear windows (FW+RW) or FW with a wind catcher (FW+WCH) for just 40 s can reduce PLD concentration (C_ave) to 5% of its initial level and the intake fraction of the infector’s neighbor (IF_n) drops by 95%. Upon closing FW+RW or FW+WCH, C_ave and IF_n take over 580 s to return to the pre-opening level. Moreover, intermittent FW opening halves C_ave and IF_n within 7 min, but leads to rapid increases upon window closure. Therefore, opening FW+RW and FW+WCH intermittently have pronounced impacts on indoor PLD concentration and are applicable approaches to control respiratory disease transmission in vehicles. According to the inhaled viral dose, it is recommended to open windows when driving time is over 12 minutes to reduce infection risk. In scenarios like epidemiological surveys and risk assessments, where assessing passenger infection risk is vital, some behaviors of opening windows cannot be overlooked and necessitate extra attention.