Failure to obtain timely fire services will lead to thermal runaway phenomenon, which will seriously threaten the lives of people and property safety. In this study, a dynamic method was proposed to estimate fire service accessibility based on real-time traffic. To validate the method, a total of 110,055 POls extracted from online maps and existing 12 fire stations in Yiyang were used as fire service demand points and supply points. Travel times of fire trucks from supply points to demand points were calculated based on real-time traffic, and a total of 550,275 valid samples were obtained. The results showed that the ≥10 min fire service coverage rate in Yiyang is only 71.60 %. Based on the distribution area of demand points ≥10 min, the locations of three optimization solutions of new fire stations are proposed. By calculating the total average time (TAT) of the three optimization solutions of proposed new fire stations, it is concluded that TAT decreases by 743.99 s, 99.72 s, 92.59 s, respectively. From perspectives of safety and economy, the construction of the 10 fire stations of optimized solution I will effectively improve the regional fire service level.