Abstract: In order to better understand the impact factors of the traffic operational conditions on urban arterials, this study used the floating car data (FCD) from taxis to investigate the average speed, speed variance on 176 arterials segments in Shanghai. Multiple linear regression was used to analyze the effect of various rad and environment features including road geometry, access points, intersection canalization, volume, region and land use on speed and speed variance. The results indicated that road and environment features (e.g., length of segment, curvature degree, median separator, number of lanes, left turn or right turn protection, land use and number of access points) would significantly influence the operational conditions on the urban arterial segments.