Abstract
It is important to understand the factors affecting speed in a link to design a better roadway. The aim of this study was to identify the extend speed is influenced by geometric design elements, traffic characteristics and roadway characteristics, particularly in urban arterial streets. This study is distinct from previous research that it analysis how speed is affected in heterogeneous traffic conditions. In this study second-by-second speed data is collected using Global Positioning System (GPS). Initial analysis showed that more than 50% of time in a trip spent at the speed below 25 km/h. Time spent in both acceleration and deceleration mode was about 65%, which explains the frequent variation of speed in a trip. A multiple linear regression model was constructed to understand the causes of variation in speed. The effect of traffic characteristics and roadside characteristics has significant influence over speed when compared to geometric characteristics. Among traffic characteristics, total volume and time spent in creep mode have significant influence over speed. In roadside characteristics, both commercial and pedestrian activities were negatively correlated. Followed by lane width, pedestrian interference, stop duration and commercial landuse are significant variables. These variables explained about 48% of variations in speed. Variables such as grade level, share of slow moving vehicles, time of the day and on-street parking didnâ??t show up in the model. However, they have shown the sign of influence on speed.