Traffic flow on an urban road network is temporal in nature. Knowledge of within-day traffic flow variation is important for evaluatingalternative measures oftraffic flow improvement on urbantransportation network. Traditionally, a within-day traffic flow variation is classified as peak and off-peak periods, and is typically interpreted from graphical representation of the traffic flow patterns. But,in case of an urban form, where land-use development of various opportunities is quiet spatially varied, the peak and off-peak periods of traffic flow are not found well-defined in variation profile. In such case, a more scientific approach is required to identifythese time periods. In the present work, an attempt has been taken to classify the within-day variation in traffic flow by considering inherent temporally continuous nature of traffic flow data. The classification has been carried out using K-means clustering technique. The work has been demonstrated by considering the multi-modal urban transportation network of Bhubaneswar city, India.
Sai Kiran Annam, Bhargab Maitra and Debasis Basu
Indian Institute of Technology
General, Urban Mobility