Empirical Methods of Capacity Estimation of Urban Roads
Determination of road capacity is a major issue for transport planners. Capacity is defined as the maximum number of vehicles that can be accommodated per unit time under given condition of occurrence. Capacity studies for heterogeneous traffic situations are very complex and only limited studies undertaken. Here again there are several methods of estimation of capacity. However the major types of estimation can be classified under two broad categories as Direct Empirical Methods and Indirect Empirical (Simulation) Methods. Because of the complexity of the heterogeneous, high volume traffic on Indian urban roads, it is appropriate to model the flow parameters and adapt direct empirical methods for estimation of capacity. In this paper an attempt is made to study the fundamental details of traffic flow and evaluate the capacity of urban mid block section, particularly for a two lane divided cross section. The traffic data at ten locations in Chennai city was collected through video graphic survey. Detailed extraction of traffic headway (inter arrival time), volume and speed were made for every 5 minute time interval, covering both the peak and non peak period. Using the fundamental parameters, capacities of sections were evaluated by three methods, namely Headway method, Observed volume method and Fundamental diagram method. Detailed analysis of the arrival pattern of vehicles was made and appropriate distribution identified to get the mean headway. The Selected Maxima method assumes that the capacity state is reached during the survey period. As against the normal time slice of 15 minutes, a 5 minute time slice was considered for identifying the peak flow rate. In the third approach, a plot was made between traffic flow in PCU and the speed, to compare the relationship. Since the data was scattered, the enveloping curve technique was adopted, even though the usual pattern of speed flow curve is parabolic. The average capacity of a standardized two lane 7m) urban mid
