Hardcl Method for Defining LOS Criteria of Urban Streets

The present study aims at the divination of the speed ranges of level of service (los) categories of urban traffic facilities. free flow speed,congested travel speed,geometric,and surrounding environmental conditions are considered to define los criteria for urban street in indian context. cluster analysis is found to be a powerful tool to delineate los criteria. hard competitive learning method is used to classify large number of speed data obtained using global positioning system. six cluster validation parameters are used to classify the urban streets as well as the los categories. it can be confirmed from the above research work that the los categories for different urban street class are lower than that of the values proposed by hcm 2000 and the average travel speed of los categories expressed in percentage of the free-flow speed is lower than the values mentioned in hcm 2010. © 2017,iran university of science and technology.