User Perception of Automobile Level of Service: Tracking Traffic with GPS Enabled Mobile Phones

The Highway Capacity Manual 2010 (HCM 2010) specifies the importance of user perception in evaluating the Automobile Level of Service (ALOS). Hence, the objective of this study is set to develop a unified methodology for quantifying the ALOS on the divided urban corridors, based on the automobile user perception. To study the behaviour of automobiles at different flow conditions, speed profiles of test vehicles were collected with the in-vehicle Global Positioning System (GPS) enabled mobile phones. The ‘Speed Tracker’ application was used to record the travel data in every second along with the location coordinates. To have a wide variety of travel conditions, four divided urban corridors of length varying from 2.9 to 3.8 km were identified in the state of Kerala as the study stretches. The segmental analysis was carried out for studying the speed variation behaviour of the vehicles with corresponding flow values. The Acceleration Noise (AN) and the speed ratio (SR) were found to be the most significant measures for defining ALOS. Non-linear regression analysis was carried out to model these measures of effectiveness. k-means and fuzzy c-means (FCM) clustering algorithms were used to obtain the threshold values for ALOS. Silhouette coefficients were calculated for validating the cluster results, and the results showed that k-means algorithm is giving better results compared to FCM. This method can be used for assessing the quality of four-lane divided urban corridors incorporating the user perception.