Evaluation of accuracy of advanced traveler information and commuter behavior in a developing country

Abstract

Heterogeneous traffic in developing countries consists of motorbikes and cars as private vehicles along with various other modes of transport for public and commercial transport. Motorbikes and cars have distinct travel characteristics, having different travel times and owned by commuters from significantly different socioeconomic backgrounds. This study uses a probe motorbike and probe car to collect travel time data for numerous trips covering a total length of 2040 km, on the various roads of Karachi. This travel time data is used to evaluate the accuracy of the only publically available advanced traveler information tool, ‘Google Live Traffic’ (GLT). Results show that there is a significant difference in the accuracies of the information for these two distinct modes of private transport. The overall accuracy of travel times predicted by GLT is 80% (85% for cars and 71% for motorbikes). Trip parameters such as traffic condition and type of road had a statistically significant correlation with the accuracy of the information provided by GLT.The behavior of commuters towards the awareness and utilization of traveler information provided by GLT was evaluated by performing an econometric analysis of selected socioeconomic and demographic parameters. A detailed stated preference survey was conducted from 1200 randomly selected motorbikes riders and car drivers from various locations of Karachi, a metropolitan city with 14.9 million inhabitants from a variety of socioeconomic background. Econometric analysis shows that age, education, income, vehicle type, and occupation have a significant correlation with awareness among commuters, whereas gender is identified as an insignificant factor.