Estimating Vehicles Fuel Consumption and Emission Using COPERT Model in Apache Spark

Abstract

The prediction and estimation of fuel consumption and emission are of immense importance as they effect human health and urban sustainability. This study will help the concerned departments regarding environmental pollution generated by traffic so they can act accordingly and reduce the which consequently effect human health. In this study we are estimating vehicle fuel consumption and CO emission of vehicle which are travelling in the major cities of Pakistan. Computer Programme to Calculate Emission from Road Transport (COPERT) model is used for estimating fuel consumption in which GPS data trajectories are divided into three main categories namely mobility activity, stationary activity with engine on and stationary activity with engine off. Apache Spark is used in this study for performance gains. The results shows that the vehicle covers 10 km distance in 25 minutes with consumption of fuel is 55 g/km during peak hours while in Lahore and Karachi the consumption of fuel is 85 g/km and 90 g/km respectively with covering distance of 7.5 and 21 km in 19 and 59 minutes, respectively. The results shows that situation of traffic on Karachi roads is same during peak and off peak hours with same traffic while Lahore and Islamabad shows difference of traffic flow on the roads during peak and off peak hours.