Vehicle emission test big data to infer active vehicle ownership as an input to trip generation model

Abstract

With the rapid increment in the demand for transportation, accurately predicting future transport demand has been a critical aspect in developing countries. The trip generation model is the first step of the traditional Travel Demand Forecasting Model which provides the number of trips generated from each zone using socio-economic characteristics as inputs. Vehicle ownership, one of the pivotal inputs to develop the trip generation model, is considered in this study. Vehicle registration data obtained from the Department of Motor Traffic (DMT) may not accurately reflect the number of vehicles actively on the road. Thus, this study considered the Vehicle Emission Test (VET) data as an alternate live stream of data. The objective of this study is to develop a methodology to infer the active vehicle ownership for each Travel Analysis Zone (TAZ) by developing a distance matrix through the QGIS software. The study reveals the number of active vehicles is much less (1/3) than what was registered at DMT in the Western Province in 2019 (the year before the pandemic) highlighting the need for further investigations