Application of regression models to assess bus interior noise in Dhaka City

Abstract

ABSTRACTThe interior noise level of buses on 29 different routes in Dhaka city was assessed and multiple linear regression was used to attribute noise metrics to various bus features (bus capacity, age and heat insulation) and trip characteristics (trip length, velocity and travel time). Large variabilities were observed in the noise indices over the different bus routes (L10 : 79–90dBA, L50 : 72.9–82.2dBA, L90 : 65.8–77.3dBA). 15% of the route-segment Leq was found to be above 85dBA, the prescribed exposure threshold set by OSHA. Most noise metrics showed significantly positive association (p < 0.05) with trip velocity, bus size and bus age. This study shows that regression models can be applied to pin-point the factors affecting interior bus noise. There is an urgent need to improve the quality of bus service in Dhaka city for the sake of the bus drivers and passengers.