The objective of this study is to develop an empirical traffic noise prediction model under interrupted traffic flow conditions using two analytical the approaches, the first being the acceleration lane approach and second being the deceleration approach. The urban road network of Bangalore city has been selected as the study area. Sixteen locations are chosen in major traffic junctions of the study area. The traffic noise data collected from the study locations were analyzed separately for both acceleration and deceleration lanes when vehicles leave an intersection on a green traffic light and come to a stop on red traffic light. Based on the study, a regression noise prediction model has been developed for both acceleration and deceleration lanes.

The environmental quality of our cities is gradually degrading by an incessant growth in the number of vehicles and the ever expanding road network, resulting in the increase of road traffi c noise. Managing road traffi c noise is a challenging task for environmental managers and urban planners. Urban planners often have to rely on road traffi c noise prediction models for their assessment. A critical review of various traffi c noise studies and the number of traffi c noise prediction models cited in literature reveals that they describe the temporal and spatial distribution of traffi c noise. Most of these models are either deterministic or statistical in nature. This article presents a critical review of some of these models.

The main objective of this study is to develop a noise prediction model under uninterrupted traffi c fl ow conditions. In this study, Bangalore city in Karnataka, India, was selected as the study area. The study locations are so chosen as to represent the different zones within an urban area like residential zone, commercial zone, silent zone and heavy traffi c zone. Traffi c noise was measured using the L eq index with an A-weighted scale of decibel unit for a 1-hour period at each study location. Based on fi eld observed traffi c data, a multiple regression noise prediction model was developed by considering all major causative factors. In the process of model development, a mean standard error of 2.32 dB(A) with r 2 value of 0.82 was observed. The validation of the model was done by collecting traffi c data from Mysore city in Karnataka, India. The results of the model validation indicated that the model is accurate to 2.6 dB(A) with r 2 value of 0.78. Statistical analysis was also done using the paired t-test technique on predicted and observed noise levels. The results indicated that the t-statistical value of the model is less than the t-critical value. This means that the values predicted by the model fi t signifi cantly with the fi eld observed ones and that the independent variables used in the model provide a better explanation of the dependent variable (L eq). The model developed in this study was also compared with the Federal Highway Administration (FHWA) Traffi c Noise Model from USA and the prediction results indicated that the values obtained from the present model are in good agreement with the fi eld observed values than the FHWA model. Therefore, the present model can be used for managing urban road traffi c noise in the Indian context.

31. Government of UK (n.d.), Retrieved from https://www.gov.uk/ driving-eyesight-rules 32. Optometry Victoria Strategic Plan 2015-18 (n.d.). Optometry Australia, Retrieved from www.optometry.org.au; http://www. optometrists.asn.au/vic/practiceinfo/driving-vision-standards.aspx 33. Verma, A., Velmurugan, S. and Chakrabarty, N., Recommendations for driver licensing and traffic law enforcement aiming to improve road safety.