Traffic Noise Model Research Articles

Statistical Modelling of Traffic Noise Using Smart PLS-SEM In Urban Roads

The objectives of this study are to assess the traffic noise levels in Shah Alam, Selangor, Malaysia, and to propose traffic noise model among influential factors. Five factors and 20 attributes were measured at three locations using a sound level meter and analyzed using the PLS-SEM approach. The results of this study showed that high traffic noise levels on urban roads exceeded the maximum permissible limit of 60 dBA during the day, according to Malaysian guidelines. However, the integrated traffic noise prediction model approach has specific indirect effects: built environment -> climate conditions -> traffic flow.

Road traffic noise prediction model "ASJ RTN-Model 2023" proposed by the Acoustical Society of Japan - Part 1: Framework of the calculation model

The Acoustical Society of Japan (ASJ) has had the Research Committee on Road Traffic Noise to develop a prediction model for road traffic noise for more than 50 years. As a result of the continuous research activities, a new version of model, "ASJ RTN-Model 2023", was published in this April. It is an upgrade version of the previous model, "ASJ RTN-Model 2018", proposed in 2019. In developing the latest version, existing knowledge was widely taken into account, in particular, the sound power levels of road vehicles under acceleration running conditions and calculation methods for sound propagation were improved in wide range. This new model is introduced in three papers with the same main title in this congress. In this paper, the framework of the model, the general procedure of prediction calculation, and the outlines of sound emissions and the calculation equations of sound propagation are presented.

Identification of strategic location for environmental noise control through the application of land use regression (lur) in a neural network platform

The significant causes of rising noise pollution within urban areas lie in extensive transportation networks, industrial operations, and commercial activities. Addressing this complexity, our study employs Land Use Regression (LUR) modelling to assess the intricate relationship between land-use features and environmental noise in urban landscape. The objective of this study is to provide a framework for predicting environmental noise levels in Delhi, India. For which, Lday, Lnight, LAeq, 24h, and Ldn noise descriptors were modelled at daily resolution data from 31 fixed-site monitors. The prediction at first stage for LUR was found to be R2 = 0.53 which was validated using the n-1 method. The model was later optimized using a NN. Output from the NN is included in the GIS environment to identify the hotspot of the noise. Previously identified engineering interventions applicable to urban contexts that were listed from the literature review were allocated as per hotspots because of the modelled output. The study also explored the avenue of using Delhi-specific transport variables developed from Google travel time data for prediction improvement of the noise level. The research contributes to advancement of noise pollution assessment and management methodologies, particularly in situations with no available validated traffic noise models.

Refinement method of equivalent source power level determination for traffic noise in urban noise mapping

For traffic noise prediction, the predictive models of source power level determination commonly in use are the Technical Guidelines for Noise Impact Assessment in China or the Calculation of Road Traffic Noise model in the United Kingdom, which incorporate traffic flow and car speeds. However, the accuracy of these models is often compromised due to urban traffic's complexity, vehicle diversity, and unforeseen events. This study investigates a refinement method based on noise on-site measurement and aforementioned approaches for equivalent source power level determination for traffic noise in urban noise mapping. In the study, onsite measurements are conducted to collect noise and traffic data, and source power level calculated by refinement method is validated using measurement data. Different prediction methods are employed for measurement sites calculations. By comparisons between predictions with noise on-site measurement data, the refinement method proposed in the study demonstrates a significant accuracy enhancement in traffic noise prediction over previous models for determining equivalent source power level. This enhancement highlights the efficacy of adopting noise measurement-based methods for setting traffic noise source power levels, leading to more precise traffic noise predictions in complex urban settings, particularly in areas surrounding urban arterial roads.

The problem of obtaining vehicles speed input for road traffic noise microscopic models: a comparison between field measurements and a stochastic approach

Mitigating environmental noise harmful effects is a relevant goal in the EU "Zero Pollution Action Plan", which targets a 30% reduction in chronic annoyance from transportation noise by 2030. Since road traffic is a major noise source in urban and non-urban settings, much effort has been invested in analyzing and modeling it. To assess noise levels, numerous Road Traffic Noise Models (RTNMs) exist, each requiring multiple inputs like vehicle flows, percentage of heavy vehicles, and average speed. Specifically, when dealing with microscopic models, the single vehicle speeds are needed. To obtain them, on-site measurements offer accurate data but are time consuming and expensive. This paper explores two approaches: an alternative measurement approach, based on video processing and object detection tools, and a stochastic approach, which randomly assigns a speed to each vehicle from a distribution curve, depending on traffic conditions and vehicle category. These methodologies are used to provide input data to a road traffic noise microscopic model, whose results are compared with field measured data. This comparison will allow to get useful insights on the performances of the proposed techniques in providing reliable inputs and their potential applications in different scenarios where single vehicle measured speeds are not available.

Road Traffic Noise Levels Estimations by means of Fully Dynamic and Microscopic Approach Compared to CNOSSOS-EU Model

In the last decades, research has been dedicated to refining models for assessing traffic noise levels, supporting national authorities in addressing associated problems. Progress has shifted from basic statistical models to dynamic ones, typically composed of a Noise Emission Model (NEM) for estimating source sound power levels and a sound propagation model for assessing equivalent noise levels at specific receiver points. These models commonly integrate the average speed of traffic flow as an input variable, placing them in the macroscopic or mesoscopic realm. The CNOSSOS-EU model (reference in Europe) exemplifies this macroscopic approach. However, due to the modular structure of road traffic noise models, a NEM can be coupled with different sound propagation models, facilitating the transition from macroscopic to microscopic assessments. This paper demonstrates the coupling of a NEM with two distinct simplified sound propagation models: CNOSSOS-EU and the sum of Sound Exposure Levels (SEL) models. The latter approach showcases the potential of using individual on-road vehicle speeds as inputs, enabling microscopic noise assessments. Vehicle speed data is extracted from video records, while the goodness of the proposed approaches is tested by comparing the models' estimations with ground truth values recorded on a National Road in Portugal.

Investigating Alternative Representations of Building Structures in Traffic Noise Modeling

This paper presents the results of an investigation of various traffic noise modeling techniques with the Traffic Noise Model (TNM) 2.5 software package, with specific focus on the representation of traffic noise shielding from residences parallel to the freeway. A comprehensive traffic noise field study was undertaken in a neighborhood adjacent to Interstate 270 in Columbus, Ohio. Data collection activities occurred in three “waves” accounting for the pre-clearing, no barrier, and with barrier conditions. Measured noise levels at the study site indicated that there was some noise shielding effect of the residential structures; this effect was most readily perceptible behind the first row of residences. Separate TNM layouts were developed to represent the neighborhood’s residential structures as either a TNM building row object or as separate TNM barrier objects for each structure. The analysis found that modeling each individual residential structure as a separate TNM noise barrier object (single building façade that is closest to the freeway) produces the most accurate modeled noise levels, as compared to the building row representation or simply omitting the shielding from structures altogether. Based on these findings, it is recommended that traffic noise practitioners consider modeling each residential structure in noise-sensitive areas as TNM noise barrier objects.

Toward a dynamic national transportation noise map: Modeling temporal variability of spectral traffic noise emission levels.

The National Transportation Noise Map predicts time-averaged road traffic noise across the continental United States (CONUS) based on annual average daily traffic counts. However, traffic noise can vary greatly with time. This paper outlines a method for predicting nationwide hourly varying source traffic sound emissions called the Vehicular Reduced-Order Observation-based Model (VROOM). The method incorporates three models that predict temporal variability of traffic volume, predict temporal variability of different traffic classes, and use Traffic Noise Model (TNM) 3.0 equations to give traffic noise emission levels based on vehicle numbers and class mix. Location-specific features are used to predict average class mix across CONUS. VROOM then incorporates dynamic traffic class mix data to obtain dynamic traffic class mix. TNM 3.0 equations then give estimated equivalent sound level emission spectra near roads with up to hourly resolution. Important temporal traffic noise characteristics are modeled, including diurnal traffic patterns, rush hours in urban locations, and weekly and yearly variation. Examples of the temporal variability are depicted and possible types of uncertainties are identified. Altogether, VROOM can be used to map national transportation noise with temporal and spectral variability.

Design optimization of floor plan for public housing buildings in Hong Kong with consideration of natural ventilation, noise, and daylighting

Indoor environmental quality plays a crucial role in determining the overall quality of life. This study aims to develop a design optimization approach for the floor plan of public housing buildings with modular flat design in Hong Kong, with focus on enhancing natural ventilation, reducing noise levels, and improve daylighting conditions. The evaluation of these environmental factors was conducted using deep neural network models and a mathematically based Calculation of Road Traffic Noise model. A general floor plan representation was developed for three- and four-winged structures of public housing buildings. An optimization approach utilizing Bayesian optimization was applied to three studied cases: Hung Shing House, Hung Hei House, and Cheung Tai House. The optimization process resulted in an average 41.5 % improvement in average natural ventilation rate. The optimized building shapes effectively served as noise barriers, leading to an average reduction of 20 % in average noise levels. The existing window configurations of each unit type under the modular flat design already provided sufficient daylighting, resulting in only a minor improvement from the optimization process.

Retrospective Cost Saving Comparison Between TNM 2.5 and TNM 3.2 Barrier Analysis Results

In December of 2021, the Federal Highway Administration (FHWA) Traffic Noise Model (TNM) version 3.1 was released to the public for use on traffic noise projects and subsequently updated to version 3.2 in December 2023. However, most or all state Departments of Transportation (DOT) continue to accept and prefer the use of TNM 2.5 (released in 2004). A presentation at NoiseCon 2023 (TNM 2.5 vs TNM 3.1 Noise Level Prediction Comparison by Mukul Pal and Michael Parsons) came to the conclusion that, generally, TNM 2.5 tends to overpredict traffic noise levels in comparison to the TNM 3.1, resulting in additional impacts and subsequent additional mitigation that may not be necessary when analyzed with the updated algorithms. In this paper, the authors determine potential cost savings by migrating from TNM 2.5 to TNM 3.2 by comparing results from previously approved traffic noise projects. By determining the potential cost savings of a migration, DOTs may be better informed about whether or not they should continue to utilize TNM 2.5 or make the transition to TNM 3.2.

Applying reflections in the FHWA Traffic Noise Model

The latest version of the Federal Highway Administration (FHWA) Traffic Noise Model (TNM), version 3.2, was released early 2024 and is encouraged for use to conduct highway noise impact and noise abatement analyses. TNM 3.2 has the advantage of allowing single order reflections in addition to a parallel barrier/wall analysis. This paper presents a process to apply single order reflections considering both their sound contribution and their efficiency in modeling and calculations. The process was applied to a real highway project to evaluate sound level increases due to single reflections and to compare predicted sound levels with and without reflections to measured sound levels. Reflecting surfaces evaluated include noise walls on one side of a highway, building facades facing the highway, and sides of buildings.

Improving Traffic-noise-mitigation Strategies with LiDAR-based 3D Tree-canopy Analysis

The leaves on trees absorb road noise and serve as noise barriers. Tree structures such as tree belts and isolated trees have various methods for absorbing sounds. The depth, surface area, and noise-absorption coefficient of trees contribute to noise absorption. Therefore, this study aims to address this issue of traffic-noise pollution through the use of trees; in particular, by analyzing the noise-absorption coefficient of leaves, the surface area of the leaves, and the depths of the trees. However, the study stresses the need for 3D tree-canopy visualization to identify these factors. To achieve this, the study used LiDAR point clouds to provide accurate data for the convex hull visualizations of canopies. Additionally, a formulated equation for calculating traffic noise after absorption has been suggested by combining the traffic-noise absorption and Henk de Kluijver traffic-noise models. The study also compares the effectiveness of tree belts and isolated trees in reducing noise pollution, concluding that, below a canopy of trees, there is no noise reduction. Finally, the study has demonstrated that the number and sizes of leaves affect noise absorption, showing that noise pollution can be reduced by 1 to 3 dB(A) in the research area by using trees.

Traffic Noise Modeling under Mixed Traffic Condition in Mid-Sized Indian City: A Linear Regression and Neural Network-Based Approach

Noise pollution is a significant concern in urban settings, caused by traffic increases, urban expansion, and industrial activity. The transportation sector is a crucial contributor to overall noise pollution, particularly in India, where different vehicles ply the roads, resulting in highly fluctuating noise levels. Consequently, traffic noise modeling is essential for addressing this severe issue. The present study employs the MLR (Multiple Linear Regression) and Artificial Neural Network (ANN) approach to model and predict traffic-induced noise levels. The ANN approach outperforms the MLR technique. The architecture of the ANN model integrates different vehicle categories and average speeds as input, resulting in precise predictions. Evaluation of the model’s performance reveals an average Root Mean Squared Error (RMSE) of 0.204 and a high Coefficient of determination (R2) value of 0.93, emphasizing its accuracy. Similarly, in the case of MLR model the RMSE for the training and testing dataset are 1.55 and 1.69 dBA, respectively with R2 value of 0.84. Subsequently, sensitivity analysis highlights the substantial impact of 2-wheelers, tractors/trailers, and 3-wheelers on noise predictions. This study contributes valuable insights into noise management, urban planning, and sustainable development. It demonstrates the efficiency of the ANN approach in addressing complex noise pollution challenges, offering a path toward quieter and healthier urban environments.

Traffic noise measurement, mapping, and modeling using soft computing techniques for mid-sized smart Indian city

The present study presents an investigation into the utilization of artificial neural networks (ANN) and multiple linear regression model (MLR) for the prediction of traffic noise levels in various locations of Dhanbad city at varying intervals. Traffic noise indices measurements were carried out using sound pressure level meter (SLM). The prediction of equivalent A-weighted sound level (LAeq) and the sound level exceeding 10 percent of the time (L10) is carried out using various influencing factors such as number of cars, number of 2 wheelers, number of 3 wheelers, number of heavy vehicle, number of medium commercial vehicles, and traffic speed. The findings demonstrate the ANN model's proficiency in comparison of MLR model for providing precise prediction accuracy of traffic noise indices with a R2 of 0.94 for LAeq and 0.91 for L10. Furthermore, the frequency spectrum analysis reveals that high peaks were observed at lower frequencies ranging from 31.5 Hz to 50 Hz, middle frequencies from 500 to 800 Hz and higher frequencies from 3.5 kHz to 5 kHz. The noise maps at varying intervals revealed that most of the locations are having higher noise levels due to increase in vehicular movement. The study emphasizes on the potential significance of the proposed neural network-based prediction model in collaboration with noise mapping as a vital tool for the anticipation of traffic noise levels and the formulation of noise mitigation strategies in context of the mid-sized smart city like Dhanbad, India.

Modeling of Traffic Noise along Urban Arterials in Irbid City of Jordan

Traffic-related noise pollution is a recurring problem in large and medium-sized cities. The objective of this study is to quantify traffic-noise levels along selected urban arterials and model the generated traffic noise based on traffic and pavement characteristics. Urban arterial and collector streets in Irbid city, Jordan, were taken as a case study. The city is considered an example of a medium-sized city. 65 urban arterial and collector sections were selected to achieve the stated objective. For each section, ten noise measurements were taken using a time interval of 2.5 minutes for each observation. The statistical pass-by method was used to measure 650 externalnoise observations. In addition, data on traffic, pavement and section geometric characteristics was obtained through field measurements. The collected traffic characteristics, including traffic flow, percentage of trucks and speed in each direction of travel, were obtained. In addition to the measurement of pavement macrotexture depth, an international roughness index was measured using a smartphone application. Investigation of the collected noise data indicated that urban streets experienced high noise levels, reaching maximum and average values of 82 and 77.2 dB(A), respectively. The results of the analyses showed that an increase in each of the included traffic characteristics resulted in significantly higher noise levels. For example, an increase in speed from 35 to 55 km/h would increase noise by 2.7 dB(A). In addition, the interaction term of roughness index and pavement macrotexture depth was found to increase the generated noise. Finally, the results of the analysis indicated that both multivariate linear -and exponential-regression models are suitable to model the generated traffic noise. Each model explained approximately 54% of the noise variability. Probably, traffic composition and vehicle-power type heterogeneity might reduce the level of explained variability. Keywords: Traffic noise, Urban arterials, Pass-by method, Noise modeling, Jordan.

Coupling Different Road Traffic Noise Models with a Multilinear Regressive Model: A Measurements-Independent Technique for Urban Road Traffic Noise Prediction.

Road traffic noise is a severe environmental hazard, to which a growing number of dwellers are exposed in urban areas. The possibility to accurately assess traffic noise levels in a given area is thus, nowadays, quite important and, on many occasions, compelled by law. Such a procedure can be performed by measurements or by applying predictive Road Traffic Noise Models (RTNMs). Although the first approach is generally preferred, on-field measurement cannot always be easily conducted. RTNMs, on the contrary, use input information (amount of passing vehicles, category, speed, among others), usually collected by sensors, to provide an estimation of noise levels in a specific area. Several RTNMs have been implemented by different national institutions, adapting them to the local traffic conditions. However, the employment of RTNMs proves challenging due to both the lack of input data and the inherent complexity of the models (often composed of a Noise Emission Model-NEM and a sound propagation model). Therefore, this work aims to propose a methodology that allows an easy application of RTNMs, despite the availability of measured data for calibration. Four different NEMs were coupled with a sound propagation model, allowing the computation of equivalent continuous sound pressure levels on a dataset (composed of traffic flows, speeds, and source-receiver distance) randomly generated. Then, a Multilinear Regressive technique was applied to obtain manageable formulas for the models' application. The goodness of the procedure was evaluated on a set of long-term traffic and noise data collected in a French site through several sensors, such as sound level meters, car counters, and speed detectors. Results show that the estimations provided by formulas coming from the Multilinear Regressions are quite close to field measurements (MAE between 1.60 and 2.64 dB(A)), confirming that the resulting models could be employed to forecast noise levels by integrating them into a network of traffic sensors.

Traffic noise modelling at intersections in mid-sized cities: an artificial neural network approach.

Traffic noise has emerged as one major environmental concern, which is causing a severe impact on the health of urban dwellers. This issue becomes more critical near intersections in mid-sized cities due to poor planning and a lack of noise mitigation strategies. Therefore, the current study develops a precise intersection-specific traffic noise model for mid-sized cities to assess the traffic noise level and to investigate the effect of different noise-influencing variables. This study employs artificial neural network (ANN) approach and utilizes 342h of field data collected at nineteen intersections of Kanpur, India, for model development. The sensitivity analysis illustrates that traffic volume, median width, carriageway width, honking, and receiver distance from the intersection stop line have a prominent effect on the traffic noise level. The study reveals that role of noise-influencing variables varies in the proximity of intersections. For instance, a wider median reduces the noise level at intersections, while the noise level increases within a 50-m distance from intersection stop line. In summary, the present study findings offer valuable insights, providing a foundation for developing an effective managerial action plan to combat traffic noise at intersections in mid-sized cities.

Random effect generalized linear model-based predictive modelling of traffic noise.

Noise pollution is one of the negative consequences of growth and development in cities. Traffic noise pollution due to traffic growth is the main aspect that worsens city quality of life. Therefore, research around the world is being conducted to manage and reduce traffic noise. A number of traffic noise prediction models have been proposed employing fixed effect modelling approach considering each observation as independent; however, observations may have spatial and temporal correlations and unobserved heterogeneity. Random effect models overcome these problems. This study attempts to develop a random effect generalized linear model (REGLM) along with a machine learning random forest (RF) model to validate the results, concerning the parameters related to road, traffic and environmental conditions. Models were developed based on the experimental quantities in Delhi in year 2022-2023. Both the models performed comparably well in terms of coefficient of determination. Random forest models with R2= 0.75, whereas random effect generalized linear model had an R2= 0.70. REGLM model has the ability to quantify the effects of explanatory variables over traffic noise pollution and will be more helpful in prioritizing of resources and chalking out control strategies.

Comparison of Analytical and Machine Learning Models in Traffic Noise Modeling and Predictions

Comparison of Analytical and Machine Learning Models in Traffic Noise Modeling and Predictions

Development of Urban Traffic Noise Model for a Mid-Sized City: A Case Study of Kanpur

Development of Urban Traffic Noise Model for a Mid-Sized City: A Case Study of Kanpur