Article

Statistical Road-Traffic Noise Mapping based on Elementary

Urban Forms in Two Cities of South Korea
Phillip Kim 1, Hunjae Ryu 2, Jong-June Jeon 3 and Seo Il Chang 4,*

1 Department of Energy and Environmental System Engineering, University of Seoul, Seoul 02504, Korea;
rlavlfflq154@uos.ac.kr
2 Korean Educational Environments Protection Agency, Chungcheongbuk-do 28166, Korea;

pgryuno1@schoolkeepa.or.kr
3 Department of Statistics & Graduate School, Department of Urban Big Data Convergence,

University of Seoul, Seoul 02504, Korea; jj.jeon@uos.ac.kr

4 School of Environmental Engineering & Graduate School, Department of Urban Big Data Convergence,

University of Seoul, Seoul 02504, Korea; schang@uos.ac.kr

* Correspondence: schang@uos.ac.kr; Tel.: +82-2-6490-2865

Abstract: Statistical models that can generate a road-traffic noise map for a city or area where only
elementary urban design factors are determined, and where no concrete urban morphology, includ-
ing buildings and roads, is given, can provide basic but essential information for developing a quiet
and sustainable city. Long-term cost-effective measures for a quiet urban area can be considered at
early city planning stages by using the statistical road-traffic noise map. An artificial neural network
(ANN) and an ordinary least squares (OLS) model were developed by utilizing data on urban form
indicators, based on a 3D urban model and road-traffic noise levels from a normal noise map of city
A (Gwangju). The developed ANN and OLS models were applied to city B (Cheongju), and the
Citation: Kim, P.; Ryu, H.; Jeon, J.-J.; resultant statistical noise map of city B was compared to an existing normal road-traffic noise map
Chang, S.I. Statistical Road-Traffic of city B. The urban form indicators that showed multi-collinearity were excluded by the OLS model,
Noise Mapping based on and among the remaining urban forms, road-related urban form indicators such as traffic volume
Elementary Urban Forms in Two and road area density were found to be important variables to predict the road-traffic noise level
Cities of South Korea. 2021, 13, 2365. and to design a quiet city. Comparisons of the statistical ANN and OLS noise maps with the normal
https://doi.org/10.3390/su13042365 noise map showed that the OLS model tends to under-estimate road-traffic noise levels, and the
ANN model tends to over-estimate them.
Academic Editor: Guillermo Rey
Gozalo
Keywords: road-traffic noise; urban forms; artificial neural network; statistical noise map
Received: 3 January 2021
Accepted: 16 February 2021
Published: 22 February 2021

Publisher’s Note: MDPI stays neu-

1. Introduction
tral with regard to jurisdictional Road-, railway-, and air-traffic networks are essential for the sustainable develop-
claims in published maps and insti- ment of a city [1,2], but transport networks can also generate undesirable by-products
tutional affiliations. including fine dust and noise [3], affecting the quality of life and health of inhabitants
[4,5]. The European Union (EU) recommends generating strategic noise maps for efficient
management of traffic noise every five years, based on the END (European Noise Di-
rective) 2002/49/EC [6]. The targets of noise maps include major roads [7], railways [8],
Copyright: © 2021 by the authors.
airports [9,10], and ports [11]. Similar to the EU, the Republic of Korea (South Korea) man-
Licensee MDPI, Basel, Switzerland.
dates noise mapping in large cities, and updates every five years [12]. The noise maps are
This article is an open access article
used to establish action plans to reduce the noise exposure of inhabitants in cities [13].
distributed under the terms and con-
In South Korea, the road-traffic network is the main transportation network [14], and
ditions of the Creative Commons At-
inevitably, high population density causes a lot of residents living near the road-traffic
tribution (CC BY) license (http://cre-
ativecommons.org/licenses/by/4.0/).
infrastructure to be exposed to high levels of road-traffic noise. Therefore, road-traffic
noise is one of the most significant adverse health impacts experienced by the people of
South Korea [15]. Various methods are used to reduce road-traffic noise, including porous

Sustainability 2021, 13, 2365. https://doi.org/10.3390/su13042365 www.mdpi.com/journal/sustainability

Sustainability 2021, 13, 2365 2 of 17

asphalt, reduced speed limits, and heavy vehicle detours. As a measure for the propaga-
tion path, noise barrier walls and tunnels are used at a relatively high cost. The shapes,
material, and specifications of walls and tunnels determine the noise reduction effect. By
utilizing a noise map, if it exists, it is possible to find a combination of noise reduction
methods and specifications that lead to the best results.
Although the noise map is an efficient tool, if it does not exist already, its generation
over a wide area requires abundant GIS (geographic information system) data on roads,
buildings, traffic, and so forth, as well as considerable hardware, software, and financial
resources. At early urban planning stages, when there is only minimal information on the
population, road density, and ground space index, a normal noise map cannot be gener-
ated because no concrete data for buildings, roads, and traffic exist.
Therefore, when there is no normal noise map available, a statistical relationship be-
tween the elementary urban forms and road-traffic noise can provide useful information
for city planning, and a noise map with a low resolution can be generated from the statis-
tical model. The relationship between the road-traffic noise level and urban forms has
been studied using various statistical methods, such as land-use regression (LUR) models
[16–22], regression models [23,24], and the spatial statistical model [25]. In these cases, the
road-traffic noise levels for the statistical analyses were measured [16–22,26] or simulated
[23–25,27–30].
Aguilera et al. [16] applied LUR modeling to assess the long-term intraurban spatial
variability of road-traffic noise in three European cities. The authors used short-term noise
measurements for the three cities to develop two LUR models: one GIS-only and one Best.
The GIS-only model considered only predictor variables derived from GIS data, and the
Best model considered GIS data and variables collected while visiting the measurement
sites. The authors’ findings indicate that LUR modeling, with accurate GIS source data,
can be a promising tool for noise exposure assessment. Han et al. [17] analyzed the rela-
tionships between environmental noise and urban morphology. Socioeconomic data and
landscape data were selected as representative of the urban morphology. The environ-
mental noise, including regional environmental noise and traffic, was significantly influ-
enced by urban morphology parameters. Xie et al. [18] applied the LUR model to depict
urban environmental noise. Six land-use types and the length of three road types were
used as independent variables. The model was fitted with linear and nonlinear structures,
but there was no significant difference. Various characteristics of the area, such as traffic,
weather, and land use, were used in the LUR modeling to predict the noise level [19–22].
These studies developed the LUR models based on the noise level measured at a specific
point.
Salomons and Berghauser Pont [23] presented relationships between the spatial dis-
tribution of traffic noise in a city and traffic volume, urban density, and form. The re-
searchers investigated this by means of numeric calculations for two cities and various
idealized urban fabrics, and in the two cities, the average sound level in an urban area
decreased with increasing population and building density. Silva et al. [24] addressed the
problems of the urban environment in the form of interactions between urban forms and
urban road-traffic noise monitored using urban form indicators and models for the effects
of noise propagation forms. The authors used the compactness index (CI), porosity index
(ROS), and complexity perimeter index (fractal) of urban areas as urban form indices and
performed a correlation analysis between the average facade noise levels and each index.
The correlations were positive between urban road-traffic noise and CI and ROS but neg-
ative between urban road-traffic noise and fractals.
Ryu et al. [25] developed spatial statistical models to predict road-traffic noise in cit-
ies. Specifically, the authors used a spatial autoregressive model (SAR) and a spatial error
model (SEM), in addition to a classical linear model, to account for spatial dependence in
determining the road-traffic noise levels in relation to urban form indicators. The SAR
model had better statistical properties, although it had slightly less explanatory power
Sustainability 2021, 13, 2365 3 of 17

than that of the SEM model. The results showed that ground space index, floor space in-
dex, traffic volume, speed, road area density, and fraction of industrial area had statisti-
cally significant direct and indirect impacts on road-traffic noise levels. In addition, vari-
ous statistical analyses between noise and characteristics of the urban area, such as urban
forms and morphologies, have been performed in previous studies [26–30].
Artificial neural network (ANN) models have been used to predict road-traffic noise
at specific points with traffic flow characteristics (e.g., traffic volume, speed, and heavy
vehicle ratio). Kumar et al. [31] fitted an ANN model with measured road-traffic noise
levels, traffic volume, speed, and heavy vehicle ratios, and the predicted noise levels using
the ANN model were closer to the measured noise levels than were the levels predicted
with the ordinary least squares (OLS) model. Garg et al. [32] predicted road-traffic noise
levels (Leq, L10) with a total of eight explanatory variables: the volumes of cars, two-
wheelers, medium commercial vehicles, three-wheelers, buses, and trucks and the speeds
of heavy and light vehicles. The predicted noise levels using the ANN model were closer
to the measured noise levels than the predictions from the multiple linear regression anal-
ysis. Hamad et al. [33] fitted an ANN model predicting road-traffic noise levels in a city
with a hot climate. The ground temperature and acoustical factors, such as distance from
noise source (road), traffic volume, and speed, were used. The database was applied to a
basic statistical traffic noise model, the Ontario road noise analysis method for the envi-
ronment and transportation (ORNAMENT), and the ANN model. The predicted road-
traffic noise level from the ANN model was closer than the other models to the measured
noise level. Similar to these studies, ANN models that predict the road-traffic noise level
based on measured traffic data were developed to replace traditional road-traffic noise
prediction [34–36].
In this study, ANN and OLS (ordinary least squares) models were developed simul-
taneously to reflect the relationship between urban form indicators and road-traffic noise
levels. The road-traffic noise levels were obtained from a simulated normal noise map of
the city of Gwangju, and the urban form indicators were estimated from the 3D city model
and relevant GIS data [37]. The developed ANN and OLS models were applied to another
city, Cheongju, to generate ANN and OLS statistical noise maps, respectively. The noise
levels from the two statistical noise maps were compared with those from a normal noise
map of the city of Cheongju.

2. Materials and Methods

The study area was divided by applying a grid cell system, and the representative
values of urban form indicators and road-traffic noise levels were calculated for each grid
cell. The representative values of urban form indicators of a grid cell were extracted based
on 3D GIS data, and the representative values of the road-traffic noise level were extracted
from an existing normal noise map and verified by on-site measurements [37].

2.1. Site Selection and Data Preparation

Gwangju Metropolitan City in South Korea, which was previously noise-mapped,
was selected as the study area, and the City of Cheongju was selected as the verification
area to compare its predicted road-traffic noise levels. The noise map of Gwangju was
produced with noise mapping software, SoundPLAN [38], and includes road, railway,
and air traffic as noise sources [39]. Figure 1 shows the road-traffic normal noise map of
Gwangju. The total area of Gwangju was 501 km2 as of 2017, when the noise map was
developed, and the population was 1.46 million. The total area of Cheongju was 153 km2
as of 2013, when the noise map was developed, and the population was about 0.689 mil-
lion.
Representative values of the road-traffic noise levels and the urban form indicators
were calculated for each grid cell. As the size of a grid cell affects the statistical results, it
should be determined by a careful test of different sizes of grid cells. Salomons and Berg-
hauser Pont [23] used 250 m × 250 m grid cells to extract representative values of facade
Sustainability 2021, 13, 2365 4 of 17

noise levels, traffic volume, and urban density. Chun and Guldmann [40] used three grid
cell systems of the sizes of 120 m × 120 m, 240 m × 240 m, and 480 m × 480 m to analyze
the impacts of urban characteristics on urban heat islands. The best model was obtained
statistically from the smallest cells of 120 m × 120 m. Ryu [41] analyzed the relationship
between road-traffic noise and urban form indicators by varying the size of grid cell sys-
tems of 125 m × 125 m, 250 m × 250 m, and 300 m × 300 m and found that the grid cell
system of 125 m × 125 m best reflected the characteristics of the urban form indicators.
Since the study and verification areas for this study were Gwangju and Cheongju of
South Korea, the 125 m × 125 m grid cell system that showed the best result in Ryu [41]
was adopted, and representative values of road-traffic noise and urban form indicators
were calculated in each grid cell.
In this study, various types of data were used to generate a 3D urban model and
road-traffic noise map. Topography data were used to develop a digital elevation model
for a 3D urban model, and road and building data were also used to create the 3D urban
model. LiDAR (light detection and ranging) data were used to supplement the building
data without height and to model individual overpass roads [37]. LiDAR data were used
to increase the precision of the 3D urban model [37,42,43], and through this, accurate pro-
duction of noise maps and clear reflection of urban characteristics were achieved. Land-
use and population data were used to calculate the urban form indicators. A summary of
the input data for Gwangju and Cheongju used in this research is shown in Appendix A.

Figure 1. Road-traffic noise map of Gwangju Metropolitan City [39,44,45].

2.2. Representative Values of the Road-Traffic Noise Level and Urban Form Indicators of a Grid
Cell
Representative values of the road-traffic noise level were calculated for each grid cell
from an existing noise map [39] by averaging daytime noise levels on every facade of the
buildings that were included in the grid cell. In previous studies [23,25], population (in-
habitant)–weighted averaging was used to estimate representative values of the road-traf-
fic noise level for grid cells. Calculation of population (inhabitant)–weighted averaging
noise levels is suitable for estimating the noise exposure in residential areas. However, the
purpose of estimating the representative value of the road-traffic noise level in this study
Sustainability 2021, 13, 2365 5 of 17

was to understand the noise distribution throughout the city regardless of land use. There-
fore, the representative value of the road-traffic noise level for each grid cell should reflect
the averaged noise level of all buildings in the grid cell. As shown in Figure 2a, the grid
cell which does not contain any building was not considered for the statistical analysis,
and the numbers of valid grid cells were 11,990 for Gwangju and 4173 for Cheongju, re-
spectively. In grid cells that contained only whole buildings, such as in Figure 2b, all the
facade noise levels were used to calculate the representative value. For buildings that were
straddled over multiple grid cells, as in Figure 2c,d, only the facade noise levels contained
in each grid cell were used for the calculation. The representative value of the road-traffic
noise level, defined as L, was measured as follows:
1
≡ 10 × log { ( 10 )} , (1)
∑ ∑ ∑
where Lijk is the noise level at the facade i, floor j, and building k.

Figure 2. Estimation of the representative value of the road-traffic noise level in each grid cell: (a)
no building in a cell; (b) whole buildings in a cell; (c) a straddled building over two cells; (d) a
building straddled over two cells and a whole building in a cell.

The representative values of urban form indicators were calculated with a 3D urban
model that was used to create a normal noise map, and the indicators were selected that
could affect road-traffic noise emissions and propagation [27,28,41]. The indicators were
categorized as population-, building-, road-, and land-use-related. Population density, P,
was chosen as a population-related urban form indicator. In previous studies [17,23], the
population density showed a weak positive or negative correlation with the road-traffic
noise level. Population density is defined as follows for a grid cell:
∑ ∑ ∑
≡ , (2)

where Pijk is the number of persons at the facade i, floor j, and building k, and A is the area
of the cell.
Ground space index (GSI) and floor space index (FSI) were selected as building-re-
lated urban form indicators. GSI is the ratio of the floor area of the buildings in a cell to
the total area of the grid cell, and FSI is the ratio of the total area of the buildings in a cell
to the total area of the grid cell. Buildings in an urban area play a role, with respect to
acoustics, in reducing road-traffic noise due to reflection, diffraction, and absorption. In
the noise mapping software, input factors such as reflection number and absorption loss
of buildings are used to consider the effect of noise reduction due to the facade of build-
ings [38,46]. GSI and FSI are calculated as follows:
Sustainability 2021, 13, 2365 6 of 17

∑
≡ , (3)

where Gk is the area of building k, and A is the cell area, and:

∑
≡ , (4)

where Jk is the number of floors of building k.

Road-related urban form indicators were estimated by road-segment-area-weighted
averaging. Traffic volume (Q), heavy vehicle ratio (PH), and speed (V) were selected. Traf-
fic volume, heavy vehicle ratio, and speed are core factors that could explain the mecha-
nism of road-traffic noise emissions. The variables are the basic input factors in the math-
ematical road-traffic noise prediction models, such as RLS-90, NMPB 2008, and TNM [47–
49], and are also inputs used in noise mapping software to calculate the emission level of
road-traffic noise [38,46]. Traffic volume, heavy vehicle ratio, and speed are defined as
follows, respectively:

∑
≡ , (5)
∑

∑
≡ , (6)
∑

∑
≡ , (7)
∑
where Ri, wi, Qi, PHi, and Vi are the length, width, traffic volume, heavy vehicle ratio, and
traffic speed of road segment i, respectively. Traffic density is defined as:

≡ (8)

The road area density, Ra, and noise barrier area density, Wa, were also selected. Road
area density is the ratio of the road area in a cell to the total area of the grid cell, and noise
barrier area density is the ratio of the noise barrier area in a cell to the total area of the grid
cell. The increased road area density tends to increase road-traffic noise emission levels,
and the increased noise barrier area density tends to reduce the road-traffic noise level.
The road area density and noise barrier density were defined, respectively, as:

∑
≡ , (9)

∑
≡ , (10)

where Ri, wi, and Wi are the length and width of road segment i and the area of noise
barrier i, respectively.

Finally, land-use-related urban form indicators were calculated as the fraction of

land use in each grid cell. Various studies have been performed to analyze the relationship
between land-use characteristics and road-traffic or environmental noise [16–22]. Catego-
ries of land use in this study were residential, commercial, industrial, and green area, and
were defined, respectively, as:
Sustainability 2021, 13, 2365 7 of 17

≡ , ≡ , ≡ , ≡ , (11)

where LR, LC, LI, LG, and A are the areas of residential, commercial, industrial, green use,
and a cell, respectively. The land-use urban form indicators were assumed to satisfy the
following condition:

1= + + + . (12)

Figures 3 and 4 show the grid maps of representative values for L, P, FSI, and V in
Gwangju and Cheongju. The grid maps of representative values for other urban form in-
dicators are presented in supplement A.

(a) (b)

(c) (d)

Figure 3. Grid maps of: (a) L (road-traffic noise level) in Gwangju; (b) L (road-traffic noise level) in Cheongju; (c) P (population den-
sity) in Gwangju; (d) P (population density) in Cheongju [44,45].
Sustainability 2021, 13, 2365 8 of 17

(a) (b)

(c) (d)

Figure 4. Grid maps of: (a) FSI (floor space index) in Gwangju; (b) FSI (floor space index) in Cheongju; (c) Q (traffic volume) in
Gwangju; (d) Q (traffic volume) in Cheongju. [44,45].

3. Results and Discussion

In this section, the relationship between the road-traffic noise level and urban form
indicators was analyzed statistically by the ANN and OLS methods. The developed mod-
els were also applied to another city, Cheongju, to generate statistical noise maps which
were compared to a normal noise map of Cheongju.

3.1. ANN and OLS Model Development

The data set for Gwangju was divided by 2 to 1 into the training and test sets, respec-
tively. The training data set was used to analyze the relationships between road-traffic
noise and the urban form indicators and to create an ANN model to predict the road-
traffic noise levels with the urban form indicators.
Among the urban form indicators defined in Section 2.2, independent explanatory
variables were selected to be used to develop the OLS and ANN models. The fraction of
green area, LG, was excluded because the degree of freedom of land-use-related urban
form indicators was 3, as defined in Equation (12). The urban form indicators Dt and LI,
which showed multi-collinearity, were excluded. Table 1 shows the result for the multi-
collinearity test after the urban form indicators Dt and LI were excluded, and the remain-
ing ten indicators were used as the explanatory variables for the OLS and ANN models.
Sustainability 2021, 13, 2365 9 of 17

Table 1. Multi-collinearity test results for urban form indicators.

Urban Form
P GSI FSI Q PH V Dt Ra Wa LR LC LI LG
Indicators
VIF
(Variance In- 3.94 2.41 5.00 1.57 3.46 4.98 - 2.46 1.09 1.71 1.48 - -
flation Factors)

The results for the OLS model are presented in Appendix B. Among the ten indicators,
seven statistically significant urban form indicators in the OLS model were found: FSI, Q,
PH, V, Ra, LR, and LC. The urban form indicators showing statistical significance in common
with Ryu et al. [25] were Q, V, Ra, and LR. FSI was a significant variable in Salomons and
Berghauser Pont [23] and in this analysis. The variables related to road area, Ra for this
study, were significant in this and previous studies [16,19]. The population-related varia-
bles were not significant in this study or in Aguilera et al. [16].
In optimizing the process of the ANN model, the control parameters of the ANN
model were adjusted until the difference between the noise levels from the normal noise
map and those from the ANN model was minimized. R [50] contains various packages
that can be used to optimize the ANN model, and for this study, the nnet package [51],
which is specialized to optimize the single-hidden layer ANN model, was used. The nnet
package can adjust various control parameters to develop the ANN model, and among
them, the number of hidden nodes and the decay parameter were adjusted to examine the
prediction accuracy. The accuracy was determined by calculating the root mean square
error (RMSE) using the difference between the existing and predicted answers. The RMSE
for the ANN model was calculated while adjusting the number of hidden nodes from 1 to
50, and the decay parameters were tested for each number of hidden nodes from e−2 to e7,
as well as 0. The thirty-three hidden nodes and decay parameter, e−1, that had the lowest
RMSE were applied to the final ANN model. Figure 5 shows the RMSE test results. As an
activation function, linout [51] was selected. The maxit [51] that determines the number of
iterations was set to 200. Figure 6 shows the final ANN model, with 10 input nodes for
urban form indicators, 33 hidden nodes, and 1 output node.

(a) (b)

Figure 5. Minimal root mean square error (RMSE) investigation to optimize the artificial neural
network (ANN) model parameters: (a) number of hidden nodes; (b) decay parameter (number of
hidden nodes = 33).
Sustainability 2021, 13, 2365 10 of 17

Figure 6. Artificial neural network model.

3.2. Statistical Noise Mapping by the ANN and OLS Models

The ANN and OLS models, which were developed from the dataset of Gwangju,
were applied to the original city, Gwangju, to test the models. The correlation coefficient
and coefficient of determination for predicted road-traffic noise levels between the normal
noise map and the ANN model were 0.70 and 0.49, respectively, and for the predicted
levels between the OLS model and the noise map were 0.66 and 0.44, respectively. These
findings demonstrate that the statistical noise map created by using the ANN model was
slightly more accurate than the OLS map. Figure 7a,b show the ANN and OLS statistical
grid maps for the road-traffic noise level in Gwangju. The two statistical noise maps are
comparable with the grid map of the road-traffic noise level from the normal noise map
in Figure 3a. It was found that the OLS model tends to under-predict the road-traffic noise
level, while the ANN model tends to over-predict it. The corresponding results of the
noise-exposed population from the two models are presented in Appendix C.
The ANN and OLS models were applied to Cheongju, and the predicted road-traffic
noise levels were compared to the levels predicted by a normal noise map. The correlation
coefficient and coefficient of determination for predicted road-traffic noise levels in
Cheongju between the ANN model and the normal noise map were 0.66 and 0.44, respec-
tively, and the correlation coefficient and coefficient of determination between the OLS
model and the normal noise map were 0.70 and 0.49, respectively. Unlike for Gwangju,
the correlation coefficient and coefficient of determination by the OLS model were higher
than those from the ANN model.
Figure 8a,b show the statistical grid maps for road-traffic noise levels in Cheongju
using the ANN and OLS models, respectively. The two statistical noise maps were com-
parable with the grid map of representative values of road-traffic noise levels from the
normal noise map in Figure 3b. As in Gwangju, the OLS model tended to under-predict
the road-traffic noise level, while the ANN model tended to over-predict it. The corre-
sponding results of the noise-exposed population from the two models are presented in
Appendix C.
Sustainability 2021, 13, 2365 11 of 17

(a) (b)
Figure 7. Statistical road-traffic noise maps of Gwangju: (a) ANN model; (b) ordinary least squares (OLS) model. [44,45].

(a) (b)
Figure 8. Statistical road-traffic noise map of Cheongju: (a) ANN model; (b) OLS model [44, 45].

3.3. Importance of Urban Form Indicators in the ANN Model

The effects of urban form indicators on road-traffic noise prediction in the ANN
model were tested by excluding urban form indicators one by one. Ten test ANN models
were developed using only nine urban form indicators, excluding one urban form indica-
tor, and the excluded urban form indicator was the subject of a test. RMSE differences
between each test model and the complete ANN model, which are shown in Figure 6,
were compared. The larger the difference in the RMSE, the more important the excluded
indicator was. Figure 9 shows the RMSE differences for each urban form indicator.
This RMSE difference analysis showed that traffic volume in the traffic-related urban
form indicators was the most important among all urban form indicators, as in previous
research [20,25,26], where it was found to be statistically significant. In Aguilera et al. [16],
however, traffic volume was not found to be an important variable, but the variables re-
lated to road area and length were significant in this analysis and Aguilera et al. [16].
While the heavy vehicle ratio in the traffic-related urban form indicators in this study was
similar to the truck flow variable in Aguilera et al. [16], truck flow was the third important
variable among 28 variables in their study [16], but the heavy vehicle ratio was just the
fifth most important among 10 variables in this study. Road-related variables such as traf-
fic volume, road area, or speed were commonly important or significant variables in pre-
vious research [16,17,19,25,26] and in this study.
Sustainability 2021, 13, 2365 12 of 17

In Ryu et al. [25], where similar urban form indicators to this study were adopted,
GSI (ground space index) was more significant than FSI (floor space index), but our anal-
ysis showed that FSI was more important than GSI. In Salomons and Berghauser Pont [23],
the increases of GSI and FSI caused road-traffic noise levels to decrease. In Aguilera et al.
[16], where a LUR model was developed, the average height of buildings at measurement
sites was statistically significant in only one city among the three cities that were studied.
The proportion of open space (porosity) and shape of the urban patch (CI) showed posi-
tive correlations with road-traffic noise, but the complexity of the perimeter of an urban
area (fractal) showed a negative correlation with road-traffic noise in Silva et al. [24].
In the RMSE difference analysis, LC (fraction of commercial area) was the third most
important variable among 10 urban form indicators, but in Sieber et al. [19], three land-
use-related variables (household density, commercial land use, and industrial land use)
were selected as predictor variables, and commercial and industrial land-use variables
were not statistically significant. In Chang et al. [20], green land and agriculture-use areas
were included with another four variables for the road-traffic noise level prediction. In
Ryu et al. [25], where similar urban form indicators to our study were adopted, LI (fraction
of industrial area) was statistically significant, but LC was not statistically significant.

Figure 9. RMSE differences between the complete and test ANN models.

4. Conclusion
ANN and OLS models were developed to generate statistical road-traffic noise maps
based on elementary urban form indicators. The cities of Gwangju and Cheongju, which
already have normal noise maps, were selected as the study and verification areas, respec-
tively. The study area was divided into a set of square grid cells, and the representative
values of the road-traffic noise level and urban form indicators were estimated for each
cell. The ANN and OLS models were fitted by using the dataset of Gwangju. The ANN
model was fitted by optimizing the number of hidden nodes and decay parameters.
The ANN and OLS models which were developed based on the city of Gwangju were
applied to the city of Cheongju, and statistical noise maps were generated. The predicted
Sustainability 2021, 13, 2365 13 of 17

road-traffic noise levels from the statistical models were compared to those from the nor-
mal noise maps to verify the statistical models’ applicability. The predictability of the
ANN and OLS models was slightly different, but it was found that the OLS model tended
to under-estimate noise levels and the ANN model tended to over-estimate noise levels.
The road-related urban form indicators, such as traffic volume and road area density,
were found to be important in the prediction of the road-traffic noise level by the ANN
model.
The limitation of the statistical noise maps is that only one representative value of
road-traffic noise level can be predicted for an area of a grid cell. The normal noise map
can give the noise levels of the building facades and grids with adjustable resolutions, but
the statistical noise maps, which are generated by any representative or averaged values
of the urban form indicators, need a minimal area to extract them. Therefore, the statistical
noise maps can only predict the noise level of a certain area, such as a grid cell.
In some previous studies which predicted environmental phenomenon such as road-
traffic noise [25] and urban heat islands [40] based on urban form indicators of the grid
cell system, the spatial dependency of neighboring cells was considered simultaneously
with a target cell through a spatial statistical model. In this study, only urban form indi-
cators of one target cell were considered independently to develop the ANN model, but
considering the effects of neighboring cells would be useful in a future study of the con-
volutional neural network (CNN) model, which is used for image recognition and classi-
fication.

Supplementary Materials: The following are available online at www.mdpi.com/2071-

1050/13/4/2365/s1, Figure S1. Grid maps of representative values for road-traffic noise level: (a)
Gwangju; (b) Cheongju, Figure S2. Grid maps of representative values for population density: (a)
Gwangju; (b) Cheongju, Figure S3. Grid maps of representative values for ground space index: (a)
Gwangju; (b) Cheongju, Figure S4. Grid maps of representative values for floor space index: (a)
Gwangju; (b) Cheongju, Figure S5. Grid maps of representative values for traffic volume: (a)
Gwangju; (b) Cheongju, Figure S6. Grid maps of representative values for heavy vehicle ratio: (a)
Gwangju; (b) Cheongju, Figure S7. Grid maps of representative values for traffic speed: (a) Gwangju;
(b) Cheongju, Figure S8. Grid maps of representative values for traffic density: (a) Gwangju; (b)
Cheongju, Figure S9. Grid maps of representative values for road area density: (a) Gwangju; (b)
Cheongju, Figure S10. Grid maps of representative values for noise barrier density: (a) Gwangju; (b)
Cheongju, Figure S11. Grid maps of representative values for fraction of residential area: (a)
Gwangju; (b) Cheongju, Figure S12. Grid maps of representative values for fraction of commercial
area: (a) Gwangju; (b) Cheongju, Figure S13. Grid maps of representative values for fraction of in-
dustrial area: (a) Gwangju; (b) Cheongju, Figure S14. Grid maps of representative values for fraction
of green area: (a) Gwangju; (b) Cheongju, Figure S15. Statistical road-traffic noise maps of Gwangju:
(a) ANN model; (b) OLS model, Figure S16. Statistical road-traffic noise map of Cheongju: (a) ANN
model; (b) OLS model.
Author Contributions: Conceptualization, H.R. and P.K.; methodology, J.-J.J.; data curation, H.R.;
writing—original draft preparation, P.K.; writing—review and editing, S.I.C. All authors have
read and agreed to the published version of the manuscript.
Funding: This work was supported by the Basic Science Research program through the National
Research Foundation of Korea (NRF), funded by the Ministry of Education (NRF-
2017R1D1A1B03035253) for Phillip Kim and Hunjae Ryu. This work was also supported by the
2020 Research Fund of the University of Seoul for Seo Il Chang.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data sharing is not applicable to this article.
Conflicts of Interest: The authors declare no conflicts of interest.
Sustainability 2021, 13, 2365 14 of 17

Appendix A

Table 1. Summary of the Gwangju and Cheongju input data used in this research.

City Data Type Parameter Provider Production Year Access Type

Topography Elevation National Geo- 2016 Public

Noise barrier graphic Infor-
LiDAR Point cloud mation Institute 2007 Proprietary

Volume Gwangju Metro-

Vehicle Speed politan Police 2016 Proprietary
Type Agency

Gwangju Metro-
Road Network 2016 Proprietary
politan City Office
Gwangju
Footprint
Gwangju Metro-
Building Elevation 2016 Public
politan City Office
Building use

Population for “- Korean National

Population 2015 Public
dong” Statistical Office

National Geo-
Land-use
Land use graphic Infor- 2017 Public
classification
mation Institute

Chungcheong-
Topography Elevation buk-do Provincial 2007 Proprietary
Government
Cheongju City
Volume
Police Agency
Vehicle Speed 2007 Proprietary
and Cheongju
Type
City Government
Chungcheong-
Road Network buk-do Provincial 2009 Proprietary
Cheongju Government

Footprint
Cheongju City
Building Elevation 2009 Proprietary
Government
Building use
Korean National
Population for “- Statistical Office Public and Propri-
Population 2009
dong” and Cheongju etary
City Government
Land-use Cheongju City
Land use 2007 Proprietary
classification Government
Sustainability 2021, 13, 2365 15 of 17

Appendix B

Table A2. Results for the OLS model.

Variable Estimate Std. Error t Pr (｜t｜)

(Intercept) 47.4 0.281 168.937 <2 × 10−16 ***
P –1.61 10.8 –1.485 0.138
GSI 1.25 0.922 1.352 0.176
FSI 2.48 0.289 8.6 <2 × 10−16 ***
Q 2.26 × 10−3 2.45 × 10−4 9.229 <2 × 10−16 ***
PH 0.366 5.17 × 10−2 7.093 1.42 × 10−12 ***
V 0.141 8.35 × 10−3 16.932 <2 × 10−16 ***
Ra 28.1 2.12 13.246 <2 × 10−16 ***
Wa 1.4 11.3 0.124 0.902
LR −1.31 0.314 –4.171 3.06 × 10−5 ***
LC −4.56 0.44 –10.375 <2 × 10−16 ***
Significant codes: ‘***’ 0.001, ‘ ’ 1. Residual standard error: 8.267 on 7982 degrees of freedom. Multiple R-squared: 0.4446,
Adjusted R-squared: 0.4439. F: 639 on 10 and 7982 DF, p: < 2.2 × 10−16.

Appendix C

Table A3. Proportions of the noise-exposed population according to prediction error in Gwangju.

Prediction Error between Noise Map and Statistical Noise Map (dB(A))
Model
<–10 –10 ~ –5 –5 ~ –3 –3 ~ 0 0~3 3~5 5 ~ 10 <10
ANN 1.9% 9.8% 10.0% 25.5% 28.6% 11.6% 11.2% 1.3%
OLS 3.1% 12.1% 9.9% 21.5% 26.6% 12.5% 12.7% 1.5%

Table A4. Proportions of the noise-exposed population according to prediction error in Cheongju.

Prediction Error between Noise Map and Statistical Noise Map (dB(A))
Model
<–10 –10 ~ –5 –5 ~ –3 –3 ~ 0 0~3 3~5 5 ~ 10 <10
ANN 0.4% 2.9% 3.1% 9.4% 20.2% 17.5% 35.6% 10.9%
OLS 2.3% 8.7% 6.5% 14.6% 23.2% 16.9% 23.9% 4.0%

References
1. Ogryzek, M.; Adamska-Kmieć, D.; Klimach, A. Sustainable transport: An efficient transportation network-case study. Sustain-
ability 2020, 12, 1–14, doi:10.3390/su12198274.
2. Kowalczyk, C.; Kil, J.; Kurowska, K. Dynamics of development of the largest cities—Evidence from Poland. Cities 2019, 89, 26–
34, doi:10.1016/j.cities.2019.01.018.
3. Stansfeld, S.A. Noise effects on health in the context of air pollution exposure. Int. J. Environ. Res. Public Health 2015, 12, 12735–
12760, doi:10.3390/ijerph121012735.
4. WHO. Health Effects and Risks of Transport Systems: The HEARTS Project; WHO: Copenhagen, Denmark, 2006.
5. WHO. Burden of Disease from Environmental Noise: Quantification of Healthy Life Years Lost in Europe; WHO: Copenhagen, Den-
mark, 2011.
6. EC. 2002/49/EC of the European parliament and the Council of 25 June 2002 relating to the assessment and management of
environmental noise. Off. J. Eur. Commun. L 2002, 189, 2002.
7. Arana, M.; San Martin, R.; San Martin, M.L.; Aramendía, E. Strategic noise map of a major road carried out with two environ-
mental prediction software packages. Environ. Monit. Assess. 2010, 163, 503–513, doi:10.1007/s10661-009-0853-5.
8. Bunn, F.; Zannin, P.H.T. Assessment of railway noise in an urban setting. Appl. Acoust. 2016, 104, 16–23,
doi:10.1016/j.apacoust.2015.10.025.
9. Iglesias-Merchan, C.; Diaz-Balteiro, L.; Soliño, M. Transportation planning and quiet natural areas preservation: Aircraft over-
flights noise assessment in a National Park. Transp. Res. Part D Transp. Environ. 2015, 41, 1–12, doi:10.1016/j.trd.2015.09.006.
Sustainability 2021, 13, 2365 16 of 17

10. Gagliardi, P.; Fredianelli, L.; Simonetti, D.; Licitra, G. ADS-B system as a useful tool for testing and redrawing noise manage-
ment strategies at Pisa Airport. Acta Acust. 2017, 103, 543–551, doi:10.3813/AAA.919083.
11. Nastasi, M.; Fredianelli, L.; Bernardini, M.; Teti, L.; Fidecaro, F.; Licitra, G. Parameters affecting noise emitted by ships moving
in port areas. Sustainability 2020, 12, 1–17, doi:10.3390/su12208742.
12. Licitra, G.; Ascari, E.; Fredianelli, L. Prioritizing process in action plans: A review of approaches. Curr. Pollut. Rep. 2017, 3, 151–
161.
13. Ministry of Environment, Korea. Noise and Vibration Control Act; Ministry of Environment, Korea: Sejong City, Korea, 2020.
14. The Korea Transport Institute, Korea. 2017 National Traffic Survey and Database System Operation and Maintenance; The Korea
Transport Institute: Sejong City, Korea, 2017. (In Korean)
15. Ruiz-Padillo, A.; Ruiz, D.P.; Torija, A.J.; Ramos-Ridao, Á. Selection of suitable alternatives to reduce the environmental impact
of road traffic noise using a fuzzy multi-criteria decision model. Environ. Impact Assess. Rev. 2016, 61, 8–18,
doi:10.1016/j.eiar.2016.06.003.
16. Aguilera, I.; Foraster, M.; Basagaña, X.; Corradi, E.; Deltell, A.; Morelli, X.; Phuleria, H.C.; Ragettli, M.S.; Rivera, M.; Thomasson,
A.; et al. Application of land use regression modelling to assess the spatial distribution of road traffic noise in three European
cities. J. Expo. Sci. Environ. Epidemiol. 2015, 25, 97–105, doi:10.1038/jes.2014.61.
17. Han, X.; Huang, X.; Liang, H.; Ma, S.; Gong, J. Analysis of the relationships between environmental noise and urban. Environ.
Pollut. 2018, 233, 755–763, doi:10.1016/j.envpol.2017.10.126.
18. Xie, D.; Liu, Y.; Chen, J. Mapping Urban environmental noise: A land use regression method. Environ. Sci. Technol. 2011, 45,
7358–7364, doi:10.1021/es200785x.
19. Sieber, C.; Ragettli, M.S.; Brink, M.; Toyib, O.; Baatjies, R.; Saucy, A.; Probst-Hensch, N.; Dalvie, M.A.; Röösli, M. Land use
regression modeling of outdoor noise exposure in informal settlements in Western Cape, South Africa. Int. J. Environ. Res. Public
Health 2017, 14, doi:10.3390/ijerph14101262.
20. Chang, T.Y.; Liang, C.H.; Wu, C.F.; Chang, L. Te Application of land-use regression models to estimate sound pressure levels
and frequency components of road traffic noise in Taichung, Taiwan. Environ. Int. 2019, 131, 104959, doi:10.1016/j.en-
vint.2019.104959.
21. Liu, Y.; Goudreau, S.; Oiamo, T.; Rainham, D.; Hatzopoulou, M.; Chen, H.; Davies, H.; Tremblay, M.; Johnson, J.; Bockstael, A.;
et al. Comparison of land use regression and random forests models on estimating noise levels in five Canadian cities. Environ.
Pollut. 2020, 256, 113367, doi:10.1016/j.envpol.2019.113367.
22. Margaritis, E.; Kang, J.; Aletta, F.; Axelsson, Ö. On the relationship between land use and sound sources in the urban environ-
ment. J. Urban Des. 2020, 25, 626–642, doi:10.1080/13574809.2020.1730691.
23. Salomons, E.M.; Berghauser Pont, M. Urban traffic noise and the relation to urban density, form, and traffic elasticity. Landsc.
Urban Plan. 2012, 108, 2–16, doi:10.1016/j.landurbplan.2012.06.017.
24. Silva, L.T.; Oliveira, M.; Silva, J.F. Urban form indicators as proxy on the noise exposure of buildings. Appl. Acoust. 2014, 76,
366–376, doi:10.1016/j.apacoust.2013.07.027.
25. Ryu, H.; Park, I.K.; Chun, B.S.; Chang, S. Il Spatial statistical analysis of the effects of urban form indicators on road-traffic noise
exposure of a city in South Korea. Appl. Acoust. 2017, 115, 93–100, doi:10.1016/j.apacoust.2016.08.025.
26. Lu, X.; Kang, J.; Zhu, P.; Cai, J.; Guo, F.; Zhang, Y. Influence of urban road characteristics on traffic noise. Transp. Res. Part D
Transp. Environ. 2019, 75, 136–155, doi:10.1016/j.trd.2019.08.026.
27. Wang, B.; Kang, J. Effects of urban morphology on the traffic noise distribution through noise mapping: A comparative study
between UK and China. Appl. Acoust. 2011, 72, 556–568, doi:10.1016/j.apacoust.2011.01.011.
28. Hao, Y.; Kang, J. Influence of mesoscale urban morphology on the spatial noise attenuation of flyover aircrafts. Appl. Acoust.
2014, 84, 73–82, doi:10.1016/j.apacoust.2013.12.001.
29. Margaritis, E.; Kang, J. Relationship between urban green spaces and other features of urban morphology with traffic noise
distribution. Urban For. Urban Green 2016, 15, 174–185, doi:10.1016/j.ufug.2015.12.009.
30. Margaritis, E.; Kang, J. Relationship between green space-related morphology and noise pollution. Ecol. Indic. 2017, 72, 921–933,
doi:10.1016/j.ecolind.2016.09.032.
31. Kumar, P.; Nigam, S.P.; Kumar, N. Vehicular traffic noise modeling using artificial neural network approach. Transp. Res. Part
C Emerg. Technol. 2014, 40, 111–122, doi:10.1016/j.trc.2014.01.006.
32. Garg, N.; Mangal, S.K.; Saini, P.K.; Dhiman, P.; Maji, S. Comparison of ANN and Analytical Models in Traffic Noise Modeling
and Predictions. Acoust. Aust. 2015, 43, 179–189, doi:10.1007/s40857-015-0018-3.
33. Hamad, K.; Ali Khalil, M.; Shanableh, A. Modeling roadway traffic noise in a hot climate using artificial neural networks. Transp.
Res. Part D Transp. Environ. 2017, 53, 161–177, doi:10.1016/j.trd.2017.04.014.
34. Mansourkhaki, A.; Berangi, M.; Haghiri, M.; Haghani, M. A neural network noise prediction model for Tehran urban roads. J.
Environ. Eng. Landsc. Manag. 2018, 26, 88–97, doi:10.3846/16486897.2017.1356327.
35. Ali Khalil, M.; Hamad, K.; Shanableh, A. Developing Machine Learning Models to Predict Roadway Traffic Noise: An Oppor-
tunity to Escape Conventional Techniques. Transp. Res. Rec. 2019, 2673, 158–172, doi:10.1177/0361198119838514.
36. Nourani, V.; Gökçekuş, H.; Umar, I.K. Artificial intelligence based ensemble model for prediction of vehicular traffic noise.
Environ. Res. 2020, 180, 108852, doi:10.1016/j.envres.2019.108852.
37. Park, T.; Kim, M.; Jang, C.; Choung, T.; Sim, K.; Seo, D.; Chang, S. Il The Public Health Impact of Road-Traffic Noise in a Highly-
Populated City, Republic of Korea: Annoyance and Sleep Disturbance. Sustainability 2018, 10, 2947, doi:10.3390/su10082947.
Sustainability 2021, 13, 2365 17 of 17

38. SoundPLAN. Available online: https://www.soundplan.eu/en/software/soundplannoise/ (accessed on 17 December 2020)

39. Environmental Policy Division of Gwangju Metropolitan City. Development of noise map of Gwangju metropolitan city and
construction of management system. 2017. Available online: http://www.prism.go.kr/homepage/entire/retrieveEntireDe-
tail.do;jsessionid=0D1120636A3E2B1D5D45CD2083309810.node02?cond_research_name=&cond_re-
search_start_date=&cond_research_end_date=&research_id=6290000-201800019&pageIndex=3&leftMenuLevel=160 (in Ko-
rean) (accessed on 17 December 2020).
40. Chun, B.; Guldmann, J.M. Spatial statistical analysis and simulation of the urban heat island in high-density central cities.
Landsc. Urban Plan. 2014, 125, 76–88, doi:10.1016/j.landurbplan.2014.01.016.
41. Ryu, H. Spatial statistical Analysis of Road-traffic noise for supporting urban planning in strategic stage. Ph.D. Thesis, Univer-
sity of Seoul, Seoul, Korea, 2017. (In Korean)
42. Seong, J.C.; Park, T.H.; Ko, J.H.; Chang, S.I.; Kim, M.; Holt, J.B.; Mehdi, M.R. Modeling of road traffic noise and estimated human
exposure in Fulton County, Georgia, USA. Environ. Int. 2011, 37, 1336–1341, doi:10.1016/j.envint.2011.05.019.
43. Gulliver, J.; Morley, D.; Vienneau, D.; Fabbri, F.; Bell, M.; Goodman, P.; Beevers, S.; Dajnak, D.; J. Kelly, F.; Fecht, D. Develop-
ment of an open-source road traffic noise model for exposure assessment. Environ. Model. Softw. 2015, 74, 183–193,
doi:10.1016/j.envsoft.2014.12.022.
44. ESRI World Topographic Map. Available online: https://www.arcgis.com/home/item.html?id=30e5fe3149c34df1ba922e6f5b
bf808f (accessed on 21 January 2021).
45. ESRI World Hillshade Available online: https://www.arcgis.com/home/item.html?id=1b243539f4514b6ba35e7d995890db1d (ac-
cessed on 20 January 2021).
46. CadnaA. Available online: https://www.datakustik.com/products/cadnaa/cadnaa/ (accessed on 17 December 2020).
47. Ullrich, S. Annahmen zu den Fahrzeugemissionen in den Richtlinien für den Lärmschutz an Straßen (RLS-90). Str. Autob. 1991,
42, 189–191.
48. Besnard, F.; Hamet, J.-F.; Duc, E.; Guizard, V.; Fürst, N.; Doisy, S.; Lelong, J.; Dutilleux, G. Road Noise Prediction—1 Calculating
Sound Emissions from Road Traffic; Setra: Bagneux, France, 2011.
49. Traffic Noise Model. Available online: https://www.fhwa.dot.gov/environment/noise/traffic_noise_model/ (accessed on 16 De-
cember 2020).
50. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria,
2013.
51. Venables, W.N.; Ripley, B.D. Modern Applied Statistics with S, 4th ed.; Springer: New York, NY, USA, 2002.