Urbanization: Smart Urban Transit System using Artificial Intelligence

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 02 | Feb 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 2787
Urbanization: Smart Urban Transit System
using Artificial Intelligence
Dilip Jain1, Siddhesh Kadam2, Aditya Mahimkar3, Prof. Vaishali Yeole4
1,2,3B.E. Student, Computer Department,
4Project Guide, Smt. Indira Gandhi College of Engineering Navi Mumbai, Maharashtra, India.
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Urban areas are home to more than half of the
world's population, and around 2.5 billion people are
projected to be added in the next 3 decades. The way we
construct new cities is at the core of our existence, from
climate crises to economic vibrancy, to our well-being and
sense of community. Bringing "smartness" to the town and
the township has been the vision of various research and
policies addressing issues with the existing system and
providing a vantage to sustainable development for a
healthier, happier, and happening tomorrow. We explore
these studies and use publicly available data to build an
intelligent pipeline of amalgamated algorithms to produce a
blueprint for smart city development.
Key Words: City, Urban, Urbanization, Planning, People,
Population, Pollution, Transportation, Artificial Intelligence.
1. INTRODUCTION
Smart cities emphasize the capacity to build a secure,
happier, and more sustainable environment that is centred
around people's well-being through improving
accessibility, transit, healthcare, waste and inconvenience
reduction, and social and economic quality. It is based on
the notion that citizens make up the city and not the other
way round. Smart Cities utilize techniques that have
worked in other locations to meet genuine local needs. To
support smart city growth, cities must build quadruple-
helix collaborative ecosystems capable of merging open
innovation, bottom-up development activities, and local
entrepreneurship in a civic framework.
Transportation has an impact on the shape and liveability
of cities, as well as their socioeconomic, cultural, and
environmental features. The efficiency of public
transportation is one of the most obvious and crucial
aspects of the transportation system for inhabitants. It is
vital to construct an ideal transit network and transit
service in order to deliver smart services.
McKinsey & Company gathered comments from two
sources to address the issue of what are the most
significant components of urban transportation systems:
experts and a poll of locals. According to both
professionals and citizens, the most important component
of transportation systems is safety, which is followed by
efficiency, affordability, availability, convenience, and
ecological sustainability [1].
Availability, Affordability, Convenience, Efficiency,
Sustainability (AACES) transit is what we must focus on.
Solutions need to be resilient, and this paper advocates the
use of technology to produce smart results for residents in
order to increase economic growth and improve people's
quality of life while maintaining environmental balance by
using both historical and future trends to propose a set of
research-backed principles and practices for a better,
greener, and more sustainable future by using both
historical and future trends to propose a set of research-
backed principles and practices for a better, greener, and
more sustainable future.
2. LITERATURE SURVEY
Todd Litman in his paper, ‘Determining Optimal Urban
Expansion, Population and Vehicle Density, and Housing
Types for Rapidly Growing Cities’, examines the economic,
social, and environmental consequences of numerous
urban development characteristics such as population and
vehicle density, housing type, highway design and
management, and recreation facility accessibility [2].
a) Unrestricted cities are surrounded by an abundance of
lower-valued land. They have a lot of room to grow.
b) Semi-constrained cities have a limited ability to
expand. Their development policies should include a
combination of infill development and modest
expansion on major corridors.
c) Constrained cities cannot expand much.
d) In spatially constrained cities, optimal concentrations
can be above 80 residents per hectare; however, as urban
expansion is limited, optimal densities rise, optimal
vehicle ownership rates fall, and a greater share of
housing should be multi-family. Cities should provide a
variety of housing and transportation options that
respond to consumer demands, particularly affordable
housing inaccessible, multimodal neighbourhoods, and
affordable travel modes, with pricing or roadway
management that favour resource-efficient modes, as well
as convenient access to parks and recreational facilities, in
order to be efficient and equitable. Unrestricted cities are
surrounded by an abundance of lower-valued land. They
have a lot of room to grow.

e) According to a study published in the Journal of Global
Economics, the ideal population density is 300 people per
square kilometre.
f) 32 homes per hectare. [3]
Table 1: Optimal Urban Expansion, Densities and
Development Policies [2].
Effective transportation systems must be constructed
which fulfil the present needs and serve future purposes
as well. Increased job prospects, a consolidated market,
better pay, and increased individual wealth are all factors
that have attracted individuals to cities. For a long time,
these factors have been a driving force behind urban
expansion. Based on the World Urbanization Prospects:
2018 report by the United Nations Population Division,
the World Bank estimated annual global urban population
growth at 1.81% in 2020 [4]. By 2045, an estimated 1.5
times increase in global urban population is expected [5].
But this global trend shows that the growth in the area of a
city is comparatively slower than the population. The
increasing urban demands and rapid urbanization
aggravate poor air and water quality, insufficient
availability of water, huge waste generation and energy
demands.
Figure 1: Population Growth Worldwide [5]
Climate change has a wide range of consequences for
people's lives and health. It puts the pillars of human
health in jeopardy - clean air, safe drinking water,
nutritious food, and safe shelter - and has the potential to
undo decades of global health gains. Malnourishment,
malaria, diarrhoea, and heat stress are anticipated to
cause an additional 250 000 fatalities per year between
2030 and 2050 as a consequence of climate change. By
2030, the direct health damage costs are expected to reach
between USD 2-4 billion per year. Even 1.5°C of global
warming is not regarded as safe; every tenth of a degree of
warming will have a significant impact on people's lives
and health [6].
3. PUBLIC HEALTH AND ENVIRONMENT
Improved health can be achieved by reducing greenhouse
gas emissions through better transportation, dietary, and
resource alternatives, particularly through reduced air
pollution. Transportation accounted for 29 percent of the
United States’ total emissions of 6.4 billion metric tons of
carbon dioxide equivalent in 2017 [7]. Transportation and
its carbon release make for a large and complicated matter.
Commutation is a daily travel habit and thus a crucial
research factor in urban geography, sociology and planning
since it may represent both the standard of living and the
effective and efficient use of urban space. Cities and
metropolitan areas serve as hubs for a multitude of
activities, many of which necessitate rapid and efficient
people and commodities movement.
A San Francisco based urban planner, urban designer and
architect Peter Calthorpe advocates pedestrian-friendly
streets and human-scale neighborhoods. Walking is a low-
impact, not expensive, non-polluting and by-far the
healthiest mode of commutation. Adults should engage in
mild-intensity strenuous cardiovascular aerobic activity,
such as a brisk walk, for at least 30 minutes five days a
week (or a total of 2 hours and 30 minutes), according to
the US Department of Health and Human Services' 2008
Physical Activity Guidelines for Americans [8]. Walking
Factor Unconstr
ained
Semi-
constrained
Constrain
ed
Growth
pattern
Expand as
needed
Expand less
than
population
Minimal
expansion
Regional
density
(residents /
ha)
20-60
40-80
80 +
Vehicle
ownership
(per 1,000
pop.)
300-400 200-300 < 200
Intersection
density per
sq. km.
40+ 60+ 80+
Portion of
land in road
10-15% 15-20% 20-25%

improves cardiovascular and pulmonary fitness, manages
hypertension, high cholesterol, joint and muscle discomfort
and also diabetes, and controls body fat. It also enhances
balance, muscle endurance and strength with stronger
bones.
According to the World Health Organization, each
individual should have access to at least 9 square meters of
urban green space [9, 10], as long as it is accessible [11],
safe [12], and functional [13]. In 2016, Mercer's Quality of
Living Survey named Vienna the most livable city in the
world, with 120 square meters of urban green space for its
1.7 million residents. Singapore, the world's third densest
metropolis, offers each of its citizens about 66 square
meters of urban green space. It's vital since the most livable
city is one with enough green space for its citizens [14–16].
4. VISION ZERO
Vision Zero is a global movement aimed at eliminating
traffic-related fatalities and casualties through a systematic
approach to road safety. This strategy is founded on the
notion that traffic injuries and deaths are both intolerable
and preventable.
Oslo, Norway's capital, has been designated a "Car-Free
City". Many new pedestrian paths have been added, and
one street has been converted to a mixed-use area. Two-
thirds of the space in the city is dedicated to pedestrians
and other municipal activities. Throughout the city,
playgrounds for children and families have been created.
[17]
Oslo proves that car-free cities are not beyond reality. A
city with no cars, a strong public transport network, and a
supportive community can exist, making its citizens
healthy and happy. More green space, less pollution, a
more aesthetically pleasing environment, fewer accidents
and casualties, better health, a raised sense of community
and togetherness are all achievable with vision zero.
5. PUBLIC TRANSPORTATION
Transportation accounted for 29 percent of the United
States' total emissions of 6.4 billion metric tonnes of
carbon dioxide equivalent in 2017, according to the U.S.
Department of Transportation (US DOT) [7] and its carbon
release make for a large and complicated matter.
The Transit system is often referred to as a city's "lifeline".
High-density movements necessitate the use of high-
capacity means of transportation like the bus, light rail, and
metro, which are less expensive, more energy-efficient, and
take up significantly less space than private automobiles.
Automobiles, unlike public transportation, are only
available to those who own and are capable of operating
them.
As Peter Calthorpe speaks [18] about the importance of
pedestrian-friendly walkable cities, focusing on Available,
Affordable, Efficient, Convenient, Sustainable (AACES)
Transit systems must be our prime focus. A rich metro
network is highly essential in a city going car-free. Optimal
metro connectivity is where the network is strong and as
quickly as possible, but also ensuring no extreme linkage.
To promote walking, we consider an average of 10,000
steps per day and the U.S. Department of Health and
Human Services recommended 30 minutes brisk walk, five
days per week, as a benchmark for the minimum number
of steps a normal individual should cover daily, and divide
the city into circles of the radius of 0.8 km which is 10-
minute walking distance (3 to 4 miles per hour is typical
for most people. [19]) or 1049 steps using a conversion
factor value 1312.33595801. The centre (approximately.
based on 3E explained in section 7. Algorithm) of each of
these circles serves as a metro station. Thus, the distance
between two adjacent stations comes out to be 1.6 km.
To deal with the inner gaps (which we call “packing left-
outs”) due to circle packing, which increases the station
reachability distance for a person outside the 0.8 km radius
of three adjacent stations, we reduce the diameter to 1.4
km. This is a significant finding backed by existing metro
systems around the globe. According to the data analysis
we performed on these metro systems, the average
distance between two stations is 1.2 km [20].
6. URBAN LAYOUT
We studied various urban layouts and decided to use the
grid plan and fused grid plan. A fused grid is a scaled-up
version of a grid layout. Long, wide streets cut through the
blocks to make transportation and navigation easier. In
addition, each block was shaped into an octagon with
chamfered corners to give the effect of being 'cut off'.
A. Grid Plan
Grid Plan also known as Grid Street Plan, Gridiron Plan is a
type of urban planning technique where each street and
road intersect at right angles thus forming a grid.
 Advantages
o Cost-Effective as the cost is inversely proportional
to the width of the street.
o Maximum land use thus solving the packing
problem.
o Lower vehicle speeds due to frequent intersections
result in fewer accidents.
 Disadvantages
o Uneven landscapes restrict the model.
o Less rainwater retention capacity (high as 30%).

o The frequency of intersections is a problem for
pedestrians, bicycles, and people with physical
limitations.
o Grid is considered to be the least safe with respect
to other layouts.
B. Fused Grid
Fused Grids are a variation of Grid Layout with each grid
larger in size and the grid consisting of cul-de-sacs or
crescents. Each grid forms precincts or quadrants of the
size of 40 acres or 400 m2.
 Advantages
o Increase in walkability compared to the
conventional grid.[22]
o Maximum land use thus solving the packing
problem.
o Neighbourhood streets exhibit low traffic volume
thus falling in noise and air pollution.[23]
o Traffic Congestion and the time delay are low
during peak hours.
 Disadvantages
o Three-way intersection rather than a four-way
intersection, thus being less safe. [24-26]
C. Concentric Zone Model
The Concentric Zone Model is one of the pioneering
models to illustrate urban social structures. The model
includes different zones in concentric circles such as
Business District at the very centre, Factory Zone,
Transition Zone mixed of residential and commercial uses,
better quality middle class homes, and Commuter Zone at
the outer circle of the city.
 Advantages
o One of the simplest models to be designed.
o Stress on economic development.
 Disadvantages
o The model does not fit the polycentric cities.
o Intercity Transportation is difficult.
The project uses data acquired from Google Earth Engine,
implementing a machine learning model for urban layout
classification and Open Street Maps data for existing site
edifices.
A. Geographic Coordinate and Landscape
The very first step includes fetching the landscape
coordinates of the user’s interest. The extraction process is
carried out by Google Earth Engine API [27]. It provides the
location geometry in the form of ee.geometry data type.
Using this we extract the coordinates of the location and its
bounding boxes for further planning and usage. As per the
World Bank data [28] on population growth, we have
considered the global average 2.5 %. This has been
considered the bounding box of the city landscape for
foresight expansion.
B. Examine Existing Edifice
While building a city, there can be existing settlements,
forest areas, water bodies, and industries. For such
situations, we either remove it if it is not contributing to
the city or preserve it if it supports the sustainable mission.
For identifying existing structures, the OSMnx library [29]
comes in handy. OSMnx is a Python library for obtaining
OpenStreetMap data and modeling, projecting, visualizing,
and analyzing real-world street networks and other
geospatial geometries using OpenStreetMap data.
Once data is fetched, we categorize each data variation
separately. These locations are classified into ones that can
be discarded and others that are to be preserved. We
create a city-masked image by omitting those regions with
edifices. This image is essential for Rail Network
Generation (explained ahead).
C. TAGEE
TAGEE [30], an acronym for Terrain Analysis in Google
Earth Engine, helps in calculating terrain attributes such as
slope, aspect, and curvatures, for different directions and
geographical extents. These attributes helped in
understanding the terrain design and structure and its
suitability for efficient urban layout. The landscape is
divided into smaller sections using the Cut Polygon
algorithm. We have taken into consideration the terrain
analysis of the complete landscape and along with it the
terrain analysis of smaller sections of the landscape. Using
this algorithm, we have analyzed 18 major cities in the
world and used the extracted data to train our machine
learning model, explained in the next section.
D. Urban Layout Classification
Urban Layout is a very important aspect of Urbanization.
The urban layout of the landscape area is decided based on
its altitude, geographic conditions, and other terrain-
related attributes using a machine learning model and data
generation using the ResNet model.
The Cut Polygon algorithm explained earlier is also used
for this process as well. For each small geometry of the
landscape, the pre-trained ResNet model [31] is used to
7. ALGORITHM

figure out the urban layout of 18 cities into smaller
sections. The whole data includes around 88 features that
can lead to multidimensionality problems. Thus, we used
PCA or Principal Component Analysis for reducing features
with similar trends and correlations. On applying PCA, 88
features boiled down to 13 features. Next, we applied
different machine learning algorithms to our dataset.
Amongst, Random Forest Classifier proved to be efficient
with an accuracy of 64%.
When the user’s landscape is provided, it will be divided
into small sections with respect to the bounding region of
the landscape. For each section, an urban layout - grid or
radial is predicted using the Random Forest Classifier
discussed earlier. Using the predicted data, an image is
created with colors describing each urban layout. With the
help of the computer vision technique, centroids of each
color are calculated. These centroid values will be passed
on to the next phase of city road generation.
E. Generative Tensor Field XStreamlines
Tensor fields are the basic building blocks of road network
generation. Use of tensor fields is a procedural method for
the generation of roads. A tensor field is a continuous
function that associates a tensor T with every point p = (x,
y) ∈ R2. Eigenvectors decide the curve of the road. Major
and minor eigenvectors are always perpendicular in the
plane. XStreamlines are a very important concept that
describes the curve along with the eigenvector fields. The
frontend UI, City Generator Simulator [32] [33], we
referred to is a city layout generation simulator that
generates imaginary maps with no on-field points and
attributes taken into consideration. It has provided a clear
picture of the project. We have made remarkable changes
as per our needs as our outputs are based on a range of
different parameters as discussed in earlier sections.
As the City Generator Simulator provides a imaginary city
map, we have normalized the whole coordinate system
with respect to the bounding box of the landscape. The
centroid values we have calculated in the last section for
urban layouts are used for center, size, and decay value.
The decay value denotes which centroid will be more
powerful and have more impact on nearby tensor fields
when two or more centroids are in the proximity of each
other.
While researching the project, we came across GCPN,
Graph Convolutional Policy Network [34]. GCPN uses a
graph-based structure to optimize the provided goals with
respect to a given pre-defined set of rules. Though the
GCPN paper has a research study in chemistry and drug
discovery, a similar use case can be applied to road
network generation but due to a very handful of content is
available on Graph Neural Network we are still
understanding the concepts and will apply the same to the
project in the future.
F. Rail Network Generation
Along with the road network, the railway network system
is an important part of urban infrastructure. While
planning the rail network, we encountered a few studies
and derived a value of 1.6 kilometers for the distance
between two stations. Thus, for generating a rail network
we had stationary and fixed nodes and now the challenge
was to connect these nodes. We first had thought of
connecting each node to all other nodes and we’ll be doing
pruning to reduce complexity. Unfortunately, this method
is computationally not viable as in a fully connected graph
of 100 nodes, the edges would be around 5000 and will
increase exponentially. Thus, we come up with a solution of
using graph-based algorithms i.e., Dijkstra’s Algorithm and
MST Algorithm.
Dijkstra’s algorithm is used to find the shortest path
between the source node and the destination node based
on some parameters such as distance or weights.
MST or Minimum Spanning Tree algorithm connects all the
vertices together of the input graph. The solution graph
avoids any cycle in the graph and provides an optimum
solution for our problem.
So, as we have the fixed nodes on the graph, we cluster the
nodes into individual clusters. Now we select points in each
cluster with respect to other clusters based on the
Euclidean distance. So, for n clusters, there can be 1 node
for all (n-1) clusters or (n-1) nodes, each for each cluster.
For inter-cluster connections, we are using Dijkstra’s
algorithm. Now considering the connections within the
cluster, we are applying the MST algorithm for efficient
intra-cluster connections.
Another point to consider here is what if the node is
plotted in place with some existing structure that can’t be
displaced such as a forest or historic place. For such a
situation, we are using the city masked image mentioned in
the Examine Existing Edifice section. This image will help
in deciding whether to consider a point if it is in the core of
the forest or shift that point, if placed on the edge of the
forest or the historic place.
Keeping in mind the future expansion of the city, we have
placed a few stations outside the user’s region of interest
within its bounding box.
While researching rail network generation, we came across
Link Prediction using Graph Neural Network [35]. Link
Prediction is helpful in predicting the edges between nodes
based on parameters such as the number of adjacent
nodes, the weight of nodes, adjacent edges, transitive
edges, etc. We are working towards this by incorporating it
into Urbanization.

7. IMPLEMENTATION
On covering the different algorithms used in Urbanization,
let’s look at the user implementation of it.
1. User Input
Users are provided with 3 different input options such as
latitude-longitude of the location, search bar for the nearby
region, and upload geojson file type of the location.
Figure 2: Urbanization UI
Next user can plot his region of interest on the earth engine
embedded map. Along with the region of interest its
bounding box is also captured and stored.
The coordinates and bounding box values are normalized
between 0 and 1 for each migration to other sections of the
project.
Figure 3: User Input on Earth Engine Map
2. 3E - Examine Existing Edifice
Looking for existing settlements or structures and creating
a city-masked image required for Rail Network Generation.
Figure 4: City-Masked Image representing 3E
The white color indicates some kind of settlement, high
altitude terrain, waterbody, or forest area. While the black
region is one where we are going to plot our city.
3. TAGEE and Urban Layout Classification
Figure 5: Urban Layouts Heatmap
Now for the bounding box, the terrain details and
attributes are extracted using TAGEE. This data is further
used for predicting urban layout. Ahead we are generating
an image describing urban layout distribution. Orange,
blue, and green denote grid, organic and radial layouts.
Centroids of each layout are stored in JSON data type.
4. GTX - Generative Tensor Field XStreamlines
As discussed in the earlier section about GTX, it plots a
landscape wide tensor field as shown in the figure. The red
dots on the tensor field are the centroids calculated for
urban layout prediction. These dots will form the center for
each layout and can influence the tensor fields in its
proximity.
Figure 6: Tensor Field with urban layout centroids

5. Optimized Blueprint Cartography / Output
Optimized based on different parameters taken into
consideration right from the beginning, the following city
blueprint is generated using Urbanization. It has included
all types of buildings, major and minor roads, parks and
forest, and water bodies with appropriate and efficient
road designing and building placement.
Figure 7: City Blueprint generated using Urbanization
6. Rail Network Generation / Output
Figure 8: Nodes with their centroids for rail network
As explained earlier in Algorithms section, we have plotted
the centroid of each cluster. Based on these centroids
Dijkstra and MST algorithms are applied for the final
result presented below. Currently we have demonstrated
the algorithm on a example set in prototype phase and we
are working on its full-scale deployment into
Urbanization.
Figure 9: Rail Network Generation in Prototype Phase
8. CONCLUSIONS
To summarise the findings of the study, the problem of
unplanned city transits is a huge setback for the urban
cities, and it must be addressed as soon as possible.
Urbanization is an online application that aids the urban
planners to design and plan transit systems of city with
more sustainability. Urbanization - Smart City Smart
Planner employs Generative Tensor field Xstreamlines
(GTX). Urban Planners using Urbanization, provide the
geographical area. The data is sent on to the GIS for area
specifications and GTX for road planning and micro blocks
designing. Taking into account all the elements stated
above regarding technology implementation and
ecologically soundness, Urbanization builds a blueprint for
the region provided and suggests some guidelines that can
benefit the planner. Urbanization does not replace the role
of urban planner but instead it aims at reducing the burden
and designing more sustainably.
REFERENCES
[1] https://www.mckinsey.com/~/media/McKinsey/Bus
iness%20Functions/Sustainability/Our%20Insights/
Elements%20of%20success%20Urban%20transporta
tion%20systems%20of%2024%20global%20cities/U
rban-transportation-systems_e-versions.ashx
[2] World Conference on Transport Research - WCTR
2016 Shanghai. 10-15 July 2016 Determining Optimal
Urban Expansion, Population and Vehicle Density, and
Housing Types for Rapidly Growing Cities Todd
Litman https://www.vtpi.org/WCTR_OC.pdf
[3] https://www.thelancet.com/journals/lanplh/article/
PIIS2542-5196(17)30119-5/fulltext
[4] Urban population growth (annual %) | Data
(worldbank.org)
[5] https://www.worldbank.org/en/topic/urbandevelop
ment/overview#1

[6] Climate change and health (who.int)
[7] Sources of Greenhouse Gas Emissions | US EPA
[8] https://health.gov/sites/default/files/2019-
09/paguide.pdf
[9] World Health Organization (2010). Urban Planning,
Environment and Health: From Evidence to Policy
Action. From
http://www.euro.who.int/__data/assets/pdf_file/000
4/114448/E93987.pdf?ua=1. Accessed on April 22,
2016.
[10] Morar, T., Radoslav, R., Spiridon, L. C., & Păcurar, L.
(2014). Assessing pedestrian accessibility to green
space using GIS.
[11] Takano, T., Nakamura, K., & Watanabe, M. (2002).
Urban residential environments and senior citizens’
longevity in megacity areas: the importance of
walkable green spaces. Journal of epidemiology and
community health, 56(12), pp. 913–918.
[12] Frumkin, H. (2003). Healthy places: exploring the
evidence. American journal of public health, 93(9), pp.
1451–1456.
[13] Singh, V. S., Pandey, D. N., & Chaudhry, P. (2010).
Urban forests and open green spaces: Lessons for
Jaipur, Rajasthan, India. RSPCB Occasional Paper, 1,
pp. 1–23.
[14] Jim, C. Y., & Chen, S. S. (2003). Comprehensive green
space planning based on landscape ecology principles
in compact Nanjing city, China. Landscape and Urban
Planning, 65(3), pp. 95–116
[15] Baharash, Bagherian (2015). Liveable Cities: Greening
for Success from
https://www.linkedin.com/pulse/liveable-cities-
greening-success-baharash -bagherian?trk=pulse-det-
nav_art. Accessed on March 20, 2015.
[16] Healthy Parks Healthy People Central (Unspecified).
Urban planning and the importance of green space in
cities to human and environmental health. From
http://www.hphpcentral.com/article/urban-
planning-and-theimportance-of-green-space-in-cities-
to-human-and-environmental-health. Accessed on July
20, 2016. Healthy Parks Healthy People Central
(Unspecified). Urban planning and the importance of
green space in cities to human and environmental
health. From
http://www.hphpcentral.com/article/urban-
planning-and-theimportance-of-green-space-in-cities-
to-human-and-environmental-health. Accessed on July
20, 2016.
[17] https://en.wikipedia.org/wiki/Oslo
[18] TED Talk -
https://www.youtube.com/watch?v=IFjD3NMv6Kw
[19] https://www.healthline.com/health/exercise-
fitness/average-walking-speed
[20] http://mic-ro.com/metro/table.html
[21] Jane McIntosh, The Ancient Indus Valley: New
Perspectives; ABC-CLIO, 2008; ISBN 978-1-57607-
907-2 ; pp. 231, 346.
[22] Sun, J. & Lovegrove, G. (2009). Research Study on
Evaluating the Level of Safety of the Fused Grid Road
Pattern, External Research Project for CMHC, Ottawa,
Ontario
[23] IBI Group 2007 – Comparing Current and Fused Grid
Neighbourhood Layouts for mobility, infrastructure
and emissions costs, Canada Mortgage and Housing
Corporation
[24] Dumbaugh, Eric; Rae, Robert (2009). "Safe Urban
Form: Revisiting the Relationship Between
Community Design and Traffic Safety". Journal of the
American Planning Association. 75 (3): 309–329.
[25] Marks, Harold (1957). "Subdividing for traffic safety".
Traffic Quarterly. 11 (3): 308–325.
[26] Lovegrove, Gordon R; Sayed, Tarek (1 May 2006).
"Macro-level collision prediction models for
evaluating neighbourhood traffic safety". Canadian
Journal of Civil Engineering. 33 (5): 609–621.
[27] https://earthengine.google.com/
[28] https://data.worldbank.org/indicator/SP.URB.TOTL.I
N.ZS
[29] Boeing, Geoff. (2017). OSMNX: New Methods for
Acquiring, Constructing, Analyzing, and Visualizing
Complex Street Networks. Computers Environment
and Urban Systems. 65. 126-139.
10.1016/j.compenvurbsys.2017.05.004.
[30] Safanelli, J.L.; Poppiel, R.R.; Ruiz, L.F.C.; Bonfatti, B.R.;
Mello, F.A.O.; Rizzo, R.; Demattê, J.A.M. Terrain
Analysis in Google Earth Engine: A Method Adapted
for High-Performance Global-Scale Analysis. ISPRS Int.
J. Geo-Inf. 2020, 9, 400. DOI:
https://doi.org/10.3390/ijgi9060400
[31] Chen W, Wu AN, Biljecki F (2021): Classification of
Urban Morphology with Deep Learning: Application
on Urban Vitality. Computers, Environment and Urban
Systems 90: 101706

[32] Guoning Chen, Gregory Esch, Peter Wonka, Pascal
Müller, and Eugene Zhang. 2008. Interactive
procedural street modeling. ACM Trans. Graph. 27, 3
(August 2008), 1–10.
DOI:https://doi.org/10.1145/1360612.1360702
[33] https://probabletrain.itch.io/city-generator
[34] You, Jiaxuan, Bowen Liu, Zhitao Ying, Vijay Pande, and
Jure Leskovec. "Graph convolutional policy network
for goal-directed molecular graph generation."
Advances in neural information processing systems
31 (2018).
[35] Muhan Zhang and Yixin Chen. 2018. Link prediction
based on graph neural networks. In Proceedings of the
32nd International Conference on Neural Information
Processing Systems (NIPS'18). Curran Associates Inc.,
Red Hook, NY, USA, 5171–5181
DILIP JAIN
BE Computer Engineering, SIGCE
SIDDHESH KADAM
ADITYA MAHIMKAR
PROF. VAISHALI YEOLE
Computer Engineering Dept, SIGCE
BIOGRAPHIES

Urbanization: Smart Urban Transit System using Artificial Intelligence

More Related Content

Similar to Urbanization: Smart Urban Transit System using Artificial Intelligence (20)

More from IRJET Journal (20)

Recently uploaded (20)

Urbanization: Smart Urban Transit System using Artificial Intelligence