sustainability

Article
Predictive Machine Learning Algorithms for Metro Ridership
Based on Urban Land Use Policies in Support of
Transit-Oriented Development
Aya Hasan AlKhereibi 1 , Tadesse G. Wakjira 2 , Murat Kucukvar 1 and Nuri C. Onat 3, *

1 Industrial and Systems Engineering, College of Engineering, Qatar University, Doha P.O. Box 2713, Qatar
2 Civil and Architectural Engineering, College of Engineering, Qatar University, Doha P.O. Box 2713, Qatar
3 Qatar Transportation and Traffic Safety Center, College of Engineering, Qatar University,
Doha P.O. Box 2713, Qatar
* Correspondence: [email protected]

Abstract: The endeavors toward sustainable transportation systems are a key concern for planners
and decision-makers where increasing public transport attractiveness is essential. In this paper, a
machine-learning-based predictive modeling approach is proposed for metro ridership prediction,
considering the built environment around the stations; it is in the best interest of sustainable transport
planning to ultimately contribute to the achievement of Sustainable Development Goals (UN-SDGs).
A total of twelve parameters are considered as input features including time of day, day of the
week, station, and nine types of land use density. Hence, a time-series database is used for model
development and testing. Several machine learning (ML) models were evaluated for their predic-
tive performance: ridge regression, lasso regression, elastic net, k-nearest neighbor, support vector
regression, decision tree, random forest, extremely randomized trees, adaptive boosting, gradient
boosting, extreme gradient boosting, and stacking ensemble learner. Bayesian optimization and grid
search are combined with 10-fold cross-validation to tune the hyperparameters of each model. The
performance of the developed models was validated based on the test dataset using five quantita-
tive performance measures. The results demonstrated that, among the base learners, the decision
Citation: AlKhereibi, A.H.; Wakjira,
tree showed the highest performance with an R2 of 87.4% on the test dataset. KNN and SVR were
T.G.; Kucukvar, M.; Onat, N.C.
the second and third-best models among the base learners. Furthermore, the feature importance
Predictive Machine Learning
investigation explains the relative contribution of each type of land use density to the prediction of the
Algorithms for Metro Ridership
metro ridership. The results showed that governmental land use density, educational facilities land
Based on Urban Land Use Policies in
Support of Transit-Oriented
use density, and mixed-use density are the three factors that play the most critical role in determining
Development. Sustainability 2023, 15, total ridership. The outcomes of this research could be of great help to the decision-making process
1718. https://doi.org/10.3390/ for the best achievement of sustainable development goals in relation to sustainable transport and
su15021718 land use.

Academic Editor: Jingxu Chen

Keywords: sustainable transportation; metro ridership; time series models; machine learning; urban
Received: 9 November 2022 planning; land use policy; sustainable development
Revised: 18 December 2022
Accepted: 22 December 2022
Published: 16 January 2023

1. Introduction
Sustainable transport includes the application of the sustainable development con-
Copyright: © 2023 by the authors.
cept in the process of planning and development of transport infrastructure. Refs. [1–3]
Licensee MDPI, Basel, Switzerland. suggested that sustainable transportation systems evolve the application of sustainable
This article is an open access article development planning strategies; for instance, transportation sustainability is defined as
distributed under the terms and “meeting the current mobility and transport needs nevertheless not compromising the
conditions of the Creative Commons future generations to meet those needs”. According to OECD, Taiwan Transport Institute,
Attribution (CC BY) license (https:// and ref. [4], sustainable transport must be able to meet long-term and simultaneous envi-
creativecommons.org/licenses/by/ ronmental, social, and economic needs and impacts. There are three aspects of sustainable
4.0/). transport: environmental, social, and economic. Environmental aspect requires taking

Sustainability 2023, 15, 1718. https://doi.org/10.3390/su15021718 https://www.mdpi.com/journal/sustainability

Sustainability 2023, 15, 1718 2 of 20

into account the external effects of the transport process; the social aspect also requires
considering the interests of residents at different socio-economic levels while improving
transport; economic aspects require efficient use and conservation of resources [1].
In this respect, to sustain the public transportation resources, planners should address
the transportation systems operations with respect to the built environment around sta-
tions; in other words, they are expected to assure the highest efficiency of the transport
service by increasing transport accessibility to as many residents as possible [4]. The trade-
offs between transportation operation needs and built environment considerations could
easily be solved by having high-density urban areas centered on transit service, yet an
acceptable economic aspect should be maintained [5]. While the socio-economic attribute
of sustainable development can be improved through transport infrastructure through
logistics and multimodal capacity planning, intra-regional accessibility of the transport
infrastructure promotes economic dimensions and other social interactions. Accessibility
plays an undeniable role in urban areas and also ensures infrastructure efficiency combined
with the multimodality of public transport [2].
With the rapid development of urbanization, the metro has become one of the main
drivers of public transport due to its various advantages, such as high efficiency, high capac-
ity, convenience, etc. [1]. During the planning phase of urban transport and construction,
it is important to examine many urban indicators from a systemic point of view to cover
urban-transport phenomena, such as density, ridership, and accessibility. Metro ridership
at the station level is an important factor in determining the size of stations and access
to facilities. Various components of the urban system (e.g., land use and socioeconomic
aspects) require the kind of modeling presented in this paper, which could be of help to
accurately estimate and predict the number of passengers, as well as analyze influencing
factors. Recently, with the development of artificial intelligence and computational capa-
bilities, machine learning (ML) techniques have gained considerable attention owing to
their ability to effectively determine the relationship between the response variable and
its predictors in a complex system [6]. In spite of their great efficacy, the literature lacks
the application of machine learning models to the metro transportation system ridership,
considering the built environment.
To this end, this study mainly focuses on metro transportation system ridership on
a station level, and the land use density component of the urban system. The research
contributes to both theory and practice as follows. The proposed methods examined
in this research enable a time series prediction of metro ridership, considering the built
environment and transportation sustainability. On the practical aspect, the examined model
outcomes are applied to the case of Doha Metro and could be similarly applied to similar
regions and cases. Furthermore, the outcomes of this research have numerous implications
for both transportation operation and urban planning policymakers, which significantly
bridge the gaps between theory and practical urban-transport models.

1.1. Public Transportation Sustainability

Integrated, strategic, and society-supported policies are required for the process of
shifting toward sustainable transportation systems and behavior [7–9]. As per refs. [10,11],
sustainable transportation measures do not seem to be restricted to mobility standards
wherever the majority of transportation studies comprise. Sustainable transportation has to
be thought of from an exceedingly additional holistic vision, therefore social, environmental,
and economic impacts of high vehicle dependency as a transportation mode alternative
may be given [3]. The study by [12] highlighted the importance of a holistic method that
comprises institutional reforms, changes in land use (urban fabric), and economic incentives
over individual technology solutions (vehicle-based) with a distorted perspective to achieve
sustainable transport goals. It is well known that urban transport must play a central
role in the development of today’s urban structures that occupy huge areas and require
excessive travel to meet basic needs, worker travel, etc. Qatar has very limited expertise
with public transport and transit-oriented systems, resulting in additional resistance to
Sustainability 2023, 15, 1718 3 of 20

changes from a vehicle-oriented society to a TOD [13]. Moreover, ref. [14] in their study
examined the spatiotemporal heterogeneity in the nonlinear influence of transit-oriented
development (TOD) on metro ridership. The findings have led to a better understanding of
the spatiotemporal heterogeneity in the nonlinear influence of TOD on metro ridership.
In addition to the current macro-level literature, a number of survey-based studies
have provided overlapping results, as they have recognized anomalies in the existing
paradigm. Ref. [15] pointed out that the influences of travel time and financial costs on
modality alternatives are independent of land use influences. In addition to the infrastruc-
ture of the city, demographic information, and the choice of transportation mode could be
a matter of higher cognitive processing on the part of citizens, and this decision is plagued
by psychological and emotional patterns. Ref. [2] accepts this as true, with ref. [3] thereon
stating that urban fabric and transportation interactions may be accomplished by observing
past tendencies in the urban fabric and a stripped-down upsurge in the ridership of public
transportation. Despite the inflated federal investments in public transportation in several
societies, such as the state of Qatar, the shared idea, implicit assumptions, and perceptions
push right up against public transportation participation.

1.2. Transport-Related Sustainable Development Goals (SDGs)

According to Sustainable Development Agenda 2030, sustainable transport systems
offer the world access to sustainable, reliable, affordable, and up-to-date energy facilities,
resilient and quality infrastructure, and strategies that provide highly helpful capacities that
can strengthen the foundations of the country’s economy [16]. However, twelve targets are
enclosed within the 2030 Agenda for sustainable development, which is associated with
transportation (Figure 1), and five of them are directly associated with transportation, whereas
seven of them are indirectly associated with it. Moreover, the five targets that are directly
associated with transportation are road safety (Target 3.6), energy efficiency (Target 7.3), sus-
tainable infrastructure (Target 9.1), urban access (Target 11.2), and fuel subsidies (Target 12.c).
However, this highlights that sustainable transport is critical to accelerating the accom-
plishment of sustainable development goals. In other words, the common relationship
between land use and transportation is seen in (SDG 9, 11, and 12). For example, travel
demand is stricken by land use, whereas the patterns of land use are conspicuously wedged
by transportation networks [17]. Thus, the connection between land use and transporta-
tion should be thought of in addressing urban planning efficiently, regarding sustainable
development [17–19]. Transportation and land use have usually had a major role in the physi-
cal and economic development of contemporary cities. However, sustainable development
advantages urban planners in supporting environmental, social, and economic goals, and
managing infrastructures intellectually within the town [20,21]. Thus, for the exact purpose of
accomplishing sustainable urban planning in cities, some approaches are established, similar
to the Transit-Oriented Development (TOD) planning approach, which has been evidenced to
be quite flourishing among varied existing projects [22].

1.3. Ridership Issues for Public Transportation

Considerable research work has been carried out with the aim of increasing the use of
public transport; ref. [1,7] stated that, typically, seven key issues need to be addressed for
transport to be sustainable in urban:
• Congestion in many urban areas is becoming more prolonged and intense. Urban
speeds decline by 5% per decade, and congestion increases with city size [23].
• Increased air pollution has led to exceeding the national standards of air quality
recommended by the (WHO, 2021) in many cities. Air pollution effect has a significant
effect on health, reduces visibility, and destroys local structures and ecosystems,
reducing the quality of urban life.
• Traffic noise has an impact on the entire life of the city. Ref. [24] estimated that about
15% of people in developed countries are most likely to be exposed to noise pollution,
mainly from vehicles, trucks, and overnight deliveries.
Sustainability 2023, 15, 1718 4 of 20

• Degradation of the urban landscape due to new road construction and new vehicles on
the roads, the historic buildings demolition, the reduction of open spaces, and urban
sprawl [7].
• Space use by a vehicle enables the movement of car drivers but decreases other
accessibility, as traffic routes turned out to be obstacles due to the fact that parked cars
create difficulties for pedestrians, cyclists, and people who have disabilities, which
leads to a predominance of traffic in urban environments.
• Global warming is the result of the usage of fossil fuels. Currently, transportation
systems account for 25% of CO2 emissions, and this level is increasing in both com-
parative and absolute terms. Moreover, transportation is entirely dependent on fuel,
which is a non-renewable source of energy.
• Furthermore, transportation contributes to change in the urban fabric of the city, thus,
development and land use factors should be added to the list above.
• Vehicle use has contributed to the city’s decentralization, combined with efficient
public transport. This significantly increases commuting time and the development of
more dispersed movement patterns with a center in the city center, thereby increasing
dependence on cars and reducing opportunities for developing an efficient public
transport community; transport has become a driving force of change.
• Industry displacement and globalization (for instance, the information economy)
have led to newfangled patterns of distribution and an increase in freight traffic
at the global, regional, and local levels. Technologies and solutions that could be
of help in promoting the relatively most efficient use of space and eliminating the
total additional land allocated for the developments are highly needed. There is
a sort of common agreement on these issues, and largely the variety of strategies
existing is known, but progress has been made, and it’s slow to integrate sustainability
into everyday solutions.
• With the intensive transformation of the built environment in the state of Qatar from a
traditional mixed-use, high-density fabric to inaccessible, car-oriented superblocks of
the newly developed cities and closed communities on the periphery, urban mobility
is noticeably affected. Along with the use of retail/commercial and office space, they
benefit greatly from the country’s sustainability [25]. Undoubtedly, based on the
Transit Oriented Development (TOD) consensus, metro transport in the state of Qatar
promises rapid and ambitious growth and is expected to become one of the main
methods of connecting different districts. TOD is an auspicious tool that enhances
greener mobility by switching from cars to the metro, encouraging short-distance
walking, and using local public transport for long distances [25], and it can slow
sprawl through the compact development of urban areas [26,27]. With the motivation
to promote rail transport systems, studies looking at the factors influencing the number
of passengers in transit have become interesting, but the impact of local communities
on the number of passengers in transit is surprisingly different from the various
contexts of the city.
a. Machine Learning-Based Ridership Prediction for Public Transportation
In general, optimization models are used for the planning of the transit network, as
well as the optimum transfer, the orientation of the route, and, therefore, the ridership [28].
Ref. [29] outlined an empirically supported genetic algorithm to enhance the performance
of the present networks by plummeting the vehicles while not penalizing the typical travel
time. In this sense, it’s been shown that heuristic algorithms are helpful in resolving
numerous transport problems; for example, the matter of the road trafficker, the routing
of multi-depot vehicles, or the shuttle network design [30–32], to simply call a couple of
examples. Ref. [33] planned a computational tool for optimizing massive routes of transit
networks that minimize transfers and optimize route openness through improved coverage
of service. They tested the tool and functioned it to a significant and accurate network
optimization drawback to the city of Miami. Ref. [34] investigated a three-tier hierarchic
optimization method for resolving the mass transit network development problems that are
 Sustainability 2023, 15, 1718 5 of 20

typical of versatile large-scale mass transit choices and the exploitation of new technologies
for passenger-vehicle communication. Ref. [35] deliberated on accessibility, which is a vital
component of service delivery. Access sometimes has an approximate service associated
with the cost, while access is expounded to the adequacy of the transit system to induce
individuals from wherever they approach to where they are in an inexpensive period. In
Australia, ref. [36] established a hybrid coverage model to concurrently expand access to
services and, moreover, increase public transportation accessibility. The scholar suggested
that the operation aspect of transport planning is the spatial effects of the service. Ref. [37]
estimated rural mass transit traveler models that reply to public demand employing types
Sustainability 2023, 15, x FOR PEER REVIEW 4 of 22
of variables and derived helpful data for correct planning transport services that respond
to local demand.

Figure 1. Transport-related sustainable development goals.

Figure 1. Transport-related sustainable development goals.

1.3. Ridership Issues

Ref. [38] for Public
addressed Transportation
the requirement to quantify the trade-off amongst ridership and
theConsiderable
way to enableresearch
public transport
work has planning and choices
been carried regarding
out with the aim theofbalance between
increasing the use
these priorities. This becomes a lot of vital given the participation of non-experts
of public transport; ref. [1,7] stated that, typically, seven key issues need to be addressed in
the democratic planning of transportation
for transport to be sustainable in urban: [39,40]. In such a sense, the incorporation of
people participation, quantitative analysis, and GIS methods, particularly Multi-Criteria
• Decision
Congestion
Analysisin (MCDA),
many urban areastoisbebecoming
is proving more prolonged
useful in promoting and intense.
joint and technically Urban
sound
speeds decline by 5% per decade, and congestion increases with city
decisions [28]. In addition, though newer models are utilized to support the planning and size [23].
• design
Increased air DRST
of recent pollution has [32,41],
services led to exceeding the national
there’s an absence standards
of special analyzes ofand
air quality
ways to rec-
ommended
integrate typicalby the (WHO,
regular 2021) in many
public transportation cities.
with Airon-demand
flexible pollution effect hasThis
services. a significant
might
beeffect
taken on
intohealth,
consideration as a primary stage in the planning of integrated,
reduces visibility, and destroys local structures and ecosystems, mounted, and re-
versatile transport services supported
ducing the quality of urban life. by indicators of accessibility and social justice. It
• ought to be
Traffic noted
noise that
has analthough
impact onaccessibility
the entireandlifeequity
of thedo notRef.
city. embrace all of the essential
[24] estimated that about
operation and design features of public transportation (e.g., economical and operative
15% of people in developed countries are most likely to be exposed to noise pollution,
public transportation intermeshed towards sustainability) [42].
mainly from vehicles, trucks, and overnight deliveries.
• Degradation of the urban landscape due to new road construction and new vehicles
on the roads, the historic buildings demolition, the reduction of open spaces, and
urban sprawl [7].
• Space use by a vehicle enables the movement of car drivers but decreases other ac-
cessibility, as traffic routes turned out to be obstacles due to the fact that parked cars
Sustainability 2023, 15, 1718 6 of 20

b. Research Motivation and Knowledge Gap

Referring to recent studies discussed in Section 1.3(a), the previous studies have
approached the ridership problem mainly from a purely operational perspective, utilizing
several optimizations and regression models. To the best extent of the author’s knowledge,
no previous study has investigated the public transport ridership problem using urban
fabric using machine learning-enhanced models. Thus, the current research is aimed to
set up a comprehensive understanding of the public transportation operation considering
built environment characteristics toward a Transit-Oriented Development, ending with the
SDGs agenda achievement. A time-series data was utilized to examine simple machine
learning techniques that may or may not capture the relationship between the response
(ridership) and its predictors (land use densities), which could dictate the use of advanced
techniques. Overall, the research attempts to develop machine learning-enhanced models
that have the ability to accurately predict metro ridership concerning the built environment
around the stations.

2. Methods
2.1. Data Collection
The State of Qatar is located on the east coast of the Persian Arabian Gulf. It is
connected to the Arabian Peninsula and borders Saudi Arabia to the south. Qatar has
been shaped by rapid globalization and urbanization. For the past five decades, notable
economic change from a fishing and pearl collecting-based economy to a thriving and
varied economy based on the production and export of natural gas/oil has been observed.
Today, Qatar has almost 13% of the total international supply of natural gas reserves [43].
The country’s growth and hosting of several mega-events, such as the 2006 Asian Games
and the 2022 World Cup, led to intensive urban growth and development. The country
witnessed a significant increase in population over the past two decades, from a projected
urban population of 492 hundred thousand in 2000 to more than 2.4 million people in
2020. Consequently, the government confronted major challenges in the management of
the growth, transport, infrastructure, accommodations, and preservation of the ecosystems.
The changes included the construction of a sophisticated road network, including
road extensions and a ring road/expressway system. The traditional low-rise housing
was almost abandoned for modern residential villas, typically three-story and high-rise
apartments [44–46]. In parallel with its rapid urbanism growth. Sustainable development
of transport systems is a core diminution in Qatar National Vision (QNV) 2030, which
includes plans to develop a 300 km metro system with four lines and 98 stations connecting
the international airport, stadiums, and urban areas [47]. In 2019, three lines and 38 stations
are in operation.

2.2. Data Preprocessing

Figure 2 outlines a simplified procedure followed in this study. Firstly, a database
of ridership is collected and pre-processed. Time-series data for metro ridership were
collected from 38 different metro stations in the State of Qatar. The database was then
randomly divided into train and test sets that consist of 80% and 20% of the entire dataset,
respectively, as shown in Figure 2. The train set was used for the development and
validation of the models, while the test set was used for the final appraisal of the models.
A total of twelve white box and black box machine learning models were run to examine
their predictive capabilities for the given problem. Furthermore, five different statistical
performance metrics have been used for the evaluation of the predictive performance of
the ML models and select the best-performing model among the twelve models considered
in this study (Figure 2). Figure 3 demonstrates the total ridership distribution on weekdays
and weekends. As shown in Figure 3, ridership during the weekdays is significantly higher
than that for the weekend from 05:00 AM to 9:00 AM. Particularly, the ridership on Friday
(the first weekend day in the State of Qatar) is zero between 05:00 AM and 12:00:00 AM;
however, it was higher than on all other days after 3:00 PM (Figure 3). The peak ridership
 performance metrics have been used for the evaluation of the predictive performance of
the ML models and select the best-performing model among the twelve models consid-
ered in this study (Figure 2). Figure 3 demonstrates the total ridership distribution on
weekdays and weekends. As shown in Figure 3, ridership during the weekdays is signif-
icantly higher than that for the weekend from 05:00 AM to 9:00 AM. Particularly, the rid-
Sustainability 2023, 15, 1718 7 of 20
ership on Friday (the first weekend day in the State of Qatar) is zero between 05:00 AM
and 12:00:00 AM; however, it was higher than on all other days after 3:00 PM (Figure 3).
The peak ridership on each day was observed between 05:00 to 6:00 PM, after which it
on each day
declines was
to less observed
than 10,000 between 05:00
at 11:00 PM onto 6:00 PM,and
Thursday after whichand
Friday it declines
at 10:00 to
PM less
onthan
Sat-
10,000 at 11:00 PM
urday and weekdays. on Thursday and Friday and at 10:00 PM on Saturday and weekdays.

bility 2023, 15, x FOR PEER REVIEW 8 of 21

Figure 2. Research Framework.
Figure 2. Research Framework.

Figure 3. Distribution of ridership on weekdays and weekends.

Figure 3. Distribution of ridership on weekdays and weekends.
The input features comprised a total of twelve parameters including time of day, day
The input
of the features comprised
week, station, and landa use
total of twelve
density for openparameters
space andincluding time of (indoor-
recreation facilities day, day
outdoor),
of the week, religious
station, facilities,
and land educationfor
use density facilities, retail/commercial,
open space residential,
and recreation special
facilities use,
(indoor-
transportation, and government facilities. The distribution of the ridership
outdoor), religious facilities, education facilities, retail/commercial, residential, special for the complete
use, transportation, and government facilities. The distribution of the ridership for the
complete dataset is shown in Figure 4 regardless of the land-use density. During the pre-
processing stage, the database shown in Figure 4 is normalized based on Equation (1) be-
low:
 Figure 3. Distribution of ridership on weekdays and weekends.

The input features comprised a total of twelve parameters including time of day, day
Sustainability 2023, 15, 1718 of the week, station, and land use density for open space and recreation facilities (indoor-
8 of 20
outdoor), religious facilities, education facilities, retail/commercial, residential, special
use, transportation, and government facilities. The distribution of the ridership for the
complete dataset is shown in Figure 4 regardless of the land-use density. During the pre-
dataset is shown in Figure 4 regardless of the land-use density. During the pre-processing
processing stage, the database shown in Figure 4 is normalized based on Equation (1) be-
stage, the database shown in Figure 4 is normalized based on Equation (1) below:
low:
j j
j Xi − Xmin
Xi, n = ,j = j
= 1,
, j ,= 1, 2, 2,. .…. ,, M;; i =
=1,1,2,2,…., . . , N (1)
(1)
Xmax − Xmin
where is the ith observation of jth input features and , is its corresponding nor-
j j
where X is
malizedi value,the ith observation
and of jth
areinput features and
the maximum Xi,minimum
and n is its corresponding
values of thenormalized
th input
j j
features,
value, Xmax and represents thethe
Xmin are total observation
maximum andnumbers,
minimum andvaluesisof the number
the of input
jth input fea-
features,
N represents the total observation numbers, and M is the number of input features.
tures.

Figure 4. Ridership
Figure 4. Ridership data
data was
was used
used for
for model
model development.
development.

Figure
Figure 55 represents
represents the the diversity
diversity of
of land
land use
use within
within the
the 800
800 mm catchment
catchment area
area at
at the
the
38
38 metro
metro stations.
stations. ItItisisclear
clearfrom
fromthis
thisfigure
figurethat
thatthe
themajority
majorityofof
the land
the use
land in in
use this area
this areais
High-Density Residential land use with a wide mix of offices, commercial, and recreation
zones. The developed model utilized the data within the catchment areas at each station to
predict the ridership at each metro station.

2.3. Predictive Models

2.3.1. Regularization
A regularization is an extended form of linear regression with an aim to enhance
the generalization ability of the model. In contrast to linear regression, which computes
the parameters estimates by minimizing the RMSE of the predictions, a regularization
parameter that punishes the models with multiple model parameters is introduced in
regularization, as illustrated below:

min( RMSE + ε(λ, β)) (2)

βi

where λ controls the trade-off between bias-variance.

Ridge regression (RR), elastic net (EN), and lasso (short for least absolute shrinkage
and selection operator) regression are the foremost usually used regularization algorithms.
In ridge regression, the regularization function is taken into account because the total of
the sq. of the constants, ε(λ, β) = λ ∑i β2i , whereas the absolute values of the coefficient
are employed in lasso regression, ε(λ, β) = λ ∑i | β i |. The penalty in lasso regression tends
to set the model parameters to zero, while RR tends to scale back the absolute values
 Sustainability 2023, 15, x FOR PEER REVIEW 9 of 21
Sustainability 2023, 15, 1718 9 of 20

is High-Density Residential land use with a wide mix of offices, commercial, and recrea-
oftion
each parameter
zones. of the models.
The developed The penalty
model utilized function
the data within thein the elastic
catchment netatiseach
areas given
sta- by:
ε(tion
λ, βto
)=predict β2i +
λ1 ∑ithe λ2 ∑i | β iat
ridership |. each metro station.

Figure 5. Land use in the stations’ catchment area.

Figure 5. Land use in the stations’ catchment area.

2.3. Predictive
2.3.2. K-Nearest Models
Neighbor
2.3.1. Regularization
The K-nearest neighbor (KNN for short) is among the nonparametric supervised ML
algorithms that can beis used
A regularization to solveform
an extended classification as well as
of linear regression regression
with an aim toproblems.
enhance theThe
predicted outputability
generalization in KNN is evaluated
of the model. Inas the mean
contrast of the regression,
to linear predictionswhich from the K data points
computes the
parameters
nearest to theestimates by minimizing
query point. Provided athe RMSE data
training of theset {( xi , yi )}aiN=regularization
predictions, 1 and a query param-
point Xq ,
eter
the that punishes
K-nearest the models
neighbor entails with multiple modelofparameters
the determination the pointsisinintroduced
the trainingin dataset
regulari-that
zation,
are as illustrated
nearest below:
to Xq [2]. The prediction of Xq is then computed as the weighted average of the
predictions from the K observations nearest to+Xq(. , )
min (2)
2.3.3.
whereSupport Vector
controls the Regression
trade-off between bias-variance.
Support vector regression
Ridge regression (SVR)
(RR), elastic netutilizes
(EN), andkernel functions
lasso (short fortoleast
mapabsolute
the datashrinkage
into a wide-
and selectionspace,
dimensional operator)
where regression are the foremost
linear separation usuallyDProvided
is conceivable. used regularization algo- set
aEtraining data
{( xi , yi )} i=1 e R × R, with N data points, where xi e X = xi , . . . , xi ⊆ R arethe
rithms. InNridge regression,
Q the regularization function is taken into
1 account
Q because
Q input
total of the sq. of the constants, ( , ) = ∑ , whereas the absolute values of the coef-
parameters Q and yi is the response, SVR the function f(x) in Equation (3).
ficient are employed in lasso regression, ( , ) = ∑ | |. The penalty in lasso regression
tends to set the model parameters to 0zero,
f ( x ) = ∑ αi − αi K ( xwhile RR tends to scale back0 the absolute values


i , x ) + b subject to αi , αi e [0, C ] (3)
of each parameter of the models.
ie SV The penalty function in the elastic net is given by:
( , )= ∑ + ∑ | |.
where C denotes the regularization parameter, K ( xi , x ) is the kernel function, SV represents
support vectors, b is the bias, and αi and αi0 are multipliers of Lagrangians of the inferior
and superior SVs.
Sustainability 2023, 15, 1718 10 of 20

2.3.4. Decision Tree

The decision tree, also known as the classification and regression tree (CART), itera-
tively splits the feature space into D disjoint and distinct regions: { R1 , R2 , . . . , R D } based
on different measures, such as the mean squared error Gini Index [48] (L. Breiman et al.,
1984) for regression and classification problems, respectively. The decision tree algorithm
comprises three main components: (a) the topmost node referred to as the root node,
(b) intermediate or internal nodes, and (c) the leaf or terminal node. Hence, the optimum
decision tree model can be found by tuning the maximum depth of the tree, the mini-
mum number of samples needed to split the internal node, and the minimum sample
number required to be at the leaf node. Bayesian optimization is combined with a cross-
validation technique to find the optimum values of the hyperparameters of each ML model
in this study.
The CART algorithm is easy to visualize and interpret and not sensitive to outliers,
but it has low generalization capability and is associated with high bias and variance. As a
result, an ensemble of several CARTs can be used to enhance its generalization ability, as
presented in the subsequent sections.

2.3.5. Bagging Ensemble: Random Forest, Extremely Randomized Trees

Bootstrap aggregation (Bagging) ensemble [49] (Leo Breiman, 2001) is one of the tree-
based ensemble models that aggregate multiple decision tree models in parallel to form
a strong model. Random forest (RF) and extremely randomized trees or extra trees (ET)
are the two commonly utilized examples of bagging ensemble. RF independently trains
T number of decision trees on randomly chosen samples with replacement referred to as
bootstrap samples. The final prediction is then determined as the mean of the prediction
from the T base learners (CART). Extremely randomized trees or extra trees (ET) is an
additional form of bagging ensemble proposed by [50] Geurts et al., (2006). In ET, each base
learner (CART) fits the entire training data set, unlike RF, which uses a bootstrap sample to
construct CART. Similar to RF, the extremely randomized trees algorithm divides nodes
with random subsets of input features, but the ET algorithm arbitrarily chooses an optimal
split [50] (Geurts et al., 2006).
To find optimum RF and ET models, the following hyperparameters should be opti-
mized: (a) the number of estimators or decision trees, (b) the minimum sample leaf, and
(c) the maximum input features number required to split a node.

2.3.6. Boosting Ensemble: Adaptive Boosting, Gradient Boosting Trees, xgBoost

Boosting ensemble sequentially aggregates several base learners, typically CART in
order to form a strong model. Adaptive Boosting (AdaBoost), gradient boosting trees (GBT),
and extreme gradient boosting (xgBoost) are the three commonly used types of boosting
ensemble. Provided a training set {( xi , yi ), i = 1, 2, . . . , N }, the AdaBoost algorithm [51]
(Drucker, 1997) performs the following using T number of predictors (CART):
n o
i. Initially, assign equal weights wi (1) = 1/N to each training example.
ii. For t = 1 to T, perform the following:
(a) Train the weak learner ht : x → y
(b) Compute the prediction from ht : yi (t) ( xi ), i = 1, . . . , N
 
(c) Determine the loss of model ht : Li (t) = Lt yi − yi (t) ( xi ) , where Lt e [0, 1]
is the loss function. Linear, square, or exponential loss functions can be used
as provided in Equations (4)–(6), respectively.

yi − yi (t) ( xi )
(t)
Li = , i = 1, . . . , N (4)
max yi − yi (t) ( xi )
Sustainability 2023, 15, 1718 11 of 20

2
yi − yi (t) ( xi )
(t)
Li = , i = 1, . . . , N (5)
max yi − yi ( p) ( xi )
|yi −yi (t) ( xi )|
( )
(t) max |yi −yi ( p) ( xi )|
Li = 1−e , i = 1, . . . , N (6)
(d) Determine the mean of the loss:

n
L(t) = ∑ t =1 L i ( t ) wi ( t ) , Lt e [0, 1] (7)

(e) Update the values of the weights:

(t)
wi ( t +1) = wi ( t ) β t 1− L i (8)

L(t)
βt = (9)
1 − L(t)
The final prediction of the response is computed as the final cumulative predictions
using T trees.
Gradient boosted trees as a more general form of boosting ensemble allows the usage of
arbitrary loss functions, unlike AdaBoost. Provided a training set {( xi , yi ), i = 1, 2, . . . , N },
GBT performs the following:
(a) Start model with constant value: Fo ( x ) = arg min ∑iN=1 L(yi , γ) in which L, yi and γ
γ
represent loss function, observed response value, and predicted response value.
(b) For t = 1 to T (where T denotes the total number of estimators):
h i
∂L(yi ,F ( xi ) )
i. Determine git = − ∂F ( x )
, i = 1, . . . , N which is the negative
i F ( x )= Ft−1 ( x )
gradient descent.
Train CART ht ( x ) using xi , git iN=1 as training examples.


ii.
iii. Compute the value of γt by solving γt = arg min ∑i L(yi , Ft−1 ( xi ) + γht ( xi )).
γ
iv. Update the model: Ft ( x ) = Ft−1 ( x ) + γt ht ( x ).
(c) Finally, output FT ( x ).
Another variation of boosting ensemble, extreme gradient boosting (xgBoost) is intro-
duced in 2016 by [52] (Chen & Guestrin, 2016) as an effective implementation of gradient
boosting. A regularization term is introduced in the objective function of xgBoost for
the purpose of minimizing the complexity of the model and overcoming the problem of
overfitting, as given in Equation (10) [52] (Chen & Guestrin, 2016).

N   t
∑L yi , ŷi (t) + ∑ Ω( f i ) (10)
i =1 i =1

1 K
Ω( f ) = γD + λ ∑ w2j (11)
2 j =1

The first term in Equation (10) denotes the loss function, while the second term denotes
the complexity of the model. Moreover, λ and j represent the regularization parameters,
D is the number of leaves, and γ and w j denote the complexity and weight of each leaf.

2.4. Model Performance Evaluation

Five different statistical performance metrics are considered to assess the predictive
performance of the ML models and select the best-performing model among the twelve
models considered in this study. The statistical metrics include: (a) the root mean squared
error (RMSE), (b) the mean absolute error (MAE), (c) the coefficient of determination (R2 ),
Sustainability 2023, 15, 1718 12 of 20

(e) the index of agreement (IA), and (f) the Kling-Gupta efficiency (KGE), as given by
Equation (12) through Equation (16). Low error (RMSE and MAE), and high IA, R2 , and
KGE demonstrate the best predictive model. The KGE as an improvement of Nash-Sutcliffe
efficiency considers three important measures: namely, correlation, bias, and variability
in a more balanced way, as given in Equation (16). KGE = 1 shows a perfect agreement
between the actual and predicted ridership. The same is true for R2 and IA.

1 N
N ∑ i =1 i
MAE = |y − ŷi | (12)
v
u1 N
u
RMSE = t ∑ (yi − ŷi )2 (13)
N i =1

2
∑iN=1 (yi − ŷi )
R2 = 1 − 2
(14)
∑iN=1 (yi − y)
2
∑in=1 (yi − ŷi )
IA = 1 − 2
, 0 < IA ≤ 1 (15)
∑in=1 (|yi − y| + |ŷi − y|)
q
KGE = 1 − (r − 1)2 + (α − 1)2 + ( β − 1)2 (16)
where yi is the ith observed value of output variable, ŷi is the ith the estimated value
of the output variable, y is the average value of the output variable, r is the linear cor-
relation between the actual and predicted response variable, α = σy /σy is a measure of
variability, β = µy /µy , is a bias term, σy and σy are the standard deviation of the pre-
dicted and actual response variable, and µy and µy are the mean of the predicted and
actual response variable.

3. Results and Discussion

As discussed, twelve ML algorithms were examined with the objective to propose the
best model that can accurately predict the response variable (ridership). The hyperparame-
ters of each model were tuned using grid search and Bayesian optimization combined with
10-fold cross-validation. The optimized values of the hyperparameters are presented in
Table 1 for all ML models except SE. The SE model combined the optimized ADB, GBT, and
xgBoost models.

Table 1. Optimized hyperparameters of ML models.

Models Parameters
RR alpha = 0.1
LR alpha = 0.001
EN alpha = 0.001
KNN Number of neighbors = 2
SVR Kernel = RBF, C = 95, ε = 0.01, gamma = ‘auto’
CART Maximum depth = 10, maximum features = 7
Number of estimators = 15, maximum features = 11, maximum
RF
depth = 13, minimum sample leaf = 1, minimum sample split = 2
Number of estimators = 20, maximum features = 12, maximum
ET
depth = 13, minimum sample leaf = 1, minimum sample split = 2
Base learner = CART, number of estimators = 30, learning rate =
ADB 0.25, maximum depth of tree for base learner = 12, maximum
features for base learner = 11
Number of estimators = 150, maximum depth = 8, learning rate =
GBT
0.25, subsample = 0.8, maximum features = 12
Number of estimators = 145, learning rate = 0.15, subsample = 0.5,
xgBoost maximum depth = 11, colsample by node = 1.0, colsample by
level = 1.0, colsample by tree = 1
Sustainability 2023, 15, 1718 13 of 20

Sustainability 2023, 15, x FOR PEER REVIEW 14 of 21

With the tuned optimal hyperparameters and the database, the prediction performance
of each model was evaluated and compared against the other models. First, the simplest ML
algorithms, namely,TableRR, lasso regression,
2. Performance measures.and EN, are investigated for ridership prediction.
Figure 6a–c represents the scatter plots for the actual ridership (Ractual ) against the predicted
Train Set Test Set
Models ridership (Rpredicted ) using these models. Meanwhile, the five RMSE
quantitative performance
MAE RMSE R2 AI KGE MAE R2 AI KGE
RR measures
231 of the models
336 on
24.6both the trained
60.71 dataset
28.75 and
247 the tested
363 dataset
27.5 are shown
60.91 in29.18
LR Table 231
2. It can be339observed 23.4from 55.80
Figure 6a–c and Table
22.45 246 2 that the
368 simplest
25.5 models
55.63 (EN,22.76
EN LR, RR)
231are unable337 to capture
24.1 the relationship
58.09 between246
25.20 the independent
365 variables
26.6 and the25.54
58.07
KNN response,52.0 which97.2 dictated93.7 98.35
the use of advanced 95.39 80.7 Among
ML models. 169 the base
84.4 learners
95.80 that90.77
SVR 117
are investigated 204 72.1 CART
in this study, 90.37
showed70.51 131
the best predictive 235
performance69.7 followed
89.11 by66.55
CART 43.9 81.7 95.5 98.85 96.81 82.4 151 87.4 96.75
KNN and SVR, respectively, on the train as well as test, as can be observed in Figure 6a–f93.54
RF 24.4 43.5 98.7 99.68 97.27 51.5 91.9 95.4 98.77 94.56
ET
and Table
17.0
2. This is shown
26.6 99.5
in Figure
99.88
7, which
98.09
represents
52.7
the values
86.4
of the
95.9
performance
98.90 93.44
ADB metrics 7.38for each15.27
base learner99.8 on the training99.61
99.96 and test datasets.
48.9 For instance,
82.4 96.3the coefficient
99.04 97.16
GBT of determination,
1.08 R2 , the value
1.39 100 of CART
100 was99.99
87.4% on32.3
the test dataset
52.8 which
98.5 was99.60
reduced96.43
xgBoost to 27.5%,
3.13 25.5%,4.27
26.6%, 84.4%,100 and100
67.7% in99.93
RR, LR, EN,
31.2 KNN, 49.3
and SVR,98.7respectively,
99.65 as96.92
SE listed2.25
in Table 2.3.10 100 100 99.80 28.8 46.5 98.8 99.69 96.93

Figure 6. Predicted versus

Figure observed
6. Predicted ridership
versus based
observed on single
ridership models.
based on single models.

Table 2. Performance measures.

Train Set Test Set

Models
MAE RMSE R2 AI KGE MAE RMSE R2 AI KGE
RR 231 336 24.6 60.71 28.75 247 363 27.5 60.91 29.18
LR 231 339 23.4 55.80 22.45 246 368 25.5 55.63 22.76
EN 231 337 24.1 58.09 25.20 246 365 26.6 58.07 25.54
KNN 52.0 97.2 93.7 98.35 95.39 80.7 169 84.4 95.80 90.77
SVR 117 204 72.1 90.37 70.51 131 235 69.7 89.11 66.55
CART 43.9 81.7 95.5 98.85 96.81 82.4 151 87.4 96.75 93.54
RF 24.4 43.5 98.7 99.68 97.27 51.5 91.9 95.4 98.77 94.56
ET 17.0 26.6 99.5 99.88 98.09 52.7 86.4 95.9 98.90 93.44
ADB 7.38 15.27 99.8 99.96 99.61 48.9 82.4 96.3 99.04 97.16
Figure 7. Performance measures for single models.
GBT 1.08 1.39 100 100 99.99 32.3 52.8 98.5 99.60 96.43
xgBoost 3.13 4.27 100 100 99.93 31.2 49.3 98.7 99.65 96.92
SE 2.25 3.10 100 100 99.80 28.8 46.5 98.8 99.69 96.93
Sustainability 2023, 15, 1718 14 of 20

Figure 6. Predicted versus observed ridership based on single models.

Figure 7.
Figure 7. Performance
Performance measures
measures for
for single
single models.
models.

The performance of the CART model can be improved with the use of ensemble
learners, as can be observed in Figure 8, which shows the scatter plots of the actual and
predicted ridership based on the ensemble models. As noted in the figure, the ensemble
learners showed good prediction performance with R2 ≥ 95.4% for both the test and train
datasets. Among the five bagging and boosting algorithm-based ensemble models (i.e.,
RF, ET, ADB, GBT, and xgBoost), xgBoost exhibited the best predictive performance in
terms of all the performance measures, as can be seen in Figure 9, which compares the
ensemble models in terms of the five performance metrics on the train as well as test sets.
Moreover, all ensemble models showed a better predictive performance compared with
that of the base learners (see Figures 6 and 8 and Table 2). The performance of xgBoost was
improved with the use of a stacking ensemble (SE) in which the best three models (xgBoost,
GBT, and ADB) were stacked using linear SVR as a meta-model. Overall, via comparing
the five quantitative performance measures, the proposed stacking ensemble was the best
predictive
Sustainability 2023,model
15, x FOR PEER the largest R2 , KGE, and IA and lowest RMSE AND MAE relative
withREVIEW 16 of to
22

the other models, as shown in Figures 6 and 8, and Table 2.

Figure 8. Predicted versus observed

Figure ridership
8. Predicted basedridership
versus observed on ensemble
based onmodels.
ensemble models.
Sustainability 2023, 15, 1718 15 of 20

Figure 8. Predicted versus observed ridership based on ensemble models.

Sustainability 2023, 15, x FOR PEER REVIEW 16 of 21

Feature Importance Analysis

Figure 9.The Figure
effect 9.measures
of the
Performance Performance measures
input parameters
for ensemble for ensembleusing
is models.
investigated models.Permutation Feature Im-
portance combined with the proposed gradient tree boosting ensemble model. The feature
Feature Importance
importance Analysis
explains the relative contribution of each type of land use density to the pre-
diction of the total metro
The effect of the input ridership by assigning
parameters a score tousing
is investigated each input parameter;
Permutation in otherImpor-
Feature
words, these findings of the feature importance are used by transport-urban planners for
tance combined with the proposed gradient tree boosting ensemble model. The feature
land use allocation in the newly developed areas to maximize public transportation use.
importance explains the relative contribution of each type of land use density to the pre-
Figure 10 depicts the relative importance of several types of land use densities based on
diction of the total metro ridership by assigning a score to each input parameter; in other
the results of the Permutation Feature Importance and gradient tree boosting model. The
words, these findings
governmental land use ofdensity,
the feature importance
educational areland
facilities used useby transport-urban
density, planners
and mixed-use
fordensity
land use allocation in the newly developed areas to maximize public transportation
are the three factors that play the most significant role in the determination of the
use. Figure
total 10 depicts
ridership and arethetherelative importance
most dominant of several
parameters, typesinofFigure
as shown land 10.
use densities based
on the results of the
In contrast, the Permutation
retail/commercialFeature Importance
land use density hasand the gradient tree boosting
least influence on the pre-model.
Thediction
governmental land useThese
of total ridership. density, educational
findings facilities
demonstrate land use
the nature of thedensity, andofmixed-use
occupiers the
three most
density important
are the land uses
three factors thatthat tend
play thetomost
use metro transit,
significant such
role inasthethe governmental of the
determination
and
total educational
ridership andsectors.
are the most dominant parameters, as shown in Figure 10.

Figure
Figure 10.10. Featureimportance
Feature importance analysis
analysisbased
basedonon
gradient treetree
gradient boosting.
boosting.

4. Model Implications
In contrast, on Policymaking land use density has the least influence on the
the retail/commercial
Thisofsection
prediction illustrates the
total ridership. relationship
These findings between the implications
demonstrate the nature of of the
the occupiers
prediction of the
urban-transport model investigated in this research and each related
three most important land uses that tend to use metro transit, such as the governmental Sustainable Devel-
opment Goal and
and educational sectors. pursues to make proposals for converting the transportation sector to a
more “sustainable” one. Several SDGs rely on transportation to achieve their goals. Trans-
portation
4. Model might not show
Implications on aPolicymaking
large role in a destination, yet for the achievement of the des-
tination and its objectives, transportation is essential and performs as an important “ena-
ThisHence,
bler”. section illustrates
it has been shown the that
relationship between
if one ignores the implications
sustainable transportation,ofit the prediction
becomes
urban-transport
much more problematic to meet most of the proposed SDGs. Figure 11 presents the urban-Devel-
model investigated in this research and each related Sustainable
opment Goal and pursues
transport-related SDGs, andtothe make proposals
following sectionfor converting
suggests policythe transportation
implications sector to
that could
a more “sustainable”
be in helping to achieveone.theseSeveral SDGs
goals, from thisrely on transportation
research study perspective: to achieve their goals.
Transportation might
SDG3: Ensure not show
healthy a large
lives and role
promote in a destination,
well-being for all of allyet
agesfor the achievement of the
The integrated and significantly planned transport networks have the probability to
enhance road safety using human-centered planning, transit-oriented development, or
traffic improvement actions, such as speed maps. In developing countries, measures to
build dimensions in road construction and transport policy and their execution are crucial
elements in attaining this goal. To conclude the well-planned road network and the insur-
Sustainability 2023, 15, 1718 16 of 20

destination and its objectives, transportation is essential and performs as an important

“enabler”. Hence, it has been shown that if one ignores sustainable transportation, it be-
comes much more problematic to meet most of the proposed SDGs. Figure 11 presents the
urban-transport-related SDGs, and the following section suggests policy implications that
could be in helping to achieve these goals, from this research study perspective:
SDG3: Ensure healthy lives and promote well-being for all of all ages
The integrated and significantly planned transport networks have the probability
to enhance road safety using human-centered planning, transit-oriented development,
or traffic improvement actions, such as speed maps. In developing countries, measures
to build dimensions in road construction and transport policy and their execution are
crucial elements in attaining this goal. To conclude the well-planned road network and the
insurance of pedestrian safety will eventually be of great help in orienting people toward
walkability and cycling for the best practice of TOD.
SDG7: Ensure access to affordable, reliable, sustainable, and modern energy for all
For the best achievement of Goal 7.3, the improvement of the fuel efficiency of vehicles
will be crucial, as vehicle-oriented transport is still high in developing countries. Addition-
ally, the decrease in private vehicle use by augmentation of public transport will efficiently
achieve this goal. Thus, the assurance of the lowest energy consumption could be achieved
by increasing public transportation ridership such as metro transit.
SDG9: Build resilience infrastructure, promote inclusive and sustainable industrialization, and
foster innovation
Urban connectivity and traffic development play a decisive role in the achievement
of sustainable industrialization. Reliable and resilient transportation infrastructure is an
indispensable component for resilience to disturbance globally. Reliable transportation
can increase social and economic resilience and meet the need for safety and emergency
response. The aptitude of society, commerce, and economy to get ready for and recover
from a disaster- such as the COVID-19 pandemic. The significance of public transport infras-
tructure in pre-and post-disaster evacuation is becoming ostensible and could save lives.
SDG11: Make cities and human settlement inclusive, safety, resilience and sustainable
Transport secures accessibility to all socio-economic needs of citizens. Dense ur-
ban planning could be of help in reducing the long travel time, in combination with an
integrative and effective public transportation system that has better options for active
mobility, more integrated transportation systems and land use, improving access to public
transportation, and contributing to environmentally sustainable development.
SDG12: Ensure sustainable consumption and production pattern
The application of green processes and technologies in the transportation sector and
logistics will be an essential part of the sophisticated strategies for better sustainable forms
of demand and supply of goods and services across the economy. Metro transit is a great
example of green transport that could significantly contribute to sustainable transport, the
environment, and the economy.
SDG13: Take urgent action to combat climate change and its impact
Sustainable transportation systems could be of help in the considerable potential for
mitigation for minimizing the effects of climate change. Inclusive sustainable transport
systems should be pursued, including land use integration with transport planning. More-
over, the greening vehicle and infrastructure construction could be of help in the reduction
of greenhouse gas emissions in the transportation industry, including preparedness, pro-
tection, response, and recovery. Climate change also has negative effects on the transport
infrastructure—for example, extreme heat cracks on streets and winding train tracks or
highways that are washed away by extreme weather events.
Sustainability 2023, 15, 1718 17 of 20

Figure 11. Urban-Transport direct and indirect related SDGs.

Sustainability 2023, 15, 1718 18 of 20

5. Conclusions and Recommendations

This paper investigates the use of machine learning-enhanced models, including
the simplest model and advanced models for predicting metro ridership, considering
the built environment around stations. A total of twelve parameters are considered as
input features, including time of day, day of the week, station, and nine types of land
use density, including land use density for open space and recreation (indoor-outdoor)
facilities, religious facilities, education facilities, retail/commercial, residential, special use,
transportation, mixed-use, and government land use densities. A time-series database is
used for model development and testing. Twelve different ML models were evaluated for
their predictive performance: ridge regression, lasso regression, elastic net, support vector
machine, K-nearest neighbor, random forest, decision tree, extremely randomized trees,
adaptive boosting, gradient boosting, extreme gradient boosting, and stacking ensemble
learner. The model’s performance was validated using the test set using five quantitative
performance measures: (a) mean absolute error, (b) root mean squared error, (c) coefficient
of determination, (d) agreement index, and (e) Kling Gupta efficiency. Based on the findings
of this research, the following conclusions could be drawn:
• The simple machine learning techniques: particularly, RR, EN, and lasso regression
cannot capture the relationship between the predictors and response variable (rider-
ship), which dictated the use of advanced techniques.
• Among the base learners, classification and regression trees showed the highest per-
formance, with an R2 of 87.4% on the test dataset. The KNN and SVR were the second
and third-best models among the base learners.
• The implementation of CART-based ensemble models considerably enhanced the pre-
diction capability of the CART model. All ensemble models showed good prediction
performance with R2 ≥ 95.4 on the train as well as in test sets.
• Overall, the stacking ensemble provided the best predictive performance. Thus, it
is selected as the best model for predicting metro ridership with respect to the built
environment around the stations.
In addition, this study presented the implications of the proposed urban-transport
predictive ML model on SDGs and identified various strategies that could be applied to
better achieve the SDGs. The discussion recommends that comprehensive sustainable
transport solutions, such as metro transport, should be pursued, including the integration
of land use and transport planning, prioritization of public transport, and non-motorized
transport/active transport, as well as green vehicle and infrastructure construction, can help
reduce greenhouse gas emissions from the transportation industry, including preparedness,
protection, response, and recovery. Climate change also has negative effects on the transport
infrastructure; thus, further studies are needed to investigate the correlation between
effective and sustainable urban transport systems and climate change effects.

Author Contributions: Conceptualization, A.H.A., T.G.W., M.K. and N.C.O.; Methodology, A.H.A.,
T.G.W., M.K. and N.C.O.; Validation, N.C.O.; Formal analysis, A.H.A. and T.G.W.; Investigation,
A.H.A. and N.C.O.; Data curation, A.H.A., T.G.W. and M.K.; Supervision, M.K. and N.C.O. All
authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.
Sustainability 2023, 15, 1718 19 of 20

References
1. Chan, W.C.; Ibrahim, W.H.W.; Lo, M.C.; Suaidi, M.K.; Ha, S.T. Sustainability of public transportation: An examination of user
behavior to real-time gps tracking application. Sustainability 2020, 12, 9541. [CrossRef]
2. Castanho, R.A.; Behradfar, A.; Vulević, A.; Naranjo Gómez, J.M.N. Analyzing Transportation Sustainability in the Canary Islands
Archipelago. Infrastructures 2020, 5, 58. [CrossRef]
3. Ercan, T.; Onat, N.C.; Tatari, O.; Mathias, J.D. Public transportation adoption requires a paradigm shift in urban development
structure. J. Clean. Prod. 2017, 142, 1789–1799. [CrossRef]
4. Reisi, M.; Aye, L.; Rajabifard, A.; Ngo, T. Transport sustainability index: Melbourne case study. Ecol. Indic. 2014, 43, 288–296.
[CrossRef]
5. Wey, W.M.; Huang, J.Y. Urban sustainable transportation planning strategies for livable City’s quality of life. Habitat Int. 2018,
82, 9–27. [CrossRef]
6. Mangalathu, S.; Hwang, S.H.; Choi, E.; Jeon, J.S. Rapid seismic damage evaluation of bridge portfolios using machine learning
techniques. Eng. Struct. 2019, 201, 109785. [CrossRef]
7. Banister, D. Sustainable urban development and transport —A eurovision for 2020. Transp. Rev. 2000, 20, 113–130. [CrossRef]
8. May, A.D. Urban Transport and Sustainability: The Key Challenges. Int. J. Sustain. Transp. 2013, 7, 170–185. [CrossRef]
9. Politis, I.; Georgiadis, G.; Nikolaidou, A.; Kopsacheilis, A.; Fyrogenis, I.; Sdoukopoulos, A.; Verani, E.; Papadopoulos, E. Mapping
travel behavior changes during the COVID-19 lock-down: A socioeconomic analysis in Greece. Eur. Transp. Res. Rev. 2021, 13, 21.
[CrossRef]
10. Karjalainen, L.E.; Juhola, S. Urban transportation sustainability assessments: A systematic review of literature. Transp. Rev. 2021,
41, 659–684. [CrossRef]
11. Litman, T.A. Comprehensive Transport Planning Framework—Best Practices for Evaluating All Options and Impacts; Victoria Transport
Policy Institute: Victoria, BC, Canada, 2012; pp. 1–59. Available online: https://[email protected] (accessed on 7
November 2022).
12. Litman, T.; Burwell, D. Issues in sustainable transportation. Int. J. Glob. Environ. Issues 2006, 6, 331–347. [CrossRef]
13. AlKhereibi, A.; AlSuwaidi, M.; Al-Mohammed, R.; Pokharel, S.; Ayari, M. An integrated urban-transport smart growth model
around metro stations: A case of Qatar. Transp. Res. Interdiscip. Perspect. 2021, 10, 100392. [CrossRef]
14. Su, S.; Wang, Z.; Li, B.; Kang, M. Deciphering the influence of TOD on metro ridership: An integrated approach of extended
node-place model and interpretable machine learning with planning implications. J. Transp. Geogr. 2022, 104, 103455. [CrossRef]
15. Zhang, Y.; Haghani, A. A gradient boosting method to improve travel time prediction. Transp. Res. Part C Emerg. Technol. 2015,
58, 308–324. [CrossRef]
16. Singh, Y.J.; Lukman, A.; Flacke, J.; Zuidgeest, M.; van Maarseveen, M.F.A.M. Measuring TOD around transit nodes—Towards
TOD policy. Transp. Policy 2018, 56, 96–111. [CrossRef]
17. Shirzadi Babakan, A.; Taleai, M. Impacts of transport development on residence choice of renter households: An agent-based
evaluation. Habitat Int. 2015, 49, 275–285. [CrossRef]
18. Motieyan, H.; Mesgari, M.S. Towards sustainable urban planning through transit-oriented development (A case study: Tehran).
ISPRS Int. J. Geo-Information 2017, 6, 402. [CrossRef]
19. Singh, Y.J.; Flacke, J.; van Maarseveen, M. Planning for Transit Oriented. In GIS in Sustainable Urban Planning and Management;
CRC Press: Boca Raton, FL, USA, 2018; Volume 2002, pp. 267–282.
20. Jain, A.; Ong, S.P.; Hautier, G.; Chen, W.; Richards, W.D.; Dacek, S.; Cholia, S.; Gunter, D.; Skinner, D.; Ceder, G.; et al. Commentary:
The materials project: A materials genome approach to accelerating materials innovation. APL Mater. 2013, 1, 011002. [CrossRef]
21. Türe, C. A methodology to analyse the relations of ecological footprint corresponding with human development index: Eco-
sustainable human development index. Int. J. Sustain. Dev. World Ecol. 2013, 20, 9–19. [CrossRef]
22. RCervero, R.; Dai, D. BRT TOD: Leveraging transit oriented development with bus rapid transit investments. Transp. Policy 2014,
36, 127–138. [CrossRef]
23. Archer, J.; Fotheringham, M.; Symmons, N.; Corben, B.J. The Impact of Lowered Speed Limits in Urban/Metropolitan Areas; Transport
Accident Commission: Geelong, Australia, 2008.
24. Wrótny, M.; Bohatkiewicz, J. Traffic noise and inhabitant health—A comparison of road and rail noise. Sustainability 2021, 13, 7340.
[CrossRef]
25. An, D.; Tong, X.; Liu, K.; Chan, E.H.W. Understanding the impact of built environment on metro ridership using open source in
Shanghai. Cities 2019, 93, 177–187. [CrossRef]
26. Cervero, R. Linking urban transport and land use in developing countries. J. Transp. Land Use 2013, 6, 7–24. [CrossRef]
27. Khan, J.; Hrelja, R.; Pettersson-Löfstedt, F. Increasing public transport patronage—An analysis of planning principles and public
transport governance in Swedish regions with the highest growth in ridership. Case Stud. Transp. Policy 2021, 9, 260–270.
[CrossRef]
28. Giuffrida, N.; le Pira, M.; Inturri, G.; Ignaccolo, M. Addressing the public transport ridership/coverage dilemma in small cities: A
spatial approach. Case. Stud. Transp. Policy 2021, 9, 12–21. [CrossRef]
29. Bielli, M.; Caramia, M.; Carotenuto, P. Genetic algorithms in bus network optimization. Transp. Res. Part C Emerg. Technol. 2002,
10, 19–34. [CrossRef]
Sustainability 2023, 15, 1718 20 of 20

30. Roy, A.; Manna, A.; Maity, S. A novel memetic genetic algorithm for solving traveling salesman problem based on multi-parent
crossover technique. Decis. Mak. Appl. Manag. Eng. 2019, 2, 100–111. [CrossRef]
31. Barma, P.S.; Dutta, J.; Mukherjee, A. A 2-opt guided discrete antlion optimizationalgorithm for multi-depotvehicle routing
problem. Decis. Mak. Appl. Manag. Eng. 2019, 2, 112–125. [CrossRef]
32. Calabrò, G.; Inturri, G.; le Pira, M.; Pluchino, A.; Ignaccolo, M. Bridging the gap between weak-demand areas and public transport
using an ant-colony simulation-based optimization. Transp. Res. Procedia 2019, 45, 234–241. [CrossRef]
33. Zhao, Y.; Huang, P.; Long, G.; Yuan, Y.; Sun, Y. Influence of Fluid Viscous Damper on the Dynamic Response of Suspension Bridge
under Random Traffic Load. Adv. Civ. Eng. 2020, 2020, 1857378. [CrossRef]
34. Pagès, L.; Jayakrishnan, R.; Cortés, C.E. Real-time mass passenger transport network optimization problems. Transp. Res. Rec. J.
Transp. Res. Board 2006, 1964, 229–237. [CrossRef]
35. Murray, A.T.; Davis, R.; Stimson, R.J.; Ferreira, L. Public transportation access. Transp. Res. D. Transp. Environ. 1998, 3, 319–328.
[CrossRef]
36. Murray, A.T. A Coverage Model for Improving Public Transit System Accessibility and Expanding Access. Ann. Oper. Res. 2003,
123, 143–156. [CrossRef]
37. Mattson, J. Estimating Ridership of Rural Demand–Response Transit Services for the General Public. Transp. Res. Rec. J. Transp.
Res. Board 2017, 2647, 127–133. [CrossRef]
38. Walker, J. Purpose-driven public transport: Creating a clear conversation about public transport goals. J. Transp. Geogr. 2008,
16, 436–442. [CrossRef]
39. Le Pira, M.; Ignaccolo, M.; Inturri, G.; Pluchino, A.; Rapisarda, A. Modelling stakeholder participation in transport planning. Case
Stud. Transp. Policy 2016, 4, 230–238. [CrossRef]
40. Gonzalez-Urango, H.; Inturri, G.; le Pira, M.; García-Melón, M. Planning for Pedestrians with a Participatory Multicriteria
Approach. J. Urban Plan. Dev. 2020, 146, 05020007. [CrossRef]
41. Inturri, G.; Giuffrida, N.; le Pira, M.; Fazio, M.; Ignaccolo, M. Linking Public Transport User Satisfaction with Service Accessibility
for Sustainable Mobility Planning. SPRS Int. J. Geo-Inf. 2021, 10, 235. [CrossRef]
42. Inturri, G.; Fiore, S.; Ignaccolo, M.; Caprì, S.; le Pira, M. “You study, you travel free”: When mobility management strategies meet
social objectives. Transp. Res. Procedia 2020, 45, 193–200. [CrossRef]
43. Major, M.D.; Tannous, H.O.; Al-Thani, S.; Hasan, M.; Khan, A.; Salaheldin, A. Macro and micro scale modelling of multi-modal
transportation spatial networks in the city-state of Doha, Qatar. In Proceedings of the 56th ISOCARP Virtual World Planning
Congress, Online, 8 November 2020–4 February 2021. [CrossRef]
44. Furlan, R.; Almohannadi, M. Light rail transit and land use in Qatar: An integrated planning strategy for Al-Qassar’s TOD.
Archnet-IJAR Int. J. Arch. Res. 2016, 10, 170–192. [CrossRef]
45. Salama, A.M.; Wiedmann, F. Demystifying. Doha: On Architecture and Urbanism in an Emerging City; Qatar University: Doha, Qatar, 2016.
46. Furlan, R.; Petruccioli, A.; Major, M.D.; Zaina, S.; Zaina, S.; Al Saeed, M.; Saleh, D. The urban regeneration of west-bay, business
district of Doha (State of Qatar): A transit-oriented development enhancing livability. J. Urban Manag. 2019, 8, 126–144. [CrossRef]
47. Furlan, R.; Al-Mohannadi, A. An Urban Regeneration Planning Scheme for the Souq Waqif Heritage Site of Doha. Sustainability
2020, 12, 7927. [CrossRef]
48. Breiman, L.; Ihaka, R. Nonlinear discriminant analysis via scaling and ACE. In Davis One Shields Avenue Davis; Department of
Statistics, University of California: Berkeley, CA, USA, 1984.
49. Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [CrossRef]
50. Geurts, P.; Ernst, D.; Wehenkel, L. Extremely randomized trees. Mach. Learn. 2006, 63, 3–42. [CrossRef]
51. Drucker, P.F. The future that has already happened. Harv. Bus. Rev. 1997, 75, 20–22.
52. Chen, T.; Guestrin, C. Xgboost: A scalable tree boosting system. In Proceedings of the 22nd Acm Sigkdd International Conference
on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016; pp. 785–794.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.