Content uploaded by Rahul Raj Singh
Author content
All content in this area was uploaded by Rahul Raj Singh on Nov 24, 2024
Content may be subject to copyright.
Dynamic Transportation Problem
Rahul Raj Singh, Kartikesh Thakur, Saurabh Kasaudhan, Gurwinder
Singh
Chandigarh University, Mohali, India
Corresponding author: Rahul Raj Singh, Email: rahulrajsingh2503@gmail.com
Rising urbanisation and population expansion present diculties for urban transportation
networks. All of these constraints worsen trac-related problems: unforeseen congestion,
poor route planning, and underuse of resources. Dynamic Transportation Optimisation
(DTO) technologies have consequently become intriguing means to increase the sustainabil-
ity and eciency of metropolitan transportation networks. Dynamic programming (DP),
Simulated Annealing (SA), Ant Colony Optimisation (ACO), Swarm Intelligence (SI), Ge-
netic Algorithms (GA), Reinforcement Learning (RL), Machine Learning Models (ML), Ge-
ographic Information Systems (GIS), and Integer Linear Programming (ILP) are just a few
of the wide range of approaches DTO methods apply. These technologies generate com-
plete trac analysis and dynamic routing solutions by using real-time data from security
cameras, trac sensors, and GPS devices. This paper gives a complete evaluation of DTO ap-
proaches coupled with assessments of their usefulness, performance, and constraints in the
framework of urban transportation. By way of a detailed comparison analysis, the study in-
tends to highlight the advantages and disadvantages of every method, so supporting stake-
holders in choosing and deploying DTO solutions customised to unique urban transport
concerns. Furthermore highlighted in light of developments in articial intelligence, big
data analytics, and predictive modelling is the future prospects of DTO techniques. The
integration of these technologies within present urban contexts aims to develop resilient
transportation networks, thereby contributing to the continuing discourse on the evolution
of urban mobility solutions. The subsequent sections of this article delve into the intricacy
of each DTO technique, oering insights into their capabilities, actual implementations,
and projected repercussions on urban transportation networks. By combining empirical
evidence with theoretical frameworks, the research presents a holistic understanding of
the role and signicance of DTO techniques in dening the future of urban mobility.
Keywords: Dynamic transportation optimization, urban mobility, transportation manage-
ment, DTO methodologies, comparative analysis, articial intelligence, big data analytics.
2024. In Mukesh Saraswat & Rajani Kumari (eds.), Applied Intelligence and Comput-
ing, 145–156. Computing & Intelligent Systems, SCRS, India. DOI: https://doi.org/10.
56155/978-81-955020-9-7-16
1 Introduction
The dynamic transportation problem poses a substantial difficulty in today's metropolitan
environments, as rapid swings in traffic circumstances necessitate adaptable and effective route-
planning solutions. This objective comprises designing an ideal transportation network model that can
respond to real-time traffic dynamics, therefore empowering drivers to make informed decisions and
maximize the utilization of transportation assets. Key objectives include minimising journey time,
distance, and toll expenditures while optimizing the overall cost incurred by drivers. Addressing this
challenge is crucial not merely for enhancing driver happiness but also for lowering congestion,
minimising air pollution, and optimizing fuel usage, thereby improving the overall efficiency of
transportation systems.
The integration of real-time traffic data with Geographic Information Systems (GIS) is the cornerstone
of real-time dynamic transportation optimization. However, the availability and accessibility of real-
time traffic data are still limited, giving a major obstacle in adopting efficient solutions. This project
proposes to study strategies for real-time dynamic transportation optimization, employing advanced
algorithms and control behavior models to dynamically update traffic information and give optimal
routing solutions to drivers in real time.
This research dives at numerous routing algorithms and traffic forecasting methodologies, attempting
to offer insights into the dynamic transportation sector. Additionally, a detailed case study of an
Intelligent Transport System (ITS) created in the United States will be studied, illustrating the practical
application of real-time traffic data in dynamically updating transportation routes. Through this
inquiry, the paper seeks to contribute to the enhancement of real-time dynamic transportation
optimization and provide the framework for more efficient and sustainable transportation systems in
urban environments.
2 Literature Review
2.1 Real-Time Traffic Information
Real-time traffic data is crucial for solving the rising transportation problem. It gives rapid information
on traffic conditions, congestion levels, and travel lengths, which are crucial for creating ideal transit
routes [1]. Real-time data sources comprise GPS-enabled automobiles, traffic sensors, and mobile
applications; yet, issues pertaining to data quality, coverage, and integration continue [2]. The scarcity
of comprehensive and trustworthy real-time traffic data causes substantial challenges in building
efficient transportation optimisation systems [3]. Zhan and Noon [4] propose that obstacles like data
delay and restricted coverage might greatly impair the effectiveness of dynamic transportation models.
Sheffi [15] underscores the usefulness of equilibrium analysis in urban transport networks, underlining
the necessity of real-time data to sustain effective traffic flow and reduce congestion. Notwithstanding
breakthroughs, the efficient incorporation of real-time input into optimisation models remains a
persistent challenge.
2.2 Dynamic Transportation Challenge
The dynamic transportation issue is a tough optimization challenge that focuses on adjusting
transportation routes in response to real-time traffic dynamics [5]. Key objectives include reducing
travel time, distance, and cost, while boosting system efficiency and user satisfaction [6]. Urban areas,
typified by high traffic density and variable demand patterns, provide significant problems for
optimisation [7]. Traditional approaches may fail to account for dynamic occurrences, leading to the
advent of adaptive algorithms capable of incorporating real-time data [8]. Ichoua, Gendreau, and
Potvin [9] think that tackling dynamic transportation difficulties requires a multidisciplinary method
that integrates operations research, computer science, and transportation engineering. Gendreau et al.
Rahul Raj Singh, Kartikesh Thakur, Saurabh Kasaudhan, Gurwinder Singh
146
[17] elaborate on this, exhibiting how stochastic vehicle routing may be applied to dynamic scenarios
where real-time traffic data might impact route decisions. Additionally, Vlahogianni, Karlaftis, and
Golias [16] stress that short-term traffic forecasting plays a crucial role in dynamic systems, assisting in
the anticipation of traffic changes and permitting more responsive transportation solutions.
2.3 Existing Solutions
Existing transportation optimization methods, including Dynamic Programming (DP), Simulated
Annealing (SA), Ant Colony Optimization (ACO), Swarm Intelligence (SI), Genetic Algorithms (GA),
Reinforcement Learning (RL), Machine Learning (ML) models, Geographic Information Systems
(GIS), and Integer Linear Programming (ILP), demonstrate partial efficacy but struggle with scalability
and real-time adaptability [10]. For example, while Reinforcement Learning has demonstrated
potential, its computational complexity inhibits its real-time utilisation [11]. Geographic Information
Systems provide tremendous geographic analytic capabilities, but integrating real-time traffic data
remains problematic [12]. According to Bräysy and Gendreau [18], vehicle routing issues, notably those
with time limits, demand route planning and local search algorithms that can react to real-time traffic
changes. Similarly, Eksioglu, Vural, and Reisman [19] present a full assessment of vehicle routing
challenges, categorising various solutions but highlighting that real-time integration continues to be a
substantial hurdle. Kumar and AbouRizk [20] further stress the significance of simulation in hybrid
vehicle routing systems, suggesting that such systems may benefit from more seamless integration with
real-time traffic data to better decision-making and route optimization.
As urban transportation networks evolve, the necessity for novel technologies that can efficiently
exploit real-time data to dynamically improve transit routes develops. Gavalas and Konstantopoulos
[14] provide unique ways to bridge the difference between typical optimization methodologies and the
significant issues experienced by urban mobility, opening the door for more intelligent and flexible
transportation systems. Additionally, Sheffi [15] underlines the need of utilising mathematical
programming approaches to establish equilibrium in urban traffic systems, further stressing the
demand for real-time data integration to assure optimal flow and alleviate congestion.
Figures
The figures illustrate the comparison of numerous optimization methodologies employed in handling
the dynamic transportation problem.
Figure 1. Heat map comparing Optimization Techniques
Applied Intelligence and Computing
147
Figure 1 illustrates a heat map that visualizes the comparison of multiple optimization methodologies
based on four factors: Problem Size, Complexity, Cost, and Customer Satisfaction. Each cell in the heat
map reflects the rating of a strategy for a certain factor, ranging from 1 (lowest) to 5 (highest). The color
intensity affects the rating, with deeper hues signifying higher ratings. This graphic aids in
understanding the strengths and limits of each technique across numerous evaluation criteria.
Figure 2. Comparison of Optimization Techniques by Factor
Figure 2 consists of a series of bar charts that highlight the comparison of numerous optimization
approaches across distinct factors: Problem Size, Complexity, Cost, and Customer Satisfaction. Each
subplot represents one component, with methodologies shown on the x-axis and ratings on the y-axis.
The height of each bar correlates to the rating of a technique for a specific factor, allowing for an easy
comparison of processes for each evaluation criterion. This image allows the assessment of approach
efficacy across several areas continuously, assisting in decision-making for dynamic transportation
optimization strategies.
Figure 3. Radar Chart of Optimization Techniques
Figure 3 is a radar chart that displays the performance of several optimization tactics across multiple
criteria: Efficiency, Cost, Scalability, Adaptability, and Effectiveness. Each strategy is represented by a
Rahul Raj Singh, Kartikesh Thakur, Saurabh Kasaudhan, Gurwinder Singh
148
colored line, with the length and shape of the line signifying its performance relative to other
approaches. The image gives a visual comparison of the benefits and limits of each technique across
multiple evaluation variables, enabling stakeholders to make knowledgeable choices regarding the
selection of optimization techniques for dynamic transportation challenges. The area inside the blue
shaded region indicates the average performance across all strategies, offering additional insights into
overall performance trends.
3 Methodology
3.1 Data Collection
Data collection is the initial stage in tackling the dynamic transportation dilemma. Real-time traffic
data, covering information on traffic flow, speed, and congestion levels, is obtained from varied sources
such as GPS-enabled vehicles, traffic sensors, and transportation companies. The gathering technique
comprises merging and processing data streams from many sources to acquire a thorough knowledge of
prevailing traffic conditions.
3.2 Data Processing
Upon collected, the raw traffic data undergoes processing to extract useful attributes and insights. This
stage encompasses data cleaning, standardization, and transformation to insure consistency and
compatibility across numerous datasets. Advanced data processing techniques like as machine learning
algorithms may be utilised to find patterns, abnormalities, and trends in the traffic data, giving more
exact modelling and analysis.
3.3 Modelling the Transportation Problem
Modelling the transportation problem entails creating mathematical and computer models to capture
the dynamic interplay between traffic flow, transportation routes, and user preferences. Various
optimization approaches, including linear programming, genetic algorithms, and reinforcement
learning, can be employed to construct robust models capable of dynamically altering transportation
routes based on real-time traffic data. The modelling technique comprises creating target functions,
limitations, and decision variables to maximize critical metrics such as trip time, distance, and cost
while meeting user preferences and system limits. Additionally, the integration of geographic
information systems (GIS) with network analysis tools increases spatial representation and
visualization of transportation networks, aiding in the comprehension and analysis of model outputs.
4 Real-Time Traffic Data Analysis
4.1 Traffic Patterns and Trends
Analysis of real-time traffic data gives crucial insights into traffic patterns and trends, enabling the
detection of recurring congestion hotspots, peak traffic hours, and variable traffic volumes. By studying
historical traffic data, researchers can find long-term trends and seasonal swings in traffic flow,
supporting preemptive initiatives to alleviate congestion and optimize transport routes.
4.2 Impact of Traffic Congestion
Traffic congestion has far-reaching repercussions for transportation efficiency, environmental
sustainability, and economic development. Through data analytic approaches, the detrimental
Applied Intelligence and Computing
149
repercussions of congestion, such as increased travel time, fuel consumption, and vehicle emissions,
can be estimated and assessed. Moreover, real-time traffic data analysis enables the construction of
congestion mitigation solutions, including traffic signal optimization, lane management, and dynamic
route directing systems, geared at optimizing traffic flow and eliminating congestion-related delays.
Figure 4. Heatmap Showing Congestion on different Routes
Figure 4 shows a heatmap illustrating congestion levels on different routes within an urban
transportation network. This figure provides a visual representation of the real-time traffic data,
highlighting areas of high congestion in darker shades. By analyzing this heatmap, stakeholders can
identify the most congested routes and understand traffic flow patterns. This information is crucial for
dynamic transportation optimization, as it helps in developing strategies to alleviate congestion and
improve overall traffic management. The insights gained from this visualization can guide the
implementation of adaptive routing solutions that respond to real-time traffic conditions, ultimately
enhancing the efficiency of urban transportation systems.
Figure 5. Impact of Traffic Congestion on Delivery Tim
Figure 5 demonstrates the influence of traffic congestion on delivery time through a complete
depiction. This chart depicts the link between increasing levels of congestion and the resulting delays in
delivery timetables. By assessing this relationship, it becomes evident how increasing traffic congestion
can considerably degrade the punctuality of deliveries, resulting to longer transit times and associated
delays in supply chain activities. This knowledge is useful for devising strategies to decrease
congestion-related delays, such as enhancing route planning or introducing dynamic traffic
management systems. The insights from this figure can help in maximising the efficiency of delivery
Rahul Raj Singh, Kartikesh Thakur, Saurabh Kasaudhan, Gurwinder Singh
150
systems, boosting customer contentment, and lowering the total expenses linked with delayed
deliveries.
4.3 Predictive Modelling
Predictive modelling examines real-time traffic data to forecast future traffic conditions and anticipate
probable congestion scenarios. By utilising machine learning algorithms and time series analytic
methodologies, researchers may construct prediction models capable of anticipating traffic flow,
congestion levels, and journey times with high accuracy. These predictive algorithms enable
transportation authorities and commuters to proactively plan and change their travel routes to avoid
congestion and minimize travel delays. Additionally, predictive modelling permits the creation of
dynamic traffic management systems that dynamically modify traffic signals, lane configurations, and
speed limits to optimize traffic flow and mitigate congestion in real time.
4.4 Visualization of Transportation Data
In order to adequately study and grasp different dimensions of transportation logistics, we employed
several visualization methods. Firstly, a Sankey diagram highlighted the distribution of gasoline
expenses across different fuel types, providing insights into cost allocation. Subsequently, a bar chart
presented transportation expenses by mode, displaying the comparative charges connected with
different transit options. Furthermore, a scatter plot showed the relationship between vehicle speed
and fuel economy, assisting in finding appropriate speed ranges for cost-effective transportation.
Lastly, a line graph depicted the link between delivery times and distances, offering insights into the
time-distance trade-offs inherent in transportation operations.
Additionally, these visualizations act as vital tools for stakeholders to find possible areas for
development and optimization inside transportation networks. By visually portraying detailed data,
decision-makers can better grasp patterns, trends, and linkages, leading to more informed strategy
development and resource allocation. Moreover, the use of visualization tools enhances transparency
and communication among stakeholders, enabling collaborative problem-solving and innovation in
transportation logistics. Overall, the introduction of visualization tools boosts the effectiveness and
efficiency of transportation management, ultimately leading to the creation of more robust and
responsive transportation networks.
Figure 6. Fuel Costs Distribution
Applied Intelligence and Computing
151
Figure 6 presents a pie chart that illustrates the distribution of fuel expenses among numerous fuel
types, including Diesel, Petrol, CNG, and Electricity. Each part of the graphic indicates the proportion
of the overall fuel expenses allotted to a certain fuel type. By presenting this distribution, the figure
offers insights into the economic impact of different fuel sources on the whole transportation network.
This information is helpful for understanding how fuel choices effect operational expenses and can lead
approaches for improving fuel usage. Additionally, evaluating this distribution helps identify regions
where alternative fuels or more efficient energy sources could be used to cut total transportation costs
and promote sustainable practices within the network.
Figure 7. Transportation Costs by Mode
Figure 7 depicts a bar chart that offers a comparative assessment of transportation expenses related
with numerous modes of transportation, including Road, Rail, Air, and Sea. Each bar shows the cost
incurred for carrying products via a particular mode, giving for an easy comparison of expenses across
different transportation methods. By reviewing this chart, stakeholders can identify which modes are
more cost-effective and appreciate the financial ramifications of adopting each mode for goods
movement. This information is crucial for making intelligent judgements on determining the most
efficient and affordable shipping route, thereby increasing overall logistics and supply chain
management. The comparative structure of this graphic assists in assessing the trade-offs between
different transportation modes, such as cost against speed or environmental impact, enabling a
comprehensive approach to transportation planning.
Figure 8. Fuel Efficiency vs. Vehicle Speed
Rahul Raj Singh, Kartikesh Thakur, Saurabh Kasaudhan, Gurwinder Singh
152
Figure 8 illustrates a scatter plot that visualizes the link between vehicle speed and fuel economy. Each
data point reflects a distinct combination of vehicle speed and the accompanying petroleum cost. This
graphic helps detect patterns and trends in how speed influences fuel usage, providing insights into the
most cost-effective speed ranges for transportation. By evaluating this scatter plot, stakeholders can
discover optimal speed ranges that minimize fuel costs while maintaining effective transport
operations. Understanding this link is critical for devising ways to minimise petroleum usage, reduce
transportation expenses, and boost overall efficiency within the transportation network. This research
can lead the deployment of pace management methods that contribute to more sustainable and cost-
effective transportation networks.
5 Result & Discussion
5.1 Comparative Analysis of DTO Techniques
The comparison analysis identified considerable disparities in the performance and usability of
multiple Dynamic Transportation Optimization (DTO) approaches across distinct urban transportation
circumstances. Techniques such as Genetic Algorithms (GA) and Reinforcement Learning (RL) showed
robust performance in intricate, dynamic scenarios, efficiently adjusting to changing conditions.
Conversely, Simulated Annealing (SA) and Integer Linear Programming (ILP) exhibited limitations in
scalability and flexibility, straining to manage real-time alterations in traffic patterns. Machine
Learning Models (ML) demonstrated positive results in anticipating traffic patterns and optimizing
routing systems, yet their effectiveness was heavily dependent on the quality and volume of training
data. The analysis indicates the necessity for continuing refinement and adjustment of these strategies
to suit the rising demands of urban transportation networks.
5.2 Impact on Urban Mobility Efficiency
The use of DTO concepts has led to large increases in urban mobility efficiency. Key benefits include
reduced commute times, lower congestion, and optimal resource utilisation. By exploiting real-time
data, DTO systems can dynamically adjust routing methods, divert traffic from congested regions, and
optimize traffic signal timings. Case studies and simulations have underscored the potential of DTO
approaches to reduce traffic congestion and increase the reliability of urban transportation networks,
eventually leading to more efficient and sustainable urban mobility solutions.
5.3 Future Prospects and Emerging Trends
The future combination of DTO approaches with new technologies holds huge promise for considerably
increasing urban mobility management. Innovations in artificial intelligence, big data analytics, and
autonomous vehicle technology present new possibilities for improving transportation networks and
enhancing service quality for passengers. Predictive modeling and autonomous operations are likely to
play more essential roles in refining route planning and reducing traffic congestion in urban settings.
Embracing these developing tendencies will be vital for expanding the capabilities and effect of DTO
systems.
5.4 Challenges and Considerations
Despite the benefits, the widespread deployment of DTO techniques meets various obstacles, including
data privacy concerns, infrastructural limits, and governmental restraints. Protecting the security and
privacy of real-time traffic data is vital, especially with the growth of IoT devices and smart city
projects. Additionally, scaling DTO systems to service growing urban populations and evolving mobility
needs entails enormous investment in infrastructure and technology. Addressing these difficulties will
be critical for the successful development and operation of DTO systems.
Applied Intelligence and Computing
153
5.5 Ethical and Social Repercussions
As DTO techniques grow more crucial to urban mobility, it is important to examine their greater ethical
and social ramifications. Ensuring fair access to transportation services, limiting environmental
damage, and increasing social inclusion are essential concerns in the design and operation of DTO
systems. Moreover, minimising algorithmic bias and guaranteeing fairness in decision-making
processes are vital for promoting trust and acceptability among various groups. These elements will be
important to establishing DTO systems that are not simply technologically proficient but also socially
responsible and inclusive.
5.6 Optimization Using OpenRouteService API
Optimization of vehicle routes using the OpenRouteService API comprises leveraging geographic data
and vehicle attributes to construct routes that reduce journey time and resource utilisation. By
identifying the vehicles' beginning locations, capacity constraints, and the destinations (jobs) to be
visited, the OpenRouteService API delivers efficient routing possibilities. The suggested routes are
represented on a map using Folium, giving a clear picture of appropriate vehicle paths. This strategy
enhances route planning and resource allocation, benefiting many applications like logistics,
transportation, and urban mobility management. The integration of OpenRouteService API indicates
the possibility for new tools to improve operational efficiency and effectiveness.
Figure 9. Route Optimization using Open Route Service API
Figure 9 demonstrates route optimization using the OpenRouteService API. The visualization
demonstrates how the API assesses geographic data and vehicle parameters to create optimized routes.
Each route displayed on the map reflects the most efficient path for a vehicle, taking into consideration
metrics such as distance, journey time, and resource usage. By leveraging the OpenRouteService API,
the system can dynamically compute and show ideal routes based on real-time data and predefined
limits.
This statistic highlights the efficacy of the OpenRouteService API in enhancing route planning and
resource allocation. The ability to present best routes in real-time improves in understanding how
various routes compare in terms of efficiency, allowing for better decision-making in logistics and
transportation management. The map provides a visual picture of the ideal vehicle tracks, highlighting
the API's capacity to improve operational efficiency and streamline transportation processes.
Rahul Raj Singh, Kartikesh Thakur, Saurabh Kasaudhan, Gurwinder Singh
154
6 Conclusion
In conclusion, the study offered stresses the major influence of Dynamic Transportation Optimization
(DTO) methods in tackling the complex issues inherent in urban transportation. By blending real-time
data analysis with advanced computing methodologies, DTO systems offer feasible options for
improving transportation efficiency and boosting the quality of urban life. The capacity to dynamically
adjust routing and resource allocation in reaction to real-time traffic circumstances allows for
significant savings in trip time, congestion, and operational expenses. This skill is crucial for creating
sustainable transportation networks that can adapt to the dynamic character of urban surroundings,
eventually leading to smoother traffic flow and lower environmental impact.
Nevertheless, completely achieving the promise of DTO techniques requires overcoming several
hurdles, including technological restrictions, ethical considerations, and social repercussions.
Addressing these challenges includes not only improving the underlying technology but also ensuring
that solutions are egalitarian, safe, and inclusive. Future developments in DTO should focus on
harnessing emerging technologies, such as artificial intelligence and big data analytics, while
encouraging collaborative techniques to urban mobility management. By stressing these traits, DTO
systems may grow to meet the expanding demands of urban regions, providing smarter, more
responsive transportation networks that boost overall efficiency and resilience.
References
[1] F. B. Zhan and C. E. Noon, "Shortest path algorithms: An evaluation using real road networks,"
Transportation Science, vol. 32, no. 1, pp. 65-73, 1998.
[2] H. J. Miller and S. L. Shaw, Geographic Information Systems for Transportation: Principles and
Applications. Oxford University Press, 2001.
[3] X. Wang and J. Liu, "Real-time traffic management in intelligent transportation systems: A survey,"
Journal of Transportation Systems Engineering and Information Technology, vol. 8, no. 2, pp. 1-13, 2008.
[4] F. Dion and H. Rakha, "Evaluation of dynamic traffic signal timing optimization models," Transportation
Research Record, vol. 1925, no. 1, pp. 167-176, 2005.
[5] S. Ichoua, M. Gendreau, and J. Y. Potvin, "Diversion issues in real-time vehicle dispatching,"
Transportation Science, vol. 34, no. 4, pp. 426-438, 2000.
[6] K. G. Zografos and K. N. Androutsopoulos, "A decision support system for integrated hazardous materials
routing and emergency response decisions," Transportation Research Part C: Emerging Technologies, vol.
16, no. 6, pp. 684-703, 2008.
[7] R. W. Hall, "Route choice and advanced traveler information systems: The state of the art," Transportation
Research Part C: Emerging Technologies, vol. 4, no. 2, pp. 89-112, 1996.
[8] M. Gendreau and J. Y. Potvin, "Dynamic vehicle routing and dispatching," in Fleet Management and
Logistics, pp. 115-126, Springer, Boston, MA, 2010.
[9] D. Gavalas and C. Konstantopoulos, "Intelligent transportation systems: Applications and challenges,"
Transportation Research Part C: Emerging Technologies, vol. 11, no. 6, pp. 409-421, 2003.
[10] M. Dorigo and L. M. Gambardella, "Ant Colony System: A cooperative learning approach to the traveling
salesman problem," IEEE Transactions on Evolutionary Computation, vol. 1, no. 1, pp. 53-66, 1997.
[11] P. Vansteenwegen, W. Souffriau, and D. Van Oudheusden, "The orienteering problem: A survey,"
European Journal of Operational Research, vol. 209, no. 1, pp. 1-10, 2011.
[12] R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction. MIT Press, 2018.
[13] M. Batty, Predicting the Complexity of Dynamic Systems: Implications for Geographic Information
Systems. Springer Science & Business Media, 1997.
[14] A. Kesting, M. Treiber, and D. Helbing, "Adaptive cruise control design for active congestion avoidance,"
Transportation Research Part C: Emerging Technologies, vol. 16, no. 6, pp. 668-683, 2008.
[15] Y. Sheffi, Urban Transportation Networks: Equilibrium Analysis with Mathematical Programming
Methods. Prentice-Hall, 1985.
Applied Intelligence and Computing
155
[16] E. I. Vlahogianni, M. G. Karlaftis, and J. C. Golias, "Short-term traffic forecasting: Overview of objectives
and methods," Transport Reviews, vol. 24, no. 5, pp. 533-557, 2004.
[17] M. Gendreau, G. Laporte, and R. Séguin, "Stochastic vehicle routing," European Journal of Operational
Research, vol. 88, no. 1, pp. 3-12, 1996.
[18] O. Bräysy and M. Gendreau, "Vehicle routing problem with time windows, Part I: Route construction and
local search algorithms," Transportation Science, vol. 39, no. 1, pp. 104-118, 2005.
[19] B. Eksioglu, A. V. Vural, and A. Reisman, "The vehicle routing problem: A taxonomic review," Computers &
Industrial Engineering, vol. 57, no. 4, pp. 1472-1483, 2009.
[20] A. Kumar and S. AbouRizk, "Role of simulation in hybrid vehicle routing systems," Journal of Construction
Engineering and Management, vol. 130, no. 1, pp. 55-66, 2004.
Rahul Raj Singh, Kartikesh Thakur, Saurabh Kasaudhan, Gurwinder Singh
156
