Farhana Murtaza Choudhury
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Farhana Murtaza Choudhury has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Zhifeng Bao
The University of Queensland, Brisbane, Australia
9 shared publications - J. Shane Culpepper
The University of Queensland, Brisbane, Australia
8 shared publications - Timos Sellis
Swinburne University of Technology, Melbourne, Australia
8 shared publications - Egemen Tanin
The University of Melbourne
5 shared publications - Stephan Winter
The University of Melbourne
5 shared publications - Hamid Bagha
Department of Infrastructure Engineering, The University of Melbourne, Parkville, Melbourne, Australia
4 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 27 (11) , 14713-14726
Dynamic Configuration of On-Street Parking Spaces for Enhanced Traffic Flow Using Reinforcement Learning
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 27 (11) , 14713-14726
Dynamic Configuration of On-Street Parking Spaces for Enhanced Traffic Flow Using Reinforcement Learning
- Open Access
- 17 October 2026
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
Modeling the Spatiotemporal Implications of Autonomous Vehicles’ Empty Cruising on Parking Dynamics and Traffic Flow
Autonomous vehicles’ (AVs) ability to reposition themselves can transform parking dynamics. We focus on empty cruising as an alternative to short-term parking. This capability affects parking space occupancy and traffic flow because AVs use road space instead of parking spaces. This research models the spatiotemporal implications of empty cruising on parking dynamics and traffic flow in a mixed environment of AVs and conventional vehicles (CVs). The contributions of this research compared with existing research include: (1) analyzing how empty cruising influences parking dynamics and traffic flow across diverse urban zones; (2) examining its temporal effect on parking search and traffic congestion; and (3) assessing how CVs are affected when this strategy is adopted by varying numbers of AVs. The methodological framework of this research is based on agent-based modeling to investigate the spatiotemporal implications of AVs empty cruising at different adoption rates, compared with a full CV environment. The findings of this research show that when empty cruising is performed to address short-term parking demand of up to 45 min, it can reduce overall parking search by approximately 2.5%, with central business district (CBD) zones experiencing around 33% higher reduction compared with suburban Traffic Analysis Zones. However, this reduction in parking search ratio exacerbates congestion, increasing travel time in the CBD zones by around 56% and in suburban zones by around 21%. The insights of this research can guide policy formulation for balanced use of road and parking infrastructure in the age of AVs.
- 17 October 2026
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
Open AccessModeling the Spatiotemporal Implications of Autonomous Vehicles’ Empty Cruising on Parking Dynamics and Traffic Flow
Autonomous vehicles’ (AVs) ability to reposition themselves can transform parking dynamics. We focus on empty cruising as an alternative to short-term parking. This capability affects parking space occupancy and traffic flow because AVs use road space instead of parking spaces. This research models the spatiotemporal implications of empty cruising on parking dynamics and traffic flow in a mixed environment of AVs and conventional vehicles (CVs). The contributions of this research compared with existing research include: (1) analyzing how empty cruising influences parking dynamics and traffic flow across diverse urban zones; (2) examining its temporal effect on parking search and traffic congestion; and (3) assessing how CVs are affected when this strategy is adopted by varying numbers of AVs. The methodological framework of this research is based on agent-based modeling to investigate the spatiotemporal implications of AVs empty cruising at different adoption rates, compared with a full CV environment. The findings of this research show that when empty cruising is performed to address short-term parking demand of up to 45 min, it can reduce overall parking search by approximately 2.5%, with central business district (CBD) zones experiencing around 33% higher reduction compared with suburban Traffic Analysis Zones. However, this reduction in parking search ratio exacerbates congestion, increasing travel time in the CBD zones by around 56% and in suburban zones by around 21%. The insights of this research can guide policy formulation for balanced use of road and parking infrastructure in the age of AVs.
- Open Access
- 17 January 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
- Vol. 17 (1) , 1-26
OCP: Proactive Optimal Charging Planning for Electric Vehicles
Due to the limited driving range, insufficient charging facilities, and time-consuming recharging, optimizing charging routes for electric vehicles (EVs) presents unique challenges compared to conventional vehicles. The time and location of EV charging during a trip not only affect an individual EV’s travel time but also influence others, as queues may form at charging station(s). This issue is at large seen as a significant constraint for uplifting EV sales in many countries. In this study, we introduce a novel EV Route Planning problem, which involves two parts: (i) finding the fastest route with recharging for an EV routing request. We model the problem as a new graph problem and prove its NP-hardness. We propose an innovative two-phase algorithm that efficiently traverses the graph to identify the optimal charging route for each EV. (ii) We find routes with minimized travel time for an EV while strategically avoid charging stations and time to recharge at those stations which can lead to minimized travel time for upcoming EVs. For this purpose, we introduce the concept of an “influence factor” to guide heuristic decisions. Our results demonstrate that this method reduces total travel time by 50% compared to the state-of-the-art on real-world datasets, with benefits becoming more significant as the number of EVs on the road increases.
- 17 January 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
- Vol. 17 (1) , 1-26
Open AccessOCP: Proactive Optimal Charging Planning for Electric Vehicles
Due to the limited driving range, insufficient charging facilities, and time-consuming recharging, optimizing charging routes for electric vehicles (EVs) presents unique challenges compared to conventional vehicles. The time and location of EV charging during a trip not only affect an individual EV’s travel time but also influence others, as queues may form at charging station(s). This issue is at large seen as a significant constraint for uplifting EV sales in many countries. In this study, we introduce a novel EV Route Planning problem, which involves two parts: (i) finding the fastest route with recharging for an EV routing request. We model the problem as a new graph problem and prove its NP-hardness. We propose an innovative two-phase algorithm that efficiently traverses the graph to identify the optimal charging route for each EV. (ii) We find routes with minimized travel time for an EV while strategically avoid charging stations and time to recharge at those stations which can lead to minimized travel time for upcoming EVs. For this purpose, we introduce the concept of an “influence factor” to guide heuristic decisions. Our results demonstrate that this method reduces total travel time by 50% compared to the state-of-the-art on real-world datasets, with benefits becoming more significant as the number of EVs on the road increases.
- Open Access
Dynamic Configuration of On-Street Parking Spaces using Multi Agent Reinforcement Learning
With increased travelling needs more than ever, traffic congestion has become a major concern in most urban areas. Allocating spaces for on-street parking, further hinders traffic flow, by limiting the effective road width available for driving. With the advancement of vehicle-to-infrastructure connectivity technologies, we explore how the impact of on-street parking on traffic congestion could be minimized, by dynamically configuring on-street parking spaces. Towards that end, we formulate dynamic on-street parking space configuration as an optimization problem, and we follow a data driven approach, considering the nature of our problem. Our proposed solution comprises a two-layer multi agent reinforcement learning based framework, which is inherently scalable to large road networks. The lane level agents are responsible for deciding the optimal parking space configuration for each lane, and we introduce a novel Deep Q-learning architecture which effectively utilizes long short term memory networks and graph attention networks to capture the spatio-temporal correlations evident in the given problem. The block level agents control the actions of the lane level agents and maintain a sufficient level of parking around the block. We conduct a set of comprehensive experiments using SUMO, on both synthetic data as well as real-world data from the city of Melbourne. Our experiments show that the proposed framework could reduce the average travel time loss of vehicles significantly, reaching upto 47%, with a negligible increase in the walking distance for parking.
Open AccessDynamic Configuration of On-Street Parking Spaces using Multi Agent Reinforcement Learning
With increased travelling needs more than ever, traffic congestion has become a major concern in most urban areas. Allocating spaces for on-street parking, further hinders traffic flow, by limiting the effective road width available for driving. With the advancement of vehicle-to-infrastructure connectivity technologies, we explore how the impact of on-street parking on traffic congestion could be minimized, by dynamically configuring on-street parking spaces. Towards that end, we formulate dynamic on-street parking space configuration as an optimization problem, and we follow a data driven approach, considering the nature of our problem. Our proposed solution comprises a two-layer multi agent reinforcement learning based framework, which is inherently scalable to large road networks. The lane level agents are responsible for deciding the optimal parking space configuration for each lane, and we introduce a novel Deep Q-learning architecture which effectively utilizes long short term memory networks and graph attention networks to capture the spatio-temporal correlations evident in the given problem. The block level agents control the actions of the lane level agents and maintain a sufficient level of parking around the block. We conduct a set of comprehensive experiments using SUMO, on both synthetic data as well as real-world data from the city of Melbourne. Our experiments show that the proposed framework could reduce the average travel time loss of vehicles significantly, reaching upto 47%, with a negligible increase in the walking distance for parking.
- 28 November 2025
- Published by Springer Nature in Transportation
- p. 1-34
Melbourne’s parking dynamics and traffic flow: a dataset for microscopic agent-based simulation
- 28 November 2025
- Published by Springer Nature in Transportation
- p. 1-34
Melbourne’s parking dynamics and traffic flow: a dataset for microscopic agent-based simulation
- Open Access
- 3 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 77-89
DeepMDV: Global Spatial Matching for Multi-depot Vehicle Routing Problems
- 3 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 77-89
Open AccessDeepMDV: Global Spatial Matching for Multi-depot Vehicle Routing Problems
- Open Access
GrASP: A Generalizable Address-based Semantic Prefetcher for Scalable Transactional and Analytical Workloads
Data prefetching--loading data into the cache before it is requested--is essential for reducing I/O overhead and improving database performance. While traditional prefetchers focus on sequential patterns, recent learning-based approaches, especially those leveraging data semantics, achieve higher accuracy for complex access patterns. However, these methods often struggle with today's dynamic, ever-growing datasets and require frequent, timely fine-tuning. Privacy constraints may also restrict access to complete datasets, necessitating prefetchers that can learn effectively from samples. To address these challenges, we present GrASP, a learning-based prefetcher designed for both analytical and transactional workloads. GrASP enhances prefetching accuracy and scalability by leveraging logical block address deltas and combining query representations with result encodings. It frames prefetching as a context-aware multi-label classification task, using multi-layer LSTMs to predict delta patterns from embedded context. This delta modeling approach enables GrASP to generalize predictions from small samples to larger, dynamic datasets without requiring extensive retraining. Experiments on real-world datasets and industrial benchmarks demonstrate that GrASP generalizes to datasets 250 times larger than the training data, achieving up to 45% higher hit ratios, 60% lower I/O time, and 55% lower end-to-end query execution latency than existing baselines. On average, GrASP attains a 91.4% hit ratio, a 90.8% I/O time reduction, and a 57.1% execution latency reduction.
Open AccessGrASP: A Generalizable Address-based Semantic Prefetcher for Scalable Transactional and Analytical Workloads
Data prefetching--loading data into the cache before it is requested--is essential for reducing I/O overhead and improving database performance. While traditional prefetchers focus on sequential patterns, recent learning-based approaches, especially those leveraging data semantics, achieve higher accuracy for complex access patterns. However, these methods often struggle with today's dynamic, ever-growing datasets and require frequent, timely fine-tuning. Privacy constraints may also restrict access to complete datasets, necessitating prefetchers that can learn effectively from samples. To address these challenges, we present GrASP, a learning-based prefetcher designed for both analytical and transactional workloads. GrASP enhances prefetching accuracy and scalability by leveraging logical block address deltas and combining query representations with result encodings. It frames prefetching as a context-aware multi-label classification task, using multi-layer LSTMs to predict delta patterns from embedded context. This delta modeling approach enables GrASP to generalize predictions from small samples to larger, dynamic datasets without requiring extensive retraining. Experiments on real-world datasets and industrial benchmarks demonstrate that GrASP generalizes to datasets 250 times larger than the training data, achieving up to 45% higher hit ratios, 60% lower I/O time, and 55% lower end-to-end query execution latency than existing baselines. On average, GrASP attains a 91.4% hit ratio, a 90.8% I/O time reduction, and a 57.1% execution latency reduction.
- Open Access
- 5 August 2025
- Published by Springer Nature in Research Square
Melbourne’s Parking Dynamics and Traffic Flow: A Dataset for Microscopic Agent-based Simulation
Private vehicles park 95% of the time, prompting urban planners and policymakers to implement policies to manage parking supply and demand in city centres efficiently. Understanding the parking dynamics and its correlation with traffic flow is essential for effective urban infrastructure planning and policy formulation. This paper introduces a comprehensive dataset that includes real-world data on the road network and parking spaces, with traffic flow generated based on demographic information from the City of Melbourne. The dataset enables the investigation of the correlation between parking dynamics and traffic flow across city zones with heterogeneous parking demand and traffic flow characteristics. Designed for use with the Simulation of Urban MObility (SUMO) application, this dataset serves as a baseline to simulate and analyze parking dynamics and traffic flow at a microscopic level. Additionally, the paper provides the code necessary to generate similar datasets for other geographical locations or to adjust the parking dynamics and traffic flow parameters. Our dataset follows the FAIR principles (Findable, Accessible, Interoperable, and Reusable), ensuring that it is easy to locate, access, integrate with other data, and apply in various research contexts. This resource can significantly benefit researchers in transport and urban planning by facilitating the setup of real-world simulations of parking dynamics and traffic flow, thereby enabling a deeper understanding and more informed decision-making in urban infrastructure management.
- 5 August 2025
- Published by Springer Nature in Research Square
Open AccessMelbourne’s Parking Dynamics and Traffic Flow: A Dataset for Microscopic Agent-based Simulation
Private vehicles park 95% of the time, prompting urban planners and policymakers to implement policies to manage parking supply and demand in city centres efficiently. Understanding the parking dynamics and its correlation with traffic flow is essential for effective urban infrastructure planning and policy formulation. This paper introduces a comprehensive dataset that includes real-world data on the road network and parking spaces, with traffic flow generated based on demographic information from the City of Melbourne. The dataset enables the investigation of the correlation between parking dynamics and traffic flow across city zones with heterogeneous parking demand and traffic flow characteristics. Designed for use with the Simulation of Urban MObility (SUMO) application, this dataset serves as a baseline to simulate and analyze parking dynamics and traffic flow at a microscopic level. Additionally, the paper provides the code necessary to generate similar datasets for other geographical locations or to adjust the parking dynamics and traffic flow parameters. Our dataset follows the FAIR principles (Findable, Accessible, Interoperable, and Reusable), ensuring that it is easy to locate, access, integrate with other data, and apply in various research contexts. This resource can significantly benefit researchers in transport and urban planning by facilitating the setup of real-world simulations of parking dynamics and traffic flow, thereby enabling a deeper understanding and more informed decision-making in urban infrastructure management.
- 5 May 2025
- Published by Springer Nature in World Wide Web
- Vol. 28 (3) , 1-26
Identifying key nodes for enhancing community stability in bipartite networks
- 5 May 2025
- Published by Springer Nature in World Wide Web
- Vol. 28 (3) , 1-26
Identifying key nodes for enhancing community stability in bipartite networks
- 31 March 2025
- Published by Springer Nature in Knowledge and Information Systems
- Vol. 67 (7) , 5619-5649
Efficient stable community search in temporal signed graphs
- 31 March 2025
- Published by Springer Nature in Knowledge and Information Systems
- Vol. 67 (7) , 5619-5649
Efficient stable community search in temporal signed graphs