Melike Baykal-Gürsoy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Melike Baykal-Gürsoy has a total of 56 co-authors. Here are the top co-authors by number of shared publications:
Andrey Garnaev
5 shared publications- Harold Vincent Poor
Princeton University
5 shared publications - Abdolmajid Yolmeh
Department of Industrial Engineering, Iran University of Science and Technology, Tehran, Iran
4 shared publications Zhe Duan
Department of Management Science, School of Management, Xi’An Jiaotong University, China
4 shared publicationsAndrew Reed Benton
Department of Industrial and Systems Engineering, Rutgers University, New Brunswick, NJ
3 shared publicationsAbdolmajid Yolmeh
UPS, Louisville, KY, USA
2 shared publications
Latest Publications
- 4 December 2025
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2680 (7) , 711-730
Probabilistic Deep Learning for Traffic Density Prediction
Real time, accurate predictions of recurrent and nonrecurrent traffic congestion are essential for optimizing transportation systems and ensuring a smooth user experience. Traditional models often focus on long-term point estimates, limiting their use in scenarios requiring short-term predictions or probabilistic assessments (e.g., traffic signal optimization, dynamic tolling, and emergency response). This study explores probabilistic deep learning (DL) for real time traffic density distribution prediction. This study demonstrates that an adapted multi-quantile recurrent neural network (MQRNN), termed MQRNN-monotonic, outperforms traditional time series methods, particularly when handling nonrecurrent disruptions. A novel loss function is introduced to address quantile crossing issues, ensuring valid distributional predictions. Experiments on two highway data sets show that probabilistic DL for traffic density prediction yields well-calibrated and sharp dynamic traffic congestion distributions. This study offers a promising new approach to real time traffic density forecasting, paving the way for transportation systems that respond faster and smarter to changing road conditions, making traffic smoother, more sustainable, and more predictable for everyone.
- 4 December 2025
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2680 (7) , 711-730
Probabilistic Deep Learning for Traffic Density Prediction
Real time, accurate predictions of recurrent and nonrecurrent traffic congestion are essential for optimizing transportation systems and ensuring a smooth user experience. Traditional models often focus on long-term point estimates, limiting their use in scenarios requiring short-term predictions or probabilistic assessments (e.g., traffic signal optimization, dynamic tolling, and emergency response). This study explores probabilistic deep learning (DL) for real time traffic density distribution prediction. This study demonstrates that an adapted multi-quantile recurrent neural network (MQRNN), termed MQRNN-monotonic, outperforms traditional time series methods, particularly when handling nonrecurrent disruptions. A novel loss function is introduced to address quantile crossing issues, ensuring valid distributional predictions. Experiments on two highway data sets show that probabilistic DL for traffic density prediction yields well-calibrated and sharp dynamic traffic congestion distributions. This study offers a promising new approach to real time traffic density forecasting, paving the way for transportation systems that respond faster and smarter to changing road conditions, making traffic smoother, more sustainable, and more predictable for everyone.
- Open Access
- 23 December 2024
- Published by Cambridge University Press (CUP) in Probability in the Engineering and Informational Sciences
- Vol. 39 (2) , 188-216
Completion times of jobs on two-state service processes and their asymptotic behavior
We consider the task completion time of a repairable server system in which a server experiences randomly occurring service interruptions during which the server works slowly. Every service-state change preempts the task that is being processed. The server may then resume the interrupted task, it may replace the task with a different one, or it may restart the same task from the beginning, under the new service-state. The total time that the server takes to complete a task of random size including interruptions is called completion time. We study the completion time of a task under the last two cases as a function of the task size distribution, the service interruption frequency/severity, and the repair frequency. We derive closed form expressions for the completion time distribution in Laplace domain under replace and restart recovery disciplines and present their asymptotic behavior. In general, the heavy tailed behavior of completion times arises due to the heavy tailedness of the task time. However, in the preempt-restart service discipline, even in the case that the server still serves during interruptions albeit at a slower rate, completion times may demonstrate power tail behavior for exponential tail task time distributions. Furthermore, we present an $M/G/\infty$ queue with exponential service time and Markovian service interruptions. Our results reveal that the stationary first order moments, that is, expected system time and expected number in the system are insensitive to the way the service modulation affects the servers; system-wide modulation affecting every server simultaneously vs identical modulation affecting each server independently.- 23 December 2024
- Published by Cambridge University Press (CUP) in Probability in the Engineering and Informational Sciences
- Vol. 39 (2) , 188-216
Open AccessCompletion times of jobs on two-state service processes and their asymptotic behavior
We consider the task completion time of a repairable server system in which a server experiences randomly occurring service interruptions during which the server works slowly. Every service-state change preempts the task that is being processed. The server may then resume the interrupted task, it may replace the task with a different one, or it may restart the same task from the beginning, under the new service-state. The total time that the server takes to complete a task of random size including interruptions is called completion time. We study the completion time of a task under the last two cases as a function of the task size distribution, the service interruption frequency/severity, and the repair frequency. We derive closed form expressions for the completion time distribution in Laplace domain under replace and restart recovery disciplines and present their asymptotic behavior. In general, the heavy tailed behavior of completion times arises due to the heavy tailedness of the task time. However, in the preempt-restart service discipline, even in the case that the server still serves during interruptions albeit at a slower rate, completion times may demonstrate power tail behavior for exponential tail task time distributions. Furthermore, we present an $M/G/\infty$ queue with exponential service time and Markovian service interruptions. Our results reveal that the stationary first order moments, that is, expected system time and expected number in the system are insensitive to the way the service modulation affects the servers; system-wide modulation affecting every server simultaneously vs identical modulation affecting each server independently. - 12 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 18 (15566013) , 846-858
Power Allocation for Cooperative Jamming Against a Strategic Eavesdropper Over Parallel Channels
This paper considers a friendly interferer allocating jamming power to eavesdropping channels to increase the level of secrecy of a wireless network. The friendly interferer has access to limited power, while the eavesdropper may not have the ability to attack all channels simultaneously. When all channels used for secret communication are under the threat of eavesdropping attacks, the optimal power allocation policy results from solving a convex optimization problem. In this case, the optimal policy is unique and can be obtained via a water-filling scheme. When the eavesdropper cannot attack all channels, the eavesdropper should behave strategically and may select the targets probabilistically. We propose a non-zero-sum game that helps the friendly interferer predict and concentrate on the targets selected by the eavesdropper. Under certain conditions, we prove that there exists a unique Nash equilibrium (NE) strategy pair, which has a threshold type structure. We provide conditions under which the eavesdropper’s equilibrium strategy is deterministic. We devise a strategy iteration algorithm to compute an equilibrium power allocation strategy. We present examples showing that the game-theoretic power allocation strategy performs better than the conservative power allocation strategy that assumes every channel to be under attack.
- 12 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 18 (15566013) , 846-858
Power Allocation for Cooperative Jamming Against a Strategic Eavesdropper Over Parallel Channels
This paper considers a friendly interferer allocating jamming power to eavesdropping channels to increase the level of secrecy of a wireless network. The friendly interferer has access to limited power, while the eavesdropper may not have the ability to attack all channels simultaneously. When all channels used for secret communication are under the threat of eavesdropping attacks, the optimal power allocation policy results from solving a convex optimization problem. In this case, the optimal policy is unique and can be obtained via a water-filling scheme. When the eavesdropper cannot attack all channels, the eavesdropper should behave strategically and may select the targets probabilistically. We propose a non-zero-sum game that helps the friendly interferer predict and concentrate on the targets selected by the eavesdropper. Under certain conditions, we prove that there exists a unique Nash equilibrium (NE) strategy pair, which has a threshold type structure. We provide conditions under which the eavesdropper’s equilibrium strategy is deterministic. We devise a strategy iteration algorithm to compute an equilibrium power allocation strategy. We present examples showing that the game-theoretic power allocation strategy performs better than the conservative power allocation strategy that assumes every channel to be under attack.
- 11 November 2022
- Published by Springer Nature in Annals of Operations Research
- Vol. 320 (1) , 493-507
A decomposable resource allocation model with generalized overarching protections
- 11 November 2022
- Published by Springer Nature in Annals of Operations Research
- Vol. 320 (1) , 493-507
A decomposable resource allocation model with generalized overarching protections
- 19 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 23 (8) , 13000-13010
How Random Incidents Affect Travel-Time Distributions
We present a novel analytical model to approximate the travel-time distribution of vehicles traversing a freeway corridor that experiences random quality of service degradations due to non-recurrent incidents. The proposed model derives the generating function of travel times in closed-form using clearance time, incident frequency and severity, and other ordinary traffic characteristics. We validate the model using data from a freeway corridor where weather events and traffic accidents serve as the principal causes of service degradation. The resulting model is equivalent in performance to widely used methodologies while uniquely providing a clear connection on how incidents affect travel time distribution. With this connection, the model readily yields travel time reliability measures for alternative roadway behaviors, providing crucial information for long-term planning.
- 19 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 23 (8) , 13000-13010
How Random Incidents Affect Travel-Time Distributions
We present a novel analytical model to approximate the travel-time distribution of vehicles traversing a freeway corridor that experiences random quality of service degradations due to non-recurrent incidents. The proposed model derives the generating function of travel times in closed-form using clearance time, incident frequency and severity, and other ordinary traffic characteristics. We validate the model using data from a freeway corridor where weather events and traffic accidents serve as the principal causes of service degradation. The resulting model is equivalent in performance to widely used methodologies while uniquely providing a clear connection on how incidents affect travel time distribution. With this connection, the model readily yields travel time reliability measures for alternative roadway behaviors, providing crucial information for long-term planning.
- 17 September 2021
- Published by Springer Nature
- p. 25-37
A Friendly Interference Game in Wireless Secret Communication Networks
- 17 September 2021
- Published by Springer Nature
- p. 25-37
A Friendly Interference Game in Wireless Secret Communication Networks
- 1 February 2021
- Published by Elsevier in European Journal of Operational Research
- Vol. 289 (1) , 338-349
Weighted network search games with multiple hidden objects and multiple search teams
- 1 February 2021
- Published by Elsevier in European Journal of Operational Research
- Vol. 289 (1) , 338-349
Weighted network search games with multiple hidden objects and multiple search teams
- 20 August 2020
- Published by Taylor & Francis in Terrorism and Political Violence
- Vol. 34 (7) , 1386-1414
Behavior Associations in Lone-Actor Terrorists
Terrorist attacks carried out by individuals have significantly accelerated over the last twenty years. This type of lone-actor (LA) terrorism stands as one of the greatest security threats of our time. While the research on LA behavior and characteristics has produced valuable information on demographics, classifications, and warning signs, the relationship among these characteristics is yet to be addressed. Moreover, the means of radicalization and attacking have changed over decades. This study conducts an a-posteriori analysis of the temporal changes in LA terrorism and behavioral associations in LAs. We initially identify twenty-five binary behavioral characteristics of LAs and analyze 190 LAs. Next, we classify LAs according to ideology first, incident-scene behavior (determined via a virtual attacker-defender game) secondly, and, finally, the clusters obtained from the data. In addition, within each class, statistically significant associations and temporal relations are extracted using the A-priori algorithm. These associations would be instrumental in identifying the attacker’s type and intervening at the right time. The results indicate that while pre-9/11 LAs were mostly radicalized by the people in their environment, post-9/11 LAs are more diverse. Furthermore, association chains for different LA types present unique characteristic pathways to violence and after-attack behavior.
- 20 August 2020
- Published by Taylor & Francis in Terrorism and Political Violence
- Vol. 34 (7) , 1386-1414
Behavior Associations in Lone-Actor Terrorists
Terrorist attacks carried out by individuals have significantly accelerated over the last twenty years. This type of lone-actor (LA) terrorism stands as one of the greatest security threats of our time. While the research on LA behavior and characteristics has produced valuable information on demographics, classifications, and warning signs, the relationship among these characteristics is yet to be addressed. Moreover, the means of radicalization and attacking have changed over decades. This study conducts an a-posteriori analysis of the temporal changes in LA terrorism and behavioral associations in LAs. We initially identify twenty-five binary behavioral characteristics of LAs and analyze 190 LAs. Next, we classify LAs according to ideology first, incident-scene behavior (determined via a virtual attacker-defender game) secondly, and, finally, the clusters obtained from the data. In addition, within each class, statistically significant associations and temporal relations are extracted using the A-priori algorithm. These associations would be instrumental in identifying the attacker’s type and intervening at the right time. The results indicate that while pre-9/11 LAs were mostly radicalized by the people in their environment, post-9/11 LAs are more diverse. Furthermore, association chains for different LA types present unique characteristic pathways to violence and after-attack behavior.
- 1 June 2020
- Published by Elsevier in European Journal of Operational Research
- Vol. 283 (3) , 942-953
Mean-variance analysis of the newsvendor problem with price-dependent, isoelastic demand
- 1 June 2020
- Published by Elsevier in European Journal of Operational Research
- Vol. 283 (3) , 942-953
Mean-variance analysis of the newsvendor problem with price-dependent, isoelastic demand