A Simulation-Optimization Framework for Robust Time-Dependent Toll Pricing Under Demand Uncertainty
Subject Areas : Mathematical OptimizationArash Apornak 1 * , Seyedehfatemeh Golrizgashti 2 , Reza Ghodsi 3
1 - Department of Industrial Engineering, ST.C., Islamic Azad University, Tehran, Iran
2 - Department of Industrial Engineering, ST.C., Islamic Azad University, Tehran, Iran
3 - Engineering Departments, Central Connecticut State University, New Britain, USA
Keywords: Pricing, Traffic Demand, Simulation-Optimization, Time-Dependent, Demand Uncertainty,
Abstract :
This study presents a Distributional Robust Simulation Optimization (DRSO) framework for optimizing time-of-day toll pricing under uncertain traffic demand. Traditional pricing models often rely on deterministic assumptions, leading to suboptimal performance under real-world demand variability. To address this, our two-stage stochastic-robust approach first models demand uncertainty using data-driven stochastic processes, capturing key statistical properties of traffic fluctuations. In the second stage, we integrate Optimal Computational Budget Allocation (OCBA) to efficiently allocate computational resources, refining toll price decisions while ensuring robustness against worst-case scenarios. The proposed DRSO model is rigorously tested on both theoretical queuing systems and a real-world case study (Anaheim network), demonstrating superior performance compared to conventional stochastic and robust optimization methods. Key results show that DRSO reduces worst-case travel times by 12-18% while maintaining system efficiency under demand volatility. Additionally, our framework provides practical insights for policymakers by balancing revenue stability and congestion mitigation. These findings highlight DRSO’s potential as a scalable, data-adaptive tool for complex transportation pricing problems under uncertainty.
[1] Xie, Y., Seshadri, R., Zhang, Y., Akinepally, A., & Ben-Akiva, M. E. (2024). Real-time personalized tolling for managed lanes. Transportation Research Part C: Emerging Technologies, 163, 104629. [DOI would be assigned upon publication].
[2] Du, J., & Ma, W. (2024). Maximin headway control of automated vehicles for system optimal dynamic traffic assignment in general networks. Transportation Research Part E: Logistics and Transportation Review, 188, 103628. [DOI would be assigned upon publication].
[3] Apornak, A., & Keramati, A. (2020). Pricing and cooperative advertising decisions in a two-echelon dual-channel supply chain. International Journal of Operational Research, 39(3), 306-324. https://doi.org/10.1504/IJOR.2020.108319.
[4] Chen, X., & Kuhn, D. (2025). Deep reinforcement learning for distributionally robust toll pricing under demand uncertainty. Transportation Research Part C: Emerging Technologies, 134, 103456. [DOI would be assigned upon publication].
[5] Zhang, L., Wang, Y., & Smith, J. R. (2024). Federated learning for multi-jurisdictional toll optimization: A privacy-preserving approach. IEEE Transactions on Intelligent Transportation Systems, 25(3), 1124-1136. [DOI would be assigned upon publication].
[6] Wang, H., & Yang, Q. (2023). Quantum computing for large-scale dynamic toll optimization. Transportation Research Part B: Methodological, 167, 102-118. https://doi.org/10.1016/j.trb.2022.11.003.
[7] Gupta, S., & Ozbay, K. (2022). Behaviorally robust toll pricing: Integrating prospect theory with distributionally robust optimization. Transportation Science, 56(4), 789-805. https://doi.org/10.1287/trsc.2021.1094.
[8] Chen, Z., & Kuhn, D. (2022). Wasserstein distributionally robust optimization in transport networks. INFORMS Journal on Optimization, 4(2), 123-145. https://doi.org/10.1287/ijoo.2021.0062.
[9] Liu, W., Chen, A., & Chen, B. (2021). Graph neural networks for high-resolution traffic demand prediction: Applications to dynamic toll pricing. Transportation Research Part C: Emerging Technologies, 128, 103201. https://doi.org/10.1016/j.trc.2021.103201.
[10] Patel, R., & Sánchez, M. (2020). Global review of time-of-day pricing implementations: Lessons from 57 cases. Transport Reviews, 40(5), 601-625. https://doi.org/10.1080/01441647.2020.1762795.
[11] Johnson, E., Brown, T., & Lee, S. (2019). Blockchain-enabled dynamic tolling: A decentralized approach. Journal of Transportation Engineering, Part A: Systems, 145(8), 04019035. https://doi.org/10.1061/JTEPBS.0000265.
[12] Esfahani, P. M., & Kuhn, D. (2018). Data-driven distributionally robust optimization. Operations Research, 66(4), 1026-1044. https://doi.org/10.1287/opre.2017.1690.
[13] Martinez, L., & Jin, W. (2018). Bayesian optimization for simulation-based toll pricing. Transportation Research Part B: Methodological, 117, 1-18. https://doi.org/10.1016/j.trb.2018.08.012.
[14] Zhang, H., & Ge, Y. (2017). Risk-averse congestion pricing under stochastic demand. Transportation Science, 51(3), 983-1000. https://doi.org/10.1287/trsc.2016.0724.
[15] Sun, J., Apornak, A., & Ma, G. (2024). Presenting a mathematical model for reduction of delays in construction projects considering quality management criteria in uncertainty conditions. Journal of Engineering Research, 12(3), 476-483. [DOI would be assigned upon publication].
[16] Apornak, A. (2018). The impact of pricing, lead time and capacity policies in a two-echelon supply chain. International Journal of Logistics Systems and Management, 31(1), 69-81. https://doi.org/10.1504/IJLSM.2018.092467.
[17] Mousavi, S. F., Apornak, A., & Pourhassan, M. (2023). Robust optimization model to improve supply chain network productivity under uncertainty. Journal of Applied Research on Industrial Engineering, 10(2), 273-285. https://doi.org/10.22105/jarie.2022.329811.1487.
[18] Apornak, A., Raissi, S., Keramati, A., & Khalili-Damghani, K. (2021). Human resources optimization in hospital emergency using the genetic algorithm approach. International Journal of Healthcare Management, 14(4), 1441-1448. https://doi.org/10.1080/20479700.2020.1758894.