Michael Levin - Research

Transportation network design

Where should new roads be built, or capacity added to existing roads, to minimize traffic congestion? Network design problems are frequently encountered by practitioners and researchers in traditional contexts (roads, public transit) and emerging technologies (automated vehicles, electric vehicle charging stations) . This problem is difficult due to the Braess paradox, where drivers trying to minimize their individual travel times cause congestion to increase after a new road is added. Mathematically, it is a bi-level nonlinear optimization problem that can be discrete or continuous. The bi-level structure makes it non-convex and NP-hard.

I have been developing globally optimal solution algorithms that compute a converging sequence of lower bounds using a relaxation, compared with upper bounds from solving traffic assignment for specific leader solutions. Some methods can solve network design on networks with ~900 links (Berlin Mitte Center and Anaheim) in less than 1 hour.

Related papers

M.W. Levin & D. Rey (2026). A spatial branch-and-price-and-cut algorithm for finding globally optimal solutions to the continuous network design problem. Transportation Research Part E: Logistics and Transport Review, 211, 104870.

Max-pressure intersection control

Max-pressure intersection control models the network as a Markov decision process; queue lengths are the state, traffic signals are the control variables, and stochasticity comes from entering demand and turning ratios. Rather than solve it numerically (which would be computationally intractable due to the curse of dimensionality), max-pressure control proves that an analytical pressure-based adaptive signal control results in a positive recurrent Markov chain if at all possible. Equivalently, network outflow = inflow for max-pressure control if it could be achieved by any signal timing. Furthermore, max-pressure control is decentralized; each intersection optimizes its own timing based on the surrounding vehicle queues.

Picture of intelligent intersection control

Related papers

H. Liu, V. Gayah, & M. W. Levin (2024). A max pressure algorithm for traffic signals considering pedestrian queues. Transportation Research Part C: Emerging Technologies, 169, 104865.
T. Xu, S. Barman, & M. W. Levin (2024). Smoothing-MP: A novel max-pressure signal control considering signal coordination to smooth traffic in urban networks. Transportation Research Part C: Emerging Technologies, 166, 104760.
T. Xu, Y. Bika, & M. W. Levin (2024). Ped-MP: A pedestrian-friendly max-pressure signal control policy for city networks. ASCE Journal of Transportation Engineering, Part A: Systems, 150(7), 04024028.
S. Barman & M. W. Levin (2023). Throughput properties and optimal locations for limited deployment of max-pressure controls. Transportation Research Part C: Emerging Technologies, 150, 140105.
M. W. Levin (2023). Max-pressure traffic signal timing: a summary of methodological and experimental results. ASCE Journal of Transportation Engineering, Part A: Systems, 149(4), 03123001. (2025 Alfred Noble Prize)
J. Robbennolt, R. Chen, & M. W. Levin (2022). Microsimulation study evaluating the benefits of cyclic and non-cyclic max-pressure control of signalized intersections. Transportation Research Record, 03611981221095520.
T. Xu, S. Barman, M. W. Levin, R. Chen, & T. Li (2022). Integrating public transit signal priority into max-pressure signal control: methodology and simulation study on a downtown network. Transportation Research Part C: Emerging Technologies, 138, 103614.
S. Barman & M. W. Levin (2022). Performance evaluation of modified cyclic max pressure controlled intersections in realistic corridors. Transportation Research Record, 03611981211072807.
M. W. Levin, J. Hu, & M. Odell (2020). Max-pressure signal control with cyclical phase structure. Transportation Research Part C: Emerging Technologies, 120, 102838.
V. Dixit, D. Jayakumar Nair, S. Chand, & M. W. Levin (2020). A simple crowdsourced delay-based traffic signal control. PLOS ONE, 15(4), e0230598.
R. Chen, J. Hu, M. W. Levin, & D. Rey (2020). Stability-based analysis of autonomous intersection management with pedestrians. Transportation Research Part C: Emerging Technologies, 114, 463–483.
M. W. Levin, D. Rey, & A. Schwartz (2019). Max-pressure control of dynamic lane reversal and autonomous intersection management. Transportmetrica B: Transport Dynamics, 7(1), 1693–1718.
D. Rey & M. W. Levin (2019). Blue phase: Optimal network traffic control for legacy and autonomous vehicles. Transportation Research Part B: Methodological, 130, 105–129.

Transportation and the environment

28% of U.S. greenhouse gas emissions were caused by transportation in 2016. New vehicle technologies such as autonomous vehicles and intelligent transportation systems could reduce the environmental impacts in several ways. Technologies that reduce congestion also improve the fuel efficiency. Autonomous vehicles could support alternative fuels such as electric vehicles through empty repositioning to improve access to recharging. Reservation-based intersection controls could reduce acceleration/deceleration cycles at intersections.

Related papers

S. He, S. Wang, Y. Shao, Z. Sun, & M. W. Levin (2026). A connectivity-based real-time traffic prediction considering lane-changing maneuvers with application to eco-driving control of electric vehicles. IEEE Transactions on Vehicular Technology, 75(1), 168–181.
M. W. Levin & D. Rey (2025). Charging station location and fleet sizing for shared autonomous electric vehicles using Benders' decomposition. Networks and Spatial Economics, 25, 681–716.
M. Zamanpour, S. He, M. W. Levin, Z. Sun (2025). Incorporating lane-change prediction into energy-efficient speed control of connected autonomous vehicles at intersections. Transportation Research Part C: Emerging Technologies, 171, 104968.
R. Cotta Antunez & M. W. Levin (2024). How does eco-routing affect total system emissions? City network predictions from user equilibrium models. IEEE Transactions on Intelligent Transportation Systems, 25(2), 1408–1417.
W. Sun, S. Wang, Y. Shao, Z. Sun, & M. W. Levin (2022). Energy and mobility impacts of connected autonomous vehicles with co-optimization of speed and powertrain on mixed vehicle platoons. Transportation Research Part C: Emerging Technologies, 142, 103764.
R. Shah, Nezamuddin, & M. W. Levin (2020). Supply-side network effects on mobile-source emissions. Transport Policy, 98, 21–34.
R. Chen, T. Zhang, & M. W. Levin (2020). Effects of variable speed limit on energy consumption with autonomous vehicles on urban roads using modified cell transmission model. ASCE Journal of Transportation Engineering, Part A: Systems, 146(7), 0000379.
M. W. Levin, E. Jafari, R. Shah, & S. D. Boyles (2017). Network-based model for predicting the effect of fuel price on transit ridership and greenhouse gas emissions. International Journal of Transportation Science and Technology, 6(4), 272–286.
M. W. Levin, H. Fritz, & S. D. Boyles (2017). On optimizing reservation-based intersection controls. IEEE Transactions on Intelligent Transportation Systems, 18(3), 505–515.
M. W. Levin, M. Duell, & S. T. Waller (2014). Effect of road grade on network wide vehicle energy consumption and eco-routing. Transportation Research Record: Journal of the Transportation Research Board, (2427), 26–33.

Non-motorized traffic

Government agencies often spend far more effort on transportation systems designed for motor vehicles, but pedestrians and cyclist are traffic that needs to be supported too. How can network methods be used to understand pedestrian and cyclist behavior? How should proposed methods and algorithms for traffic control and transportation design consider non-motorized traffic?

Related papers

M. W. Levin, J. Nie, & J. Grzesiak (2025). Does traveling to school before sunrise affect whether elementary school students walk or bike? Results from stated and revealed preference surveys in Minnesota. Transport Policy, 171, 476–486.
S. Barman, M. W. Levin, R. Stern, & G. Lindsey (2025). Efficient pedestrian and bicycle traffic flow estimation combining mobile-sourced data with route choice prediction. Transportation Research Part C: Emerging Technologies, 173, 105046.
H. Liu, V. Gayah, & M. W. Levin (2024). A max pressure algorithm for traffic signals considering pedestrian queues. Transportation Research Part C: Emerging Technologies, 169, 104865.
T. Xu, Y. Bika, & M. W. Levin (2024). Ped-MP: A pedestrian-friendly max-pressure signal control policy for city networks. ASCE Journal of Transportation Engineering, Part A: Systems, 150(7), 04024028.
T. Tao, G. Lindsey, R. Stern, & M. W. Levin (2024). The use of crowdsourced mobile data in estimating pedestrian and bicycle traffic: a systematic review. Journal of Transport and Land Use, 17(1), 41–65.

Maziar Zamanpour and Jianshe Guo demonstrate a system to warn drivers to brake to avoid running a red light, using traffic signal timing broadcast by an intersection in Scott County.

Connected/automated vehicle control

Vehicle-to-infrastructure and vehicle-to-vehicle connectivity provides extra information about traffic conditions and interactive communications. When combined with new methodologies to integrate vehicle control with digital communications, automated vehicles can be optimized to improve traffic flow or energy consumption. For example, connected/automated platoons can achieve freeway free flow speeds at high densities. Even without vehicle automation, connectivity-based driver guidance can improve safety by warning drivers about traffic conditions.

Related papers

S. He, M. Zamanpour, J. Guo, M. W. Levin, & Z. Sun (2025). A traffic prediction-based individualized driver warning system to reduce red light violations. ASCE Journal of Transportation Engineering, Part A: Systems, 152(3), 04025149.
M. Zamanpour, S. He, M. W. Levin, Z. Sun (2025). Incorporating lane-change prediction into energy-efficient speed control of connected autonomous vehicles at intersections. Transportation Research Part C: Emerging Technologies, 171, 104968.
W. Sun, S. Wang, Y. Shao, Z. Sun, & M. W. Levin (2022). Energy and mobility impacts of connected autonomous vehicles with co-optimization of speed and powertrain on mixed vehicle platoons. Transportation Research Part C: Emerging Technologies, 142, 103764.
Z. Li, M. W. Levin, X. Qu, & R. Stern (2023). A network traffic model for the control of autonomous vehicles acting as moving bottlenecks. IEEE Transactions on Intelligent Transportation Systems, 24(9), 9004–9015.
M. W. Levin & D. Kang (2023). A multiclass link transmission model for a class-varying capacity and congested wave speed. ASCE Journal of Transportation Engineering, Part A: Systems, 149(10), 04023096.
S. Wang, M. W. Levin, & R. Stern (2023). Optimal feedback control law for automated vehicles in the presence of cyberattacks: a min-max approach. Transportation Research Part C: Emerging Technologies, 153, 104204.
S. Wang, M. Shang, M. W. Levin, & R. Stern (2023). A general approach to smoothing nonlinear mixed traffic via control of autonomous vehicles. Transportation Research Part C: Emerging Technologies, 146, 103967.
S. Wang, R. Stern, & M. W. Levin (2022). Optimal control of autonomous vehicles for traffic smoothing. IEEE Transactions on Intelligent Transportation Systems, 23(4), 3842–3852.
Y. Shao, M. W. Levin, S. D. Boyles, & C. Claudel (2021). Semi-analytical computation of solutions to the Lighthill-Whitham-Richards equation with switched triangular diagrams: application to variable speed limit control. IEEE Transactions on Automation Science and Engineering, 19(1), 473–485.
S. Chakraborty, D. Rey, M. W. Levin, & S. T. Waller (2021). Freeway network design with exclusive lanes for automated vehicles under endogenous mobility demand. Transportation Research Part C: Emerging Technologies, 133, 103440.
C. L. Melson, M. W. Levin, B. E. Hammit, & S. D. Boyles (2018). Dynamic traffic assignment of cooperative adaptive cruise control. Transportation Research Part C: Emerging Technologies, 90, 114–133.
R. Patel, M. W. Levin, & S. D. Boyles (2016). Effects of autonomous vehicle behavior on arterial and freeway networks. Transportation Research Record: Journal of the Transportation Research Board, (2561), 9–17.
M. W. Levin & S. D. Boyles (2016). A multiclass cell transmission model for shared human and autonomous vehicle roads. Transportation Research Part C: Emerging Technologies, 62, 103–116.
M. W. Levin & S. D. Boyles (2015). Effects of autonomous vehicle ownership on trip, mode, and route choice. Transportation Research Record: Journal of the Transportation Research Board, (2493), 29–38.

Shared autonomous vehicles

Instead of autonomous vehicles being owned privately, they might instead be used as part of a ridesharing service. Without the need to pay a driver, the cost may be competitive to personal vehicles. Shared autonomous vehicles create many different challenges, such as how to match passengers to vehicles, and how to coordinate vehicle route choice. If the fleet of shared autonomous vehicles is sufficiently large, they may contribute significantly to congestion.

Related papers

M. W. Levin & D. Rey (2025). Charging station location and fleet sizing for shared autonomous electric vehicles using Benders' decomposition. Networks and Spatial Economics, 25, 681–716.
T. Xu, M. Cieniawski, & M. W. Levin (2024). FMS-Dispatch: A fast maximum stability dispatch policy for shared autonomous vehicles including exiting passengers under stochastic travel demand. Transportmetrica A: Transport Science, 20(3), 10.1080/23249935.2023.2214968.
J. Robbennolt & M. W. Levin (2023). Maximum throughput dispatch for shared autonomous vehicles including vehicle rebalancing. IEEE Transactions on Intelligent Transportation Systems, 24(9), 9871–9885.
M. W. Levin (2022). A general maximum-stability dispatch policy for shared autonomous vehicle dispatch with a characterization of the stable region of demand. Transportation Research Part B: Methodological, 163, 258–280.
D. Kang & M. W. Levin (2021). Maximum-stability dispatch policy for shared autonomous vehicles. Transportation Research Part B: Methodological, 148, 132–151.
P. Venkatraman & M. W. Levin (2021). A congestion-aware tabu search heuristic to solve the shared autonomous vehicle routing problem. Journal of Intelligent Transportation Systems, 25(4), 343–355.
M. W. Levin, M. Odell, S. Samarasena, & A. Schwartz (2019). A linear program for optimal integration of shared autonomous vehicles with public transit. Transportation Research Part C: Emerging Technologies, 109, 267–288.
R. Chen & M. W. Levin (2019). Dynamic user equilibrium of mobility-on-demand system with linear programming rebalancing strategy. Transportation Research Record: Journal of the Transportation Research Board, (2673), 447–459. (Winner, 2019 Ryuichi Kitamura Award.)
M. W. Levin (2017). Congestion-aware system optimal route choice for shared autonomous vehicles. Transportation Research Part C: Emerging Technologies, 82, 229–247.
M. W. Levin, S. D. Boyles, K. Kockelman, & T. Li (2017). A general framework for modeling shared autonomous vehicles, with dynamic ride-sharing and dynamic traffic assignment application. Computers, Environment, and Urban Systems, 64, 373–383.

Autonomous vehicles and travel demand

Autonomous vehicles permit hands-free driving, and vehicles could even travel empty. Travelers may be willing to drive more, live farther from work, or avoid parking costs by sending their vehicle on empty trips to park elsewhere. Overall, autonomous vehicles could cause increases in the number and length of trips, resulting in significant increases in travel demand.

NCTCOG website

Related papers

S. Wang, Z. Li, & M. W. Levin (2022). Optimal policy for integrating autonomous vehicles into the auto market. Transportation Research Part C: Emerging Technologies, 143, 103821.
D. Kang, F. Hu, & M. W. Levin (2022). Impact of automated vehicles on traffic assignment, mode split, and parking behavior. Transportation Research Part D: Transport and Environment, 104, 103200.
S. Wang, M. W. Levin, & R. Caverly (2021). Optimal parking management of connected autonomous vehicles: a control-theoretic approach. Transportation Research Part C: Emerging Technologies, 124, 102924.
M. W. Levin, E. Wong, B. Nault-Maurer, & A. Khani (2020). Parking infrastructure design for repositioning autonomous vehicles. Transportation Research Part C: Emerging Technologies, 120, 102828.
M. W. Levin, H. Smith, & S. D. Boyles (2019). A dynamic four-step planning model of empty repositioning trips for personal autonomous vehicles. ASCE Journal of Transportation Engineering, Part A: Systems, 145(5).
M. W. Levin & S. D. Boyles (2015). levi. Transportation Research Record: Journal of the Transportation Research Board, (2493), 29–38.

Signal-free control

Instead of conventional traffic signals, automated vehicles can use individual communications with intersection controllers to avoid conflicts. Vehicles use wireless communications to request an intersection movement at a specific time and speed from the intersection manager. The intersection manager simulates requests and accepts a non-conflicting subset. Vehicle connectivity is required for coordination, and automation is needed for precise adherence to trajectories.

Jack Olsson demonstrating optimized signal-free control in AIMSUN.

Related papers

S. Wang, A. Mahlberg, & M. W. Levin (2023). Optimal control of automated vehicles for autonomous intersection management with design specifications. Transportation Research Record, 03611981221109166.
D. Kang, Z. Li, & M. W. Levin (2022). Evasion planning for autonomous intersection control based on an optimized conflict point control formulation. Journal of Transportation Safety & Security, 14(12), 2074–2110.
J. Olsson & M. W. Levin (2020). Integration of microsimulation and optimized autonomous intersection management. ASCE Journal of Transportation Engineering, Part A: Systems, 146(9), 0000407.
R. Chen, J. Hu, M. W. Levin, & D. Rey (2020). Stability-based analysis of autonomous intersection management with pedestrians. Transportation Research Part C: Emerging Technologies, 114, 463–483.
D. Rey & M. W. Levin (2019). Blue phase: Optimal network traffic control for legacy and autonomous vehicles. Transportation Research Part B: Methodological, 130, 105–129.
M. W. Levin, D. Rey, & A. Schwartz (2019). Max-pressure control of dynamic lane reversal and autonomous intersection management. Transportmetrica B: Transport Dynamics, 7(1), 1693–1718.
E. Lukose, M. W. Levin, & S. D. Boyles (2019). Incorporating insights from signal optimization into reservation-based intersection controls. Journal of Intelligent Transportation Systems, 23(3) 250–264.
M. W. Levin & D. Rey (2017). Conflict-point formulation of intersection control for autonomous vehicles. Transportation Research Part C: Emerging Technologies, 85, 528–547.
M. W. Levin, H. Fritz, & S. D. Boyles (2017). On optimizing reservation-based intersection controls. IEEE Transactions on Intelligent Transportation Systems, 18(3), 505–515.
M. W. Levin, S. D. Boyles, & R. Patel (2016). Paradoxes of reservation-based intersection controls in traffic networks. Transportation Research Part A: Policy and Practice, 90, 14–25.
M. W. Levin & S. D. Boyles (2016). A multiclass cell transmission model for shared human and autonomous vehicle roads. Transportation Research Part C: Emerging Technologies, 62, 103–116.
M. W. Levin & S. D. Boyles (2015). Intersection auctions and reservation-based control in dynamic traffic assignment. Transportation Research Record: Journal of the Transportation Research Board, (2497), 35–44.

Report abuse