Abstract
We propose an incentive-based traffic demand management policy to alleviate traffic congestion on a road stretch that creates a bottleneck for commuters. The incentive targets electric vehicle (EV) owners by proposing a discount on the energy price used to charge their vehicles if they are flexible in their departure time. We show that, with a sufficient monetary budget, it is possible to completely eliminate traffic congestion and we compute the optimal discount. We also analyze the case of a limited budget, where congestion cannot be completely eliminated, computing the policy that minimizes congestion and estimating the level of inefficiency for different budgets. Numerical simulations corroborate our theoretical findings, providing practical advice for policy design.
Introduction
Traffic congestion at bottlenecks—such as bridges, tunnels, or narrow highway segments—is a persistent urban challenge, leading to significant economic waste and environmental pollution. Traditional "hard" management policies, like congestion pricing or tolls, are often effective but face significant political opposition and are criticized for being regressive.
With the rise of Electric Vehicles (EVs), a new lever for influence has emerged: charging. This paper explores an "Incentive-Based Traffic Demand Management" (IBTDM) approach. The goal is to determine if offering energy discounts to EV owners who are flexible with their departure times can effectively "flatten the curve" of rush-hour traffic without the need for mandatory tolls.
Modeling the EV Commuter
The study extends the classic Vickrey Bottleneck Model to incorporate the unique characteristics of EV ownership. In this framework, commuters choose their departure time to minimize a total cost function that balances travel delay (time spent in a queue) against "schedule delay" (the cost of arriving earlier or later than desired).
Two critical extensions are introduced:
- Joint Decision Making: Commuters decide both when to enter the bottleneck and how long to spend at a participating charging station.
- Diminishing Perceived Utility: The model accounts for the psychological reality of "range anxiety." The first few kilowatt-hours (kWh) of charge are perceived as more valuable for peace of mind than the final kWh. This creates an "inefficiency gap" between the policy maker's expenditure and the utility perceived by the user.
Designing the Incentive Policy
The core of the IBTDM policy is the distribution of the energy discount $p(t)$ over time. To eliminate congestion, the highest incentives are not placed at the peak of the rush hour. Instead, the optimal policy provides higher discounts at the "tails" of the travel period to pull commuters away from the peak arrival time.
As shown in the figure above, when the budget is limited, the optimal strategy involves a "discontinuity": providing zero incentive during the absolute center of the rush hour and focusing funds on the edges where they can most effectively shift behavior.
Results: Slashing Congestion
The theoretical findings were validated through numerical simulations involving 9,000 commuters and a bottleneck capacity of 60 vehicles per minute. The results demonstrate that the IBTDM approach is remarkably efficient.
Key findings include:
- Congestion Elimination: With a sufficient budget (calculated at ~$37,400 for the test scenario), the queue is entirely removed (the green line in the figure above).
- High Impact at Low Cost: Even a modest budget of $8,660 (less than 25% of the "ideal" budget) reduced the Total System Travel Time (TSTT) by 70%.
- Diminishing Returns: The first dollars spent are the most effective. As the budget increases, the marginal gain in traffic reduction decreases, suggesting that even small-scale incentive programs can yield significant societal benefits.
Conclusion
This research provides a robust analytical framework for using EV charging as a tool for traffic management. By leveraging the flexibility of EV owners through price signals, authorities can significantly reduce bottleneck congestion. While the current model assumes homogeneous agents and 100% EV penetration, it lays the groundwork for more complex multi-modal strategies that combine "soft" incentives with traditional infrastructure management.
Summary updated