A Game-Theoretic Framework for Intelligent EV Charging Network Optimisation in Smart Cities

Abstract

The transition to Electric Vehicles (EVs) demands intelligent, congestion-aware infrastructure planning to balance user convenience, economic viability, and traffic efficiency. We present a joint optimisation framework for EV Charging Station (CS) placement and pricing, explicitly capturing strategic driver behaviour through coupled non-atomic congestion games over road networks and charging facilities. From a Public Authority (PA) perspective, the model minimises social cost, travel times, queuing delays and charging expenses, while ensuring infrastructure profitability. To solve the resulting Mixed-Integer Nonlinear Programme, we propose a scalable two-level approximation method, Joint Placement and Pricing Optimisation under Driver Equilibrium (JPPO-DE), combining driver behaviour decomposition with integer relaxation. Experiments on the benchmark Sioux Falls Transportation Network (TN) demonstrate that our method consistently outperforms single-parameter baselines, effectively adapting to varying budgets, EV penetration levels, and station capacities. It achieves performance improvements of at least 16% over state-of-the-art approaches. A generalisation procedure further extends scalability to larger networks. By accurately modelling traffic equilibria and enabling adaptive, efficient infrastructure design, our framework advances key intelligent transportation system goals for sustainable urban mobility.

Figures (4)

• Figure 1: Sioux Falls Network: Solving PA model using JPPO-DE framework for peak-hours results in optimised EV Charging network- Reflecting traffic flow of drivers on links in equilibrium, charging fee and number of chargers in each node and different zones in terms of site rental & maintenance costs $(T)$. The numbers on the links and nodes correspond to their respective identifiers.
• Figure 2: Total Social Cost of EV drivers VS. Budget within 3 Optimisation Approaches: JO consistently minimises total social cost across all budget levels. At the optimal budget of $B=169$, JO reduces costs by 16% compared to PlO and 19% compared to PrO, due to its ability to avoid over-installation in low-demand areas and overpricing in low-cost zones.
• Figure 3: Average Social Cost and Required Chargers vs. $\boldsymbol{\mu} \& \boldsymbol{\alpha}$: Left: As the charger service rate $\mu$ increases, assuming fixed $e$, queueing time drops, reducing the number of chargers needed and charging fees. Right: For each 1% increase in $\alpha$, about 13 more chargers are required. Although the average social cost per EV rises gradually with greater penetration, expanding infrastructure alone may be inadequate. Additional measures, such as demand-side incentives, may be necessary to sustain system efficiency.
• Figure 4: Network Resilience Response to CS Failure: Lower social costs emerged in the network when dynamic JPPO-DE was applied to adapt prices, compared to the fixed price model.