Smart Pricing Strategy Balances EV Charging Load Across Cities
As electric vehicles (EVs) surge in popularity across urban centers worldwide, one of the most pressing challenges for city planners and energy providers is managing the spatial distribution of fast charging demand. Uneven charging patterns often lead to congestion at popular stations while others remain underutilized, creating inefficiencies for users, operators, and the power grid. To address this issue, a team of researchers from Southwest Jiaotong University and State Grid Sichuan Economic Research Institute has developed a novel dynamic pricing mechanism that leverages cooperative game theory to balance fast charging loads across urban networks.
Published in the April 10, 2024 issue of Automation of Electric Power Systems, the study introduces a comprehensive framework that integrates real-world travel behavior, user preferences, and strategic pricing to optimize the spatial distribution of EV charging. Led by Yang Shuai, Dai Chaohua, Guo Ai, and Ye Shengyong, the research proposes a system where charging stations within a region collaborate under the coordination of a distribution system operator (DSO) to dynamically adjust prices based on real-time demand and congestion levels.
The core of the proposed mechanism lies in its ability to shift user behavior without significantly increasing individual charging costs. By modeling the complex interplay between travel patterns, charging station selection, and pricing incentives, the researchers have created a solution that benefits all stakeholders: EV drivers experience shorter wait times, charging station operators see increased revenue, and the power grid operates more efficiently.
Understanding the Problem: The Hidden Inefficiencies of Fast Charging
While much attention has been paid to time-of-use pricing strategies that encourage off-peak charging, the spatial dimension of charging demand has received less focus. Traditional approaches often fail to address the fact that two charging stations located just a few kilometers apart can experience vastly different levels of congestion. This imbalance stems from a combination of factors, including the location of stations near commercial hubs, office districts, or tourist attractions, as well as user preferences for convenience over cost.
When drivers arrive at a busy station, they may face long wait times, leading to frustration and reduced satisfaction. Some may choose to abandon charging altogether, while others may drive further than necessary to find an available spot, increasing traffic and energy consumption. From the operator’s perspective, underutilized stations represent lost revenue, while overloaded stations may require costly upgrades to meet demand.
The researchers note that existing studies have primarily focused on either temporal load shifting or individual station optimization, often overlooking the interactions between multiple station operators within a region. This gap is particularly significant in markets where third-party operators manage charging infrastructure, as competition can lead to suboptimal outcomes for the system as a whole.
A New Approach: Cooperation Over Competition
The proposed solution flips the traditional competitive model on its head by encouraging cooperation among charging station operators. Instead of each station setting prices independently to maximize its own profit, the DSO coordinates a regional alliance where stations work together to achieve a common goal: load balance.
This cooperative game-theoretic approach is grounded in the principle that collective action can yield greater benefits than individual efforts. By forming a coalition, stations can strategically adjust their prices to redirect users from congested areas to underutilized ones, thereby increasing overall system efficiency.
The mechanism operates in real time, with the DSO collecting data on charging station occupancy, expected vacancy times, and user charging requests. When a user submits a charging request through the DSO’s interface, they can specify their preferences, such as willingness to detour, sensitivity to price, or tolerance for waiting. The DSO then uses this information, along with real-time congestion data, to calculate optimal pricing for each station.
The pricing strategy is designed to incentivize users to make choices that benefit the entire network. For example, during peak hours, a heavily congested station might see its price increase slightly, while nearby stations with available capacity might offer lower rates. This price differential encourages some users to detour to less busy stations, reducing wait times at the congested site and increasing utilization at the quieter ones.
Modeling Real-World Behavior: From Travel Patterns to User Preferences
A key strength of the proposed model is its integration of real-world data to simulate user behavior accurately. The researchers used anonymized ride-hailing data from Chengdu, China, to construct an origin-destination (OD) matrix that captures the probability of travel between different functional zones, such as residential areas, business districts, and entertainment hubs.
By applying inverse geocoding to the trip data, the team was able to map travel patterns to specific land-use categories and time periods. This allowed them to model not only where people are going but also when they are likely to need charging. For instance, the model accounts for the fact that drivers are more likely to charge after returning from work or during midday breaks, and that their destination preferences vary throughout the day.
To further refine the model, the researchers incorporated a modified Huff model to simulate charging station selection. The Huff model, originally developed for retail location analysis, predicts the probability that a customer will choose one store over another based on factors like distance, attractiveness, and price. In this context, the model was adapted to include variables such as travel time, queue length, station amenities, and service fees.
Users are assigned weights for each factor based on their preferences, allowing the model to capture the diversity of decision-making in the real world. Some users may prioritize minimizing travel distance, while others may be more sensitive to price or willing to wait longer for a station with better facilities. By allowing users to set their own preferences, the system becomes more personalized and responsive to individual needs.
The Role of the Distribution System Operator
Central to the success of this mechanism is the role of the DSO as a neutral coordinator. Unlike traditional models where pricing is set by individual operators or centralized authorities without input from the market, this approach empowers the DSO to act as a facilitator of cooperation.
The DSO does not dictate prices but rather computes an equilibrium solution through an iterative algorithm that considers the strategic interactions between stations. Each station submits its pricing strategy, and the DSO evaluates the resulting load distribution. Through repeated iterations, the system converges on a set of prices that optimizes load balance while respecting constraints such as minimum and maximum price levels and user cost thresholds.
To ensure fairness and stability, the researchers employ the Shapley value method to distribute the additional revenue generated by the cooperative alliance. The Shapley value, a concept from cooperative game theory, allocates gains based on each participant’s marginal contribution to the coalition. In this case, stations that play a critical role in balancing the load—such as those that absorb excess demand during peak periods—receive a larger share of the benefits.
This approach ensures that all participants have an incentive to join the alliance, as each station is guaranteed to earn at least as much as it would operating independently. The researchers emphasize that this collective rationality is essential for maintaining long-term cooperation and preventing free-riding behavior.
Simulation Results: A Win-Win for All Stakeholders
To validate their approach, the team conducted a simulation study based on a real-world scenario in Chengdu, involving five fast charging stations and 1,500 EVs. The results demonstrate significant improvements across multiple metrics.
First, the variance in charging station utilization—a measure of load imbalance—was reduced by 86.8% compared to the baseline scenario where stations operated independently. This indicates a much more even distribution of charging demand across the network.
Second, user satisfaction improved markedly. Average waiting times at the busiest stations dropped from over 10 minutes to less than 2 minutes, representing a reduction of more than 80%. This not only enhances the user experience but also increases the number of vehicles that can be served during peak periods.
Third, charging station revenues increased by 4.1%, from 8,640 yuan to 8,996.6 yuan per day. This gain comes primarily from attracting users who would have otherwise postponed or canceled charging due to long wait times. By offering lower prices at underutilized stations, the system captures this latent demand, benefiting both operators and users.
Importantly, these improvements were achieved without significantly increasing the average cost per user. The researchers note that while some users may pay slightly more to avoid detours, others benefit from lower prices at less congested stations. Overall, the average cost remained stable, ensuring that the system remains equitable and accessible.
Implications for Urban Mobility and Energy Systems
The implications of this research extend far beyond the immediate improvement of charging station operations. By optimizing the spatial distribution of EV charging, the proposed mechanism contributes to broader goals of urban sustainability and energy efficiency.
Reduced congestion at charging stations means less idling and lower emissions, both from vehicles waiting in line and from the power grid during peak demand periods. More balanced load distribution also reduces stress on local distribution networks, potentially delaying or avoiding costly infrastructure upgrades.
Moreover, the cooperative pricing model could serve as a blueprint for other shared mobility services, such as bike-sharing or scooter-sharing systems, where spatial imbalances are also a challenge. The principles of demand shaping through dynamic pricing and user incentives are applicable across a range of urban mobility contexts.
For policymakers, the study offers a practical framework for regulating and incentivizing EV charging infrastructure. Rather than imposing top-down mandates, governments could encourage the formation of regional charging alliances and provide the necessary data infrastructure to support real-time coordination.
Future Directions and Challenges
While the simulation results are promising, the researchers acknowledge several challenges that must be addressed before the system can be implemented at scale. One key issue is data privacy: collecting and processing detailed user location and charging behavior data raises legitimate concerns about surveillance and misuse. The authors suggest that robust data anonymization and user consent mechanisms will be essential for public acceptance.
Another challenge is scalability. The current model assumes a relatively small number of stations and users, but in large metropolitan areas, the computational complexity of coordinating thousands of charging points could be prohibitive. Future work will need to explore distributed computing architectures and machine learning techniques to handle larger-scale deployments.
Additionally, the model relies on users being willing to follow the DSO’s recommendations. While the study assumes full compliance, in reality, some users may ignore pricing signals or have rigid travel patterns that limit their flexibility. Incorporating behavioral economics insights could help improve the model’s predictive accuracy.
Finally, the long-term impact of such a pricing system on market competition remains an open question. While cooperation improves system efficiency, it could also reduce price competition among operators, potentially leading to higher overall prices. Regulatory oversight may be needed to ensure that the benefits of cooperation are passed on to consumers.
Conclusion: A Smarter Way to Charge
As cities around the world accelerate their transition to electric mobility, the need for intelligent, adaptive charging infrastructure becomes increasingly urgent. The research by Yang Shuai, Dai Chaohua, Guo Ai, and Ye Shengyong offers a compelling vision of how technology, economics, and user behavior can be integrated to create a more efficient and equitable charging ecosystem.
By shifting the paradigm from competition to cooperation, their dynamic pricing mechanism demonstrates that it is possible to balance charging loads, reduce wait times, and increase revenues—all without burdening users with higher costs. As EV adoption continues to grow, solutions like this will be essential for ensuring that the charging experience keeps pace with the vehicles themselves.
The study, published in Automation of Electric Power Systems (DOI: 10.7500/AEPS20230614001), represents a significant step forward in the intelligent management of urban energy systems. It underscores the importance of interdisciplinary research that combines engineering, economics, and data science to solve complex real-world problems. As cities strive to become smarter and more sustainable, innovations like this will play a crucial role in shaping the future of transportation.
Yang Shuai, Dai Chaohua, Guo Ai, Ye Shengyong, Automation of Electric Power Systems, DOI: 10.7500/AEPS20230614001