⚡ electrical engineering

Flexible Electric Vehicle Charging with Karma

This paper proposes a non-monetary karma economy using online auctions to fairly and efficiently manage real-time flexible Electric Vehicle charging with limited capacity, demonstrating that a Stationary Nash Equilibrium exists and outperforms benchmark schemes by balancing deadline adherence with user urgency.

Original authors: Ezzat Elokda, Ruiting Wang, Karl H. Johansson, Angela Fontan

Published 2026-04-09

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ezzat Elokda, Ruiting Wang, Karl H. Johansson, Angela Fontan

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you work in a busy office building with a shared parking lot. There are only a few electric car chargers, but hundreds of employees with electric vehicles. Everyone needs to charge their car before they leave for the day, but the power grid can't handle everyone plugging in at 9:00 AM.

If we use money to solve this, the company might say, "Whoever pays the most gets to charge first." This is unfair: the rich employee gets a full battery, while the employee on a tight budget leaves with a dead car, even if they have a crucial meeting in an hour.

If we use first-come, first-served, the person who arrives at 7:00 AM gets to charge all day, while the person who arrives at 4:00 PM with a dying battery and a flight to catch gets nothing.

This paper proposes a third way: The Karma System.

The Core Idea: A Closed-Loop Token Economy

Instead of money, the charging station uses a digital currency called "Karma." Think of Karma like a "good citizen" token that you can't buy, sell, or trade. You can only earn it by waiting your turn, and you spend it when you really need to charge.

Here is how the daily cycle works:

The Daily Allowance: Every morning, everyone starts with a certain amount of Karma tokens.
The Auction: When you arrive at the charging station, you place a "bid" using your Karma tokens. You decide how many tokens you are willing to "spend" to charge right now.
- If you are in a rush (high urgency), you bid high.
- If you have plenty of time, you bid low or zero.
The Winner: The system gives the limited charging spots to the people with the highest bids.
The Payback: The tokens you spent don't vanish. They are collected into a big pot and redistributed equally to everyone in the system at the end of the day.

The Magic Trick: Because the tokens come back to you, you aren't actually "losing" anything in the long run. You are just borrowing from your future self to help your present self. This creates a fair, self-sustaining loop where no one gets rich or poor; everyone just manages their "patience" differently.

The "Game" of Strategy

The paper uses complex math (called a Dynamic Population Game) to figure out the best strategy for everyone. It turns out, the system naturally teaches users to be smart:

The "Rich" in Karma: If you have a lot of tokens, you might save them for a rainy day.
The "Urgent" User: If you have a flight in one hour and your battery is low, you will naturally bid all your tokens because the cost of not charging (missing your flight) is higher than the cost of spending your tokens.
The "Patient" User: If you have all day, you bid zero. You wait, let the urgent people go first, and then you charge later.

Why is this better than the other methods?

The researchers ran simulations to test this against two common methods:

First-Come, First-Served (FCFS): Like a line at a coffee shop.
Earliest Deadline First (EDF): Like a hospital triage where the person with the soonest appointment goes first.

The Results:

FCFS is chaotic. People with urgent needs get stuck behind people who arrived early but aren't in a rush.
EDF is better at meeting deadlines, but it ignores urgency. It treats a person who needs a charge in 1 hour the same as someone who needs it in 10 minutes.
The Karma System is the "Goldilocks" solution. It balances the two. It ensures that people with tight deadlines get charged, but it also prioritizes the most urgent cases (like the person with the flight) over the merely "on-time" cases.

The Big Picture

The paper proves that this system is mathematically stable. It won't collapse; it will find a "Nash Equilibrium," which is a fancy way of saying "a state where everyone is playing their best possible strategy, and no one wants to change their behavior."

In simple terms:
Imagine a crowded room with one door.

Money: The richest person walks out first.
First-Come: The person who stood in line the longest walks out first.
Karma: Everyone gets a ticket. If you are in a huge rush, you spend your tickets to jump the line. If you aren't in a rush, you save your tickets. At the end of the day, everyone gets their tickets back. The result? The people who really need to leave get to leave, the system stays fair, and no one feels cheated.

This paper suggests that by using "Karma" instead of cash, we can manage the transition to electric vehicles fairly, ensuring that the grid doesn't crash and that the person with the most urgent need gets the power they require.

1. Problem Statement

The rapid electrification of transport poses significant challenges to power grids and charging infrastructure due to limited capacity, physical power limits, and energy scarcity. While monetary pricing mechanisms (dynamic pricing, auctions) have been proposed to manage this scarcity, they suffer from three critical flaws:

Social Rejection: They are often unacceptable in communal settings (e.g., workplace charging).
Low Responsiveness: Many users do not react to small price variations due to the cognitive burden of planning.
Inequity: Monetary mechanisms allocate resources based on "willingness to pay," discriminating against users with lower "ability to pay."

The paper addresses the need for a fair, efficient, and non-monetary scheme to manage the intertemporal allocation of limited EV charging capacity. The goal is to balance meeting private trip deadlines with prioritizing high-urgency trips without using money.

2. Methodology

The authors propose a Karma Economy, a non-monetary token system where users bid for charging slots using non-tradable "karma" tokens. The system is modeled as a Dynamic Population Game (DPG).

A. The Karma Mechanism

Token System: Users are endowed with karma tokens. Tokens are non-tradable and flow in a closed loop.
Auction Process:
- At discrete time intervals (e.g., hourly), an online auction is held.
- Eligible users (present at the facility with SOC < 80%) place a bid based on their current karma balance.
- The highest bidders are admitted to charge.
- Redistribution: Winners pay their bid amount. The total collected karma is redistributed uniformly to all users (regardless of participation) to maintain a closed, sustainable economy.
User Dynamics: Users have private attributes:
- State of Charge (SOC): Current battery level.
- Deadline ( $t_d$ ): Time by which the trip must start.
- Desired SOC ( $s_d$ ): Target charge level.
- Urgency ( $u$ ): A private metric representing the cost of missing the deadline.

B. Game-Theoretic Formulation

The system is modeled as a collection of discounted, infinite-horizon Markov Decision Processes (MDPs) coupled through a social state.

State Space ( $x$ ): Includes global time-of-day, presence status, private deadline, desired SOC, urgency, current SOC, and karma balance.
Social State ( $d, \pi$ ):
- $d$ : The distribution of individual states across the population.
- $\pi$ : The symmetric policy (probability of bidding $b$ given state $x$ ).
Payoff Function: Users incur a cost proportional to their urgency ( $u$ ) for every time step they remain at the station after their deadline has passed.
State Transitions:
- SOC: Increases if the user wins the auction (based on a threshold bid derived from the population's bid distribution).
- Karma: Decreases by the bid amount if charging; increases by the redistributed share if not.
- Time/Deadline: Time advances; deadlines decrement once the user arrives.
Discounting: A state-dependent discount factor is applied only at the end of the day to handle the infinite horizon while respecting daily cycles.

C. Theoretical Guarantee

The paper proves the existence of a Stationary Nash Equilibrium (SNE). By ensuring the immediate payoff and state transition functions are continuous (using a continuous approximation for the auction threshold), the authors guarantee that a stable equilibrium strategy exists where no user can improve their long-term payoff by unilaterally changing their bidding strategy.

3. Key Contributions

Extension of Karma DPGs: The authors extend previous karma economy formulations to a complex EV charging setting. This introduces novel intra-day SOC dynamics and private trip deadlines alongside urgency, which were not present in prior transportation applications (like highway priority lanes).
Existence Proof: They demonstrate that despite the added complexity of intra-day dynamics and deadline constraints, a Stationary Nash Equilibrium (SNE) is guaranteed to exist.
Non-Monetary Fairness: The scheme achieves equity by relying entirely on non-monetary participation, ensuring that resource allocation is based on private needs (urgency/deadline) rather than financial status.
Algorithmic Implementation: They utilize an evolutionary dynamics-inspired algorithm to compute the SNE numerically.

4. Results

Numerical analysis was conducted under two scenarios: Moderate Scarcity (1/3 of users can charge at once) and High Scarcity (1/4 of users can charge at once). The Karma scheme was compared against two benchmarks:

FCFS (First-Come-First-Serve): Prioritizes arrival order.
EDF (Earliest-Deadline-First): Prioritizes the earliest deadline (assuming truthful reporting).

Key Findings:

Average Payoff: The Karma scheme achieved the highest average user payoff, outperforming both FCFS and EDF.
Wait Times:
- FCFS had the highest average wait time (poor performance).
- EDF had the lowest average wait time (best at meeting deadlines) but ignored urgency.
- Karma fell between the two regarding raw wait times but provided a superior balance.
Urgency Handling: When conditioning on urgency, the Karma scheme drastically reduced wait times for high-urgency users compared to EDF. EDF treats all deadlines equally, whereas Karma allows high-urgency users to bid more aggressively (spending more karma) to secure charging, effectively prioritizing them without monetary cost.
Threshold Bids: In high scarcity, threshold bids were higher and persisted longer throughout the day, reflecting the increased competition for limited slots.

5. Significance

This paper presents a viable alternative to monetary pricing for managing EV charging infrastructure. Its significance lies in:

Equity: It solves the fairness issue of monetary systems by decoupling resource access from wealth.
Efficiency: It leverages the "truthful bidding" property of karma economies, where users bid according to their true private needs (urgency and deadlines) because they are effectively "playing against their future selves."
Scalability: The DPG framework allows for the analysis of large populations without tracking individual interactions, making it computationally feasible for real-world deployment.
Future Outlook: The authors suggest future work could involve automating bidding based on the SNE policy, handling heterogeneous user types (different battery sizes, home charging access), and coupling EV charging with other resource allocations like traffic or household energy.

In conclusion, the Karma economy offers a robust, fair, and efficient mechanism to manage the temporal scarcity of EV charging, successfully balancing the competing objectives of meeting deadlines and prioritizing urgent needs.