By Dave DeFusco

A research team led by Dr. David Li, director of the Katz School’s M.S. in Data Analytics and Visualization, introduced a novel game theory-driven framework that optimizes reward policies in the sharing economy, boosting user engagement and platform sustainability, at the 59th Annual Conference on Information Science and Systems at Johns Hopkins University in March.

In their study, “A Dynamic Framework for Optimizing Reward Policies in the Sharing Economy,” Dr. Li and co-authors Chengkun Yao and Cheng Li, both Katz School master’s students, as well as Angela Li, a researcher in the Applied Mathematics & Physics Department at Stony Brook University, apply Nash equilibrium models to maximize resource efficiency while keeping platforms profitable.

The sharing economy has revolutionized traditional industries, offering technology-driven platforms that facilitate temporary access to goods and services. Whether through ride-sharing apps, short-term home rentals or freelance service marketplaces, this economic model fosters resource efficiency, economic sustainability and additional income opportunities. However, sustaining these platforms requires active participation from service providers, which is heavily influenced by the incentives offered.

“Static reward models, where incentives remain fixed regardless of market conditions, often fail to adapt to evolving user behaviors and demand fluctuations,” said Dr. David Li, senior author of the study. “Our study addresses the limitations of such models by introducing a dynamic framework that adjusts rewards in real-time, maximizing engagement while keeping costs under control.”

At the heart of this framework is the application of game theory, which provides insights into strategic interactions among users. By employing Nash equilibrium principles, the study models how users respond to different incentive structures. The researchers propose that instead of a one-size-fits-all reward model, platforms should implement adaptive policies that dynamically adjust incentives based on real-time market conditions and user participation trends.

For instance, in a grocery delivery platform scenario, where customers place orders and gig workers fulfill them, reward optimization plays a crucial role. When demand surges and workers are scarce, increased incentives can attract more shoppers, ensuring order fulfillment. Conversely, when supply exceeds demand, reducing incentives prevents unnecessary expenditure while maintaining efficiency.

“To implement this adaptive model, we formulated the reward allocation problem as an optimization challenge,” said Chengkun Yao, a co-author and student in the M.S. in Digital Marketing and Media. “Using a dynamic programming algorithm, the framework recursively selects which users to incentivize, aiming to maximize platform efficiency and profitability while adhering to budget constraints.”

By leveraging historical data and machine learning predictions, the framework determines the optimal allocation of incentives to balance supply and demand efficiently. The iterative nature of the model allows platforms to continuously refine their strategies based on observed market responses.

“The findings from these simulations reinforced the model’s ability to adapt to varying market conditions,” said Cheng Li, a co-author of the study and student in the M.S. in Data Analytics and Visualization. “The results demonstrated diminishing returns at high participation levels, indicating that beyond a certain point, additional incentives fail to yield proportional increases in engagement. This insight underscores the importance of targeting high-impact users—those who contribute significantly to platform activity at a minimal cost.”

The framework also proved effective under tight budget constraints. By strategically reallocating rewards, it ensured optimal engagement even with limited financial resources. This adaptability makes the approach particularly valuable for emerging platforms seeking to scale operations sustainably.

From a business perspective, implementing an optimized reward system based on game theory principles offers multiple benefits:

• Maximized Engagement: Users are more likely to participate when incentives are tailored to real-time demand fluctuations.
• Cost Efficiency: Platforms can minimize unnecessary spending while maintaining service quality.
• Scalability: The dynamic nature of the framework allows seamless adjustments as platforms grow since platform activities are region-specific, optimization can be performed independently for each region. This decentralized approach allows for large-scale parallel computing in a cloud-based infrastructure.
• Fairness and Transparency: Equitable reward distribution fosters trust among participants, enhancing long-term retention.

“The integration of game theory into incentive design has the potential to redefine engagement strategies across the sharing economy,” said Angela Li, lead author of the study. "By transitioning from static to dynamic reward policies, platforms can unlock new levels of efficiency, profitability and user satisfaction, ensuring sustainable growth in an increasingly competitive market.”