Abstract
Crowdsourcing can be applied to the Internet-of-Things (IoT) systems to provide more scalable and efficient services to support various tasks. As the driving force of crowdsourcing is the interaction among participants, various incentive mechanisms have been proposed to attract and retain a sufficient number of participants to provide a sustainable crowdsourcing service. However, there exist some gaps between the modeled entities or markets in the existing works and those in reality: 1) dichotomous task valuation and workers' punctuality, and 2) crowdsourcing service market monopolized by a platform. To bridge those gaps of such impractical assumption, we model workers' heterogeneous punctuality behavior and task depreciation over time. Based on those models, we propose an Expected Social Welfare Maximizing (ESWM) mechanism that aims to maximize the expected social welfare by attracting and retaining more participants in the long-term, i.e., multiple rounds of crowdsourcing. In the evaluation, we modeled the continuous competition between the ESWM and one of the existing works in both short-term and long-term scenarios. Simulation results show that the ESWM mechanism achieves higher expected social welfare and platform utility than the benchmark by attracting and retaining more participants. Moreover, we prove that the ESWM mechanism achieves the desirable economic properties: individual rationality, budget-balance, computational efficiency, and truthfulness.
Introduction
Traditional IoT crowdsourcing models often rely on two binary, or "dichotomous," assumptions that fail to capture real-world complexities. First, task valuation is typically treated as a step function where value drops instantly to zero after a deadline. In practice, data value often depreciates gradually. Second, worker punctuality is often modeled as binary (on-time or absent), ignoring the stochastic nature of submission times. Furthermore, most existing research assumes a monopoly platform, whereas real-world markets involve multiple platforms competing for a shared pool of participants. This paper introduces the Expected Social Welfare Maximizing (ESWM) mechanism to address these gaps by prioritizing long-term participant retention over short-term platform profit.
The ESWM Mechanism
The ESWM mechanism is designed to operate in a competitive environment where users can migrate between platforms based on perceived utility. The mechanism follows a structured three-stage workflow to ensure efficiency and fairness.
Modeling Reality
Instead of binary deadlines, ESWM utilizes a quadratic depreciation function to model how task value fades over time. Worker behavior is captured using a truncated normal distribution, allowing the platform to calculate the probability of task completion within specific timeframes. This allows the platform to treat a worker's reliability (punctuality) as a critical resource alongside their bid price.
Workflow and Selection
As illustrated in the system architecture, the mechanism processes participant information through three distinct steps:
- Winner Selection: Using greedy algorithms (WRSA and WWSA), the platform selects requesters and workers based on a tunable ratio of value-to-depreciation and cost-to-punctuality.
- Matching: The platform pairs winning requesters with winning workers to maximize expected social welfare.
- Pricing: A dynamic pricing model adjusts final fees and rewards based on actual submission times, ensuring the mechanism remains budget-balanced and truthful.
Performance Evaluation
The ESWM mechanism was evaluated against a state-of-the-art truthful benchmark in a simulated competitive market. A key focus was the trade-off between computational efficiency and optimality. While the Hungarian algorithm provides optimal matching, its cubic complexity makes it unsuitable for large-scale IoT deployments.
As shown in the results, the greedy approach used in ESWM offers a massive improvement in scalability. For 500 requesters, the Hungarian algorithm required nearly 11 hours to compute, while the ESWM's greedy matching finished in approximately 1 second. Although the Hungarian algorithm achieves slightly higher social welfare, the computational overhead is prohibitive for real-time IoT applications.
In long-term simulations, the "Small Profits" philosophy of ESWM proved superior. By taking a smaller platform cut to attract more users, ESWM maintained a growing user base, while the benchmark platform's utility eventually collapsed as participants migrated to the more favorable ESWM ecosystem.
Conclusion
The ESWM mechanism successfully bridges the gap between theoretical crowdsourcing models and practical IoT requirements. By accounting for task depreciation and worker punctuality, and by adopting a long-term strategy for participant retention, it achieves higher social welfare and platform utility in competitive markets. Future work will explore dynamic parameter tuning using machine learning to further adapt to shifting market conditions.
Summary updated