Abstract
We study the optimal usage-based pricing problem in a resource-bounded network with one profit-maximizing service provider and multiple groups of surplus-maximizing users. We first analytically derive the optimal pricing mechanism that the service provider maximizes the service provider’s revenue under complete network information. Then we consider the incomplete information case, and propose two incentive compatible pricing schemes that achieve different complexity and performance tradeoff. Finally, by properly combining the two pricing schemes, we can show that it is possible to maintain a very small revenue loss (e.g., 0.5 % in a two-group case) without knowing detailed information of each user in the network.
Introduction
In the telecommunications industry, pricing models have evolved significantly from traditional flat-fee structures to more nuanced usage-based pricing. While flat-fee models offer simplicity, they often lead to inefficient resource utilization and missed revenue opportunities for Service Providers (SPs). Usage-based pricing, where users pay per unit of data (e.g., per GB), allows SPs to better manage network congestion and capture value based on user demand. However, implementing an optimal pricing strategy is fraught with challenges, primarily the tradeoff between system complexity and revenue performance, and the inherent information asymmetry between the provider and its users.
This paper addresses these challenges by developing a mathematical framework to optimize revenue in resource-constrained networks. The authors explore how an SP can strategically differentiate prices among various user groups to maximize profit while considering the practical limitations of information availability.
Methods
The interaction between the SP and the network users is modeled as a two-stage Stackelberg game. In the first stage, the SP acts as the leader, setting prices or price-quantity menus to maximize total revenue. In the second stage, users act as followers, deciding their resource consumption to maximize their individual surplus—defined as the utility derived from the resource minus the cost paid to the SP.
The research investigates three primary pricing schemes based on information availability:
- Complete Price Differentiation (CP): The SP has perfect information about each user's willingness to pay and sets a unique price for each group.
- Single Pricing (SP): The SP sets a uniform price for all users, regardless of their valuation.
- Partial Price Differentiation (PP): The SP groups users into a limited number of price tiers (clusters).
To solve these problems, the authors utilize convex optimization techniques to transform non-convex revenue functions into manageable forms. For the PP scheme, they propose a polynomial-time clustering algorithm with $O(I^J)$ complexity, where $I$ is the number of user groups and $J$ is the number of tiers, to find the optimal grouping of users.
Key Insights and Ideas
The "Effective Market" Threshold
A critical finding is that when network resources are finite, the SP should not necessarily serve every user. Instead, there exists an "effective market" threshold. The SP maximizes revenue by focusing resources on high-value users and potentially excluding those with a low willingness to pay. This ensures that scarce resources are allocated where they generate the highest marginal revenue.
Reversing the Volume Discount
Under incomplete information, where the SP does not know which user belongs to which group, they must design incentive-compatible (IC) price menus. To prevent high-value users from masquerading as low-value users to get cheaper rates, the SP may need to implement what looks like a "reverse volume discount." In this scenario, users who consume larger quantities of resources are charged a higher unit price. This serves as a screening mechanism, forcing high-valuation users to reveal their type through their usage patterns.
Consecutive Partitioning
For the clustering algorithm used in partial price differentiation, the authors prove a "consecutive partitioning" property. This means that in an optimal multi-tier system, users should always be grouped by adjacent willingness-to-pay levels. An optimal strategy will never bundle the highest and lowest payers together while keeping middle-tier payers separate.
Key Results
The theoretical models were validated through extensive numerical simulations:
- Significant Revenue Gains: Moving from a single-price model to complete price differentiation can increase revenue by up to 50%, particularly when high-value users represent a small minority of the total population.
- Efficiency of Few Tiers: The "sweet spot" for complexity vs. performance is often reached with very few tiers. In a 5-group scenario, moving from one to just two price tiers captured nearly 15% more revenue, reaching within 2.4% of the theoretical maximum possible with five distinct prices.
- Information Parity: The study identifies conditions under which an SP can achieve the same revenue with hidden user identities as they would with perfect information, provided the differences between user groups are sufficiently pronounced.
Limitations and Future Work
While the framework provides robust insights, it operates under the assumption of a monopoly provider. In competitive markets, the ability to differentiate prices would be influenced by the strategies of rival ISPs. Additionally, while the core results utilize logarithmic utility functions, different types of network traffic (e.g., low-latency gaming vs. background updates) might necessitate different utility models. Future research could also extend this static model into a dynamic one, where pricing tiers adjust in real-time based on fluctuating network congestion levels.
Summary updated