Choice Modeling and Pricing for Scheduled Services

2025-12-24

AI Summary

This paper addresses the dynamic pricing problem in appointment-based services featuring multi-tiered, multi-temporal, and multi-window substitutable options (e.g., varying time slots, prices, and capacity levels). We propose a unified framework integrating hierarchical discrete choice modeling with dynamic pricing. Our approach innovatively employs decision trees to drive interpretable market segmentation and segment-specific parametric choice models—explicitly incorporating reference price effects. We further design an efficient heuristic algorithm for scalable pricing optimization. An A/B test conducted on an Amazon business line demonstrated a 19% improvement in core metrics; the solution was fully deployed starting Q4 2023, enabling rapid iteration of new services. To our knowledge, this is the first work to jointly integrate interpretable segmentation, behavior-aware choice modeling, and scalable pricing optimization—significantly enhancing demand forecasting accuracy and revenue performance in complex appointment settings.

Technology Category

• Search and Optimization: Algorithm Configuration
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Advertising auctions, pricing, markets, and exchanges

Abstract

We describe a novel framework for discrete choice modeling and price optimization for settings where scheduled service options (often hierarchical) are offered to customers, which is applicable across many businesses including some within Amazon. In such business settings, the customers would see multiple options, often substitutable, with their features and their prices. These options typically vary in the start and/or end time of the service requested, such as the date of service or a service time window. The costs and demand can vary widely across these different options, resulting in the need for different prices. We propose a system which allows for segmenting the marketplace (as defined by the particular business) using decision trees, while using parametric discrete choice models within each market segment to accurately estimate conversion behavior. Using parametric discrete choice models allows us to capture important behavioral aspects like reference price effects which naturally occur in scheduled service applications. In addition, we provide natural and fast heuristics to do price optimization. For one such Amazon business where we conducted a live A/B experiment, this new framework outperformed the existing pricing system in every key metric, increasing our target performance metric by 19%, while providing a robust platform to support future new services of the business. The model framework has now been in full production for this business since Q4 2023.

Problem

• Develops a discrete choice model for scheduled service pricing
• Optimizes prices using segmentation and behavioral reference effects
• Enhances performance metrics through live A/B testing validation

Innovation

• Decision trees segment market for scheduled services
• Parametric discrete choice models capture reference price effects
• Fast heuristics optimize pricing with 19% metric improvement