ShortageSim: Simulating Drug Shortages under Information Asymmetry

TL;DR

ShortageSim introduces an LLM-powered, multi-agent framework to study drug shortages under information asymmetry, modeling FDA regulators, manufacturers, and buyers as strategic actors responding to regulatory signals. The framework uses a two-stage Analyze-Decide pipeline with modular prompts to capture heterogeneous interpretations of announcements and market conditions. Experiments on a real FDA shortage dataset demonstrate that ShortageSim better reproduces historical trajectories than a zero-shot baseline and can quantify policy effects, such as how proactive FDA alerts influence stockpiling and investment. This work provides a flexible testbed for counterfactual policy evaluation and opens public data to advance research on supply chain resilience under information asymmetry.

Abstract

Drug shortages pose critical risks to patient care and healthcare systems worldwide, yet the effectiveness of regulatory interventions remains poorly understood due to information asymmetries in pharmaceutical supply chains. We propose \textbf{ShortageSim}, addresses this challenge by providing the first simulation framework that evaluates the impact of regulatory interventions on competition dynamics under information asymmetry. Using Large Language Model (LLM)-based agents, the framework models the strategic decisions of drug manufacturers and institutional buyers, in response to shortage alerts given by the regulatory agency. Unlike traditional game theory models that assume perfect rationality and complete information, ShortageSim simulates heterogeneous interpretations on regulatory announcements and the resulting decisions. Experiments on self-processed dataset of historical shortage events show that ShortageSim reduces the resolution lag for production disruption cases by up to 84\%, achieving closer alignment to real-world trajectories than the zero-shot baseline. Our framework confirms the effect of regulatory alert in addressing shortages and introduces a new method for understanding competition in multi-stage environments under uncertainty. We open-source ShortageSim and a dataset of 2,925 FDA shortage events, providing a novel framework for future research on policy design and testing in supply chains under information asymmetry.

Figures (11)

• Figure 1: ShortageSim Architecture. The framework models the drug supply chain dynamics through three agent roles: FDA regulator, buyer consortium, and $n$ competing manufacturers, coordinated by an Environment module. Each agent operates under information asymmetry (shown in Information Availability) and follows a two-stage LLM pipeline (Analyze $\rightarrow$ Decide). The timeline illustrates the sequential decision process from supply disruption through market reaction (details in Sequential Decision Timeline), capturing realistic stakeholder responses to regulatory announcements during drug shortage events.
• Figure 2: Effect of market competition on supplier investment (average supply per period)
• Figure 3: ShortageSim Web Interface
• Figure 4: Manufacturer Agent - Collector & Analyst. System and user instructions.
• Figure 5: Manufacturer Agent - Decision Maker. System and user instructions.