Using AI to detect panic buying and improve products distribution amid pandemic.

Area of Science:

• Supply Chain Management
• Artificial Intelligence
• Consumer Behavior

Background:

• The COVID-19 pandemic caused widespread panic buying, leading to stockouts of essential goods.
• Retailers lacked the technical infrastructure to manage sudden demand surges effectively.

Purpose of the Study:

• To develop an AI-powered framework for detecting demand anomalies and improving product distribution.
• To enhance retailers' strategic response capabilities during supply chain disruptions.

Main Methods:

• Utilized a data-driven framework integrating internal and external data sources.
• Applied AI models for anomaly detection and developed a prescriptive analytics simulation tool.
• Validated models with over 15 million observations across three product categories.

Main Results:

• The anomaly detection model successfully identified panic-buying-related demand surges.
• The prescriptive tool demonstrated a 56.74% increase in essential product accessibility during the March 2020 panic-buying event.
• External data integration improved model predictability and interpretability.

Conclusions:

• The developed AI framework provides a systematic solution for managing demand volatility.
• Retailers can strategically enhance essential product distribution and availability during uncertain periods.
• The study highlights the importance of data integration and AI in modern supply chain resilience.