Insights from data scientist Hatim Kagalwala, ET CIO

Between 2021 and 2023, the world witnessed just how vulnerable global supply chains can be. Factory shutdowns during the pandemic, geopolitical tensions, and large-scale logistical disruptions exposed the fragility of these interconnected networks. As the ripple effects impacted nearly every industry, it became clear that a smarter, more adaptive approach was urgently needed. For Hatim Kagalwala—a data science expert who has led machine learning initiatives at Amazon and American Express—the answer lies not only in technological innovation but in deeply understanding and responding to customer behavior.

In 2024, he authored a pivotal study titled “Using Machine Learning to Model Supply Chain Management.” In it, Kagalwala emphasized that true resilience starts with visibility, adaptability, and informed decision-making. His research demonstrated how machine learning models can simulate complex supply networks, enabling businesses to test various scenarios and proactively prepare for disruptions. However, he believes that ML’s greatest strength lies in its ability to monitor and respond to customer demand in real time.

“Data science isn’t just about algorithms—it’s about understanding customer behavior and solving business problems in a reliable, responsible, and impactful way,” he explains.

Rather than focusing solely on backend logistics, Kagalwala has developed forecasting models that leverage customer data—such as purchase history, browsing behavior, and seasonal trends—to align inventory with anticipated demand. This minimizes overstock, prevents costly shortages, and ensures products are available when and where customers need them. To him, effective supply chain management isn’t just about speed—it’s about delivering the right products at the right time, based on a deep understanding of customer needs.

The global supply chain crisis exposed the limitations of traditional forecasting systems, which relied on static assumptions and struggled to adapt to sudden, large-scale disruptions. These systems failed to anticipate rapid shifts in consumer demand, and just-in-time inventory strategies collapsed under pressure. This, Kagalwala argues, is precisely where machine learning offers a critical advantage—by enabling more adaptive, responsive, and data-driven decision-making.

Machine learning algorithms can estimate demand using real-time data—such as weather changes, social trends, or regional events. They also have the capability to detect potential disruptions across vast networks, including shipping delays or supplier inconsistencies, before they escalate. When these insights are combined with dynamic routing and real-time logistics optimization, businesses can respond to unexpected events faster and more intelligently.

Equally important is risk modeling. By analyzing geopolitical developments, financial indicators, and supplier histories, machine learning can uncover hidden vulnerabilities in supply networks, allowing companies to adjust proactively. “Resilient supply chains,” Kagalwala notes, “are customer-centric and data-driven. They must be able to sense, learn, and adapt.”

Looking ahead, he envisions supply chains evolving into dynamic ecosystems—responsive, decentralized, and deeply attuned to customer needs. His ongoing research explores how real-time customer data can be integrated into machine learning frameworks, ensuring that technology doesn’t just optimize for speed but also for accuracy and relevance.
Through this fusion of data science and human insight, Hatim Kagalwala is helping redefine what resilience means in today’s volatile world. His work charts a new path for supply chains—making them smarter, more adaptable, and always connected to the people they serve.

• Published On Jun 2, 2025 at 08:11 PM IST

Join the community of 2M+ industry professionals

Subscribe to our newsletter to get latest insights & analysis.

Download ETCIO App

• Get Realtime updates
• Save your favourite articles

Scan to download App