Data-Driven Supply Chain Risk Management - Nieuws

Data-Driven Supply Chain Risk Management
Thomas De Backker and Robert Boute
KU Leuven and Flanders Make

In recent years, businesses have faced significant challenges from disruptive events such as COVID-19, geopolitical conflicts, and logistical bottlenecks such as the Suez Canal blockage. Their unpreparedness towards disruptions has led to supply delays, inventory shortages, unmet customer demand, damage to reputation and various indirect costs. To prevent potential losses next time, companies must proactively build resilience.

To support this proactive approach, KU Leuven has developed a data-driven tool to enable companies to learn from past performance and use this information for strategic decisions on supply chain resilience. The tool segments components based on their exposure to disruption risk. This facilitates resilience by design, both in the design of their supply chain and their products.

Analogous to a traditional risk matrix that categorizes risks based on their likelihood and severity, our Component Risk Matrix classifies components according to the likelihood of stock disruptions and the ensuing downtime to recover from them. By harnessing advanced machine learning techniques applied to historical stock disruptions and component attributes within a specific product group, we can predict both the frequency and duration of these stock breaks for each component.

We have successfully validated our approach to data from a large industrial manufacturer. The resulting risk matrix delineates distinct zones based on frequency and severity (see left figure), providing insights into the factors influencing component criticality. These insights can then inform strategic decisions regarding resilience strategies (see quadrants on the right figure).

This research is supported by Flanders Make, the strategic research centre for the manufacturing industry in Flanders, Belgium.

If you would like to learn more about our research or if you have interesting viewpoints to share, please get in touch with Thomas De Backker (thomas.debackker@kuleuven.be) or Robert Boute (robert.boute@kuleuven.be).