How we developed a data-driven forecasting model using Palantir Foundry to optimize spare parts inventory and improve availability for a leading global MedTech company.
A leading medical technology company faced high spare parts consumption, stockouts, and inefficient inventory management at overseas locations. The company struggled with multiple operational challenges that impacted service reliability.
Key issues included inaccurate demand forecasting leading to frequent stockouts and emergency orders, high inventory costs due to overstocking of slow-moving parts, lack of visibility into spare parts consumption patterns, and long lead times affecting service reliability.
The company needed a data-driven supply chain strategy to balance availability and cost efficiency across their global operations.
Our project followed a structured Analyze – Optimize – Implement framework:
Analyzed spare parts consumption data to identify patterns. Developed a predictive model using Palantir Foundry for demand forecasting. Applied machine learning algorithms to differentiate between fast- and slow-moving parts. Significantly improved accuracy of spare parts demand prediction through advanced analytics.
Optimized reorder points and safety stock levels, reducing unnecessary inventory. Established automated alerts for critical stock levels, preventing service disruptions. Streamlined ordering processes to reduce lead times and emergency orders.
Integrated dashboards to provide real-time supply chain visibility. Conducted training sessions for supply chain managers on the new forecasting system. Defined a long-term governance model to ensure continuous improvement and sustainable operations.
Improved forecast accuracy by over 50% via ML-based demand forecasting and supply planning.
Identified €600M in untapped sales potential and reduced RfQ processing time by 20%.
3-phase analysis enabling data-driven investment decision with 20% model accuracy improvement.
