Client Background
A leading U.S.-based multinational delivering aviation aftermarket services, including component sales, aircraft maintenance, and financing for major airlines.
Client Need
Manage diverse demand patterns across 300,000+ aircraft parts with varying requirements
Overcome limitations of traditional forecasting in handling complex demand
Optimize inventory levels while addressing challenges in procurement timing
Improve sourcing decisions influenced by lead times and inventory costs
Implement advanced solutions to manage intermittent demand effectively
Solution
Data Mining and Cleansing: Processed large volumes of customer data to detect anomalies, resolve consistency issues, and prepare high-quality datasets for analysis
Demand Signal Analysis: Consolidated and compared quantities requested, quoted, invoiced, and ordered to build accurate demand visibility across channels
Interchangeable Part Logic: Managed different part combinations and applied interchangeable part logic
Developed Forecasting Models: Built traditional, ensemble, and advanced models (LSTM, Transformers, AWS Cronos, Time GPT, Patch TST, TimesFM, Moirai) to enhance prediction accuracy
Automation: Enabled real-time forecasting with automated model selection based on data characteristics
Realized Benefits
Achieved 26% demand coverage with an RMSSE score below 1, ensuring accurate forecasting
Applied five evaluation metrics—RMSE, RMSSE, MASE, Tracking Signal, and MAE—for thorough model assessment
Reduced inventory costs by 20–30% through interchangeable parts logic and alignment with forecasts
Developed custom metrics to improve forecasting accuracy by capturing range and error more precisely
Tools & Technologies
Python
AWS
Amazon SageMaker
Open AI
Trending Success Stories
Ready to Innovate with Us?
Let’s Talk!
Connect with us on social media
Write to us at
[email protected]