Canada
Puerto Rico
USA
Belgium
France
UK
India
Client Background
The client is a global aerospace company specializing in aircraft and engine services for all major US airlines. Innova works with the digital innovation group driving initiatives in simplifying their supply chain and operations.
Client Need
Customer has a total of 300K+ aircraft parts which has different type of demand patterns–Intermittent, lumpy, erratic and smooth. Considering the demand nature of the parts, traditional forecasting techniques were not that effective.
Without proper forecasting in place, there were challenges in maintaining the right inventory levels. Considering the lead times and cost of keeping inventory, procurement team find it challenging on which part to source at what time.
Customer was looking for advanced solution which works well with intermittent demand patterns.
Solution
Worked with business stakeholders for in depth data analysis to understand the challenges with data, demand patterns etc.
Collaborated with business to define the target demand calculation as there is lot of data inconsistencies w.r.t. Quantity requested, Quantity quoted, Quantity invoiced, and Quantity ordered
Used Interchangeable part logic to consolidate the demand and have different part combinations. Reduced the variability in demand at part combination level.
Experimented with both traditional, ensembled and advanced deep learning models for different demand patterns.
Built models which generate range forecast for each part using LSTM, Transformers, GRU, Cronos, Time GPT, Patch TST.
Realized Benefits
Built an MVP that covers 26% of demand and has an RMSSE score less than 1
Tried out different evaluation metrics which include RMSE, RMSSE, NFB, MSE, MAE
Built custom evaluation metrics for range and error calculation
Saved 20-30% of the inventory cost by proposing a model on interchangeable parts logic and planning the inventory based on the forecast
Tools & Technologies
Python
AWS
OpenAI
Trending Success Stories
Modernizing Enterprise Data Warehouse for Smarter & Faster Healthcare Insights for a Leading Value-Based Caregiver
Implementing Data Lakehouse for a Leading Biotech Major for Faster, Deeper Insights
Reducing Legacy Data Costs While Migrating 12,000+ Tables to a Cloud-Native Platform
Modernizing Epic and EHR on Azure for Scalable, Secure Healthcare Operations
Ready to Innovate with Us?
Let’s Talk!
Connect with us on social media
Write to us at
[email protected]